[HOT] HOT Digest, Vol 144, Issue 7
Michael Montani
michael.montani at un.org
Tue Mar 1 08:16:14 UTC 2022
Dear all,
thank you for mentioning us on this thread. For sake of clarification, we are mapping no more buildings in Mogadishu, as the Virtual Operation Center (VOC) pilot project ended last September https://tasks.hotosm.org/projects/11003 . The project output was presented at HOT Summit 2021.
Anyway, we keep a close eye on Mogadishu as it is one of the areas (and actually the most imporant one) of interest for UNSOS peacekeepers in Somalia, and we daily fetch OSM data for our operational activities. Among the features we look after the most, surely we have landuse and landcover relations (the residential area of Mogadishu was indeed broken by an activity related to swm, and corrected by us), as well as highways and buildings.
About the usage of private imagery, this seems not to be forbidden but it's a hot topic in discussions with the OSM community at large, and it is always suggested to open it to the public when possible (especially in places of interest as it may the capital city of Somalia). Thank you to HOT and World Bank to manage that! UN Mappers are going to use it when needed.
We are anyway a bit concerned about the HOT projects related to swm. It happens to us to see many errors<https://www.openstreetmap.org/#map=19/2.04394/45.35332> (roads that become not straight, unsquared buildings) coming from swm ( https://www.openstreetmap.org/way/5055097#map=18/2.04283/45.35256 ) and despite we are happy to validate and have a look from time to time to Mogadishu city, it's a bit difficult for us to cope up with continuously upcoming issues.
When mapping buildings in Mogadishu for VOC, knowing the harsh situation for beginner mappers, our approach was to create many tasks aligned with roads and including 2-3 islands of buildings in them, to have mappers interacting the least possible with roads. We also gave good instructions, including squaring of buildings, mapping of footprints at the base and not at the roof of the buildings and so on. You can have a look to our previous projects https://tasks.hotosm.org/projects/11003 https://tasks.hotosm.org/projects/10881 https://tasks.hotosm.org/projects/10869 (we were mapping only areas of high priority identified by our security, but nonetheless big ones). At the end of each project, our internal mappers proceed with validation (e.g. mass alignment of buildings with JOSM, square the unsquared and add more details to simplified buildings). So far, we never received a complaint.
I hope you may take inspiration from our approach and ensure consistent validation in Mogadishu!
Thanks,
--
Michael Montani
GIS Consultant, Client Solutions Delivery Section
Service for Geospatial, Information and Telecommunications Technologies
United Nations Global Service Centre
United Nations Department of Operational Support
________________________________
From: Immaculata Mwanja via HOT <hot at openstreetmap.org>
Sent: Monday, February 28, 2022 11:54 AM
To: hot at openstreetmap.org <hot at openstreetmap.org>
Cc: Justus Muhando <justus at spatialcollective.com>
Subject: Re: [HOT] HOT Digest, Vol 144, Issue 7
Hi all,
Some of the issues mentioned were brought to our attention last year when we were working with YouthMappers from the Somali National University (SNU) to digitize the buildings. This was part of an initiative to strengthen the YM chapter in Mogadishu, but after we started the digitization, we realized that Mogadishu is a very complicated for beginner mappers therefore, we engaged advanced mappers from OSM Uganda while YMs from SNU continued to familiarize themselves with easier projects. Validation has always been done by the team of advanced mappers from Spatial Collective.
We saw the diary from UNGSC and responded to all the questions/concerns that were raised. One issue was that the custom imagery used was not public, but it is now and you can find it here: https://mogadishu-tiles.hotosm.org/{zoom}/{x}/{-y}.png<https://mogadishu-tiles.hotosm.org/%7Bzoom%7D/%7Bx%7D/%7B-y%7D.png>; you can add it as TMS imagery to their JOSM and will be able to view it and compare the digitization.
We have also engaged the UNMappers who are also using digitization data in their 3D modeling activities and are on the same page; they are also now using this imagery in their activities.
I am the author/creator for the "#hotosm-project #mogadishu-swm" projects my OSM username is 'Immaculate Mwanja' and my colleague Justus OSM username 'simoa'. Please send us direct messages to OSM if you have any more questions and we will gladly answer.
Thanks,
Imma
--
Immaculata Mwanja
Global Projects Associate
Twitter: imma_mwanja<https://twitter.com/imma_mwanja>
Phone: +255 (0) 672 093 027 (WhatsApp, Signal)
+250 (0) 791 504 743 (Phone)
Timezone: UTC +02:00
Humanitarian OpenStreetMap Team
Using OpenStreetMap for Humanitarian Response & Economic Development
web<http://hotosm.org/> | twitter<https://twitter.com/hotosm> | facebook<https://www.facebook.com/hotosm> | donate
<http://hotosm.org/donate>
On Mon, Feb 21, 2022 at 2:11 PM <hot-request at openstreetmap.org<mailto:hot-request at openstreetmap.org>> wrote:
Send HOT mailing list submissions to
hot at openstreetmap.org<mailto:hot at openstreetmap.org>
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.openstreetmap.org/listinfo/hot
or, via email, send a message with subject or body 'help' to
hot-request at openstreetmap.org<mailto:hot-request at openstreetmap.org>
You can reach the person managing the list at
hot-owner at openstreetmap.org<mailto:hot-owner at openstreetmap.org>
When replying, please edit your Subject line so it is more specific
than "Re: Contents of HOT digest..."
Today's Topics:
1. Re: Project 11385 (Ralf Bernhardt)
----------------------------------------------------------------------
Message: 1
Date: Sun, 20 Feb 2022 14:51:43 +0100
From: Ralf Bernhardt <raabee at gmx.de<mailto:raabee at gmx.de>>
To: hot at openstreetmap.org<mailto:hot at openstreetmap.org>
Subject: Re: [HOT] Project 11385
Message-ID: <d275aa06-55e9-a47b-cd94-2381d01b9e57 at gmx.de<mailto:d275aa06-55e9-a47b-cd94-2381d01b9e57 at gmx.de>>
Content-Type: text/plain; charset="utf-8"; Format="flowed"
getting interested by the blog post and subsequent discussion here, I
decided to monitor ongoing developments in this area. I wrote some
comments on changesets to make participants aware that alternative
imagery to the given project imagery should also be used - no response.
I also tried to correct some buildings and the road layout west of the
airport. Now mostly ruined again by project 12111. Also interesting to
see the discussion about changset 116808167.
Guess what? After mapping and validation this area with project 12120
again, the crap is still there. How is something like this possible?
Am 13.01.22 um 21:22 schrieb john whelan:
> I think one problem is each project is separate and independent.?
> There is no requirement that says validation should be done in JOSM.?
> Technically the task manager software could be configured to only
> allow JOSM to be used for validation.
>
> The other problem is the numbers involved. There just isn't a big
> enough pool of experienced mappers who validate.
>
> Just how do you validate a project with buildings?? Much of the time
> the imagery is fuzzy.? Some mappers always use Bing imagery to do HOT
> mapping no matter?what the instructions say.? Then you get to the odd
> shapes and sizes.? Are you supposed to correct them?? It takes much
> longer to correct them than to map them in JOSM with the
> buildings_tool plugin.? This doesn't keep validators movitated to
> carry on validating.
>
> More central control?? Dunno, there are tradeoffs but certainly we
> need some sort of standards for projects.
>
> Cheerio John
>
> On Thu, Jan 13, 2022, 06:02 Frans Schutz <frans.schutz at gmail.com<mailto:frans.schutz at gmail.com>> wrote:
>
> Hello.I have checked? at random a few mappings of project 11385
> and can't say anything else but that the validation went well wrong.
> There are 2 people who validated most of them in iDeditor !!! and
> without correcting any significant errors.
> Info from? "Overview of the ResultMaps @ neis-one.org<http://neis-one.org>
> <http://neis-one.org>"
>
> Validator 1 (experience with mapping? since early December 2021)
> Quality assurance: iD editor: Resolved issues=679, Ignored
> warnings=1100
> OSMI issues (details): routing=12, tagging=8
> Osmosis issues: Level 1=0, Level 2=1, Level 3=2
>
> Validator 2 experience with mapping since early December 2021)
> Quality assurance: iD editor: Resolved issues=152, Ignored
> warnings=133
> OSMI issues (details): routing=2
> Osmosis issues: Level 1=0, Level 2=0, Level 3=0
>
> Then I am not talking about whether or not poorly mapped buildings
> were squared. Yes , changes were made, but most of them were not
> changed, while in iDeditor the errors were reported.
>
> To my opinion the best option here is to asign a few experenced
> validators to this project, put all tasks back to "more validaton
> needed" and start all over again with validating
>
> Best regards.
> Frans
>
> Op 11-1-2022 om 18:09 schreef john whelan:
>> Thank you Jiri,
>>
>> Thank you Pierre
>>
>> I just can't think of polite words to describe my reaction to the
>> news other than a plea to use JOSM and the buildings_tool plugin
>> to map buildings to avoid these sort of?problems but at least we
>> understand some of the sources.
>>
>> Cheerio John
>>
>> On Tue, Jan 11, 2022, 12:04 Pierre B?land via HOT
>> <hot at openstreetmap.org<mailto:hot at openstreetmap.org>> wrote:
>>
>> The link below points to a diary where a contributor reporte
>> bad edits in Modagisciu.
>> https://www.openstreetmap.org/user/UNGSC-DTLM-Ale_Zena/diary/398388
>>
>> Osmose Summary of errors shows that it is possible to monitor
>> a Collaborative project and assure we dont leave tons of
>> errors behind. The TM project 11385 for this area has just
>> been completed this week. Osmose errors summary for TM 11385
>> Area
>> <http://osmose.openstreetmap.fr/en/issues/open?item=xxxx&source=&class=&username=&bbox=45.279918,2.000613,45.330809,2.037097>
>> showed on 2022-01-08 :
>>
>> * 564 Highway intersecting buildings
>> * 231 Building intersections
>> * ?93 Large building intersections
>> * ?10 Long Highway underground and no tunnel
>> * and yes a lot of very imprecise buildings tracings could
>> be added to the list.
>>
>> Some errors might have been corrected since Osmose queried
>> the database 3 days ago, but
>> Highway https://www.openstreetmap.org/way/124549760 deserves
>> a mention since we see 10 highways reported as underground.
>> The higway did already exist in the database before this
>> project.? A contributor traced buildings over it, then
>> revised the highway tags, adding layer=-2.
>>
>> See this Overpass query that loads both the highway and the
>> overlapped buildings.
>>
>> http://overpass-turbo.eu/?Q=(way(id%3A124549760)%3B%0Away(around%3A0)%5B%22building%22~%22.%22%5D%3B%0Anode(w)%3B)%3B%0Aout%20meta%3B%3E%3B%20out%20meta%3B%0A&C=2.03235;45.32651;20
>> <http://overpass-turbo.eu/?Q=(way(id%3A124549760)%3B%0Away(around%3A0)%5B%22building%22~%22.%22%5D%3B%0Anode(w)%3B)%3B%0Aout%20meta%3B%3E%3B%20out%20meta%3B%0A&C=2.03235;45.32651;20>
>>
>>
>> Pierre
>>
>>
>> Le mardi 11 janvier 2022, 10 h 53 min 46 s UTC?5, Martien van
>> de Griendt <m.griendt at gmail.com<mailto:m.griendt at gmail.com>> a ?crit :
>>
>>
>> Fyi, duplicate mapped buildings can also be caused by the
>> fact that ID Editor hides buildings when zoomed out.
>>
>> So mappers start to map them unaware of the fact that they
>> are already present.
>>
>> Regards, Martien
>>
>> On Tue, Jan 11, 2022 at 12:56 PM Jiri Vlasak
>> <jiri.hubacek at gmail.com<mailto:jiri.hubacek at gmail.com>> wrote:
>>
>> A bit of clarification about the script:
>>
>> > I note there are over 3,000 duplicated or overlapping
>> HOT projects.
>>
>> The script didn't care about if a project is archived.
>> Every ID between
>> 8000 and 12000 returning some geometry was considered. I
>> improved it
>> [1], to avoid project with "ARCHIVED" status. Here are
>> the results [2].
>>
>> > The AOI-BBOX is a rough approximation of the zone
>> covered by the
>> > Project
>>
>> The script uses `areaOfInterest`, not `aoiBBOX`. Polygon
>> or Multipolygon
>> is used, therefore the overlap detection is pretty precise.
>>
>> > The only problem I see is two overlapping active
>> project at exactly
>> > the same time.
>>
>> It's unfortunate HOT tech team is not able to deliver such a
>> functionality.
>>
>> Would it be useful to provide the similar stats on daily
>> basis? If so,
>> are there any information about the projects that should
>> be included?
>>
>> [1]:
>> https://paste.sr.ht/~qeef/4100c0f7547a241f7a8b8b48c980da78e4c6d402
>> [2]:
>> https://paste.sr.ht/~qeef/c2fe540f2f41a512f3f89d5960d9beeda21417b7
>>
>> _______________________________________________
>> HOT mailing list
>> HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
>> https://lists.openstreetmap.org/listinfo/hot
>>
>> _______________________________________________
>> HOT mailing list
>> HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
>> https://lists.openstreetmap.org/listinfo/hot
>> _______________________________________________
>> HOT mailing list
>> HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
>> https://lists.openstreetmap.org/listinfo/hot
>>
>>
>> _______________________________________________
>> HOT mailing list
>> HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
>> https://lists.openstreetmap.org/listinfo/hot
>
> --
> Frans Schutz
>
>
> _______________________________________________
> HOT mailing list
> HOT at openstreetmap.org<mailto: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/20220220/e53c6d60/attachment-0001.htm>
------------------------------
Subject: Digest Footer
_______________________________________________
HOT mailing list
HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/hot
------------------------------
End of HOT Digest, Vol 144, Issue 7
***********************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20220301/edfccab0/attachment.htm>
More information about the HOT
mailing list