By the way, that was found via OpenGeoData - http://opengeodata.org/nice-blog-post-by-someone-getting-started-wit

The "bot" account is a normal account - it just lets you isolate those bulk entries from your personal contributions.

Getting GPS logs from large fleets is a wonderful opportunity - I'm thinking about how to evangelize them about it...

Yes, features must represent the current reality.

Here is how I would tag that one :

highway=unclassified
surface=dirt
motorcar=no
lanes=1

I was going to say "awesome"... And then I found that Adam did it already !

I read about other initiatives such as loaning cheap GPS receivers to bus drivers. Sounds like fleets managers are people we should be talking to, to harvest GPS tracks at a large scale.

I have noticed that in many places, in countries in which Google does not have significant commercial interest, even many villages have part of their street grid mapped. But looking a little closer, this is a partial mapping of a seemingly random subset of the grid, none of those streets have names and the precision is horrendous. To me, this looks like what some grid recognition automaton would produce if configured to only trace the streets it detects with a certainty above a certain threshold. I mentioned that idea a while ago on OSM-talk (http://www.mail-archive.com/[email protected]/msg23852.html) but it did not get much echo. But I cannot believe that human mappers would produce such bad maps.

I like the very clear schematic diagrams. But isn't there a way to distinguish between L1a and L1b ?

@Ævar - I admit I let my enthusiasm get the better of my objectivity. From a rational point of view, you are right. But maybe some exuberance in expressing feelings is a way to attract attention - which then is more receptive to the more rational argument.

Nevertheless, I object that users of Map Maker know exactly what they sign up for : many users do not understand how licensing is key to the whole system. Making noise about it is a start - it may merely be a way to wake them up, but it takes a fully awoken user to start considering things rationally.

Your announcement made me discover TopOSM and I'm in awe of its beauty. With the CORINE land cover imports going on in Europe, it is going to be magnificent there !

If they are an exact carbon copy of reality and feature a credible source, then it is acceptable... But indeed if reality diverges and no source is cited, this hints at malpractice. If you have any question, your best option is to just ask the author privately.

@Ævar : Linking to an uploaded GPS trace sounds like a good idea to me.

@Richard : The use case I had in mind was when in an area with multiple GPS traces I want to relate to a specific one that I trust because I carefully surveyed it.

Modification date does not tell you how old is the specific source you chose. But as you suggest, maybe modification date is enough.

I'm taking it slow, improving gradually what is already there. Landmark buildings with names and height and surface ways for starters.

On top of the vertical complexity, the lower levels pose a data-gathering challenge : they don't appear on satellite imagery, GPS is useless there and I don't have a decent inertial navigation system at hand... I hope I'll find source material with an accommodating license. Maybe I'll ask EPAD - the semi-public organism in charge of managing La Défense.

I wonder about how to model a station. The simplest way is of course a single node along the way. But is it advisable to model the station as an area ?

Check your URL - you put an extraneous parenthesis at the end.

JOSM lets you open a list of people working on the selected object (Alt+Maj+A). I won't blame the author publicly - but maybe you'll want to contact him to explain him how to do it the right way...

@brig Thanks - I just discovered the use of the Yahoo imagery underlay, and I agree that it is very practical for mapping my city where high resolution imagery is available. Indeed I would not have been able to achieve this level of detail with my 1/15 Hz GPS logger. The case I reported was a rural area where no high resolution imaging was available - in that case the GPS traces seem to provide better resolution, even though the roads traced from Yahoo imagery are most often correct if slightly offset.

Everything as an area - or as a volume - makes ultimate sense. But how are we going to get data at that resolution ?

Some analysis about the release of Ordnance Survey maps : http://www.systemed.net/blog/?p=145

@andrewpmk - Thanks for the tip : indeed some of the cases encountered were slight misalignment of a whole stretch and would have benefited from the method you suggest. Other cases were simply were the definition of the GPS tracks was greater than the definition of roads traced from Yahoo imagery which was otherwise correct.

@ctheile - La N1 et quelques un de ses affluents à l'ouest de Sinndia étaient légèrement à côté. Rien de grave, et j'ai faites les corrections...