Re: [OSM-talk] Awards

2009-08-23 Thread Till Harbaum / Lists
Hi,

Am Sonntag 23 August 2009 schrieb Ed Avis:
 There are plenty of unnamed streets on the map - where in the real world no
 name has been assigned by the local authority.  We could name those streets
 after top OSM contributors.
Sounds like a perfect idea to cause confusion. The whole idea of maps is
to represent the real worls a precise as possible.

Till


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


Re: [OSM-talk] Awards

2009-08-23 Thread Till Harbaum / Lists
Hi,

Am Sonntag 23 August 2009 schrieb Liz:
 plane, and 'precision' is not part of a Mercator projection. Mercator's 
 projection was for the purpose of calculating direction of travel, in 
 particular for seafarers.
a) osm by itself does not have a projection. It's the maps that project
  the osm data
b) mercartor is as precise as possible with repect to directions. 

Sure, a map cannot be perfect and you have to select those aspects
that are important for your particular application (be it angles or distances
or just the selection of objects you show).

But that all doesn't give us a reason to add artificial, misleading and useless 
information to the osm maps. 

Till

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


Re: [OSM-talk] Announce: OSM2Go map editor 0.6.13 released for Maemo, Debian, and Ubuntu

2009-02-18 Thread Till Harbaum / Lists
Hi,

Am Mittwoch 18 Februar 2009 schrieb Andrew Chadwick (email lists):
 Now *that* would be a worthwhile new feature. If you're using it on an
 n810 of course, you can use the rather excellent maemo-mapper to zoom in
 and get a view; it's still a little clunky though. Having something
 built-in would be interesting.
 
 Anybody out there have a GPL3-compatible GTK2 slippy map coordinate
 selector widget coded up in C? :) Ideally goocanvas-based

I thought about this some time ago and found this to be rather interesting:
  http://www.johnstowers.co.nz/blog/index.php/tag/osmgpsmap/

I played around with the map and wasn't 100% satisfied as not everything
worked as it should. But this may be a basis for such a feature. 

Till

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


[OSM-talk] osm2go as yet another desktop tool?

2008-12-15 Thread Till Harbaum / Lists
Hi,

osm2go is evolving faster than i expected and with first volunteers joining
development things progress even faster,

We recently implemented support for JOSMs elemstyle.xml file allowing us
to display everything very close to the josm style. However, things didn't stop
there and we have started to extend the format for e.g. dashed lines (next
will likely be some way to describe bridges as potlatch displays them).

Also Andrew wrote a mapnik like style which imho is the nicest style so far
and which has been selected to be the osm2go default style. See

http://www.harbaum.org/till/maemo/osm2go-0.6-7.jpg

So the question is: This has started as a mobile editor with focus on small
devices and ease of use. Do you think there's enough demand for yet another
desktop editor to also support binary distributions for desktop linux machines?

Till

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


Re: [OSM-talk] osm2go as yet another desktop tool?

2008-12-15 Thread Till Harbaum / Lists
Hi,

Am Montag 15 Dezember 2008 schrieb sylvain letuffe:
 I'll be happy to give it a go, if the porting cost isn't too huge.
Porting costs? What's that supposed to be?

 JOSM is allready quite good, but in liberty, there is diversity... and one 
 editors and a half doesn't look enough to me
JOSM imho isn't trivial for beginners. That may be where osm2go can help
on the desktop.

Till

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


Re: [OSM-talk] osm2go as yet another desktop tool?

2008-12-15 Thread Till Harbaum / Lists
Hi,

Am Montag 15 Dezember 2008 schrieb sylvain letuffe:
 I suppose you didn't asked that if it was allready ready for linux distro, so 
 there might be an additionnal cost such as :
 - package making
 - compilation testing
 - libraries dependencies
 etc.
You mean after writing 12.000 lines of code it may just be too much work writing
a debian control file? 

 So yes, I'm interested, but unfortunetly don't have enough knowlege to help
Ok, _that_'s a reason. 

Till

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


[OSM-talk] osm2go on the openmoko

2008-12-08 Thread Till Harbaum / Lists
This is for you openmoko users:

http://comiles.eu/~natanael/wordpress/2008/12/08/osm-on-the-mo

Can't wait to see a port to the pandora :-)

Till

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


Re: [OSM-talk] Yahoo WMS server?

2008-12-07 Thread Till Harbaum / Lists
Hi,

Am Sonntag 07 Dezember 2008 schrieb Chris Browet:
 It doesn't allow to rip their tiles, converted/scales/cropped them and
 re-serve the modified ones.
Ok, but that's exactly what the josm plugin also does. So i assume it's ok
to use such a server internally without providing them to the public.

You could even run it on the device itself.

Till

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


Re: [OSM-talk] JOSM style collection?

2008-12-07 Thread Till Harbaum / Lists
Hi,

Am Sonntag 07 Dezember 2008 schrieb Ulf Lamping:
 Seems a little more generalization of style files (between potlatch, 
 mapnik, josm, osmarender, ...) might save us all a lot of work in the 
 long run ...
Yepp, that's exactly why i made osm2go to use josm files (presets.xml and
elemstyle.xml by now), early versions used potlatch config files.

  From time to time I'm maintaining the default JOSM style and don't know 
 of any more such styles.
Now you know :-)

 Will make it much more likely, that stuff you want to see in the default 
 JOSM style will get in (hint, hint, ...) ;-)
I've added a link to the osm2go source archives containing those alternate
style files.

Osm2go uses another layer of style files (which in turn include elemstyle.xml)
so global settings for area transparencies, area border settings, icon scaling
settings etc etc can be changed.

Ciao,
  Till

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


Re: [OSM-talk] Yahoo WMS server?

2008-12-06 Thread Till Harbaum / Lists
Hi,

Am Donnerstag 04 Dezember 2008 schrieb Frederik Ramm:
 None that I'm aware of. We're going through a lot of pain with JOSM to 
 display Yahoo background images, and we wouldn't do that if there was a WMS.
I wonder why the ywms plugin is a plugin at all as it sounds like it's exaclty
the type of server i am asking for.

  Yahoos license seem to allow this
 Tell me more?
I've been told that the yahoo license permits derivative works which also
is the reason why potlatch and josm are making use of this. Wouldn't a 
server distrubuting slightly converted/scales/cropped images not also
provide just derivative work?

Till

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


[OSM-talk] Yahoo WMS server?

2008-12-04 Thread Till Harbaum / Lists
Hi,

is there a stand-alone yahoo WMS server? Yahoos license seem to allow this
and some users of my osm2go ask for yahoo images. Going through a 
wms server would be the easiest and cleanest solution i think.

Thanks,
  Till

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


Re: [OSM-talk] OSM not acceptable for geocaching.com

2008-11-27 Thread Till Harbaum / Lists
Hi,

it's restored. They say that they thought OSM was an unreliable commercial (due 
to
the request for donations) service which they don't want to have their users to 
do.

Kind of strange explanation as the purpose of OSM is pretty obvious and my cache
even said osm is for streetmaps what wikipedia is for encyclopedia. So the 
idea
that it's non-commercial basically can't be missed.

Till


Am Mittwoch 26 November 2008 schrieb Nick Black:
 Did you get anywhere with this?  Have you tried emailing the admins at
 geocaching.com to see why they removed the cache?  I'd love to hear more.
 
 Nick
 
 
 On Sat, Nov 15, 2008 at 2:02 PM, Thomas Wood [EMAIL PROTECTED]wrote:
 
  2008/11/15 Till Harbaum / Lists [EMAIL PROTECTED]:
   Hi,
  
   i have recently released a geocache which basically required you to look
  up a certain node
   in the OSM database. The position of that node was then the place where
  the geocache was
   hidden. Geocaching.com users can perhaps still read the original listing
  at:
  
  http://www.geocaching.com/seek/cache_details.aspx?guid=80a9308b-6719-485d-a0dc-846798a8cac2
 
  Through a bug in their site code, the original listing is visible
  here:
  http://www.geocaching.com/seek/cdpf.aspx?guid=80a9308b-6719-485d-a0dc-846798a8cac2
 
   Geocaching.com recently completely deleted that cache antry as they claim
  that it forces you to use a certain
   software (a web browser!!!) and a certain web service.
 
  They have un-published the listing, an event that occurs not very
  often - usually only if the reviewer who published it realises they
  made a mistake soon after.
  The specific guideline reads something like caches that require
  (unusual) third party software to be installed are not permitted,
  there's also a similar rule about cache perminance in terms of
  external resources on the net - eg hosting an mp3 on a personal
  website will not be acceptable as a part of the 'puzzle' as they have
  a habit of falling offline.
 
   This is a strange explanation as geocaches requesting you to find a
  certain image on google earth
   are pretty common. On the other hand Geocaching.com seems to have a
  business with google. This
   may be the explanation why they don't like to deal with openstreetmap. I
  really wonder if
   it's google behind this.
 
  They have business with Google as far as using their Maps API,
  publishing KML files, and using AdWords, I don't think they have any
  further links with them.
 
   This includes quite extreme behaviour on the GC.com side as they are not
  using their usual methods
   of disabling or archiving caches. Instead they reset their entire
  database with respect to this
   cache to the state before it was published. It's like they really want to
  clean all traces related to
   this geocache.
 
  The GC.com side is usually just a volunteer reviewer rather than one
  of the company's employees. As noted, caches can be removed completely
  from the site - 'unpublished' on the event of the reviewer making a
  mistake.
 
   IMHO a very interesting issue and may mean that google sees a serious
  competitor arriving ...
 
  Not in my view.
 
   Till
 
  I'm asking some contacts I have to see if I can get the full logs for
  publishing and subsequent removal of it to see if a reason is further
  given.
 
  --
  Regards,
  Thomas Wood
  (Edgemaster)
 
  ___
  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: [OSM-talk] Osm2go: mobile mapping with Nokia n800/n810

2008-11-23 Thread Till Harbaum / Lists
Hi,

thanks. I am not sure about the performance impact of the patch as it increases
the overall canvas size significantly. I'll have an eye on this as especially 
the small
maemo devices seem to fight with their low memory.

One nice thing one can do with this is to set the frisket transparency to e.g. 
0xc0.
This will give you a light view on everything outside the working area. But then
again: This has a significant performance impact which sure isn't acceptible.
But nice, anyway :-)

One of the next releases will sure use your patch as i can always remove it 
again
if there's a real drawback.

Thanks again,
  Till

Am Sonntag 23 November 2008 schrieb Andrew Chadwick (mailing lists):
 Till Harbaum / Lists wrote:
  I do have a small problem with it: I i try to zoom out and hit the zoom 
  limit, 
  the canvas slowly scrolls a tiny step top and left with every trial to zoom 
  further out.
 
 Okay, give this a spin. The attached patch applies cleanly against 
 vanilla osm2go-0.5-5 from the source .dsc/.tar.gz once again. Rather 
 than fight with the rounding errors that were causing this, I've added 
 some recentreing code to make attempts to zoom out beyond the limit 
 recentre gently pan the map towards the centre in controllable 
 increments. Once it hits the centre, all is still. You might want to add 
 a message flash at either the no-more-zoom point, the no-more-recentring 
 point, or both; should be obvious when those happen.
 
 The feel on zooming out shouldn't be much different to what currently 
 happens in vanilla osm2go actually, but now there's freedom to pan the 
 map around when a project is opened. There are a few hardcoded constants 
 that you might want to tweak for speed and prettiness on actual n8X0 
 hardware; I'll leave that to you because I don't have the SDK here right 
 now.
 



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


Re: [OSM-talk] Osm2go: mobile mapping with Nokia n800/n810

2008-11-22 Thread Till Harbaum / Lists
Hi,

i like this patch.

I do have a small problem with it: I i try to zoom out and hit the zoom limit, 
the canvas slowly scrolls a tiny step top and left with every trial to zoom 
further out.

Till

Am Samstag 22 November 2008 schrieb Andrew Chadwick (mailing lists):
 Till Harbaum wrote:
  
  If you're zoomed out beyond the extent of the current project, it's
  impossible to pan the map so that part of it is hidden beyond the canvas
  edge. This might hinder new users [...]
  ??? I don't understand that. There's the area you downloaded. You can 
  see only that and work only on that. You can never leave the active area
  by panning/scrolling and you can't place anything there.
 
 Some users (myself included) like to pan the display *before* zooming 
 in. This way there's less panning required to get to an object in the 
 corner or the working area. Therefore, when the map is zoomed all the 
 way out, it should be possible to pan the display a bit.
 
 But the best way of explaining an idea like this is code! In the 
 attached patch, I've rejigged the way panning and scrolling work in a 
 way that addresses my perceived problem. Hope this is useful.
 
  That's not how the goocanvas works. You cannot easily make display
  data outside the given working area.
 
 The attached patch should address this, for goocanvas builds only. Note 
 that now working area != display area, but that shouldn't affect uploads 
 or the semantics of the app in the slightest.
 
 It applies cleanly against osm2go-0.5-5, as found at 
 http://repository.maemo.org/extras/pool/chinook/free/source/o/osm2go/
 



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


Re: [OSM-talk] Osm2go: mobile mapping with Nokia n800/n810

2008-11-17 Thread Till Harbaum / Lists
Hi,

Am Montag 17 November 2008 schrieb Andrew Chadwick (email lists):
 There's quite a lot of stylus dragging in osm2go right now, both for
 panning and for moving objects. It's not always that reliable on my
You can already pan using the directop pad (or perhaps this isn't in the
version already published. So at least the next release will allow you 
to pan by the keys).

 the way Maemo-Mapper does it, with a single tap *or* tap+drag. Not sure
 how well that'd work in places with a lot of areas though.
That's the problem. I have plenty of examples where there's always an
area whereever you go ...

 osm2go should have a mode for following the GPS marker when it wanders
 off-screen :) Maybe turned off temporarily while you have something
I am afraid i don't understand what you want.

 At very high zoom levels, POIs can hide very short ways causing them to
 be uneditable. Is it possible to implement a maximum pixel size for POIs
 and Way widths? I can supply an example if you need one.
Hmm, that's difficult as the canvas always handles the items at their given
size. So the road widths and poi diameters just scales with the zoom.

I was already thinking about some global things to change the size of
all objects. The opposite also happens if you are in a very dense populated 
area and everything is just a bunch of thin lines.

 If you're zoomed out beyond the extent of the current project, it's
 impossible to pan the map so that part of it is hidden beyond the canvas
 edge. This might hinder new users because it creates an artificial
??? I don't understand that. There's the area you downloaded. You can 
see only that and work only on that. You can never leave the active area
by panning/scrolling and you can't place anything there.

 distinction between the behaviour when zoomed in (can pan without
 hitting a wall) and when zoomed out (cannot pan in some or all
 directions). I'd suggest drawing a shaded border of about 64px width
 around the current project, perhaps looking like JOSM's new display
 boundaries of downloaded data hatching, and using that as an outer,
 visible handle. So it'd be possible to pan to
That's not how the goocanvas works. You cannot easily make display
data outside the given working area.

 The screen real estate usage is generally very well thought out and
 makes sense to me. It'd be really cool if the dependency on Hildon,
 osso, and various other Maemo-specific parts could be relaxed for users
 of devices like Eee PCs, but I guess that's less of an immediate
 requirement :)
There's no hildon requirement at all. In fact i do 99% of all coding and
testing on the gtk linux desktop. This should run out of the box on
any gtk based machine.

 I wonder if, like gosmore, it would make sense for osm2go to do some
 sort of level-of-detail hiding for lower zoom levels, in order to speed
 up the drawing process. It might make sense to restrict editability only
 to higher zooms too, like potlatch does (IIRC).
Also that's not how the goocanvas works and it's hard to circumvent those
things.

Till


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


[OSM-talk] OSM not acceptable for geocaching.com

2008-11-15 Thread Till Harbaum / Lists
Hi,

i have recently released a geocache which basically required you to look up a 
certain node
in the OSM database. The position of that node was then the place where the 
geocache was
hidden. Geocaching.com users can perhaps still read the original listing at:
http://www.geocaching.com/seek/cache_details.aspx?guid=80a9308b-6719-485d-a0dc-846798a8cac2
 

Geocaching.com recently completely deleted that cache antry as they claim that 
it forces you to use a certain
software (a web browser!!!) and a certain web service.

This is a strange explanation as geocaches requesting you to find a certain 
image on google earth
are pretty common. On the other hand Geocaching.com seems to have a business 
with google. This
may be the explanation why they don't like to deal with openstreetmap. I really 
wonder if
it's google behind this.

This includes quite extreme behaviour on the GC.com side as they are not using 
their usual methods 
of disabling or archiving caches. Instead they reset their entire database with 
respect to this 
cache to the state before it was published. It's like they really want to clean 
all traces related to 
this geocache. 

IMHO a very interesting issue and may mean that google sees a serious 
competitor arriving ...

Till

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


[OSM-talk] Osm2go: mobile mapping with Nokia n800/n810

2008-11-12 Thread Till Harbaum / Lists
Hi,

i have been working on osm2go, my little mapping application for the nokia 
internet tablets.
The latest version has all basic functionality required for mapping and i have 
even been able
to map a small village with it.

Nokia N800/N810 users can get a ready-to-run package directly from the extras 
repository.

See http://www.harbaum.org/till/maemo/index.shtml#osm2go for details and 
download
instructions.

Regards,
   Till 

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