Re: [OSM-talk] Beware Pokemon users

2017-01-03 Thread Steve Bennett
Hey everyone,

  Just a question which no one seems to have addressed - is there any
evidence that Go is actually using a *feed* of OSM data, rather than just a
one-off dump? It's really rare, IMHO, for anyone to bother with a feed for
a project like this - so much easier to just get a planet extract, process
it, and do whatever you want with it. For the Go people to actually
constantly update from live OSM data seems like a lot of work, for not much
benefit.

If they're *not* using live data, then as soon as the Go players realise
that, then presumably the vandalism will stop.

Steve
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk


Re: [talk-au] City of Melbourne data

2016-12-18 Thread Steve Bennett
Awesome! I sit with the GIS team, so if you need any information about the
spatial data or any questions, don't hesitate.

We have quite a few identifiers of various types (asset IDs, property IDs
etc), so it may be helpful to import those to facilitate future updates?

Another couple of datasets to look at are the bar/tavern/pub and
cafe/restaurant ones. They're comprehensive lists of every establishment of
those types across the whole municipality, updated on a rolling cycle every
2 years.

Steve

On Mon, Dec 19, 2016 at 1:22 PM, Daniel O'Connor <daniel.ocon...@gmail.com>
wrote:

> I'll put my hand up for (slowly) importing addresses; have begun sketching
> out a plan in https://wiki.openstreetmap.org/wiki/Import/Catalogue/
> data.melbourne.vic.gov.au-addresses
>
> Dataset:
> https://data.melbourne.vic.gov.au/Property-Planning/
> Address-Points/a7rp-xtya
>
> Sent email re explicit permission (all datasets), will add to wiki.
>
>
> On Mon, Dec 19, 2016 at 12:07 PM, Daniel O'Connor <
> daniel.ocon...@gmail.com> wrote:
>
>> A good dataset may be address point data, if available. I know we have
>> GNAF, but we've been unable to get the explicit permission needed.
>>
>> Coverage is OK at the moment, so any import would be better as a
>> semi-manual, street by street kind of thing.
>>
>>
>> Public toilets, BBQs, Playgrounds, Park names, etc also spring to mind.
>>
>>
>> On Mon, Dec 19, 2016 at 11:55 AM, Steve Bennett <stevag...@gmail.com>
>> wrote:
>>
>>> Hi all,
>>>   I'm a long time contirbutor (user:Stevage) but a bit quiet of late.
>>> I'm now working as Senior Open Data Specialist at City of Melbourne (email:
>>> opend...@melbourne.vic.gov.au), and have just got approval for our data
>>> to be used in OpenStreetMap.
>>>
>>> So: If someone would like to email that address and formally request
>>> permission to use current and future open data in OSM, I'd be very happy to
>>> respond in the affirmative.
>>>
>>> Also: is anyone interested in importing any of our data? You can see
>>> most of our public spatial data at maps.melbourne.vic.gov.au (to
>>> actually access it, go to data.melbourne.vic.gov.au). If there is
>>> interesting data on the maps site that's not on the open data portal, we'll
>>> prioritise getting it out.
>>>
>>> Perhaps some of the street furniture datasets (bike hoops, water
>>> fountains etc) would be good starting points?
>>>
>>> Steve
>>>
>>>
>>>
>>> ___
>>> Talk-au mailing list
>>> Talk-au@openstreetmap.org
>>> https://lists.openstreetmap.org/listinfo/talk-au
>>>
>>>
>>
>
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


[talk-au] City of Melbourne data

2016-12-18 Thread Steve Bennett
Hi all,
  I'm a long time contirbutor (user:Stevage) but a bit quiet of late. I'm
now working as Senior Open Data Specialist at City of Melbourne (email:
opend...@melbourne.vic.gov.au), and have just got approval for our data to
be used in OpenStreetMap.

So: If someone would like to email that address and formally request
permission to use current and future open data in OSM, I'd be very happy to
respond in the affirmative.

Also: is anyone interested in importing any of our data? You can see most
of our public spatial data at maps.melbourne.vic.gov.au (to actually access
it, go to data.melbourne.vic.gov.au). If there is interesting data on the
maps site that's not on the open data portal, we'll prioritise getting it
out.

Perhaps some of the street furniture datasets (bike hoops, water fountains
etc) would be good starting points?

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Importing Vicmap Lite data

2015-12-14 Thread Steve Bennett
Hi Russell,

  I've been involved in a discussion with VicMap (Peter Debicki
specifically) on this issue, but I've dropped the ball (or at least am
holding on to it a bit too long).

Here's the short version:

- From OSM's perspective, we need a clear statement from VicMap that they
are ok with us importing their data into OSM and potentially re-licensing
it.
- From VicMap's perspective: they don't think we need this statement, and
they don't want to provide it, but they're ok with us importing their data
into OSM.

Yeah, it's a bit silly.

Vicmap's position on IP is pretty clear. They consider all their *data* to
be open licensed, and available for anyone to use for free. They consider
their *services* (including rendered map tiles) to be commercially valuable.

Steve

On Mon, Dec 14, 2015 at 1:17 PM, Russell Edwards 
wrote:

> Hi,
>
> I just wanted to double check before I do this as people seem to talk
> casually about using Vicmap Lite data but the OSM Wiki has a long process
> to go through with the community before doing any imports.
>
> I have downloaded "Statewide Public Land Classification boundaries,
> polygon - 1:250,000 to 1 :2 million. Vicmap Lite". It is CC
> BY-4.0-International.
>
> I want to add boundaries of some local state forests I frequent, using
> this data. My questions are
>
> * Can I just go ahead and do it?
> * I haven't done boundaries before, do I need a shared node where roads
> and waterways cross the boundary?
> * What if there is an existing overlapping area natural=wood that follows
> the aerial tree extent (which doesn't exactly match the boundary)? Leave
> them both there? Do I need natural=wood on the state forest boundary area
> to, to make it display green? What about the conflict there?
> * Anything else to watch for?
>
> I might have a look at Vicmap Hydro to get the creeks as well.
>
> Thanks in advance
>
> Russell
>
>
> ___
> Talk-au mailing list
> Talk-au@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-au
>
>
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [OSM-legal-talk] Any expert CC-BY -> ODbL negotiators?

2015-09-01 Thread Steve Bennett
On Tue, Sep 1, 2015 at 9:29 PM, Jukka Rahkonen  wrote:

>
> Have you noticed that there are already quite many Australian datasets
> including Victorian Government data listed here:
>
>
> http://wiki.openstreetmap.org/wiki/Contributors#Contributor_specific_attribution_and_disclaimer
>
>
Thanks, Jukka.  I suspect that "permission" isn't actually valid, as it
seems to extend from data.gov.au (Federal government) but most of the
datasets there are state or territory (eg, the VicMap Rivers dataset), and
are published on the relevant state/territory data portals (data.vic.gov.au,
data.act.gov.au) etc.

Steve
___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] Any expert CC-BY -> ODbL negotiators?

2015-08-31 Thread Steve Bennett
Wow, great responses everyone - much appreciated.

Simon Poole:
>Before this discussion goes off on a tangent, which version of CC-by are
they currently using?

http://creativecommons.org/licenses/by/3.0/au/

You can see this for instance at:
https://www.data.vic.gov.au/data/dataset/vicmap-hydro-connector-watercourse

Richard Fairhurst:
>In which case they've chosen the wrong licence.
>If you license your work under a permissive, attribution-only licence
>(CC-BY), then you are automatically giving permission for it to be
>relicensed under a share-alike, attribution-only licence (CC-BY-SA). You
>can't license under CC-BY and say "no-one may incorporate this data into a
>dataset with share-alike restrictions". That would defeat the point of a

Thanks for confirming that. I was a bit confused by the term in the CC-BY
deed:

>You may not apply legal terms or technological measures
<http://creativecommons.org/licenses/by/3.0/au/#> that legally restrict
others from doing anything the license permits.

Which could be interpreted to mean you can't apply CC-BY-SA which
"restricts others" from using the data and not sharing it, which "the
license" (CC-BY) "permits". But CC's wiki explicitly says this is ok:

https://wiki.creativecommons.org/wiki/Frequently_Asked_Questions#If_I_derive_or_adapt_material_offered_under_a_Creative_Commons_license.2C_which_CC_license.28s.29_can_I_use.3F

>When creating an adaptation of material under the license identified in
the lefthand column [BY], you may license your contributions to the
adaptation under one of the licenses indicated on the top row [BY-SA] if
the corresponding box is green [it is].

Richard again:
>Where "etc." means "TomTom". There are only four worldwide geodata providers.
It's hardly a slippery slope of individual permissions.

Good point. Although hypothetically there could be non-worldwide providers.


Andrew Turner:
>So a simpler route here would be to suggest "upgrading" to use CC-By 4.0?

Getting DELWP (and most likely the rest of the Victorian government) to
switch to CC-BY 4.0, let alone CC-0, would be a huge task that would
probably involve all kinds of bureaucracy, working groups, etc etc.

Richard Best:
[a long, very helpful email currently stuck in moderation]
>Perhaps helpfully, there is a New Zealand Government precedent for the
situation you're facing here. Land Information New Zealand licenses a wide
range of geospatial data under CC-BY licences. However, it has also allowed
data to be made available on OSM under the ODbL. See here for details:
http://wiki.openstreetmap.org/wiki/LINZ

>I suggest you draw the Victorian State Government's attention to the
approach taken by Land Information in New Zealand, with a view to asking
the Victorian State Government to take the same approach. I completely
understand their position of not wanting to "get into creating one-off
variations for every potential user with a preference" but I think it's
important to appreciate that we're talking here about a global project
which has decided on the ODbL, that it would be extremely difficult to
change the current regime to another one and that there is no real downside
to the Victorian State Government dual-licensing its data, once under CC-BY
and separately under the ODbL (or, actually, allowing OSMF to license under
the ODbL). It's perfectly entitled to do that and people who want to use
the data directly from the Victorian State Government's site under CC-BY
can always do so. Most significantly, dual-licensing under CC-BY (via a
government site) and the ODbL (via OSM) allows the data to be used in a
wider range of contexts and for it to be mashed up with other rich datasets
already in OSM or that may be added to OSM in the future. Allowing this
could result in cultural, environmental, economic or social benefits for
Australians. This is all entirely consistent with the rationale for open
licensing in the first place.

Wow. Hard to believe that New Zealand open licensed their data 7 years ago.
Maybe this will be the precedent we need. That's really an excellent,
well-worded summary. Would it be alright with you if I CC'ed you in my next
response to them (which they have asked to be directed to *their* legal
team...)?

Steve



On Mon, Aug 31, 2015 at 4:07 PM, Simon Poole <si...@poole.ch> wrote:

>
> Hi Steve
>
> Before this discussion goes off on a tangent, which version of CC-by are
> they currently using?
>
> Simon
>
>
>
> Am 30.08.2015 um 17:14 schrieb Steve Bennett:
>
> Hi all,
>   I've been trying to convince the state government of Victoria (southeast
> Australia) to allow their VicMap raw data to be imported into OSM. It's
> currently CC-BY, and they've told me they're happy in principle for it to
> be used this way, but they're uncomfortable making the recommended
> statement 

[OSM-legal-talk] Any expert CC-BY - ODbL negotiators?

2015-08-30 Thread Steve Bennett
Hi all,
  I've been trying to convince the state government of Victoria (southeast
Australia) to allow their VicMap raw data to be imported into OSM. It's
currently CC-BY, and they've told me they're happy in principle for it to
be used this way, but they're uncomfortable making the recommended
statement DELWP has no objections to geodata derived in part from Vicmap,
either traced from Vicmap map products, or directly from spatial extracts,
being incorporated into the OpenStreetMap project geodata database and
released under a free and open license.

Specifically, they don't think ODbL is as free and open as CC-BY, and
they don't particularly want to make a one-off statement for OSM.

The only other requirement is not to apply legal terms or technological
measures that legally restrict others from doing anything the license
permits. If the ODbL is more restrictive than our CC by Attribution this
presents a problem for OSM, not for us. My initial response is that we
wouldn't want OSM to apply a more restrictive license than ours, and in
respect of the statement Steve wants us to provide, DELWP doesn't want to
get into creating one-off variations for every potential user with a
preference - Google, HERE, etc.
...

We believe the CC by Attribution appropriate to sufficiently and equitably
provide our data to all/anyone, and if Steve is concerned he should take it
up with OSM. I'll refer it to Legal (not ours, DataVic's) if he wants to
pursue it further.


It's all getting quite subtle and possibly out of my depth. I'm not sure if
the concern is a misunderstanding about the implications of dual licensing,
a philosophical objection to free licences that impose share-alike
restrictions like ODbL, or something different.

I wonder if there are any expert licence negotiators here who might be able
to get involved in the discussion.

Thanks very much,
Steve
___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] Any expert CC-BY - ODbL negotiators?

2015-08-30 Thread Steve Bennett
Huh. Really? Did I completely misunderstand this?
http://wiki.openstreetmap.org/wiki/Import/GettingPermission

My understanding was that when you import data into OSM, you assign special
permission to the OSMF to re-license the data under ODbL, so you need more
than just CC-BY licensing to begin with. Did something change, or have I
just been mistaken for a long time?

Steve

On Mon, Aug 31, 2015 at 10:41 AM, Alex Barth a...@mapbox.com wrote:


 On Sun, Aug 30, 2015 at 2:33 PM, Stephan Knauss o...@stephans-server.de
 wrote:

 Hello Steve,

 On 30.08.2015 17:14, Steve Bennett wrote:

 I wonder if there are any expert licence negotiators here who might be
 able to get involved in the discussion.


 I'm no such expert, but they just require attribution. Did they state any
 specific way of doing so? If not, then maybe just mentioning in the wiki is
 fine for them?

 http://wiki.openstreetmap.org/wiki/Contributors


 Right. You don't need DELWP to give you any statement or permission in
 order to import their data to OpenStreetMap or derive data  for
 OpenStreetMap from their data.



 ___
 legal-talk mailing list
 legal-talk@openstreetmap.org
 https://lists.openstreetmap.org/listinfo/legal-talk


___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [talk-au] VicMap licensing - help?

2015-03-06 Thread Steve Bennett
That would be great. I randomly, through work, met with some of the VicMap
people today, but I'd rather know a bit more of the background before I
bring it up.

My understanding of the situation is they regard much of the data as
open, but they consider the Vicmap API a value-add service, hence not
open. But I don't know how they consider using the API to trace from -
could depend a bit on exact phrasing.

Steve

On Fri, Mar 6, 2015 at 1:54 PM, Nyall Dawson nyall.daw...@gmail.com wrote:

 On Wed, 4 Mar 2015 at 13:33 Steve Bennett stevag...@gmail.com wrote:
 
  Hi guys,
Can someone point me to a clear statement that copying data from
 VicMap is definitely ok? I can't find anything on the VicMap API that says
 it's open licensed, and all I can find to go on is this one email thread
 starting 10/10/13 that's a bit inconclusive and unofficial.

 As far as I know no-one has been able to obtain this statement. I
 tried a while back (maybe 12 months or more) and while everyone I
 could reach was supportive, no-one would make the final call and
 provide written approval.

 If you're interested in chasing this up I can dig up the old email
 chains and forward them on to you...

 Cheers,
 Nyall

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


[talk-au] VicMap licensing - help?

2015-03-03 Thread Steve Bennett
Hi guys,
  Can someone point me to a clear statement that copying data from VicMap
is definitely ok? I can't find anything on the VicMap API that says it's
open licensed, and all I can find to go on is this one email thread
starting 10/10/13 that's a bit inconclusive and unofficial.

Thanks,
Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Proposed import of 19 bicycle repair stations in au/nz

2015-03-03 Thread Steve Bennett
 But if anyone wants to do some armchair mapping there is lots of online
 info, for example:

 http://pcs.unimelb.edu.au/traffic-and-parking/bicycle-parking/bicycle-repair-stands.html


Fascinating - I work very close to one of those points, had no idea.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Cycling network tag

2015-01-06 Thread Steve Bennett
On Mon, Dec 15, 2014 at 9:29 PM, SomeoneElse li...@atownsend.org.uk wrote:

  The cycle tourism network that I suspect that you're referring to is
 the National Byway http://www.nationalbyway.org/welcome.asp which is a
 bit of a one-off - there are other RCNs that suit different cycling styles
 and needs.


Ah, could be. I did ride some parts of the NB. Strangely, in several places
it was very different from what was marked on OSM - I kept hitting it at
right angles.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Cycling network tag

2014-12-14 Thread Steve Bennett
On Wed, Dec 10, 2014 at 10:02 AM, Nicholas Barker nicholasbark...@gmail.com
 wrote:

 Hi Steve/Frank


[replying back to list, looks like that was your intention]



 First up IMO I wouldn't be putting the Mawson as an NCN.

 Any kind of 'cycle network'  including a 'national one' implies that it
 will be negotiable by most, if not all bicycles. This would require a good
 quality firm surface. The Mawson is sold as a long distance MTB route.
 Whilst in reality a good proportion of it is on (boring) unsealed roads
 there are many sections that would be a struggle on anything other than a
 mountain bike or sturdy hybrid.


We had 9 in our group. 4 mountain bikes, 3 touring bikes, 2 cyclocrosses,
all pretty heavily loaded. Yes, it's rough in places, and a couple of
sections at the far north were genuine mountain biking. But it's so far
from anywhere that there's basically no chance of anyone thinking oh I
know, I'll just take this beautiful bike path called the Mawson Trail from
Blinman to Quorn - OH GOD IT'S SO ROUGH :)



 I made the mistake of tagging the bicentennial as a NCN a while
 backooops.


Yeah, but large parts of the BNT aren't rideable on *anything*.


 As for routes i believe that it should only be tagged if its signed as a
 route or there is other strong evidence of it being a route...Wikipedia
 entry below

 A *route* is a customary or regular line of passage or travel, often 
 *predetermined
 and publicized*. Routes consist of paths taken repeatedly by people and
 vehicles


Sounds good to me. I'd emphasise the presence of some kind of physical
signage, even if it's a bit sparse and incomplete.



 I will only tag a LCN/RCN/NCN route if i either see:
 Signs
 Plans from authorities that 'own' the infrastructure (softcopy or
 hardcopy) - e.g. a councils 'cycling plan'
 Literature from associations that have the backing of the 'owning
 authority' - e.g. scenic routes such as the Mawson/Kidman/Goldfields Trek
 etc.


Agreed.



 I don't agree with LCN/RCN/NCN routes that have been tagged:
 apparently popular (unless it is publicized) - an example was a local
 cycling club have tagged their regular road criterium route as an
 LCNuseless to anyone who isn't in the club.


Agreed.


 for a one off event/bike ride - someone once tagged an 'organised and
 popular cycle ride' that occurred only once along the route they tagged.
 This is wrong and just cluttering the map with useless information. The
 following year the event used a completely different route anyway...
 just because the tagger thought it was a 'nice route/ride' - had one of
 these tagged through western Victoria - its still there i think.


Yeah, I think we just need to encourage these people to go to more suitable
places like bikely, mapmyride, strava...



 HOWEVER.MTB routes are a whole different kettle of fish i think which
 needs some more discussion/guidelines as the feature is used in a more
 'unofficial' way with route suggestions from users etc. I am guilty of
 tagging mtb friendly singletracks as 'routes' purely so other MTBers are
 aware that they are legal, ride-able and fun...


Also agreed. mtb=yes is one way of doing that. I would prefer that
route=mtb means there is some kind of official backing behind it, including
a rating, but I definitely your need, and I don't think it causes many
problems.



 Maybe there needs to be a different tag that isnt a route but shows up in
 opencyclemap as 'MTB suggested'


 IMHO, don't be too fixated on OpenCycleMap. The styling is pretty ugly,
the maintainer isn't especially open to feedback, and doesn't seem to care
all that much about tagging outside the UK.

A site I run, http://cycletour.org, does show mtb=yes tags (a pink halo).

https://www.dropbox.com/s/b8kjc1h5v0x8mju/Screenshot%202014-12-15%2013.19.46.png?dl=0

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Cycling network tag

2014-12-08 Thread Steve Bennett
On Mon, Dec 8, 2014 at 11:17 AM, Frank sundowne...@optusnet.com.au wrote:

 The network tag used on cycling routes .. is for example
 National Cycling Network (ncn) as in from one country to another

 In Australia we may not use this tag as per the definition .. but the
 length of the route may mean that in Europe it would be an ncn. So should
 'we' adopt the prractice of 'elevation' the network status? A similar
 practice has been used for highways.


It was quite an eye opener for me earlier this year to cycle in the UK to
discover that they really do have LCN, RCN, and NCN. And they're slightly
different from what I expected: NCN is basically a network that links towns
together, LCN and other stuff, but with the same goal of efficiently
getting from place to place. RCN is a cycle tourism network, and follows
scenic, rather than efficient, routes. (Following an NCN route is often
disappointing...)

Anyway, the practice I've generally followed (in Victoria at least) and I
think others roughly do too is:

LCN: local council routes of minor interest, particularly the on-street ones
RCN: the main network of off-road paths in Melbourne, and all non-mtb
tourism routes (rail trails etc)
NCN: major long-distance routes, like the Mawson Trail.

For my site, cycletour.org, I only render RCN, and I selectively display a
couple of long distance MTB routes.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Moderators password?

2014-12-08 Thread Steve Bennett
Yep, Ben has passed it on, thanks. It turns out the queue was all spam. :)

Steve

On Mon, Dec 8, 2014 at 9:11 AM, Charles Gregory osm.li...@chuq.net wrote:

 According to the list info page, moderators are: ben.kelley, stevagewp and
 inas66+osm, all at the same gmail.com address.  Hopefully one of the
 others should have the password?

 Regards,

 Charles


 On Mon, Dec 8, 2014 at 9:08 AM, Steve Bennett stevag...@gmail.com wrote:

 Hi,
   I haven't looked at this list in a while, but I'm getting a lot of
 moderator request(s) waiting. So I must be a moderator. I don't seem to
 have the password though. Does anyone know it? Currently seeing 13 requests
 in the queue.

 Steve


 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 https://lists.openstreetmap.org/listinfo/talk-au



___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Railways

2014-12-07 Thread Steve Bennett
On Fri, Nov 28, 2014 at 6:28 PM, Matt White mattwh...@iinet.net.au wrote:

  This particular thing really annoys me. Fair enough if the track is still
 in place - go nuts mapping it - but there are disused rail lines marked
 up even when the track hasn't been in place for 30 years, and it's only an
 historical curiosity. See the Inner Circle railway in Melbourne for
 example: http://www.openstreetmap.org/#map=19/-37.78015/144.97908

I've been tempted to nuke this disused railway for ages, especially as
 the current path doesn't actually reflect where it even use to run with any
 accuracy.

 Hi Matt,
  Didn't we (or maybe it was someone else) argue about this a year or two
ago? Please don't nuke it. There are lots of physical traces of the former
railway present, like platforms, pieces of rail, stations, infrastructure
like a former electrical substation, etc.

I don't know what you mean by only an historical curiosity. It is of
historical interest to many people, and is as real and concrete as other
historical ruins.

If you think the current geometry is inaccurate, please improve it if
possible.

Thanks,
Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


[talk-au] City of Melbourne data imports

2014-08-19 Thread Steve Bennett
Hi guys,
  I'm working with the City of Melbourne on some open data projects.
www.theage.com.au/it-pro/government-it/an-open-data-vision-20140818-105d5n.html

One thing we could do is import some of their data into OSM. For example:
- water fountains
- addresses
- businesses
- trees

I haven't done any automated imports before, but I'm sensible enough to go
through the procedures, triple check that we're not duplicating existing
data etc.

Just wanted to check what you think of this idea, and whether anyone has
worked with CoM data before?

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Vicmap data copying

2014-05-23 Thread Steve Bennett
On Tue, May 20, 2014 at 8:28 AM, Warin 61sundow...@gmail.com wrote:

 What about this confusing one: http://bit.ly/1hXvwVK

  The picnic ground/campsite is literally signed No Name, and that's how
 everyone refers to it. I have no idea what the history is. (And there's a
 corner on the way up Mount Buller called 'Unnamed corner').



 Well if you add the tag source:name then it should be clear that it is a
 real name? Rather than a description .. like Service Road, No Public
 Access ... in some ways I don't mind that in the name tag as it does
 convey information that may not be avaliable otherwise.


Yes, it's unambiguous - but still confusing. One of my friends on this
recent cycling trip down the Snowy thought the No Name on his GPS was a
mistake in OSM. Maybe in a weird case like that we should make it name=No
Name (ie, actually include the quotes), or name=No Name Picnic Ground.



 The tag unsigned come from the wiki
 http://wiki.openstreetmap.org/wiki/Proposed_features/Noname
 I don't like the finality of the tag noname as that implys there is no
 name at all... the sign may be missing .. but it may still have a name.
 Even just a local name that the locals use to idntify it.


Agreed.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Vicmap data copying

2014-05-19 Thread Steve Bennett
On Mon, May 19, 2014 at 4:22 PM, David Bannon dban...@internode.on.netwrote:


 No, no Steve, I worded my last letter really badly and totally apologise
 if I unintentionally offended anyone. My comment related specifically to
 your line -


Ok, no worries :)



  Yeah. I'm still deciding what to do about places where Vicmap shows a
  track in the bush that can't be seen on any imagery - probably because
  the vegetation is too dense.

 I meant leave the 'grey' areas to the survey people. There are many
 roads (and particularly tracks) that cannot been seen clearly on the
 imagery, and many more where some parts cannot be seen. I'd rather the
 people working with imagery or other non (recent) survey data such as
 Vic Maps did not make educated guesses but go and have a look, or ask
 some else to go and have a look.


Yeah, it's a real issue. Like I said, I'm still trying to work out how best
to proceed. My general approach is to be a bit more liberal with roads that
don't go through, and more conservative with ones that do. I'm also trying
to use fixme=* to express doubt:

fixme=unverified from vicmap (ie, I can't see through the vegetation, so
I'm taking vicmap's word for it)
fixme=verify access (I can see a track, vicmap has a track, but I'm still a
little skeptical that it's public access)

Sometimes I also use highway=path rather than highway=track if I'm dubious
that the public can drive a vehicle down it.

My intention in all this is to minimise the chance that someone gets routed
down a road that is not publicly accessible, or otherwise impassable.
Personally, I think it's ok to show dead-end 4WD tracks that happen to not
be driveable, because I think the people that use those kinds of maps
expect that. But definitely willing to discuss this point, and open to all
opinions...



 I have had a road (into a new estate) removed, apparently because it did
 not show up on Bing. Very annoying to a new owner there who was
 directing tradies via OSM ! But that in no way means I don't value the
 armchair mappers contribution.


I'd suggest adding notes on to the road in question, like note=This road
was built in early 2014 and is surveyed.


 I'd just like them to double check their
 data, one way or another before committing.


To be honest, I go so much faster if I'm not doing any checking - you might
be surprised how fast I map :) It doesn't really make sense to spend 10
minutes verifying a road that I created in 10 seconds - I just take the
chance that I'm introducing a couple of errors. But mostly I'm doing stuff
out in the bush. I'm usually pretty cautious about deleting anything like
you describe. I have come across a couple where I just couldn't fathom why
someone had drawn a road somewhere, but I'll usually cross-check against a
couple of other sources.

Example: http://www.openstreetmap.org/way/172612474


 Maybe what we need is some sort of register ? The people studying
 imagery are good at picking up anomalies, differences between image and
 map. They could log it and have some local go and check ? Better than
 just jumping in.


Are there enough of us to make it worthwhile? Anyway, a better mechanism
would probably be through fixme=*, so you can go and look for fixme's in
your area at your leisure.



 You may be amused to know that some years ago, I was shocked to discover
 I had apparently built my house in the middle of the Bendigo Region
 National Park. I was waiting to get a letter telling me to move it when
 I realised someone had just followed the tree line, assuming all was
 national park. They had swept up the Park it self, the Welsford State
 Forest, Sugarloaf Conservation Park and a large number of private
 properties. A very quick check would have prevented that error.


Yeah, that seems pretty silly. Although IMHO we need a better approach to
maintaining administrative boundaries - it doesn't really make sense for
anyone to be able to move them at will, since there is a genuine authority
for each.



 I am pretty sure all we want is for the database to have accurate,
 relevant data.


You left out comprehensive and useful. I think I have a higher
tolerance for error because I want OSM to be useful and complete-ish *now*.
I use it on a regular basis for planning trips, and I can't wait a few
years for all the checking. I'd rather a pretty complete map with a few
errors which will be corrected over time.

But that's all it is - different priorities. Maybe I think 98% accuracy is
enough, whereas you want 99.5% - and someone else might want 99.95%...

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Vicmap data copying

2014-05-19 Thread Steve Bennett
On Mon, May 19, 2014 at 7:55 PM, Warin 61sundow...@gmail.com wrote:

  Humm .. there are places I've been before GPS... One example:

 I know the road is there as I've been on it. However it is now closed for
 vehicles - inside a National Park. I've mapped bits of  it into OSM as it
 may be of use to walkers. The bit I cannot 'see' with imagery I've
 connected with very apparent straight lines. I do have copyright maps of
 the area but I'm not looking at those now (they were current when I was
 there ... many years ago!). I'm not going back there, I've many other (new
 to me) places to go. Nor would I request someone to go there. Someone
 probably will go there .. but I'll leave their interest and trip up to
 them. So I'm adding stuff that I think is of use, an indication rather than
 accurate in some places .. but those bits are straight lines and anyone who
 knows the area will know that those are not 'truth'.


Yeah, someone (you?) added lots of tracks through the Victorian Alps in
very low detail. It was actually incredibly helpful, and really motivated
me to go through and improve them all - rather than starting from a blank
slate. And in certain areas, I get a real kick out of doing very high
quality aerial mapping like this: http://bit.ly/1hXv9KZ


 I've been deleting the tag name={Unnamed] and adding the tag
 unsigned=yes, in one case I 'know' the roads name .. but my source I
 remember as a street directory . So I cannot use it untill the memory fades
 a bit more. In most cases I've been past some of the roads .. and there is
 no street sign (indeed most have no power poles nor street lighting). In
 this case I think it can be changed without 'knowing' as the intent is
 clear - there is no local sign to get a name from .. at least not when the
 tag was added.

Sounds sensible.

What about this confusing one: http://bit.ly/1hXvwVK

The picnic ground/campsite is literally signed No Name, and that's how
everyone refers to it. I have no idea what the history is. (And there's a
corner on the way up Mount Buller called 'Unnamed corner').

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Vicmap data copying

2014-05-15 Thread Steve Bennett
Hi Nick,
  From Li Xia's email on 10/10/13: I had a meeting with Vicmap staff today
in regards to importing Vicmap data into OSM under the CC license. They are
very excited about the community showing interest in their data and are
have clarified that importing it is fine.

I'm not clear on whether we need to add any attribution tags, but for now
when I trace stuff from Vicmap, I just add source=vicmap.

IMHO some small scale imports may be useful, but from my comparisons, the
VicMap data is not necessarily better than OSM. It often has stuff OSM
doesn't, but sometimes that includes spurious stuff like roads that no
longer exist, never did, etc.

Steve


On Thu, May 15, 2014 at 4:55 PM, Nick Hocking nick.hock...@gmail.comwrote:

 Am I correct in saying that it is permissable to copy street names from
 the VicMap into OSM?
 Also - what about the house numbers, is that ok as well?

 I have neither the time, talent or inclination to do an import of house
 numbers, but would help out in any manual effort to add all house numbers
 for Victoria
 into OSM. Is such an import envisaged because, if so, then I wouldn't want
 to muddy the waters by starting to manually add them.

 Also - I remember someone saying that Gold Coast roade name data was
 available. Is this available yet for josm or potlach since I would love to
 get the rest
 of the roads named up there?

 Cheers
 Nick

 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 https://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Vicmap data copying

2014-05-15 Thread Steve Bennett
On Thu, May 15, 2014 at 5:52 PM, Nick Hocking nick.hock...@gmail.comwrote:

 Steve wrote

 IMHO some small scale imports may be useful, but from my comparisons, the
 VicMap data is not necessarily better than OSM. It often has stuff OSM
 doesn't, but sometimes that includes spurious stuff like roads that no
 longer exist, never did, etc.


 Thanks Steve,

 As far as importing goes, I'm only talking about house numbers (since they
 are so hard to collect by survey).


Yeah, house numbers are probably a really good example where in most places
we have zero data.


 I definitely think that road names must NOT be imported but added
 individually, where current osm data and bing imagery indicate that there
 really is a road (currently OSM unamed) there.



Yeah. I'm still deciding what to do about places where Vicmap shows a track
in the bush that can't be seen on any imagery - probably because the
vegetation is too dense.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


[talk-au] VicMap in Potlatch2

2014-05-13 Thread Steve Bennett
Hi all,
  I've just discovered you can add VicMap, Victoria's open data licensed
authoritative mapping service, into Potlatch 2. It's not at all obvious
how, so here's the answer:

In Background, click Edit then add:
http://whoots.mapwarper.net:80/tms/$z/$x/$y/WEB_MERCATOR/http://api.maps.vic.gov.au/geowebcacheWM/service/wms?VERSION=1.1.1TILED=true

To explain:
- Vicmap provides an API which is basically a weird forked version of
OpenLayers:
http://api.maps.vic.gov.au/index.php/developers-resources/javascript-api
- Digging through that you can find their actual WMS endpoint, which is
http://api.maps.vic.gov.au/vicmapapi/map/wms
- Except, that data is in EPSG 3111 projection, rather than the web
standard EPSG 3857. Their documentation says that EPSG 3857 will be
available by the end of August 2013, then no more info about it:
http://api.maps.vic.gov.au/index.php/product-information/projections-sacles/web-mercator
- Anyway it turns out the EPSG 3857 projection is available at
http://api.maps.vic.gov.au/geowebcacheWM/service/wms (I assume the WM
stands for 'web mercator', aka EPSG 3857)
- But still, it's only provided as a WMS service, which isn't directly
supported by Potlatch 2. That's why you need to use the
whoots.mapwarper.netproxying service which converts WMS to web tiles.

Enjoy.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Adding residential properties?

2013-12-10 Thread Steve Bennett
On Fri, Dec 6, 2013 at 8:31 AM, Daniel O'Connor daniel.ocon...@gmail.comwrote:

 I started out with buildings, but got a bit excited in my local area;
 getting down into trees, power lines, fences, driveways etc.

 http://www.openstreetmap.org/#map=19/-34.84928/138.52277

 Not super pretty looking.


Hi guys,
  Personally I'm not that into micromapping, but could I make the
suggestion that you use (additional) tags to make it easy to filter this
stuff out? It makes it hard to render a reasonable map if there's no
distinction between, say:
* relevant public buildings, vs private houses  (use building=house sounds
good)
* walking paths through parks, vs ordinary footpaths (use sidewalk=yes?)
* public or semi-public tennis courts (eg at a sporting facility, vs
private tennis courts in backyards (access=private?)
* similarly for swimming pools

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [OSM-talk] Bitcoin Spam

2013-12-03 Thread Steve Bennett
On Wed, Dec 4, 2013 at 9:10 AM, Frederik Ramm frede...@remote.org wrote:

 However, in the case at hand, it seems that the interest is not to
 improve OSM but instead we're just a vehicle for people to show up on

 the coinmap, a business directory for bitcoin-accepting businesses.

I don't think we should worry about people's motivations. What's the
problem here? That there are business POIs in OSM that are missing tags. It
doesn't sound all that different to me from any other data quality problem.
Either we fix the missing tags (if possible), or delete them as junk. And
if the business in question doesn't deserve a mention in OSM (eg, a mail
order place with no shop front), again, just delete it.

No?

Steve
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Long tail challenges - was: Re: Bitcoin Spam

2013-12-03 Thread Steve Bennett
On Wed, Dec 4, 2013 at 12:53 PM, Martin Koppenhoefer dieterdre...@gmail.com
 wrote:

 There are many businesses where you can't go and buy something, but still
 they should be in OSM. Think of a car manufacturer, you can't go there and
 buy a car.


Yeah but the difference is a manufacturing plant has a big physical
presence that is worth indicating on a map. A person can see it, wonder
what it is, and look at the map. And, more pragmatically, factories don't
come and go as quickly as businesses run on residential premises.

There's no way OSM really wants to be a database of every single registered
(or even unregistered!) business, surely?

Steve
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk


Re: [talk-au] Bicentennial National Trail

2013-11-30 Thread Steve Bennett
On Wed, Nov 27, 2013 at 6:30 AM, Ian Sergeant inas66+...@gmail.com wrote:

 Hi,

 It seems the point of the three relations is to identify which parts
 of the trail are accessible to which categories of users.  How do you
 intend to encapsulate that info?

 What is the basis for splitting the trail into state sections, and
 putting three relations into another reln?  I don't think relations of
 relations is well supported, and I can't see the motivation for it
 here.


Hi guys,
  I noticed the three-way duplication but assumed it was for a different
reason: so that, say, a hiking map that looks for route=hiking relations
will show the BNT, a mountain bike map that looks for route=mtb will also
show it etc. Unfortunately I think this is basically legitimate: if the
same route is a hiking, cycling and mountain biking route (and we haven't
even done horse riding yet) then it probably needs those duplicates.

(FWIW, that's a bit of an if - most of the Victorian section is pretty
useless for cycling, and not great for unsupported hiking either.)

Btw you can see both the BNT and AAWT on my map, http://cycletour.org -
just zoom in a couple of clicks.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


[talk-au] Wednesday December 11th: meetup with VicRoads

2013-11-30 Thread Steve Bennett
Hi guys,
  I'm helping organise an open data meetup with VicRoads called Meet the
data owners:

http://www.meetup.com/Datahack-Melbourne/events/152600552/

The idea is to help build relationships between data consumers
(particularly developers) and Victorian government data providers, in this
case VicRoads. It's not specifically to do with OpenStreetMap, but I
thought some people on this list might have a lot to contribute to the
question: what could we do if VicRoads made more of their road and traffic
data available?

Anyway, if you're in Melbourne on Wednesday week, please RSVP and come
along. If it's a success there will be more of these in the future with
other Vic gov departments and agencies - some spatial, some not.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] data.qld.gov.au explicit permission request

2013-11-30 Thread Steve Bennett
Hi Jason,
  Nice work - any response?

Steve


On Fri, Nov 15, 2013 at 2:49 PM, Jason Ward jasonjwa...@gmail.com wrote:

 Hi everyone,

 My apologies if this has already been posted but I've just sent off a
 request to data.qld.gov.au for explicit permission to incorporate and
 publish their applicable datasets.  I've provided them a link to the SA
 Government response to the same permission request so I hope it is just a
 formality that one of their legal eagles can complete.

 SA Link provided to them:
 http://wiki.openstreetmap.org/wiki/Attribution/sa.data.gov.au_explicit_permission

 I'm pretty new to OSM and figured I'd better not waste my effort already
 expended by not having this type of permission supplied.  I know that the
 whole dataset is CC-BY-3.0 AU so I'm not even sure if my request is
 required (but I figure there is no harm in seeking it).

 --
 Cheers,

 Jason


 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 https://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


[talk-au] Bridleways and tagging rail trails

2013-10-01 Thread Steve Bennett
Nick Barker sent me a message through openstreetmap.org - hope you don't
mind me replying here:

I've just been going through some edits and i noticed you changed the
Tyers rail trail back to a path from a bridleway and just was wondering
why.

The reason i tagged it as a bridleway as apparently on the rail trail
website it says horses are allowed. Looking at trail hierarchies i
therefore tagged it as a bridleway as

walkers only = path
 walkers and cyclists (obviously path for MTB only but its definitely
doable on a hybrid - the rail trail website err well on the side of
caution) = cycleway
 walkers cyclists and horses = bridleway. (note this is a UK/european
system as far as i know)
i asked this on the forum and people sort of agreed in the absence of any
specific aussie system this made the most sense.

Likewise for the Walhalla rail trail too.
What are you thoughts on this?

Thanks for raising this. My thoughts are that the basic tag should be as
representative as possible. There are lots of different websites, apps etc
that use openstreetmap, so we should tag whatever is most useful.
Cycling-focused maps will tend to show highway=cycleway and ignore
highway=bridleway. I don't know if there are equestrian-focused maps -
horseriding is a pretty fringe activity compared to cycling.

A bridleway is not just a path that horse riders can go on, it's a bit
unique. Typically the surface is very soft, like grass or sand, and they
often have cavalettis to stop other kinds of traffic. Although you might be
able to ride bikes on some bridleways, that wouldn't be the default
assumption. We do have a few bridleways around the place, mostly in outer
outer suburbs with established horse riding areas, like Lysterfield,
Victoria.

So, the tagging that makes the most sense to me:
highway=cycleway: a path that is designed for (non-mtb) cycling (and nearly
always caters for pedestrians too). If horses are allowed, then horse=yes
highway=bridleway: a path that is specifically designed for horse riders,
pretty much to the exclusion of others.
highway=footway: a good quality path for pedestrians
highway=path: other paths that could be used by hikers or mountain bikers,
that aren't of the same quality as a cycleway/footway.

Tagging a rail trail as a bridleway doesn't make much sense to me for a few
reasons:
1) The number of horse riders is a tiny fraction of the number of cyclists
2) Some rail trails have sections that split between a designated bridleway
(off the railway alignment) and the main rail trail. Tagging the main rail
trail as highway=cycleway, horse=yes/no (and then a separate
highway=bridleway) is clear and makes sense. Switching the main rail trail
from highway=bridleway to highway=cycleway would be kind of bizarre.

For example:
http://www.openstreetmap.org/#map=18/-37.70318/143.63900layers=N

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


Re: [OSM-talk] Who interprets semicolon in tag values?

2013-09-30 Thread Steve Bennett
On Mon, Sep 30, 2013 at 9:08 PM, NopMap ekkeh...@gmx.de wrote:

 The original example wasn't mine, but I think it is very common that a pub
 has much later opening hours than a restaurant. Having both as part of the
 same enterprise is a british speciality if I remember my visits correctly.

 And mapping them as separate POIs is the only practical and human readable
 way we currently have available, that's the point. :-)


I have to agree with this. The most common clashes I run into mapping towns
in Australia;

bakery/cafes (technically you could do amenity=cafe, shop=bakery)
general store/cafe
general store/fuel
general store/post office
pub/hotel (I tend to create two nodes, in roughly the right places)
pub/restaurant (in the country, I take the restaurant as assumed - every
pub serves meals at least some days of the week)
cafe/bar

A general solution that would solve these and other problems would be a
relation to group parts of the same business. That way, you could have the
expressiveness of precise tagging (eg, is there a bakery near here? yes.
is there a cafe near here? yes) but easy rendering: show one icon, and the
renderer can decide whether it's a cafe or bakery icon.

Steve
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Making iD the default editor on osm.org - some numbers

2013-08-25 Thread Steve Bennett
On Tue, Aug 20, 2013 at 9:36 PM, SomeoneElse li...@mail.atownsend.org.ukwrote:

 In an attempt to put some numbers to to the errors made by new mappers
 debate, I've done a count-back of new users and editors that they use for
 they area that I keep an eye on in the UK (England and bits of Wales, not
 including bits that I'm unfamiliar with such as London and the south-east)



Hi,
  Just wondering what tools you use to keep an eye on that area? I'd love
to have a better idea of what other editors are doing in my area.

Steve
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Slow TileMill rendering - Postgres using 1 core?

2013-08-25 Thread Steve Bennett
Hi Christian,
  Thanks for the reply. After further investigation the actual problem was
a missing index on planet_osm_polygon. (I'm really not sure why.)

Steve


On Sun, Aug 25, 2013 at 9:00 PM, Christian Quest cqu...@openstreetmap.frwrote:

 As far as I know, Tilemill is using mapnik which is querying postgres.
 Plain vanilla Mapnik is not doing more than one postgres query at a time
 (not multithreading queries).
 A patch made by mappy allows mapnik to multithread its pg queries. Are you
 using the exact same version of Mapnik as before ?


 2013/8/25 Steve Bennett stevag...@gmail.com

 Hi all,

 I'm running TileMill on an 8 core Ubuntu VM with 32GB of memory, on an
 OpenStack cloud. Recently, my VM was destroyed, and I rebuilt it
 (identically, I thought) on slightly different hardware (same cloud, but
 different physical infrastructure).

 The new build is much slower at rendering - a screen worth of tiles at
 zoom 13 can take around a minute. That is, with virtually the same setup,
 same data, same styles. You can see some slow tiles here:

 http://emscycletours.site44.com/mel.html

 While panning around, the 'top' command shows mostly Postgres processes
 (different from last time I had performance problems[1], when the
 bottleneck was in Mapnik). Total CPU usage hangs around 12%: ie, exactly 1
 out of 8 cores is being used.


 https://dl.dropboxusercontent.com/u/767553/GIS/Screen%20shot%202013-08-25%20at%2011.15.01%20AM.png

 top - 11:10:32 up 3 days, 36 min,  1 user,  load average: 0.06, 0.17, 0.22
 Tasks: 133 total,   4 running, 129 sleeping,   0 stopped,   0 zombie
 %Cpu(s): 11.5 us,  0.1 sy,  0.0 ni, 88.4 id,  0.0 wa,  0.0 hi,  0.0 si,
  0.0 st
 KiB Mem:  32950396 total,  7150132 used, 25800264 free,   117864 buffers
 KiB Swap:0 total,0 used,0 free,  5221356 cached

   PID USER  PR  NI  VIRT  RES  SHR S  %CPU %MEMTIME+  COMMAND
  2353 postgres  20   0 8510m 640m 635m S  41.2  2.0  23:36.57 postgres
  2354 postgres  20   0 8510m 644m 639m S  40.2  2.0  23:24.26 postgres
  2350 postgres  20   0 8510m 642m 638m S  14.0  2.0  23:19.19 postgres
  2375 postgres  20   0 8510m 643m 639m S  14.0  2.0  23:17.80 postgres
 13102 postgres  20   0 8508m 531m 527m S  13.6  1.7  13:03.21 postgres
  2355 postgres  20   0 8508m 531m 526m S  13.3  1.7  13:45.15 postgres
  2352 postgres  20   0 8510m 640m 636m S  10.0  2.0  23:31.17 postgres
  2348 postgres  20   0 8510m 644m 639m S   9.3  2.0  23:41.88 postgres
 12420 mapbox20   0 3818m 1.0g 755m S   9.3  3.2  36:48.39 nodejs
  2357 postgres  20   0 8508m 530m 526m S   7.3  1.7  13:38.57 postgres
  2356 postgres  20   0 8508m 531m 526m R   6.3  1.7  13:42.52 postgres
  2376 postgres  20   0 8508m 531m 527m S   6.0  1.7  13:35.51 postgres
 13195 postgres  20   0 8508m 531m 527m S   5.3  1.7  12:33.65 postgres
  3027 postgres  20   0 8508m 531m 527m R   3.3  1.7  13:29.06 postgres
  2349 postgres  20   0 8508m 530m 526m S   3.0  1.6  13:38.19 postgres
  2358 postgres  20   0 8508m 531m 527m S   3.0  1.7  13:44.59 postgres
26 root  20   0 000 S   0.3  0.0   0:08.64 ksoftirqd/5
  2335 postgres  20   0 8489m 2732 1340 S   0.3  0.0   1:00.48 postgres

 So, wondering if anyone has any suggestions what the problem is, or how
 to fix it? Why is Postgres apparently using only one core, even though it
 has many processes? What tools could I use to further diagnose?

 My changed Postgres settings are as follows:

 shared_buffers = 8GB
 autovacuum = on
 effective_cache_size = 8GB
 work_mem = 128MB
 maintenance_work_mem = 64MB
 wal_buffers = 1MB
 checkpoint_segments = 10

 The server is set up as described here:
 http://steveko.wordpress.com/2013/05/08/tilemill-server/

 I'm not yet using any tile cache. I will do that next, but the problem
 I'm trying to solve at the moment is very slow tile generation, not slow
 serving of rendered tiles.

 Many thanks in advance,
 Steve


 [1]
 http://gis.19327.n5.nabble.com/TileMill-performance-td5751158.html


 ___
 talk mailing list
 talk@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk




 --
 Christian Quest - OpenStreetMap France
 Un nouveau serveur pour OSM... http://donate.osm.org/server2013/

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


[OSM-talk] Slow TileMill rendering - Postgres using 1 core?

2013-08-24 Thread Steve Bennett
Hi all,

I'm running TileMill on an 8 core Ubuntu VM with 32GB of memory, on an
OpenStack cloud. Recently, my VM was destroyed, and I rebuilt it
(identically, I thought) on slightly different hardware (same cloud, but
different physical infrastructure).

The new build is much slower at rendering - a screen worth of tiles at zoom
13 can take around a minute. That is, with virtually the same setup, same
data, same styles. You can see some slow tiles here:

http://emscycletours.site44.com/mel.html

While panning around, the 'top' command shows mostly Postgres processes
(different from last time I had performance problems[1], when the
bottleneck was in Mapnik). Total CPU usage hangs around 12%: ie, exactly 1
out of 8 cores is being used.

https://dl.dropboxusercontent.com/u/767553/GIS/Screen%20shot%202013-08-25%20at%2011.15.01%20AM.png

top - 11:10:32 up 3 days, 36 min,  1 user,  load average: 0.06, 0.17, 0.22
Tasks: 133 total,   4 running, 129 sleeping,   0 stopped,   0 zombie
%Cpu(s): 11.5 us,  0.1 sy,  0.0 ni, 88.4 id,  0.0 wa,  0.0 hi,  0.0 si,
 0.0 st
KiB Mem:  32950396 total,  7150132 used, 25800264 free,   117864 buffers
KiB Swap:0 total,0 used,0 free,  5221356 cached

  PID USER  PR  NI  VIRT  RES  SHR S  %CPU %MEMTIME+  COMMAND
 2353 postgres  20   0 8510m 640m 635m S  41.2  2.0  23:36.57 postgres
 2354 postgres  20   0 8510m 644m 639m S  40.2  2.0  23:24.26 postgres
 2350 postgres  20   0 8510m 642m 638m S  14.0  2.0  23:19.19 postgres
 2375 postgres  20   0 8510m 643m 639m S  14.0  2.0  23:17.80 postgres
13102 postgres  20   0 8508m 531m 527m S  13.6  1.7  13:03.21 postgres
 2355 postgres  20   0 8508m 531m 526m S  13.3  1.7  13:45.15 postgres
 2352 postgres  20   0 8510m 640m 636m S  10.0  2.0  23:31.17 postgres
 2348 postgres  20   0 8510m 644m 639m S   9.3  2.0  23:41.88 postgres
12420 mapbox20   0 3818m 1.0g 755m S   9.3  3.2  36:48.39 nodejs
 2357 postgres  20   0 8508m 530m 526m S   7.3  1.7  13:38.57 postgres
 2356 postgres  20   0 8508m 531m 526m R   6.3  1.7  13:42.52 postgres
 2376 postgres  20   0 8508m 531m 527m S   6.0  1.7  13:35.51 postgres
13195 postgres  20   0 8508m 531m 527m S   5.3  1.7  12:33.65 postgres
 3027 postgres  20   0 8508m 531m 527m R   3.3  1.7  13:29.06 postgres
 2349 postgres  20   0 8508m 530m 526m S   3.0  1.6  13:38.19 postgres
 2358 postgres  20   0 8508m 531m 527m S   3.0  1.7  13:44.59 postgres
   26 root  20   0 000 S   0.3  0.0   0:08.64 ksoftirqd/5
 2335 postgres  20   0 8489m 2732 1340 S   0.3  0.0   1:00.48 postgres

So, wondering if anyone has any suggestions what the problem is, or how to
fix it? Why is Postgres apparently using only one core, even though it has
many processes? What tools could I use to further diagnose?

My changed Postgres settings are as follows:

shared_buffers = 8GB
autovacuum = on
effective_cache_size = 8GB
work_mem = 128MB
maintenance_work_mem = 64MB
wal_buffers = 1MB
checkpoint_segments = 10

The server is set up as described here:
http://steveko.wordpress.com/2013/05/08/tilemill-server/

I'm not yet using any tile cache. I will do that next, but the problem I'm
trying to solve at the moment is very slow tile generation, not slow
serving of rendered tiles.

Many thanks in advance,
Steve


[1]
http://gis.19327.n5.nabble.com/TileMill-performance-td5751158.html
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Bringing new life to the OSM.org front page

2013-07-23 Thread Steve Bennett
On Sun, Jul 21, 2013 at 4:16 AM, Rob Nickerson
rob.j.nicker...@gmail.com wrote:
 Finally, a few other things to note:

 * With over 1 million registered users it is impossible to please everyone.
 Please put things into perspective before replying with negative comments.
 Also consideration of the amount of development time is always appreciated.

 * OSM has numerous communication channels (too many maybe). Where would you
 stop if it was decided to post announcements to the community. Furthermore
 as the OSM Foundation did not ask for permission to send email to registered
 users they cannot therefore push notifications to all users (rather users
 will need to read the communication channels that interest them).

My suggestion: when big, highly visible changes are coming, that
non-dev community members might have an interest in, consider just
dropping a line to talk@ like this: Hi, we're working on some cool
new map controls for the front page. Check out the discussion here:
link.

(I don't think it's practical for non-developers to lurk on
development mailing lists on the off-chance that something interesting
will pop up.)

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


[OSM-talk] Application error at openstreetmap.org

2013-07-14 Thread Steve Bennett
--
Application error

The OpenStreetMap server encountered an unexpected condition that
prevented it from fulfilling the request (HTTP 500)

Feel free to contact the OpenStreetMap community if your problem
persists. Make a note of the exact URL / post data of your request.

This may be a problem in our Ruby On Rails code. 500 occurs with
exceptions thrown outside of an action (like in Dispatcher setups or
broken Ruby code)


Surprised not to see anything about it on this mailing list. Can anyone explain?

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [talk-au] Historic stations and rail

2013-07-10 Thread Steve Bennett
On Tue, Jul 9, 2013 at 1:21 PM, Nathan Van Der Meulen
natvan...@yahoo.com wrote:
 I use name=Thompson (disused) as the map user sees immediately that the
 station is no longer in use.  This may not be such a drama where the track
 is out of use as well, but when you map a station out of use on a running
 line if the station appears just like any other station it becomes
 confusing.  This is also consistent with the method used by various map
 makers.

Yeah, that maybe makes sense if a station is temporarily out of use -
can you think of an example? But generally, any renderer that supports
historic:railway=station is going to render it sufficiently
different from railway=station that it's not needed.

The main mapnik style doesn't support historic:railway=station, so
currently it will just disappear - which isn't really such a bad
thing.

PS If you're replying to digest emails, would you mind trimming off
all the other bits? That was quite a wall of text...

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


[talk-au] Historic stations and rail

2013-07-08 Thread Steve Bennett
Hi all,
  I've been working on some old train lines and stations, mostly
around Victoria, and have finally settled on a tagging scheme, based
on what's popular in taginfo. The current tagging has been pretty
inconsistent, with lots of variations like railway=station,
disused=yes, railway=station, name=Foo (former),
railway:historic=station, historic:railway=station,
building=station...

Thought I'd check whether anyone has any comments/disagreements:

Active stations (whether tourist, regional, commuter or freight):
railway=station

Disused/abandoned stations, with buildings present:
railway:historic=station

Abandoned, former station site, no buildings, little to see:
railway:historic=station_site

In all cases, the name of the station object should just be, eg, Thompson.
Not: Thompson (former)
Not: Thompson (disused)
Not: Thompson Railway Station

In addition, a station building may be tagged as:
building=station

(But one of the above should also be present - either on the building
polygon or not.)


Active rail (commuter, freight, regional):
railway=rail

Tourist rail
railway=preserved

Disused, but tracks still present and conceivably services could return:
railway=disused

Tracks removed, or state is pretty far gone.
railway=abandoned

Thoughts?

To see where we're at, I've done a quick replica of this well known
rail map of Victoria:
http://i.imgur.com/ohXHwsk.png
http://emscycletours.site44.com/rail.html

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Vicmap data released on data.gov.au

2013-07-08 Thread Steve Bennett
Hi Nyall,
  Yeah I'm quite interested to know more about this. I gather the
Spatial DataMart has been around a while, so is the difference that
they've published it on data.vic.gov.au and made it accessible to the
public? Or is it the licensing that has changed?

Might be worth having a meetup to discuss what we can do with all this data.

Steve

On Mon, Jul 8, 2013 at 1:43 PM, Nyall Dawson nyall.daw...@gmail.com wrote:

 All we need from them is a statement that a notice associated with the
 Produced Work reasonably calculated to make any Person that uses, views,
 accesses, interacts with, or is otherwise exposed to the Produced Work aware
 that Content was obtained from the Database, Derivative Database, or the
 Database as part of a Collective Database, and that it is available under
 this License (ODbL 4.3) meets the requirements of a notice reasonable to
 the medium.

 Basically, that the attribution required by the ODbL is enough. Some cities
 have viewed CC BY's  reasonable to the medium to mean every data source
 needs to be credited directly on a web map.


 Great -- this is exactly what I needed to know. I've attempted to get
 in contact with the appropriate person at data.vic.gov.au, I'll report
 back here how this goes.

 Cheers,
 Nyall Dawson

 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-au

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] surface=unsealed in 4wd/dirt road tagging

2013-07-02 Thread Steve Bennett
On Tue, Jul 2, 2013 at 8:42 PM, David Bannon dban...@internode.on.net wrote:
 Just a thought here, we'd really like the renderers to show
 unpaved/unsealed/whatever roads differently from sealed ones. In particular,
 the mapnik rendered slippery map on the OSM website

FYI, the map style I'm working on for cycle touring does make this distinction:
http://emscycletours.site44.com/map2.html#egrt

Also a quick stat for you. 165,000 highways in Australia have a
surface tag. 718,000 don't.

 They show little interest I must admit but will show even less if we point
 to a whole range of surface= tags that need that treatment. So might be
 better agreeing on one term rather than being very specific and saying
 'dirt', 'gravel'  etc. As I have mentioned before, this is a road safety
 issue as much as a presentation one.

You might be right - but on a technical front, it's no more burdensome
to show all of [unsealed, unpaved, gravel, dirt] as a dashed line
rather than just, say, unpaved.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] surface=unsealed in 4wd/dirt road tagging

2013-07-02 Thread Steve Bennett
On Wed, Jul 3, 2013 at 9:03 AM, Ian Sergeant inas66+...@gmail.com wrote:
 Surprising stat.  Especially considering paved is considered the default.

Yeah - I try to specify it wherever possible though, outside cities.

 I really like multi-level tags.

 natural=water
 water=lake

 surface=unpaved
 unpaved=gravel

 surface=paved
 paved=asphalt

 It makes it easy for people two write simple parsers without enumerating the
 options, but people who are want to parse the details to do so.

 There are a number of instances when OSM uses this type of multi-level
 tagging scheme, but it lacks any form of consistency.

Agreed on all counts. There are a lot of people who still think the
best thing to do is make up new tags on the fly and to encourage
others to do the same.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] park vs nature reserve

2013-06-25 Thread Steve Bennett
On Tue, Jun 25, 2013 at 5:42 PM, Andrew Elwell andrew.elw...@gmail.com wrote:
 Can someone point me to guidelines for where the .au distinction
 between the two lies?

Great question. For my part, I'm pretty lazy and just tag everything
as leisure=park, and hope that someone else will clean it up :) Within
cities, it will be hard to come up with clear guidelines to apply the
tags: nature reserves that are called X park, parks that are called
X reserve, parks that contain conservation reserves within them, and
vice versa.

There is some not-very-enlightening text here:
http://wiki.openstreetmap.org/wiki/Australian_Tagging_Guidelines#.28National.2C_State_etc.29_Parks

If you come up with any guidelines it would be great to document them there.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Two cycling maps I made

2013-06-19 Thread Steve Bennett
On Wed, Jun 19, 2013 at 3:16 PM, Brett Russell brussell...@live.com.au wrote:
 Looks great.  Nice to see the output from a lot of mapping effort by
 everyone in the Melbourne/Victoria area.

Yeah, we're getting to a really good level of completeness now. The
biggest shortcoming I'm noticing for this project is a lot of towns
with few/no amenities marked, including a few major ones like
Wonthaggi and Tallangatta.

 Only thing I have noticed is on the
 http://emscycletours.site44.com/map2.html the wording for the Great Divide
 Rail Trail appears twice, super imposed over each other.  Might be that the
 tittle is rendered twice?

Thanks, I've made some styling changes that have cleared that up.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Rivers that have dams on them

2013-06-18 Thread Steve Bennett
On Wed, Jun 12, 2013 at 2:31 PM, Brett Russell brussell...@live.com.au wrote:
 I used OSM on a nine day Overland Track Wall and found it very good with the
 Garmins that I was using.  Used the routable maps and found they were about
 a one kilometre in ten understated on distance due to fewer nodes recorded
 in OSM compared to the distance tracked by the GPSs.  Still very impressive
 compared to the expensive and poor quality Garmin maps.  Need to work with
 someone to get OSM bushwalking maps going as walking tracks, mountains and
 huts need to be zoomed in a long way to see thus you get “lost” in the trees
 finding them.   Also the search feature on the Garmin can be rather
 “broken”, by that items do not appear in the all POI lists but can be found
 in sub lists.  Bit more work needed by my to refine the tracks and the
 features but gradually working my way south to north refining the track.

Glad to hear it - I added the Overland Track a couple of years ago,
with the major side trips. From memory, I just beat someone else who
had also recorded the trace but hadn't uploaded it yet.

Regarding use of 'correct' tags like alpine shelters etc, I tend to
get very pragmatic and use tags like place=locality instead so they
show up. And pretty much a place like Pelion Hut *is* a locality -
it's a major landmark and navigational reference point.

What exactly do you mean by Need to work with someone to get OSM
bushwalking maps going? Going where? On a GPS? On a custom website?
In the default mapnik rendering on openstreetmap.org? Static printed
maps?

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


[talk-au] Two cycling maps I made

2013-06-18 Thread Steve Bennett
Hi all,
  Thought you might be interested to see to see two cycling maps I've
made with TileMill and OSM data:

1) Bike map of Melbourne. It really highlights the major bike routes.
The Capital City Trail is shown as a pattern of yellow dots.
http://emscycletours.site44.com/map.html

2) Cycle touring map of Victoria and rail trails. Villages/hamlets are
only shown if they have at least one
pub/cafe/restaurant/bakery/supermarket/convenience store within a 5km
radius. The size of the purple circle shows how many of those
amenities are present.

http://emscycletours.site44.com/map2.html

Happy to answer any questions about them.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Australia licence change redaction recovery..

2013-05-26 Thread Steve Bennett
On Sat, May 25, 2013 at 9:08 PM, Ian Sergeant inas66+...@gmail.com wrote:
 So, my summary would be that we've probably comprehensively remapped he
 motorways and trunk roads across the country.  We've got significantly more
 tracks, paths and residential/unclassified roads than we had before.  There
 would seem to be artifacts of extensive aerial remapping, with the lower
 percentage overall of named roads, and what I'm thinking could be a
 consequent tendency to underrate what passes for a secondary road in
 Australia.  I'd also attribute greater mapping outside of urban areas to the
 more extensive bing imagery coverage, and possibly the focus of the
 redaction process on urban areas.


Thanks very much for doing this - I've been quite curious about where
we're up to. I had guessed we were about on par - so this is good
news. I've been doing a fair bit of aerial mapping lately - not sure
whether remapping or not. I tend to be pretty conservative with road
classifications on a first pass. Later, I might look at the area and
upgrade a couple of the roads.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [OSM-talk] iD, exclusive use of tags

2013-05-24 Thread Steve Bennett
On Thu, May 23, 2013 at 11:17 PM, Martin Koppenhoefer
dieterdre...@gmail.com wrote:
 Right now when you click on other in this menu, the other tag that is
 considered principal by iD gets silently removed (e.g. if you click on a
 highway on other, the highway-tag is removed).

Yeah, this was/is a limitation of Potlatch2 that was never solved.
It's a messy reality of OSM tagging.

To those that think the answer is well, don't map like that! -
just...no. That style of mapping is perfectly appropriate in many
places and has some real advantages for data consumers (eg,
identifying roads that have tramtracks, which are a real hazard for
cyclists).

 I'd consider both of these bugs, but would like to open a discussion what
 others think about exclusive tags.

IMHO the solution is to essentially duplicate the GUI when more than
one primary tag is found. But there are some unsolved ambiguities in
the tagging, like which tag name= refers to.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] source=Google

2013-05-24 Thread Steve Bennett
On Tue, May 21, 2013 at 9:26 PM, Martin Koppenhoefer
dieterdre...@gmail.com wrote:
 What do you propose to do with source tags found on an object when you
 modify this object based on a different source?

Speaking for myself, I either replace it (if I'm replacing virtually
all the geometry) or supplement it: source=gps;Bing. If I'm tweaking
something that had no source before, I sometimes use
source=unknown;Bing.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [talk-au] OpenStreetMap in Government

2013-05-13 Thread Steve Bennett
On Sat, May 11, 2013 at 7:28 AM, Ian Sergeant inas66+...@gmail.com wrote:

 But your overall point is surely that as long as we have the basics, if some
 group of people want the extra information and are willing to gather it, and
 some other group of people want to use the information and are willing to
 render/route it, then all is good.

Yeah, absolutely. As long as we have the basics, which I'm assuming
means a couple of tags with a couple of well defined meanings (like
2WD, 4WD etc). Then people can go nuts adding extra information, as
long as it doesn't conflict.

Sometimes people think that it's better to slice up information into
lots of little objective facts, like (in the case of mountain bike
trails), width, surface, grade, etc, rather than a subjective fact
like trail rating. But in practice, it's impractical to collect that
much information, and it's impractical to combine it back into a
usable form for data consumers, so we lose twice.

 We're here to use our data in new an innovative ways, right?

Absolutely :)

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Major 4WD tracks

2013-05-10 Thread Steve Bennett
On Fri, May 10, 2013 at 1:47 PM, Li Xia lisxia1...@gmail.com wrote:
 Major 4wd tracks such as

 Birdsville track
 Old telegraph track
 Wonagatta rd
 Etc

Hi Li,
  Still not clear on what you mean by major. Do you mean important,
significant, famous...or do you mean big,well-maintained etc?
If the former, I'd think a route relation (as I described earlier),
but you'd need an authoritative source for what the route is.

PS Don't forget to reply-all.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] OpenStreetMap in Government

2013-05-10 Thread Steve Bennett
Hi all,
  This is a really interesting discussion, and thanks for the insights
about Australia vs Europe vs US. A few comments:

1) I think TileMill/MapBox will be a game changer for the rendering
guys won't listen to us problem. I suspect it will soon be much, much
easier to have lots of different map views out there, and we can
create Australian-specific maps easily. So we should continue to work
out the best tagging system and use that - even if it's not currently
supported by any rendering styles.
2) If we do use tags that are essentially unique to Australia, we
should consider still doubling up with standard tags where convenient.
If 4wd_only means you shouldn't attempt this track without a 4 wheel
drive, even if this particular section is ok, then we can still add
track_type tags to the relevant sections, if known.
3) There are decades of practice in cartography to learn from. We
might as well go with existing practice in current 4WD maps. The
standard distinctions seem to be something like 4WD/2WD/dirt/sealed,
and sometimes one more category indicating something like possibly
impassable. So no need for the 10 point roughness/tracktype scale -
it's too hard.
4) And yes, we should have simple tags that correspond to existing
cartography practice: MVO, (subject to seasonal closure) and 4WD
only.

Steve

On Fri, May 10, 2013 at 3:48 PM, David dban...@internode.on.net wrote:
 Kristy, you have spotted the problem, no clear acceptance of any one standard 
 when it comes to 4wd tracks. And while its being done a number of different 
 ways (or not done at all) we have little chance of getting the rendering 
 people to listen to us.

 In western Europe, little interest, complete lack of understanding of the 
 need. The US does have some great 4wd tracks but they are more recreational 
 in nature, you go somewhere, drive a great track and then go home. They also 
 don't understand our model of using these tracks to get to somewhere really 
 interesting !  Asia, (far) eastern Europe, get it but don't seem to want to 
 support the ideas.

 I believe (strongly) we need a multi level tag that indicates a track is 
 somewhere between a bit dodgy right through to Oh wow. That, by its very 
 nature means its subjective, you and I might well disagree with at what stage 
 a typical SUV and inexperienced driver should be warned off. We cannot help 
 that, 4wds are all different, drivers are different in their skills and 
 willingness to take risks.

  The 4wd_only tag is 'official' and was a good try. But not used very much 
 outside of Oz. And its a yes/no and life is never a yes/no situation. 
 Further, so much OSM data ends up in a psql database, one column per tag. 
 Believe it or not, psql does not like having column names start with 
 numerals. It can be worked around but I suspect that's one reason mapnik (or 
 more correctly, its slippery map) won't show 4wd_only.

 I prefer an extension to the tracktype= tag, its already widely used 
 internationally and, somewhat, rendered on the slippery map. We can add three 
 more levels to it (grade6, grade7, grade8) being possibly not suitable for 
 conventional car, 4wd stuff and 4wd extreme.

 I currently use both 4wd_only= and tracktype=

 But I would support any new, sufficiently flexible proposal.

 I don't really this a physical meet up is necessary, be surprised if we could 
 agree on a convienant location !

 David
 .

 Kristy Van Putten kristy.vanput...@gmail.com wrote:

Hi Matt,
I think your conclusions is right, that we need to put an Australian standard 
together.  It sounds like the ground work has been done (maybe even multiple 
times) but there has not been a clear acceptance of any particular schema.

How do you think we should go forward with this?  My suggestion is that we 
make a weekend of it, where we come together - where there are plenty of 
different types of 4WD tracks - and try and test the schema already made.  I 
know I am still living outside of the country, so for me this maybe hard over 
the next couple of months. I am home in July for a couple of weeks and I am 
sure I could convince someone to lend me a 4WD.  However it is winter, so it 
won't be the warmest weather! Maybe we could wait till summer?

Would anyone be keen?

Cheers



On 06/05/2013, at 4:22 PM, Matt White mattwh...@iinet.com.au wrote:

 I'm also very interested in 4wd trails - it's what 80% of my mapping 
 consists of I think (that, and house numbers in the inner north of 
 Melbourne)

 The current 4wd_only tag was one of the tags I proposed a few years ago - 
 there was a massive barney at the time over the smoothness=* and surface=* 
 tags, and all I wanted to do was mark roads that were clearly tagged as 4wd 
 only (proper 4wd as in low range, high clearance). The surface/smoothness 
 debate was interesting, but got in the way of the larger problem.

 I've come to the conclusion that the Australian mappers pretty much have to 
 go it alone in this area - what the 

Re: [talk-au] Major 4WD tracks

2013-05-09 Thread Steve Bennett
Can you give examples of major 4WD tracks? Do you mean the 4WD route
classification scheme? (I've seen some 4WD tracks near Mt Stirling
that had official signage, difficulty ratings etc). Probably you'd use
a relation like:

type=route
route=4wd
network=???
name=...


But I haven't done any research on this subject.

Steve

On Thu, May 9, 2013 at 3:30 PM, Li Xia lisxia1...@gmail.com wrote:
 Hi Mappers,

 Is there a tag that's officially recognised that can be used to highlight
 major 4WD tracks. Similar tags for hiking, cycling existing under route:xx

 Li.

 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] OpenStreetMap in Government

2013-05-05 Thread Steve Bennett
On Tue, Apr 30, 2013 at 7:29 PM, kristy van putten
kristy.vanput...@gmail.com wrote:
 On a personal note I would be interested in hearing more about the OSM
 Australia activities, and people current goals with OSM.
 I have read about the Bicentennial National Trail team, has anyone thought
 of 4WD trails in OSM?  I would also be keen to find out if there are any
 Ozzy teaching OSM to schools or scout groups etc?

Hi Kristy,
  For my part, I've done a lot of mapping in and around Melbourne, but
am now shifting attention to rural areas, particularly since Bing
imagery has improved a lot. I do a lot of cycle touring, and a bit of
hiking, and have quite an interest in having good data in OSM to
support those activities. There's also a big crossover between the
needs of 4WD-ers and the kind of cycle touring I like to do, so I'm
interested in the issues David Bannon raised. Right at the moment,
though, I'm remapping some areas along the Goulburn (southwest of
Shepparton) that got lost in the licence change.

My take on where the Australian OSM community is at is that we're
still a bit scarred from the hugely disruptive licence changeover, and
the leaving of some of the rather abrasive individuals in the process.
I'd love to see more discussion about goals for the community,
individual projects etc.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


[OSM-talk] What to do with failed (Potlatch) save changeset?

2013-04-14 Thread Steve Bennett
While trying to save a changeset in Potlatch2, I got an error message
about version conflicts, and was left with a big chunk of XML.
Potlatch appears not to be able to do anything with it, but maybe I
can still salvage something from it.

I put it here: https://dl.dropboxusercontent.com/u/767553/OSM/failedupload.osm

Assuming I'm happy to simply lose any changes where there really is a
version conflict (which I am), what can I do with it? (I don't use
JOSM at all, so would prefer to avoid that hurdle if possible...)

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] What to do with failed (Potlatch) save changeset?

2013-04-14 Thread Steve Bennett
On Sun, Apr 14, 2013 at 6:00 PM, Richard Fairhurst rich...@systemed.net wrote:
 The couple of times I've encountered this situation, I've manually edited
 out the conflicting element and then used upload.py to upload it to OSM.

 http://wiki.openstreetmap.org/wiki/Upload.py


Excellent, thanks. I'll try that.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Why do we have so many registered users with zero edits ?

2013-04-14 Thread Steve Bennett
On Sat, Apr 13, 2013 at 11:33 PM, Simon Poole si...@poole.ch wrote:

 Just a further data point. As of the end of last month we had edits from
 324'152 unique UIDs, not quite 30% of all accounts (some of the
 changesets are likely to be empty, but the number is still quite a bit
 larger than the often quoted 200'000).


Is there (or could there be?) a graph of contributions against number
of u-el-? Might be interesting also to see whether the number of users
on zero is vastly different to the number of users on one edit, two
edits etc.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] OpenSnowMap.org

2013-04-09 Thread Steve Bennett
On Sat, Apr 6, 2013 at 8:33 PM, yvecai yve...@gmail.com wrote:
 - could it be more obvious which is the start of each (downhill) run?

 Unless there a tag on the node, it would be hard. Emphasizing each way start
 node wouldn't be nice, as pistes can be made from several ways.

Well, you can already compute gradient profiles, so maybe you could
use that. And isn't there a convention that downhill routes are mapped
in the downhill direction?

 - would be great to have a summer/winter mode button (in summer mode,
 show the mountain bike trails, hiking routes etc)

 Then it's not opensnowmap anymore :)

Yeah, it's even more useful. Seriously though, most ski resorts
(around here at least) have a green and white mode. Many of the
functions of your site would also be useful for other activities at
the resort.

 What I try to achieve is controls with left clic only.

Ok, but why? All the other sites use left click to place markers, and
right click context menu to delete them. Don't reinvent GUIs if you
don't have to :)

 I take it that more
 guidance would be appreciated.

Nah, just more familiar experience. Trying to teach users how to use
your novel interface is the wrong approach, IMHO - just stick with
what they already know.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Imagery Boundary?

2013-04-09 Thread Steve Bennett
On Mon, Apr 8, 2013 at 2:53 AM, Dave Sutter sut...@intransix.com wrote:
 Creating another instance of the OSM database and server is a very
 good idea. I would propose we make the purpose of this database to
 allow people post ANY geo data that is NOT part of the base map. It
 would be an open database for general GIS data.

 Some examples of random things people could do with this database:
 - The high resolution imagery outlines discussed in this thread
 - Migratory patterns of birds (I can't find the post where someone was
 requesting where to do this...)
 - GPS tracking for running, hiking, cycling and other recreation,
 similar to Strava or MapMyRun (see
 http://wiki.openstreetmap.org/wiki/OpenSportMap)
 - GIS Management for operations like Haiti OSM team

There are very many use cases for GIS data that is useful to OSM but
doesn't meet the current OSM criteria. Perhaps so many that one
database wouldn't be enough. Some that come to mind:
- everything that fails the on the ground test (flight paths,
boundaries, designations, etc)
- statistical/population/demographic data
- project metadata like the zzz team is working within this boundary area
- subjective data (preferred cycle routes)
- historical data (explorers' routes, defunct bus routes, demolished
buildings...)

Probably some of these already have good homes (OpenAviatianMap?).

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] OpenSnowMap.org

2013-04-06 Thread Steve Bennett
On Sat, Apr 6, 2013 at 7:56 AM, yvecai yve...@gmail.com wrote:
 Pistes informations with a single clic:No more 'vector mode'. Lighter,
 better compatibility with browsers.

 Search for pistes by names:Nominatim results are augmented with a selection
 of ski pistes and lifts.

 Multi-modal routing and elevation profiles:You can climb dowhnills, take a
 lift downward and a shortcut in snowshoes. Yes !

 Forum:Yet another channel. It's more dedicated to newcomers, and aim to
 ensure the site viabilty in the long-term.

Excellent! Some feedback:
- the terrain shading is distracting and not useful at higher zoom
levels (16+, maybe 15+)
- why not show the names of the runs? (rather than having to click on them)
- the interactive mode button is not intuitive at all. I had no idea
it was a button until I happened to click Settings. Why not leave
interactive mode on all the time?
- in the mountains I looked at, the chairlift icons are often upside down...
- could it be more obvious which is the start of each (downhill) run?
- would be great to have a summer/winter mode button (in summer mode,
show the mountain bike trails, hiking routes etc)
- at low zooms (10), it's hard to find the ski resorts - they could
be much more prominent
- suggest changing mapkey to legend (or just key. Or at least
put a space in it :))
- what is the profile button meant to do? It just gives this one tip
about clicking the interactive button...
- I find the whole interactive mode clunky and not intuitive. Why
not follow the established conventions of direction finding, using a
similar interface to Google Maps, Bing Maps, or hell, OSRM? It's
pretty confusing: it took me a while to work out that it's even doing
routing. All I wanted to do was click on a run to find out more about
it.
- the mapkey doesn't explain what the little warning signs ( /!\ )
are for. Steepness?
- I also see two different coloured XC trails, but I can't see
anything in the legend that explains why?
- I love the elevation profile - would be great to be able to click on
a single run/track and see the profile for it.
- the wiki page that is linked to to explain the colour coding of ski
runs doesn't actually explain them:
http://wiki.openstreetmap.org/wiki/Proposed_features/Tag:route%3Dpiste
(Suggest you just include text descriptions, like easy, difficult
etc)
- would be great to use bezier smoothing on the ski runs if at all
possible - they look very sharp and jagged.

Anyway, overall it looks great! Don't be put off my feedback :)

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Imagery Boundary?

2013-04-06 Thread Steve Bennett
On Tue, Apr 2, 2013 at 7:47 PM, Pieren pier...@gmail.com wrote:
 It's not only for outdated outlines. As said, it is not a map feature, it's
 just for some comfort during edition (would consider the same for mapping
 party cakes). What was the easiest and most pratical solution can be
 tolerated if it is temporary or until editors provide other means like the
 plugin mentioned earlier.

Hi Pieren,
  For our information, would you mind explaining how it helps to have
these boundaries in the database? I just don't get it - if I'm
editing, and I see one of these boundaries, either: a) it lines up
with the edge of the imagery, in which case it tells me nothing new
(ie: I can already see there's no imagery!), or b) it's wrong. I
usually delete it if b :)

I'm sure I'm missing something though - what's the use case where it's helpful?

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] Imagery Boundary?

2013-04-06 Thread Steve Bennett
On Sun, Apr 7, 2013 at 2:15 AM, Janko Mihelić jan...@gmail.com wrote:
 I think this boundaries can be useful, but should be in some other database.


Are there any other appropriate databases? That is, something with the
same form (an OSM database) for stuff related to the OSM project, but
not containing actual OSM content. I'm thinking Wikipedia has talk
pages, project pages, and meta.wikimedia.org; Stack Overflow has
meta - would some kind of meta OSM database be appropriate?

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


[OSM-talk] Osm2pgsql/TileMill - how to detect whether way is in a route?

2013-03-18 Thread Steve Bennett
Hi,
  I'm trying to render a mountain bike map with TileMill. Some trails
are in route relations, eg:

3xWay: highway=path, mtb=yes
1xRelation: type=route, route=mtb, name=My Cool Trail

Osm2pgsql converts those to 4 rows: the relation is the complete trail.

What I want to do is not render any way that is also part of a route
relation - in that case I just want to render the route.

One way to fudge this is to render the way, then render the route
relation over the top, obscuring it. But that doesn't work if you want
to use line smoothing (eg, line-smooth:0.6). The route relation will
form a smooth bezier curve the intersection of two ways, but the way
rendering will form a sharp point.

So I'm just wondering if there is a way around this - some setting in
osm2pgsql, some SQL magic I could play with. The ultimate goal would
be to be able to write rules like:


#trails[highway=path][mb=yes][in_mtb_route=no] {
...
}

Thanks,
Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] TileMill performance

2013-03-04 Thread Steve Bennett
Hi again,
  I've now moved to an 8-core VM with 32Gb of RAM, which has made
certain operations faster, but processing the stylesheets and
beginning rendering still seems to be very slow.

Some numbers: from the moment I hit 'save' at zoom 13 in the centre of
Melbourne:
1. at 1:05 , the spinning icon stops (I assume this means the
stylesheets have been saved and processed)
2. at 2:15, the first tile appears
3. at 2:35, all visible tiles are rendered
.

During this process, there's almost no database activity - all queries
are handled very quickly.

Turning off every layer (so just rendering background-color):
1 0:48, spinning icon stops (tilemill-ui at just over 100% until now)
2. 1:45, first tile appears, (tilemill-tile around 100% until now).
3. 1:46 all tiles

This cluster has had problems with slow file I/O in the past, so
wonder if that's the likely result?

By contrast, a simple project with simple stylesheet saves and renders
almost instantly (1s).

The styles are here: https://github.com/stevage/stevebikemap

What I'd love to know:
1) What is happening for the first 50 seconds? Does TileMill really
take that long to parse 5 smallish stylesheets and convert them into
mapnik styles? If it sounds like a bug, I'm happy to investigate
further...
2) What exactly is happening in step 2? I don't know Mapnik at all, so
I'm just guessing that it's doing pre-computation like loading all the
objects and their styles, computing z-orders, resolving conflicts
between labels etc. Still this seems slow?
3) Why is step 2 so slow even when every layer is invisible?
4) Is there any way to increase the number of threads/processes, to
get more advantage from the 8 cores?

Hmm, with further investigation, it seems almost all the time is going
into the 'areas.mss' style. Commenting it out reduces the whole
save/render cycle to just 7 seconds (from 155!)

As an example:

#areas[zoom13][leisure=pitch],
#areas[zoom13][leisure=golf_course],
#areas[zoom13][landuse=cemetery] {
line-color:darken(green,5%);
line-width:1;
line-opacity:0.2;
polygon-fill:#bdb;
}

Rendering that one rule takes an additional 7 seconds (ie, 14 seconds
instead of 7, including saving/processing time), even at zoom 13, when
it's not even being applied. Making the #areas layer invisible had no
effect on rendering time. I guess I can further decompose these db
queries (#landuse, #leisure...), but is there anything else I can do
to speed them up? Is this normal?

Thanks again for any information,
Steve

On Wed, Feb 27, 2013 at 10:03 PM, Steve Bennett stevag...@gmail.com wrote:
 Hi,
   (First - is this the right list to discuss using TileMill? I can
 only find the MapBox support form, or gis.stackexchange.)

 I'm having some problems with TileMill rendering very slowly.
 Sometimes it seizes up altogether, until I restart it or reboot the
 server. This seems to happen particularly when I frequently interrupt
 rendering by saving the stylesheet again.

 My setup is a 2-core, 8Gb Ubuntu VM running on an OpenStack cluster.
 PostGIS (with Melbourne city data from bbbike.org), nginx for auth,
 and one shapefile. Pretty vanilla.

 I have a few questions about how to improve speed in TileMill/Mapnik:
 1) In general, what kinds of rendering rules are slow? Does the way
 you specify a rule affect the speed? (eg, is [zoom13] { #ways[...] }
 slower/faster than #ways[...][zoom13] ?)
 2) How does caching take place? It seems to me that when saving a
 stylesheet with changes, there's a long delay before anything renders,
 then subsequent small changes aren't too slow. So some layers are
 computed once then reused?
 3) Are there any easy tips for tuning the database?
 4) Or tuning TileMill/Mapnik?
 5) Watching 'top' during a render, it doesn't look like much memory is
 being used. Is there a way to trade memory for speed?
 6) Does setting a layer invisible definitely prevent it being
 computed? Sometimes I think I'm going mad...
 7) Lastly,will adding cores lead to a proportional increase in speed?

 Thanks very much in advance,
 Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] TileMill performance

2013-03-04 Thread Steve Bennett
On Mon, Mar 4, 2013 at 7:02 PM, Steve Bennett stevag...@gmail.com wrote:
 effect on rendering time. I guess I can further decompose these db
 queries (#landuse, #leisure...), but is there anything else I can do
 to speed them up? Is this normal?

Ok, yes, that's apparently what you need to do - with 21 layers, I'm
now under 7 seconds for a full render. I wish there was a more elegant
way though :/

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] TileMill performance

2013-02-28 Thread Steve Bennett
On Thu, Feb 28, 2013 at 7:27 PM, Iván Sánchez Ortega
i...@sanchezortega.es wrote:
 On Miércoles, 27 de febrero de 2013 12:03:30 Steve Bennett escribió:
 I'm having some problems with TileMill rendering very slowly.

 I suggest to have a look at Dane's slides about How to make slow maps:

 http://dbsgeo.com/foss4g2011/foss4g2011-lecture2-how-to-make-slow-maps.pdf


Nice. Couple of questions about how to avoid certain of those traps:
1) How do you control multithreading? It looks like I'm getting access
to some 8 core VMs. How should I configure TileMill/Postgres for best
performance then?
2) How do I know whether reprojection is happening?

At least I have fixed the query more than you render problem - that
did indeed improve performance significantly.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] (Off topic) Apple hiring OSM contributors in Australia?

2013-02-28 Thread Steve Bennett
On Wed, Feb 27, 2013 at 11:45 PM, Gregory nomoregra...@googlemail.com wrote:
 Why hire *away*?

As noted by the OP, this story is indeed completely off-topic, and has
nothing to do with OpenStreetMap. Apple's maps team is hiring new
staff. To pitch this as some kind of head-hunting attack on OSM
is...creative.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] TileMill performance

2013-02-28 Thread Steve Bennett
On Thu, Feb 28, 2013 at 8:05 PM, Iván Sánchez Ortega
i...@sanchezortega.es wrote:
 On Jueves, 28 de febrero de 2013 09:57:34 Steve Bennett escribió:
 Nice. Couple of questions about how to avoid certain of those traps:
 1) How do you control multithreading? It looks like I'm getting access
 to some 8 core VMs. How should I configure TileMill/Postgres for best
 performance then?

 Tilemill is not a map server. Set up a propar map server which is able to
 configure proper multithreading.

Again, I'm not trying to serve up static tiles to the public. I'm just
trying to get quicker performance for styles under development. If
that's not possible, then ok.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


[OSM-talk] TileMill performance

2013-02-27 Thread Steve Bennett
Hi,
  (First - is this the right list to discuss using TileMill? I can
only find the MapBox support form, or gis.stackexchange.)

I'm having some problems with TileMill rendering very slowly.
Sometimes it seizes up altogether, until I restart it or reboot the
server. This seems to happen particularly when I frequently interrupt
rendering by saving the stylesheet again.

My setup is a 2-core, 8Gb Ubuntu VM running on an OpenStack cluster.
PostGIS (with Melbourne city data from bbbike.org), nginx for auth,
and one shapefile. Pretty vanilla.

I have a few questions about how to improve speed in TileMill/Mapnik:
1) In general, what kinds of rendering rules are slow? Does the way
you specify a rule affect the speed? (eg, is [zoom13] { #ways[...] }
slower/faster than #ways[...][zoom13] ?)
2) How does caching take place? It seems to me that when saving a
stylesheet with changes, there's a long delay before anything renders,
then subsequent small changes aren't too slow. So some layers are
computed once then reused?
3) Are there any easy tips for tuning the database?
4) Or tuning TileMill/Mapnik?
5) Watching 'top' during a render, it doesn't look like much memory is
being used. Is there a way to trade memory for speed?
6) Does setting a layer invisible definitely prevent it being
computed? Sometimes I think I'm going mad...
7) Lastly,will adding cores lead to a proportional increase in speed?

Thanks very much in advance,
Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] TileMill performance

2013-02-27 Thread Steve Bennett
On Thu, Feb 28, 2013 at 12:27 AM, Tom MacWright t...@macwright.org wrote:
 TileMill is not designed for that kind of application (running as a live
 server with no cache), though it will work 'a bit'.

 So: it doesn't do caching - you'll want a cache. Look at CloudFront, nginx's
 cache, varnish, squid, and so on.

I might have been a bit unclear: I'm not trying to serve up
pre-computed tiles to the public. I'm just doing style development, so
rendering tiles to see what it looks like. So I don't think caching of
pre-rendered tiles is an issue?

 Tuning the database: check that you have all possible indexes installed and
 the data is in EPSG:900913.

Can you elaborate on all possible indexes? :) Where can I read up on
this stuff?

 The long-term answer is switching to something designed to be a live-server
 (mod_tile, TileStache) or rendering your tiles and serving them from MBTiles
 (with TileStache or TileStream).


Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] TileMill performance

2013-02-27 Thread Steve Bennett
On Thu, Feb 28, 2013 at 4:09 AM, AJ Ashton aj.ash...@gmail.com wrote:
 (eg, is [zoom13] { #ways[...] } slower/faster than #ways[...][zoom13] ?)

 Does your layer setup actually looks like this? ie. one 'ways' layer pulling
 in the entire planet_osm_ways table? If so this will be problematic. Unlike
 MapCSS, filtering out objects with CartoCSS will not prevent them from being
 loaded. With Mapnik/TileMill you should be querying specific thematic
 subsets of the database for each layer.

Ah. Yes, #ways is the whole table - although it's just one city.
That's probably my problem. So, I guess I should set up individual
queries (#bikepaths, #parks, ...) using PostGIS to do the  filtering
rather than Mapnik. That makes sense, thanks!

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] [Talk-us] Tagging Live indoor music venues

2013-02-25 Thread Steve Bennett
On Mon, Feb 25, 2013 at 10:13 PM, Janko Mihelić jan...@gmail.com wrote:
 I don't think separating rock music venues and concert halls is a good idea.
 They are basically the same thing, a big room where music is played.

Agreed. There are too many tags as it is. Excessive distinctions
causes a lot of pointless hair splitting arguments (should I tag this
as a concert hall or a music venue?)

OTOH, if amenity=concert_hall, amenity=music_venue and
leisure=music_venue are all in pretty similar usage, we're pretty much
free to pick one to support. I'd take amenity=music_venue, as the term
is the broadest (and happily encompasses concert halls - the reverse
isn't true). leisure=* seems wrong - those tags are generally about
sports and recreation outdoors.

 Well, when you think of it there is a difference, concert halls usually have
 chairs. Is that a good separator of the two?

Not at all - lots of other venues do too. You could argue that concert
halls are designed for the performance of classical music, but even
that will break down. It's not a distinction that needs to be made.
Can you imagine a map that has different icons for concert halls vs
music venues?

 Maybe we could invent a tag like music=rock;classical;popular that
 differentiates those halls, and tag them all with amenity=music_venue.

Sounds ok...I think in practice there will be an awful lot of fuzzy
sets. Rock music at classical venues and vice versa.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


[OSM-talk] How to obtain a small coastline polygon for TileMill?

2013-02-25 Thread Steve Bennett
Hi all,
  I'm learning TileMill with a project to render my home town of
Melbourne, which is on the southeastern coast of Australia. The only
problem I'm having is the coastline. I've tried a few things, and
nothing has really worked:

1) The default 'countries' shape that comes with TileMill
- not enough detail, misaligned with OSM data (maybe the wrong projection)

2) Cloudmade's Victoria coastline
(http://downloads.cloudmade.com/oceania/australia_and_new_zealand/australia/victoria)
- I think it's just a way, not a closed polygon. As a result, applying
a fill to it doesn't really work - bits of water don't get filled, and
bits of land do.

3) Cloudmade's victoria.poly
(http://downloads.cloudmade.com/oceania/australia_and_new_zealand/australia/victoria/victoria.poly)
- not in a format that TileMill recognises.

4) Cloudmade's whole Australia coastline:
http://downloads.cloudmade.com/oceania/australia_and_new_zealand/australia/australia.coastline.osm.bz2
- aligns perfectly, but seems to be broken:
https://dl.dropbox.com/u/767553/broken-coastline.png


5) openstreetmapdata.com's land/sea polygons
(http://openstreetmapdata.com/data/land-polygons)
- very slow to render (because it's the whole world?)
- it's off by ~25km south of the other OSM data. I used the Mercator projection

I'm using TileMill installed on an Ubuntu VM with PostGIS. The OSM
data I'm using came from here:
http://osm-metro-extracts.s3.amazonaws.com/melbourne.osm.pbf

Can someone advise the best way to get a simple coastline polygon into
TileMill? For this project, I really only need a single polygon of
Port Phillip Bay (or alternatively, a single land polygon covering
greater Melbourne).

Sorry for the newbie questions.

Thanks,
Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-talk] How to obtain a small coastline polygon for TileMill?

2013-02-25 Thread Steve Bennett
Thanks - that's perfect. I hadn't noticed the SRS box. Works great now.

Steve

On Tue, Feb 26, 2013 at 4:03 PM, AJ Ashton aj.ash...@gmail.com wrote:

 On Feb 25, 2013 11:22 PM, Steve Bennett stevag...@gmail.com wrote:

 5) openstreetmapdata.com's land/sea polygons
 (http://openstreetmapdata.com/data/land-polygons)
 - very slow to render (because it's the whole world?)
 - it's off by ~25km south of the other OSM data. I used the Mercator
 projection

 I would recommend these files (they are what we use at MapBox). TileMill's
 projection auto detection is a bit off for these files - make sure to set
 the SRS explicitly to '900913'. As for speed, make sure the file has been
 indexed with the 'shapeindex' command line utility that comes with Mapnik.
 This should speed up rendering noticeably.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [OSM-legal-talk] Question about copyrighted hiking routes in France

2013-02-22 Thread Steve Bennett
On Fri, Feb 22, 2013 at 9:20 PM, Nick Whitelegg
nick.whitel...@solent.ac.uk wrote:
 More philosophically the idea of someone claiming copyright on walking
 routes seems completely at odds with the nature of countryside walking,
 which to my mind has similar free and open values to open source software
 and data (the landowners and their Keep out signs being similar to
 proprietary licencing)


And pragmatically, there are real problems when an organisation that
establishes a route also seeks to derive income from selling
information about it. It seems logical, until your realise that the
organisation's two goals (promoting a route by disseminating
information about it; and gaining income to achieve the primary goal)
are completely contrary to each other.
The Tasmanian Trail is completely obscure because the only
information about it is in a mail-order paperback.
To walk/ride the Great Dividing Trail requires paying for four really
crappy maps. (Hopefully in two weeks' time it will be 90% OSM'ed.)
Rail Trails Australia was heading in a similar direction, but I helped
convince them that sharing information about the trails *is* their
mission.

Maybe the FFRP could be persuaded eventually? Convince them to give up
claims of copyright on the route geography, and focus on prose
descriptions, subjective details etc?

Steve

___
legal-talk mailing list
legal-talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] Question about copyrighted hiking routes in France

2013-02-21 Thread Steve Bennett
On Thu, Feb 21, 2013 at 11:22 PM, Pieren pier...@gmail.com wrote:
 First issue : it is the hiking route names themselves. For all of them
 created by the FFRP, the names are registered trademarks and cannot be
 used without permission (see question below). Second issue : the
 routes themselves are copyrighted.

Hi Pieren,
  I am also not a lawyer, but here's my two cents. First, it would be
really said if we lost the GR routes. I recently hiked about half of
the GR20, using the OSM route of course. So I don't think we should
give up easily - they're so valuable.

On the trademark front, it should be easy to establish if they have a
genuine complaint. If they do, I think we can change the names without
losing too much - even if had to call them French hiking route 20 or
something.

 Second issue : it is maybe a more specific French issue here because
 the routes themselves can be copyrighted when they are considered as
 original work. A famous case confirmed this with the IGN (publishing
 the FFRP maps) sueing a guidebook editor [5] and confirmed by the
 highest court in France (1ere chambre de la cour de cassation de
 Paris, decision of 30 june 1998 [8]. I don't know if this is the same
 in other countries but a significant part of the OSM community in
 France would consider deleting the FFRP hiking routes completely (and
 not only the trademarks mentionned in Q1).

On what basis do they claim ownership of the routes, exactly? As I
understand it, many of these routes link up lots of little trails that
had been around for decades. How did copyright get transferred from
the people who created the trails to the FFRP? Or do they claim
ownership only over new sections? Or only over a particular
representation?

Are they aware that all the data has been created independently, by
surveying the trail - not by actually copying their data?

I wonder where the exact line would be drawn - what if we didn't have
routes, but just the trails marked. But then, how would you label
such a trail - often they have no other name other than the GR number,
plus the name of the next landmark.

Presumably someone has sought French legal advice? What was it?

Steve

___
legal-talk mailing list
legal-talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/legal-talk


Re: [talk-au] Google copying from OSM maps

2013-02-14 Thread Steve Bennett
On Fri, Feb 15, 2013 at 11:32 AM, Barker, Nicholas nbar...@pb.com.au wrote:
 Not at all,  it just seemed a bit of a double standard that we are diligent 
 not to copy from other maps and only map what is on the ground...then this 
 appears on Google on their 'cycle track' layer.

 If they had mapped it from the ground properly then they would know that it 
 is a very difficult mountain bike course with big jumps and drops, not a 
 paved bike path and i doubt they would have used it in the cycle routing.

Just pointing out the obvious here - in case anyone isn't aware.
Google now supports crowdsourced mapping with the MapMaker tool, so
it's totally possible that either:
- a member of the public mapped it legitimately
- a member of the public mapped it by copying OSM

 Obviously if we copied from Google then that's bad and illegal and we should 
 be punished with big law suits...if Google copies from OSM well then that's 
 OK as it furthers the mapping cause...

Yeah that's my view - I map with the intention of helping the widest
possible range of users. Sometimes I even add stuff to Google Maps
directly! It's not everyone's view though, so we do need to respect
that.

(IMHO much better to spend time mapping than fighting Google though...)

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Cycle routes and MTB routes

2013-02-12 Thread Steve Bennett
On Tue, Feb 12, 2013 at 12:29 PM, Barker, Nicholas nbar...@pb.com.au wrote:
 As many of you will know we have a few well known long distance (100km+)
 cycle routes, some of which cross state boundaries. These routes are also
 rugged and designed to be ridden on mountain bikes or hybrids only. A few
 examples of these are:

 The Bicentennial (VIC/NSW/QLD)
 The goldfields track (VIC)
 The Mawson Trail (SA)
 The Munda Biddi (WA)

Hi Nick,

  Like John, I can only know one of these well - the Great Dividing
Trail (aka goldfields track). It's extremely rough, and very little
of it could be sensibly ridden by anything less than a mountain bike.
So, definitely MTB route only.

I'm actually intending to ride (and hence map) the entire thing in a
bit less than a month.

 As these are extended bicycle routes (albeit for MTBs or tough hybrids) are
 of national significance do they constitute a national cycling network and
 should they be mapped as such?

They would constitute a national mountain biking network, if such a
thing existed. But LCN/RCN/NCN are clearly for cycling in the sense of
a practical way of getting from A to B.

 Indeed if they do not cross state borders are the only of state significance
 and to be mapped as a RCN?

That's a harder problem. I don't think crossing state borders is
relevant. In the case of walking routes, I think the most nationally
significant routes (eg, the Overland, the Lara Pinta...) should be NWN
in order to show up prominently on route maps.

 If a cycle route is rough enough to be MTB only (an unsurfaced rail trail
 for example) does that relegate it from the LCN, RCN or NCN to just being a
 MTB route only?

Yes.

 Should these routes carry two relations, NCN and MTB routes?

I think you'd only see that situation where a section of track is part
of two distinct routes, like where a mountain bike route temporarily
uses a rail trail or something. If a significant section of an MTB
route is suitable for normal cycling, it might be worth creating an
extra relation - discuss maybe? Could happen on the BNT.

The easiest way to think of it is that MTB and cycling are two
completely distinct activities. If you can't ride a hybrid with
panniers along it, it's not a cycle route.

I haven't done it, but I understood the Mawson Trail to be dirt roads
rather than mountain biking, so would be a better candidate for an RCN
tag. Similarly, I'm pretty sure the Munda Biddi trail is intended for
all kinds of cycling.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Adding only part of a road to a route relation (was: network and route tags)

2013-02-05 Thread Steve Bennett
Also (if using Potlatch), a useful shortcut you may not know about is
Shift-R. That will copy all the relations from the last selected way
to the current one.

Steve

On Mon, Feb 4, 2013 at 6:59 AM, SomeoneElse li...@mail.atownsend.org.uk wrote:
 On 03/02/2013 12:22, David Clark wrote:


 ie I have a route (The Mawson trail) that passes along a section of a fire
 road, but it doesn't pass along the full length of the fire road. How do a I
 select only a section of the fire road (not the full length of it) so I can
 make the relationship to the route?


 You'd need to cut the fire road into 2 (or perhaps 3) and then add the
 relation to only the relevant bit.  In Potlatch (the online editor on the
 OSM website) you can do this by clicking on the node that you want to split
 it at and pressing x.

 Here's an example (not in Australia, but the same principle applies):

 http://www.openstreetmap.org/?lat=53.13304lon=-1.78357zoom=16layers=C

 Here Highfield Lane is split into two.  The eastern part is added to the
 relation for cycle route 54; the western part isn't.

 Cheers,
 Andy


 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


[OSM-talk] Which renderer for high quality, printed cycle map?

2013-01-31 Thread Steve Bennett
Hi all,
  We'd like to produce a high quality bike map, to be printed. I'm
looking at the various renderers and wondering if anyone has a
recommendation?

Requirements
- looks good (eg, labels on wiggly labels aren't too wiggly, some
ability to avoid label clashes)
- customisable rendering (MapCSS would be a bonus)
- some nice bike-friendly styles already available?
- rendering to SVG would be very useful, would give us a chance to
hand-edit the final rendering.
- Python-based is a bonus

I'm open to the idea of using existing online services, if any are
customisable enough. Something like MapOSMatic doesn't work, as the
style is fixed, and it can't render a big enough area.

Contenders so far, from glancing through the wiki:
- Mapnik, obviously. Doesn't support MapCSS, but Komap might help?
- Maperitive, also no MapCSS.
- Ceyx. Still no stable release? Hard to tell what state it's in.

Pros/cons of each?

Are there others to look at? I can't find a good list anywhere
(there's http://wiki.openstreetmap.org/wiki/Renderers_feature_list but
lots of those seem old, abandoned, etc).

Thanks,
Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [talk-au] network and route tags

2013-01-31 Thread Steve Bennett
On Thu, Jan 31, 2013 at 9:33 PM, David Clark dbcl...@fastmail.com.au wrote:
 Ok I've changed route=mtb to mtb=yes for the trail itself. That fixes the
 issue I had with the route side of this so that's great.

 Using this approach an mtb trail (singletrack) looks the same as a cycle
 path (paved commuter path). Is this correct?

Looks the same in what? Mapnik? I'm not sure - mapnik may treat
highway=path; bicycle=yes as equivalent to highway=cycleway.
They'll probably look different in opencyclemap, if and when that ever
gets updated again.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] network and route tags

2013-01-28 Thread Steve Bennett
Hi David,
  Where is this, btw? In general:

- route=lcn are for bike paths that get you somewhere useful in the
local vicinity. (We still debate exactly what LCN means in Australia)
- route=mtb are for all mountain bike trails.

Don't get hung up on any connotations you might have with a word like
route, as a native English speaker. The benefit to using route=mtb
is that they show up specially highlighted on mountain biking map
styles, which is useful and appropriate.

So, for each trail, I would:
- a route relation with route=mtb, and name=xx, and other tags as appropriate
- tag the trail itself with highway=path, name=xx,
surface=dirt/gravel, and add the route relation

Steve

On Sat, Jan 26, 2013 at 4:33 PM, David Clark dbcl...@fastmail.com.au wrote:
 Hi All,

 I am interested in a small area of trails. There are about 10 trails in a
 local reserve, all the trails are sign posted and named etc, but there is no
 actual marked route you just pick which trails you want to use to get to
 where you want to go.

 However the tagging used in OSM to me seems wrong.

 (1)
 network=lcn
 Is this correct to use?
 Should there be other tags associated with this such as network:name=x
 etc?

 (2)
 route=mtb
 All the trails are tagged with route=mtb. However there is no marked or
 recognised physical route associate with these trails. Each trails is short
 approximately 200m to 500m long so it seems to me the route tag is not
 applicable.

 Etiquette:

 If the above tags are wrong, is it ok to just delete them? These tags have
 been used in this trail area and 2 others.

 Thanks,
 David

 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [OSM-talk] Being more like Wikipedia (was: OpenStreetMap Future Look)

2013-01-09 Thread Steve Bennett
On Thu, Jan 10, 2013 at 2:53 AM, Frederik Ramm frede...@remote.org wrote:
 I'm very much an outsider to Wikimedia but if I look at how much money they
 have spent on development and how little has changed for the contributing
 user - adding a table to an article is practically as difficult now as it
 was five years ago. You sit there and wonder: How hard can it be? Hundreds
 of man-years of developer time... and still a person with average computer
 literacy cannot add a table to an article!

Hi Frederik,
  Before getting into OSM, I did a lot of work with Wikipedia: writing
articles, developing policies and guidelines, moderating the mailing
lists, various cleanup etc, mostly in 2006-8. As noted, your example
is poorly chosen: the goal of Wikipedia is to diseminate a high
quality encyclopaedia to the world's people. Letting punters create
tables easily is a low priority (and hard!), compared to all the
infrastructure of actually serving up the content, making translations
work, zillions of plugins, bots, browser support, the monster that is
the wikitext parser etc. All the developer time has produced an
enormous amount: a stable, high quality encyclopaedia that it's in the
top 10 web sites, looks good, is searchable etc etc.

 It is too simplistic, to say things like everyone wants OSM to be more like
 Wikipedia in terms of X, because you can't always separate the good from
 the bad. It's easy to say I'd like to have the kind of money that Wikimedia
 have or the popularity that Wikipedia enjoys but none of this can be had
 without a downside.

I can't speak for the money side, but I'd like OSM to be more like
Wikipedia in terms of the maturity of its community and its attitude
towards content development. Wikipedia took a firm stand that the
healthy hothouse attitude of the early days was just a passing
phase: things had to settle down, standardise, become more process
driven in order to produce high quality content. OSM has been around
enough years now for something similar to have happened, but it
hasn't. Newcomers are still encouraged to invent tags, and to ignore
the wiki, because that's just wikifiddling. Whereas Wikipedia takes
policies and guidelines seriously, has large numbers of highly
successful wikiprojects, has people who take responsibility for pretty
boring things like stub and category management, and it works. Whereas
one look at taginfo.openstreetmap.org will show you the complete chaos
that we have - and it's not getting better.

Wikipedia strives for high quality content, at the expensive of the
contributor. OSM strives for ease of use for contributors, at the
expense of content consumers. After all these years we still have no
agreement about exactly what highway=path means, dozens of very common
tags, or even sets of tags that consumers should support.

 For example, Wikipedia being as well known as it is has lead them to create
 relevance criteria - you can't create an article on a living person or a
 geographic feature, for example, unless that person or feature fulfills
 certain criteria. Wikipedians felt that this was necessary because they were
 swamped with data they considered irrelevant and un-encyclopedic. Many
 people left Wikipedia because of that (and indeed many of them are to be
 found in the ranks of OSM nowadays).

Notability. People leave Wikipedia for all kinds of reasons. Those
that leave because the content they were interested in creating wasn't
within the scope of Wikipedia were obviously on the wrong project. You
make this sound like a bad thing.

 I've heard other OSMers make fun of the
 tons of WP:xxx rules that Wikipedia has but I am sure they are not there
 because Wikipedians terribly enjoy rule-making - they probably had to be
 created in response to problems.

They were created in pursuit of a goal, and they work. Best of all
they focus debates, and move them forward. You can debate whether a
given course of action fits within existing policies and guidelines,
or you can debate whether the policy/guideline is right. But you don't
start from scratch every single time like we do in OSM debates.

Probably one reason that there are more policy/guidelines on Wikipedia
is policy writing is a closer fit with encyclopaedia writing. Whereas
geospatial types get frustrated quickly with writing text, I think.

Steve

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [talk-au] Pressing Issues (Was: Re: cities changed to towns)

2012-12-23 Thread Steve Bennett
On Fri, Dec 21, 2012 at 11:37 AM, Ian Sergeant inas66+...@gmail.com wrote:
 We also need a review of those areas mapped in the shadow of the redaction.
 I was using my Edge 800 to do cycle navigation from Ryde to Manly on the
 weekend, and quite frankly it sucked big time.  I was directed over 10 metre
 cliff drops, driveways, against one-way streets, the lot.  I'd be still
 there if I hadn't pulled out Google Maps and backtracked.  (Where the maps
 were good, the Edge 800 bicycle navigation is great, BTW).


Heh, same thing just happened to me, riding over Mt Buninyong (near
Ballarat). Mt Buninyong Rd turned out to be a private driveway, and
the only way out was down a steep, overgrown firetrail on my road
tyres. No idea if the error was related to the redaction process.

Steve

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] cities changed to towns

2012-12-11 Thread Steve Bennett
I would want place=city to refer to an urban populated area of at least
100,000 people as per http://wiki.openstreetmap.org/wiki/Key:place#Values


 I've taken to fixing errors from Geofabrik OSMI and have changed places to
 match the schema above. Whilst I find hamlet  village grate on me as
 words, they are merely code for an object to be mapped. It's only really
 issue because I speak English (Australian) and the OSM schema was developed
 in English (United Kingdom) that there is an issue. If we all spoke Finnish
 or Swahili we wouldn't be having this discussion now.


Ok, well what might be an obvious error to you is correct to someone
else. There are many OSM tags that have different meanings in different
parts of the world. It would be good to be consistent within Australia, but
it's not important whether our meaning precisely matches the meaning in the
UK or some other country.

Looking at the wiki page you cite, it's clear that those definitions are
intended as rules of thumb: Populations of villages vary widely in
different territories but will nearly always be less than 10,000 people,
often a lot less.; [Cities s]hhould normally have a population of at
least 100,000 people and be larger than nearby towns. Normally, in densely
populated areas, that is. Applying that cut off in Victoria would lead to
only Melbourne and Geelong qualifying, with Bendigo and Ballarat just
missing out.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Blind obedience following routing tools.

2012-12-10 Thread Steve Bennett
On Mon, Dec 10, 2012 at 5:48 PM, David Bannon dban...@internode.on.netwrote:


 I guess people are aware of this story, people in trouble for following
 badly constructed maps -


 http://www.abc.net.au/news/2012-12-10/apple-maps-strands-motorists-looking-for-mildura/4418400

 Looks like its to be the lead story on the abc news tonight.

 I bet we could all find examples in OSM that could cause similar
 problems.


Yep. For better or worse, OSM doesn't inspire the same trust (or blind
faith) that Apple does.

I discovered this weird error on Bing Maps yesterday:
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Blind obedience following routing tools.

2012-12-10 Thread Steve Bennett
(whoops, wrong button)
The weird error on Bing Maps:

http://binged.it/RlH1PC

The real Mt Buller can be seen 50km to the north. This error means if you
route from Mansfield to Mt Buller (less than a 1 hour drive) it takes all
the way back to Melbourne then via Licola - over 7 hours.

Steve


On Tue, Dec 11, 2012 at 9:03 AM, Steve Bennett stevag...@gmail.com wrote:

 On Mon, Dec 10, 2012 at 5:48 PM, David Bannon dban...@internode.on.netwrote:


 I guess people are aware of this story, people in trouble for following
 badly constructed maps -


 http://www.abc.net.au/news/2012-12-10/apple-maps-strands-motorists-looking-for-mildura/4418400

 Looks like its to be the lead story on the abc news tonight.

 I bet we could all find examples in OSM that could cause similar
 problems.


 Yep. For better or worse, OSM doesn't inspire the same trust (or blind
 faith) that Apple does.

 I discovered this weird error on Bing Maps yesterday:


___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] cities changed to towns

2012-12-10 Thread Steve Bennett
Hi Chris,
  Interesting topic - sadly the wiki just acknowledges the lack of an
answer. My take is that the distinction between village/town/city really
only matters for the purpose of rendering anyway - any more sophisticated
use of the data is going to use population figures to make its own decision
about how to classify towns. So I think it's ok to be a bit loose and
subjective with our definitions.

Hard for me to comment on the QLD ones. The Victorian ones are Warrnambool,
Sale and Mildura. W and M definitely sense as cities than towns. They're
major regional centres, and much more significant than towns nearby. Sale
is more lineball (although Wikipedia counts it) - nearby Bairnsdale should
be a city though.

You see the effect it has on Mapnik here:
http://www.openstreetmap.org/?lat=-36.74lon=145.31zoom=8layers=M

It's definitely wrong having Sale show up at that zoom but not Bairnsdale.
But the more I look at the Wikipedia list (counting 18 cities outside
Melbourne), the more I think it would make sense to mark all of those as
city. None of them seem out of place subjectively to me. I'm not clear on
where the list on Wikipedia was derived from though. If you compare the
list by population against the ones designated city, some omissions are
Echuca, Warragul, Bacchus Marsh, Ocean Grove-Barwon Heads. Not a big deal
though.

Steve



On Tue, Dec 11, 2012 at 9:17 AM, Chris Barham cbar...@pobox.com wrote:

 Hi,
 some Australian places have changed from cities to towns on;
 changeset was: http://www.openstreetmap.org/browse/changeset/14217241

 I've emailed to the editor to ask the source for the change as I believe
 some are now incorrect.

 I really do think Gympie, Maryborough, Warwick and Charters Towers are
 cities, and should have remained tagged as such.  Are there others, in
 other states, within this changeset that should have stayed as is?

 Wikipedia is not the best reference material I know, but they have the Qld
 ones I mentioned as cities:
 http://en.wikipedia.org/wiki/List_of_cities_in_Australia

 Additionally, I think some of the others should to be tagged cities, even
 if not officially, under the Aus tagging guidelines at:

 http://wiki.openstreetmap.org/wiki/Australian_Tagging_Guidelines#City.2C_Town_or_Village.3F

 So looking at the tagging guidelines, haven't we agreed to tag by
 population size or significance in remote areas?
 Here is a populated list of places by population for Qld that could be
 useful any discussion:
 http://www.bonzle.com/c/a?a=fsc=lgst=3cmd=sp

 Cheers,
 chas



 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] When is a road a cycle route?

2012-12-06 Thread Steve Bennett
On Thu, Dec 6, 2012 at 3:43 PM, Ian Sergeant inas66+...@gmail.com wrote:


 Yes!  we are overloading the cycle route to not simply mean this is a way
 to get from A to B but also to mean a good way to cycle there.

 So, yes, we are giving two meanings to the same tag.  Yours is simply
 navigational, and mine assures a minimum level of amenity in urban areas.

 I'd argue my use adds substantial value to a router.  Your use adds little
 value over a pure shortest route algorithm.


Hi Ian,
  I'm all in favour of recording subjective information about the
rideability of individual streets - in fact there are several other
projects out there doing this already. I don't think misusing the LCN/RCN
tags is the right way to do this though. A few reasons:
- You can't record anything about *why* this is a good route
- You can't record different levels of goodness
- You can't distinguish between this is a good route and this is
actually a signed cycle route

Using an existing tag which has an existing tag, for a novel purpose,
purely so that existing routing software will behave a certain way is
exactly the definition of tagging for the router. (Not that I believe in
chanting slogans instead of actually spelling out the argument, which is
why I've done so above.)

And yes, my use adds little value over a pure shortest route algorithm -
it was never my understanding (or intention) that LCN/RCN tags would be
used by routers. But they're used by renderers of various kinds.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] When is a road a cycle route?

2012-12-06 Thread Steve Bennett
On Thu, Dec 6, 2012 at 11:58 AM, Ben Kelley ben.kel...@gmail.com wrote:

 I think this one is an edge case. I know the sign Ian means, and in my
 judgement it does not indicate a route.


Oh, can you elaborate?


 I think the more common case (that is currently not well defined) is where
 some other map indicates a route, but there is nothing on the ground to
 show it. If we map it in OSM, are we copying the map, or mapping a known
 fact (but one that is only known from the other map)?

 Apart from copyright (which is significant here) part of the problem is
 government bodies (local and state) mapping optimistically in that they
 plan to mark out a route in some way, but didn't actually do it. In general
 I would say that map only routes are not routes in the OSM sense.
 Possibly they are suggestions.

 I think we should not map such routes in OSM due to the copyright issue
 alone, but at what point does the existence of the map imply the existence
 of the route?


If we're worried about copyright, I would be very curious to know why it's
ok to copy information from street signs, but not from other sources.
Ultimately, a bicycle route is a creative work (in a way that a street name
isn't): it wouldn't be hard to demonstrate creativity in the selection of
streets, bike paths etc that go into a route. So if we think that it's not
ok to contain such routes in OSM (IIRC the French OSM reached that
conclusion) then it doesn't matter whether they come from surveying street
signs and markings, or from maps.

Has anyone actually contacted any councils about this? I find it hard to
believe they would intentionally consider their routes copyright and have a
problem with their being copied in OSM, but they may never have considered
the issue.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] When is a road a cycle route?

2012-12-06 Thread Steve Bennett
On Fri, Dec 7, 2012 at 8:49 AM, Ian Sergeant inas66+...@gmail.com wrote:

 Fundamentally, I think it is a continuous set of navigational signs that
 should be the primary indicator of a bicycle route, preferably agreeing
 with some documentation from the relevant authority.


Ok, great - we do agree then.



 However,  we need to take care we don't get to the point where blue
 bicycle sign implies bicycle route in OSM.


I'm still not sure what these blue bicycle signs are - can you find an
example on Google Streetview?



 So, there are several reasons we should not tag these official routes in
 OSM..

 1. If they are wholly unsuitable for cycling of any form.


An example? I've never seen such a thing, so I'm finding it hard to relate.



 2. If it is a  cyclists get the hell outta here situation.  Implemented
 to remove cyclists from a way rather than give them a positive route.


Like this? http://goo.gl/maps/C0ysv

I think I'd agree. Maybe. I'm not sure. Are there many?



 3. If it is only a planned route.


Agreed.



 4. If it is outdated and a newer route has been implemented (and the old
 signs just happen to be still in place).


Agreed.



 I can only see one reason where we should tag an section that is not part
 of an official route, and that is where it logically connects two
 signposted sections, and it would be obvious to the reasonable cyclist
 after consulting any available documentation that the signs for the
 intermediate section are just omitted.


Hmm, what about this one: http://osm.org/go/uG4JzSmmU--?layers=C

In this case, there are two bike routes (both rail trails, by coincidence)
that come within 100m of meeting each other. But the signage for these
things is in general terrible, so it's not surprising there's no signage
linking the two. It just seems logical to show a link on the map (up Poath
Rd then down the little service road).

I don't think this is an issue with routes like railtrails.  They are put
 together with care.  However, this is very applicable to RTA routes and
 council routes.  Some of the RTA maps are decades old, and are really bad.
 Some of the council signposts are downright crazy.


RTA as in the NSW version of VicRoads? (Somewhat confusingly, RTA could
also be Rail Trails Australia...)



 In 95% of cases, I'm sure we would agree with what should be a cycle route
 and what shouldn't be.


Sounds like it.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] When is a road a cycle route?

2012-12-05 Thread Steve Bennett
On Tue, Dec 4, 2012 at 11:01 PM, Ian Sergeant inas66+...@gmail.com wrote:

 Generally the case, but not always.  My bicycle sign on Parramatta road
 being my best example so I'm sticking with it.  A cycle route down a narrow
 three lane road, carrying trucks who'd soon as take you out as look at
 you



 Well, I guess I'm focussed on being alive when I get to B.


These use cases are handled by routing software making good use of data
such as cycleway=lane. Using the lcn/rcn tagging system to mark
safety/suitability is simply incorrect. In your example, there's a signed
bicycle route - so we map it. To do otherwise would be like not mapping a
car park because we don't think people should park there, or not mapping a
school because we don't think it's a good one.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Historical rail lines

2012-12-03 Thread Steve Bennett
Hi Matt,
  It seems we've reached the point of simply restating our views. I don't
think yours represents consensus - but please discuss it on the main OSM
talk list if you want.

Steve


On Fri, Nov 30, 2012 at 7:10 PM, Matt White mattwh...@iinet.com.au wrote:

  Abandoned makes it sounds like there are tracks in place for the
 length of the line, just no trains running on it.

 But that's not the case - in the 4km the line used to run on there are 11
 remaining artifacts, the largest being a station building (old North
 Carlton station), the smallest being a single 4 metre track section in
 Edinburgh gardens, or the one remaining concrete pylon base. They are the
 vestigial traces that need to be mapped. As for the rest, it's a mostly a
 park now with a bike track along it (the bits that aren't are houses) ...
 and that's what it should be mapped as.



 On 30/11/2012 6:23 PM, Mark Rennick wrote:

  Matt

 ** **

 I believe abandoned railway lines should be mapped. 

 ** **

 If it is necessary to have a current physical feature to justify mapping,
 then the railway formation (cut and fill earth works) generally remain,
 particularly if the railway reserve has been retained as a rail trail, road
 or linear park.  

 ** **

 *From:* Matt White [mailto:mattwh...@iinet.com.au mattwh...@iinet.com.au]

 *Sent:* Friday, 30 November 2012 7:31 AM
 *To:* 'talk-au'

 *Subject:* Re: [talk-au] Historical rail lines

  ** **

 Right. So if I delete the mapped rail line that doesn't exist, then remap
 the individual pieces of track, the remaining point and weighbridge, three
 overhead pylon mounts, one remaining station and one cutting that remains
 as historical artifacts, then everyone is cool?

 If it exists on the ground now, it will get mapped. Otherwise, it won't.

 Matt

 On 29/11/2012 4:46 PM, Paul Norman wrote:

 Actually, the slope is slippery. People have made it about old roads.
 There are people who have mapped old roads where they have been completely
 developed over and no trace remains.

  

 Mapping the traces of an old rail line isn’t historical mapping. If there
 are currently traces there then it’s mapping the present.

  

  

 *From:* Steve Bennett [mailto:stevag...@gmail.com stevag...@gmail.com]
 *Sent:* Wednesday, November 28, 2012 7:02 PM
 *To:* Matt White
 *Cc:* talk-au
 *Subject:* Re: [talk-au] Historical rail lines

  

 On Mon, Nov 26, 2012 at 7:31 PM, Matt White mattwh...@iinet.com.au
 wrote:

 Admin boundaries are a slightly different thing - they may be intangible
 on the ground, but they are also current. We don't keep historical versions
 of admin boundaries either

 The problem with the historical thing is that to my mind, it is a slippery
 slope. There's a park near me that is currently, well, a park. But I know
 that it was previously a quarry, and then a rubbish tip/landfill, cos there
 is a sign saying so. But I certainly wouldn't tag the parks as a quarry or
 landfill, because it isn't. It's a park


 IMHO this slope is not slippery. Every time the do we map historical
 stuff debate comes up, it's always about train lines. That is, we're still
 at the top of this supposedly slippery slope, waiting to slide down.
 Somehow, train lines are different. They just are.

 To reiterate what I said before in different words: we're not mapping the
 1890 route of a long forgotten train line. We're mapping the vestigial
 traces of a former line. And I'm absolutely not proposing to record any
 information about when lines opened or closed, or were re-routed or
 whatever.


 Steve

  

 ** **



 ___
 Talk-au mailing list
 Talk-au@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] When is a road a cycle route?

2012-12-03 Thread Steve Bennett
Hi Ben,

 Thanks very much for starting this conversation - yes, it's a messy one.
Mostly because the European (and particularly UK) concept of cycle route
hasn't really existed here. But it's still worth trying to fit into because
lots of tools (especially OpenCycleMap) do support that concept.
* Normal residential street. No road markings. No signs. No maps listing
this street as a cycle route. I would say this is not a cycle route.

 * As above, but where I think this is a handy street to ride down. I would
 say this is not a cycle route.
 * As above, but where some other people also think this is a handy street
 to ride down (and in fact I saw some just the other day). Again, not a
 cycle route in the OSM sense.


Agreed.


 * As above, but there is a council map that says this street is a cycle
 route. (The map also lists other streets as cycle routes, and other streets
 do have signs, but this street does not.) I have found this to be fairly
 common. I would say this is not a cycle route.


Disagree. If it's a designated cycle route - it's a cycle route. Could you
elaborate on your reasoning?



 Tricky ones:

 * A council map says this is a cycle route, but there are no markings. In
 fact the council does not use road signs or paint to mark any of its cycle
 route. This is tricky, but I would not mark this in OSM, as the
 (copyright) map cannot be verified on the ground.


I'm not sure of the difference between this and the previous one. Is it
that in this case, there are no markings *anywhere* for the route?


 * A section of street that does not have any markings connects other
 streets that do have markings (e.g. bike symbols painted on the road).
 Cyclists commonly use this street to connect. Maps show this street as a
 cycle route. This also is tricky.


I generally mark these, because it makes the map more useful. I think it's
pedantry to leave little gaps in the map because those particular streets
don't happen to have the markings shared by the rest of the route.
Unhelpful pedantry, at that.


 * A shared use path that does not connect to any other known cycle routes.
 I would probably not mark this as a cycle route, but it depends on where it
 is.


Yeah. Sometimes I mark these as LCN, sometimes I don't. If I can infer some
sort of route thinking (ie, a series of streets or paths that connect),
I'm more likely to.


 * A section of road has a cycle lane (where the law requires cyclists to
 ride in it), but the section of road does not connect to any other known
 cycle routes. Again tricky, and it probably depends on where it is.


Personally, I don't equate bike lane with cycle route in the way that
others (notably John Henderson, below) do. Bike lanes are infrastructure.
Cycle routes are, well, routes. Quiet streets can be part of a bike route,
but not have bike lanes because they're quiet. Similarly, busy roads can
have bike lanes without being part of a bike route.



 Easier ones:

 * In states where riding on footpaths is normally not allowed, a shared
 use path that connects known (marked) cycle routes. Yes this is a cycle
 route.


What do you mean by connect here? Simply that one path joins the other
two? But yeah, probably.


 * A number of other maps show this as a cycle route. It has bikes painted
 on the road. Signs every 500m saying Cycle Route. Signs at every
 intersection with a picture of a bike, and showing the destination. Yes
 this is a cycle route.


Again, I'm not really fussed what's painted on the ground. The indication
of a cycle route, in my local council areas, is generally signs with a
little bicycle, possibly the words bicycle route and an arrow. They
frequently point down streets with no other bicycle infrastructure.
Example: http://goo.gl/maps/M7FB9

I can think of more tricky edge cases, but in general I am more concerned
 with whether some physical presence on the ground is required, as opposed
 to I thought this might be a nice street to ride my bike down.


To me, a bicycle route is much more about navigability than desirability
for cycling. That is, when you follow a bicycle route, it should be easy
to follow - based on signs, or good external (and official) documentation.
Whether it has painted bike lanes is irrelevant.

One of complications that arises, though, in inferring a route from signs
is how far you allow between signs. What if the signs are far enough apart
that there is ambiguity about which choice of streets in between is
intended? etc...

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] When is a road a cycle route?

2012-12-03 Thread Steve Bennett
On Sun, Dec 2, 2012 at 9:34 AM, John Henderson snow...@gmx.com wrote:

 I take a simplistic approach to this.  A road is a bicycle route if and
 only if it has a bicycle lane (lanes if a two-way road).


Simple, but not very helpful IMHO. cycleway=lane already captures that
information. lcn=* and route relations should capture something beyond
simple infrastructure.

Incidentally, another case for discussion: Someone a while ago added the
Golden Trail to OSM: marking a huge series of roads from near Adelaide to
Ballarat (or Castlemaine - I forget) with a route relation and rcn_ref=The
Golden Trail. There was never going to be any on the ground signage or
painting, but at the time at least there was a website and plans for
brochures etc. So I was a bit ambivalent about it.

Since then, the website has gone, and the whole concept seems to have
disappeared, so I've been progressively removing them. Still a fair chunk
remains:
http://osm.org/go/uGTLrE?layers=C

I assume we're all agreed that we wouldn't want this kind of thing in OSM?
What if there was good external documentation? What if there were signs but
no bike lanes?

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] When is a road a cycle route?

2012-12-03 Thread Steve Bennett
On Tue, Dec 4, 2012 at 2:43 PM, Ian Sergeant inas66+...@gmail.com wrote:


 We're heading towards a day when everybody will have a routing application
 on their mobile device or accessible elsewhere.  So navigation is a
 diminishing issue, and desirability for cycling is an increasing one.


Interesting thought. I don't know if I totally agree - I tend to carry a
smartphone, *and* I have a GPS mounted on the handlebars, yet neither of
those things is convenient as following actual signs or markings.



 If there is no cycling amenity of any kind, then it is just a route?  How
 does it differ from any other just by being signed?


I'm not sure I understand your question. By definition, a route is an
abstraction on top of the physical world. What route did you take to get
there - there's nothing physically distinguishing about a route.

Maybe we're getting confused by unstated end goals. My interest in routes
is having a map (much like opencyclemap) that elevates bike paths and other
signage to the level of highways for cars. Jump on the X trail, follow
that till you get to the Y trail etc.

I think maybe some other people in this thread are focussing more on a
where is good to ride use case? Definitely valuable, but is that the
primary meaning of route in OSM?


 If we start including roads with no cycling amenity, then we devalue every
 other quality cycle route we mark.  Because an end user can no longer
 expect cycle amenity from a marked cycle route they become worthless to
 most of our urban cyclist users who are looking for just that.  Of course
 amenity can come in many varied forms, so I don't mean cycle lanes.


Could you elaborate on what amenity means to you? Me, I'm assuming that
if the council has put up bicycle route signs, it's because they've
determined that that road is inherently better for bikes than some nearby
street - both because it's safer and more comfortable, and because it goes
somewhere mildly useful.



 However, I accept that things like railtrails, long distance cycle routes,
 etc are exceptions here - where even poor amenity may want to be included
 in the route.  I'm not quite sure how we distinguish these type of trails
 where people are trying to fill in the gaps, from some of the just plain
 stupid mapped/signed routes that pass for cycle routes in some council
 areas.


Well, I guess they seem stupid if you're focusing on where's good to
ride. They're totally logical and sensible if you're focusing on how do I
get to point B.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Historical rail lines

2012-11-29 Thread Steve Bennett
On Fri, Nov 30, 2012 at 7:30 AM, Matt White mattwh...@iinet.com.au wrote:

  Right. So if I delete the mapped rail line that doesn't exist, then
 remap the individual pieces of track, the remaining point and weighbridge,
 three overhead pylon mounts, one remaining station and one cutting that
 remains as historical artifacts, then everyone is cool?


Not me.



 If it exists on the ground now, it will get mapped. Otherwise, it won't.


Your line of reasoning basically goes we will only map individual
historical artefacts that are each worth mapping. The reason (IMHO) that
we map a train line like railway=abandoned is to connect lots of little
artefacts and landscape features that individually are too trivial to map.
For example, a slight embankment (normally not something we'd map), in the
context of other abandoned rail features makes sense under a
railway=abandoned. Similarly, a line of trees, or simply the absence of
development. Frequently, the corridors in which abandoned rail lines lie
are still owned by the state. Mapping the railway line makes sense, and is
meaningful to many people: Our house is on Station St, just the other side
of the old rail line - even if strictly speaking there is nothing on the
ground.

I have no objections to removing sections that have been built over.

So maybe my position is: If the former rail line still plays a part as a
landmark or in planning and development, it should be mapped.

Similarly, I'm ok with removing former stations that have completely gone
and been built over, but if their former presence is preserved in some way,
they should be mapped.

It seems we both agree on mapping *the present* but differ in how to
interpret that.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [OSM-talk] Recommendations for OSM mobile app?

2012-11-28 Thread Steve Bennett
Hi Jo,
  Yeah, OsmAnd does everything - but it's pretty complicated. Switching
between offline and online maps (a pretty basic task) is really hard (and
very hard to remember). To go to online maps you seem to have to:
1) Menu | Settings | Offline data (Download)
2) Expand Offline maps (vector)
3) Hold down on each offline map until 'deactivate' appears, which you
choose.
4) Back out to Settings, select Online maps
5) Select Online and tile maps

To then choose which one actually shows:
6) Menu | Define view | Map source...
7) Pick one

Hmm...written out it doesn't sound that bad, but it took quite a lot of
fumbling around to get that far. (I guess the problem is the OsmAnd authors
think of vector and tile maps as completely different functions,
implemented completely differently, but for the user, they're just two
variations on a theme.)

Steve


On Wed, Nov 28, 2012 at 4:56 AM, Jo winfi...@gmail.com wrote:

 I wanted to tell you a few days ago that OsmAnd does all you want. Maybe
 write a small manual page for it, for the subset of features your users
 need.

 I didn't do it then, hoping somebody else would have a better suggestion.

 Polyglot

 2012/11/24 Steve Bennett stevag...@gmail.com

 Hi all,
   I'm looking for a recommendations of mobile apps to recommend, for
 people visiting rail trails in Australia. What we* need is:

 Must have:
 1) It has to be simple to use. There are lots of powerful apps like
 OSMand, Locus etc, but they're incredibly complex, with lots of features we
 don't need.
 2) Should by default use bicycle rendering (OpenCycleMap or other), or be
 very easy to change to that.
 3) We'll need a recommendation for both iPhone and Android.

 Nice to have:
 4) Easy-to-use offline tile downloading/use would be a bonus. (Personally
 I find it pretty complex switching between offline vector rendering and
 online tiles in the apps I mentioned before).
 5) Features like food/drink near here would be a bonus.
 6) Ability to also choose Google Maps would be a bonus.
 7) Ability to route along bike paths (ie, rail trails) would be nice.

 Once we've chosen an app, we're going to have to give instructions like
 download the app, push this button to change to cycle mode, push this
 button to download tiles etc, so the simpler the better.

 Thanks,
 Steve

 * we = Rail Trails Australia

 ___
 talk mailing list
 talk@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk



___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk


Re: [talk-au] Historical rail lines

2012-11-25 Thread Steve Bennett
Hi Matt,
  The question about mapping old, historical features is much wider than
just the Australian context. I'm pretty sure the current consensus is that
we old rail lines should be mapped - even if there is not much to see on
the ground. There might be more than you think - there's a station building
(now a community hall, I think), other things too, perhaps. There are
probably other former railways about with much less to see (the Rosstown
Railway comes to mind) - at least with this one there are physical remnants
such as tracks.

So, yes, I object. Feel free to raise the issue on the main OSM talk list
though.

Steve


On Sun, Nov 25, 2012 at 5:15 PM, Matt White mattwh...@iinet.net.au wrote:

 A question for the list regarding historical/disused rail lines.

 The old inner circle rail line in Melbourne is mapped in OSM, and I'm
 unconvinced of it being a good thing. Here's a little bit of it that I can
 talk about with some local knowledge of: http://www.openstreetmap.org/?**
 lat=-37.780512lon=144.982887**zoom=18layers=Mhttp://www.openstreetmap.org/?lat=-37.780512lon=144.982887zoom=18layers=M

 Given that there is pretty much no trace of the rail line left, why are we
 mapping it? It was on the ground 30 years ago, but it certainly isn't now.

 (That said, there are some small pieces of the track remaining - where it
 crosses Rae St and Brunswick St Nth, two or three 15 metre sections + a set
 of points just north of the end of Birkenhead St (including what appears to
 be an old rail weighbridge), and a short three metre section in Edinburgh
 Gardens, and the old North Carlton station building is still there)

 If there are no complaints, I'm going to remove it. It's historical, and
 appears on old maps, but does not exist today.

 Matt

 __**_
 Talk-au mailing list
 Talk-au@openstreetmap.org
 http://lists.openstreetmap.**org/listinfo/talk-auhttp://lists.openstreetmap.org/listinfo/talk-au

___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Tagging dirt and 4x4 roads - new approach

2012-11-25 Thread Steve Bennett
* Unpaved roads are difficult to really classify the surface in terms of
anything other than dirt/sand/rock. The surface state changes over time
from smooth immediately after grading, to possibly deep
ruts/corrugations/mud after rain and wear. In this case, my personal
opinion would be to use some sort of tag like surface condition (options
being something like: maintained | uneven | degraded |
corrugated | rocky | rutted | deep_rutted, but even those change
immediately after track maintenance), with perhaps a best/worst case tag or
similar

One thought that occurs here would be to tag the *maintenance* of a track
rather than its *current state*. Some tracks are essentially never
maintained, while others are graded frequently. That, combined with the
season that you're travelling (eg, late summer vs early spring) might be
enough to make an informed decision.



 * Overall, it seems like Australia has both the special conditions
 requiring some extensions to the current 4WD/dirt road mapping data and the
 active mapping community to back it up. I don't see why we shouldn' agree
 on a handful of tagging rules for the AU conditions on this list and use
 them (assuming that they are well thought out etc). Document them nicely so
 the rest of the world can take them up, and make the rendering changes etc
 ourselves (how hard can a casing change be in the renderer? If we can do it
 an submit it to the trac system...)


AFAIK the major issue with rendering changes is resources to implement
them. So, if someone writes the code to do it, much greater chance of it
happening.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


Re: [talk-au] Historical rail lines

2012-11-25 Thread Steve Bennett
On Sun, Nov 25, 2012 at 9:28 PM, Ian Sergeant inas66+...@gmail.com wrote:


 I'm pretty sure we've reached consensus in the past that if there is
 absolutely no evidence of it on the ground - no tunnels - no cuttings - no
 tracks.  In other words there was a railway line, but now it is a shopping
 mall, then it doesn't get mapped.  We don't maintain layers of history in
 OSM right now.


Here's what the wiki says:

Abandoned - The track has been removed and the line may have been reused
or left to decay but is still clearly visible, either from the replacement
infrastructure, or purely from a line of trees around an
original cutting or embankment. Use 
railwayhttps://wiki.openstreetmap.org/wiki/Key:railway
=abandoned https://wiki.openstreetmap.org/wiki/Tag:railway%3Dabandoned.
Where it has been reused as a cycle path then add
highwayhttps://wiki.openstreetmap.org/wiki/Key:highway
=cycleway https://wiki.openstreetmap.org/wiki/Tag:highway%3Dcycleway

For the case of the Inner Circle line, there is ample evidence:
- some track, buildings etc
- large sections of reserved land (according to our map, the Linear Park
Reserve)
- a bike path (the Inner Circle Rail Trail):
https://www.railtrails.org.au/trail?view=trailid=133

I agree that where a rail line has been completely removed and sold off,
and built over, the story is a bit different. But in this case, great
effort has been expended to retain it as a feature of the landscape: hence
the park, bike path, etc. Its presence lives on much more than some
abstract representation on a map. It's completely plausible that people
would want to follow the old train line on the map - in a way that wouldn't
be the case if it had been built over by houses or shopping malls.

There are other abandoned railways that perhaps shouldn't be mapped, but
the case is pretty good for this one.

Steve
___
Talk-au mailing list
Talk-au@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-au


  1   2   3   4   5   6   7   8   9   10   >