[HOT] How to handle existing features for #1018 Nepal task

kusala nine kusala9 at googlemail.com
Thu Apr 30 16:57:01 UTC 2015


An interesting observation here. during this time HOT is experiencing a
large expansion of activity and our strength will be how we learn and adapt
will be our strength for the future. At some point in the future we should
go through all our email trails and mine the experiences people have had to
set development priorities for the future. there's many generic
crowdsourcing themes here - the importance of validation, the enthusiasm of
new users, the profound commitment of contributors at all levels and the
overall ability of a self-organising system to produce a valuable resource
during this time. Possibly a small group of contributors with a variety of
experience and skills could find the useful observations at some point -
we're all very busy now but.... just my observation. jon.

On Thu, Apr 30, 2015 at 3:29 PM, Pierre Béland <pierzenh at yahoo.fr> wrote:

> Hi Steve,
>
> I agree that we can improve. Each activation is pushing us to our limits.
> The leaders, we do not have time to look at the details. Did not have time
> either to write updates either then sending short messages on
> https://twitter.com/pierzen
> An other area whre help woudbe appreciated. I dont have time to record all
> the excellent suggestions on the list.
>
> regard
>
> Pierre
>
>   ------------------------------
>  *De :* Steve Bower <sbower at gmavt.net>
> *À :* hot at openstreetmap.org
> *Envoyé le :* Jeudi 30 avril 2015 2h07
> *Objet :* [HOT] How to handle existing features for #1018 Nepal task
>
> I'm working on project #1081 Nepal detailed mapping 2nd pass. I'm new to
> OSM but have lots of GIS experience.
>
> Not sure how best to handle existing features already in the data (either
> from the 1st pass, or predating the project)
>
> (1) The instructions say "do not trace all the paths in the fields or
> small paths of a few hundred meters that do not connect to road networks."
> If there are existing paths like that in the data, should I delete them?
>
> (2) An existing long way tagged "highway=track" appears to start as a
> track that could support motorized vehicles (2 tire tracks are visible),
> but soon becomes very difficult to distinguish and is perhaps impassible.
> I'm guessing it was traced from different imagery (I checked Bing and
> MapBox, per the instructions) - I would not have traced much of it - too
> hard to see. Should I split this long way and label the second part
> 'highway=unclassified' or similar?
>
> (3) Some small hamlets of 5-10 buildings, accessible only by paths, are
> enclosed in existing 'landuse=residential' polygons. The validation
> instructions are to confirm there are highways connecting 'residential'
> areas, and that there are 'residential' polygons around clusters of "20 or
> so houses". Should I remove the 'residential' polygons around tiny hamlets
> that are not on roads?
>
> (By the way, I read the AAAARRRRGH! thread and agree with Stacey and
> others that more detailed project instructions would be one key way to
> improve quality and consistency, especially from new users. My questions
> reflect the sort of basic examples that could be part of more detailed
> instructions. Being a new user with "fresh eyes", I could help with that -
> but that's a different thread.)
>
> Steve
>
>
> _______________________________________________
> HOT mailing list
> HOT at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/hot
>
>
>
> _______________________________________________
> HOT mailing list
> HOT at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/hot
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20150430/10e0b2ae/attachment.html>


More information about the HOT mailing list