Re: [Talk-GB] OSM feature density vs edits per features

[De Sabbata, Stefano (Dr.)]

Dear all,

I would like to warmly thank all those who have taken the time today to answer 
to my email – this has been most helpful!

If you have further thoughts, or if you like to see some aggregated stats 
mapped for Greater London, please do get in contact. I have some drafts of my 
current analysis that I am happy to share, although they are not ready to be 
published. Although, I actually did publish some very early stage map on the 
work I am doing (which OSM is part of) on Twitter some months ago.

https://twitter.com/maps4thought/status/854843894860845057

Also, last year I did publish a short paper (also a slightly different, simpler 
analysis than what I am doing now) focusing on Leicester (see link below). Any 
comment on that would helpful too. :)

http://escholarship.org/uc/item/5hc4d2q6

More recently, I collaborated on a paper titled “Geodemographic biases in 
crowdsourced knowledge websites: Do neighbours fill in the blanks?” – although 
that’s based on a relatively small and specific sample of places with alcohol 
licence, in the scope of future health policy analysis.

https://twitter.com/maps4thought/status/868064714466754560

All the best,
Stefano.


From: SK53 <sk53@gmail.com>
Date: Friday, 1 September 2017 at 18:50
To: Philip Barnes <p...@trigpoint.me.uk>
Cc: "Talk-GB@openstreetmap.org" <talk-gb@openstreetmap.org>
Subject: Re: [Talk-GB] OSM feature density vs edits per features

I suspect the problem with Leicester is simply fewer mappers and therefore much 
more higher noise to signal. Also Leicester is odd in having a very active 
mapping community early on and much less (comparatively) activity more 
recently. Although the same is probably true of both Liverpool & Manchester.
Given that active mapping communities on a per country basis are often in the 
hundreds or low thousands using a city as big as London makes sense, even 
allowing for the effect of visitors. It's also too big for a few dedicated 
mappers to do on their own.
J

On 1 September 2017 at 18:13, Philip Barnes 
<p...@trigpoint.me.uk<mailto:p...@trigpoint.me.uk>> wrote:
On Fri, 2017-09-01 at 12:04 +0100, Andy Townsend wrote:
>
> One thing you won't be short of is the sheer volume of data.  If you
> like a challenge, go for it - but I suspect that you might be better
> starting with somewhere with fewer variables if you want some
> quantifiable results to come out of it.
>
Or as you are based in my old hometown of Les-tah, why not use that as
a starting point to get to grips with OSM data. It has fewer variables
and is mostly mapped by locals. It is not such a tourist attraction as
London.

And as for learning OSM it makes sense to start with where you are, if
something is puzzling you then you can go and have look see without
traveling 100 miles.

Phil (trigpoint)

___
Talk-GB mailing list
Talk-GB@openstreetmap.org<mailto:Talk-GB@openstreetmap.org>
https://lists.openstreetmap.org/listinfo/talk-gb

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

Re: [Talk-GB] OSM feature density vs edits per features

[Philip Barnes]

On Fri, 2017-09-01 at 12:04 +0100, Andy Townsend wrote:
> 
> One thing you won't be short of is the sheer volume of data.  If you 
> like a challenge, go for it - but I suspect that you might be better 
> starting with somewhere with fewer variables if you want some 
> quantifiable results to come out of it.
> 
Or as you are based in my old hometown of Les-tah, why not use that as
a starting point to get to grips with OSM data. It has fewer variables
and is mostly mapped by locals. It is not such a tourist attraction as
London.

And as for learning OSM it makes sense to start with where you are, if
something is puzzling you then you can go and have look see without
traveling 100 miles.

Phil (trigpoint)

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

Re: [Talk-GB] OSM feature density vs edits per features

[Gregory]

Disclaimer: I suppose I'm a former member of the London OSM community, and
now only visit occasionally.

I'd add that it's worth noting OSM in London has history similar to the
city it's self. It's not a typical city/town, but many small villages that
expanded & grew with centres that then merged into each other. This is why
London has lots of different highstreets/market squares/"downtowns".

OpenStreetMappers in London worked on their local areas from a blank
canvas, and started expanding the distance from home/work that they mapped.
The mapping distance/radius continues to expand for as long as OSM interest
is maintained, or until they think it's reached/met other map areas. If
interest is maintained, it might be that they return/reset to 0-distance
and start adding different features/detail. The result can be fuzzy halos
of data/detail between different mapper's patches. It's hard to casually
spot those boundaries of detailed data, particularly because the existing
mappers are not regularly visiting the areas (due to the reasons that
created the anomalies).

A previous replier referred to "City of London" when I think they meant
"Central London". Be careful, because "City of London" is a very specific
place and isn't the center or downtown for most people/uses.

As for "Visiting OSMers" and their impact on data maintenance, there are
other cities I would pick to look into that. It would be good to look at,
and also to see if it can be measured how well a city/community has dealt
with it.

>From the center of Newcastle,
Gregory.


On 1 Sep 2017 12:29 pm, "Dan S"  wrote:

Hi

One specific possibility: I know that at least a handful of people have
been deliberately filling out London's buildings, and in un-surveyed areas
this is often by tracing building shapes from aerial views. This is
motivated by London's "patchy" coverage: areas where mappers live, plus
tourist/popular streets, are well-mapped, while other (residential etc)
areas sometimes looked blank. Maybe worth checking if the difference you
see is due to a specific object type (in particular, building=*).

Best
Dan


2017-09-01 12:01 GMT+03:00 De Sabbata, Stefano (Dr.) <
s.desabb...@leicester.ac.uk>:

> Dear all,
>
>
>
> my name is Stefano, and I am a lecturer in geography at the University of
> Leicester. Volunteered geographic information and OSM have been one of my
> research interests for a few years now. I have also done some very minor
> contributions to OSM myself, but never in London and I am not very familiar
> with the London OSM community.
>
>
>
> As I have been recently looking at some aspects data production and
> quality of OSM in London, and I was wondering whether any of you might have
> some time to help me out in interpreting some of the data… I have a couple
> of maps that can use some local expertise to make sense of them. :)
>
>
>
> In particular, I am looking at the relationship between feature density
> and number of edits per features (among other things). It looks like there
> are areas of London with consistently high density but low average number
> of edits per features. I was wondering whether this might be the result of
> local projects or mapathons organised by the community? Or might there be
> (and most probably there are) other reasons I am overlooking?
>
>
>
> If you are interested, please contact me: at s.desabb...@le.ac.uk
>
>
>
> All the best,
>
> Stefano.
>
>
>
>
>
> *Dr Stefano De Sabbata*
>
> *Lecturer in Quantitative Geography*
>
> Department of Geography,
>
> University of Leicester, University Road, Leicester, LE1 7RH, UK
>
> *t: *+44 (0)116 252 3812 <+44%20116%20252%203812>
>
> *e:* s.desabb...@le.ac.uk
>
> *w: *le.ac.uk/departments/geography/people/stefano-de-sabbata
> 
>
> *twitter: *@maps4thought 
>
>
>
> *Research Associate*
>
> Oxford Internet Institute, University of Oxford
>
> *oii.ox.ac.uk/people/desabbata *
>
> Information Geographies 
>
> Connectivity, Inclusion, and Inequality 
>
>
> [image: id:77B79125-D80D-4E72-9F70-791C2419DE28@home]
>
> Follow us on Twitter  or visit our
> Facebook  page
>
>
>
> ___
> Talk-GB mailing list
> Talk-GB@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-gb
>
>

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

Re: [Talk-GB] OSM feature density vs edits per features

[Dan S]

Hi

One specific possibility: I know that at least a handful of people have
been deliberately filling out London's buildings, and in un-surveyed areas
this is often by tracing building shapes from aerial views. This is
motivated by London's "patchy" coverage: areas where mappers live, plus
tourist/popular streets, are well-mapped, while other (residential etc)
areas sometimes looked blank. Maybe worth checking if the difference you
see is due to a specific object type (in particular, building=*).

Best
Dan


2017-09-01 12:01 GMT+03:00 De Sabbata, Stefano (Dr.) <
s.desabb...@leicester.ac.uk>:

> Dear all,
>
>
>
> my name is Stefano, and I am a lecturer in geography at the University of
> Leicester. Volunteered geographic information and OSM have been one of my
> research interests for a few years now. I have also done some very minor
> contributions to OSM myself, but never in London and I am not very familiar
> with the London OSM community.
>
>
>
> As I have been recently looking at some aspects data production and
> quality of OSM in London, and I was wondering whether any of you might have
> some time to help me out in interpreting some of the data… I have a couple
> of maps that can use some local expertise to make sense of them. :)
>
>
>
> In particular, I am looking at the relationship between feature density
> and number of edits per features (among other things). It looks like there
> are areas of London with consistently high density but low average number
> of edits per features. I was wondering whether this might be the result of
> local projects or mapathons organised by the community? Or might there be
> (and most probably there are) other reasons I am overlooking?
>
>
>
> If you are interested, please contact me: at s.desabb...@le.ac.uk
>
>
>
> All the best,
>
> Stefano.
>
>
>
>
>
> *Dr Stefano De Sabbata*
>
> *Lecturer in Quantitative Geography*
>
> Department of Geography,
>
> University of Leicester, University Road, Leicester, LE1 7RH, UK
>
> *t: *+44 (0)116 252 3812 <+44%20116%20252%203812>
>
> *e:* s.desabb...@le.ac.uk
>
> *w: *le.ac.uk/departments/geography/people/stefano-de-sabbata
> 
>
> *twitter: *@maps4thought 
>
>
>
> *Research Associate*
>
> Oxford Internet Institute, University of Oxford
>
> *oii.ox.ac.uk/people/desabbata *
>
> Information Geographies 
>
> Connectivity, Inclusion, and Inequality 
>
>
> [image: id:77B79125-D80D-4E72-9F70-791C2419DE28@home]
>
> Follow us on Twitter  or visit our
> Facebook  page
>
>
>
> ___
> Talk-GB mailing list
> Talk-GB@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-gb
>
>
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Re: [Talk-GB] OSM feature density vs edits per features

[Andy Townsend]

On 01/09/2017 10:55, Neil Matthews wrote:


Not a member of the London OSM community, but...

Also, do you think that there might also be an effect from "visiting" 
OSM contributors to an area?


I used to regularly walk to/from occasional meetings with companies in 
London and map items on the way




I'm also exactly that sort of "non-London London mapper".

With a DWG* hat on I occasionally get asked to investigate problems in 
lots of places around the world, so I've got a bit of subjective 
familiarity with "how different places get mapped", and I suspect that 
London is perhaps one of the most difficult places to undertake the 
analysis that it sounds like you're doing, for a few reasons:


o lots of history (lots of it pre OSM licence change)

o Very high feature density, so that edits to one thing can accidentally 
affect others nearby


o Very wide spectrum of different sorts of mappers ranging from people 
who just map "everything" to people who concentrate on certain sorts of 
features (who may or may not be locals - wikidata adders sometimes 
aren't, for example) and tourists using MAPS.ME.


One thing you won't be short of is the sheer volume of data.  If you 
like a challenge, go for it - but I suspect that you might be better 
starting with somewhere with fewer variables if you want some 
quantifiable results to come out of it.


Best Regards,
Andy

* OSM's Data Working Group.  For the avoidance of any doubt I'm writing 
this in an entirely personal capacity.



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

[Talk-GB] OSM feature density vs edits per features

[De Sabbata, Stefano (Dr.)]

Dear all,

my name is Stefano, and I am a lecturer in geography at the University of 
Leicester. Volunteered geographic information and OSM have been one of my 
research interests for a few years now. I have also done some very minor 
contributions to OSM myself, but never in London and I am not very familiar 
with the London OSM community.

As I have been recently looking at some aspects data production and quality of 
OSM in London, and I was wondering whether any of you might have some time to 
help me out in interpreting some of the data… I have a couple of maps that can 
use some local expertise to make sense of them. :)

In particular, I am looking at the relationship between feature density and 
number of edits per features (among other things). It looks like there are 
areas of London with consistently high density but low average number of edits 
per features. I was wondering whether this might be the result of local 
projects or mapathons organised by the community? Or might there be (and most 
probably there are) other reasons I am overlooking?

If you are interested, please contact me: at 
s.desabb...@le.ac.uk

All the best,
Stefano.


Dr Stefano De Sabbata
Lecturer in Quantitative Geography
Department of Geography,
University of Leicester, University Road, Leicester, LE1 7RH, UK
t: +44 (0)116 252 3812
e: s.desabb...@le.ac.uk
w: 
le.ac.uk/departments/geography/people/stefano-de-sabbata
twitter: @maps4thought

Research Associate
Oxford Internet Institute, University of Oxford
oii.ox.ac.uk/people/desabbata
Information Geographies
Connectivity, Inclusion, and Inequality

[id:77B79125-D80D-4E72-9F70-791C2419DE28@home]

Follow us on Twitter or visit our 
Facebook page

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