@1ec5 reported on Slack that Framadate doesn’t distinguish timezones, which is “pretty much a dealbreaker for anything in OSM that I would use it for”.

See https://framagit.org/framasoft/framadate/framadate/-/issues/158

Regarding the code collaboration platform: almost all projects in the https://github.com/openstreetmap organisation depend on some sort of CI/CD pipeline (by the way, are those the projects which are in focus here?).

I’m not 100% sure what the options are available when using Gitea, but in any case I would consider CI/CD as an absolute must have feature for any kind of new platform to switch to.

OSMF based code hosting isn’t an entirely new idea: we used to have svn.openstreetmap.org (+ trac) for quite some time before some projects started moving to Github.

OSM’s OAuth will support OAuth2 anyway in the future, which should rule out many of today’s issues with OAuth 1.0a.

For reference: https://github.com/openstreetmap/openstreetmap-website/issues/1408

Thanks for the update! I have to say it was somewhat unexpected to see this in an official newsletter ;)

That’s a pretty weird URL for what you call a “free book”. As it stands, you’re most likely linking to some pirated copy here.

Apologies, i didn’t mean to throw a random piece of overpass gibberish over the fence… :)

https://dev.overpass-api.de/blog/loop_and_group.html#for has some more details about how the “for” statement works. In the example above this boils down to “for” calculating a number of groups based on the “name” tag, and then iterate over those groups.

For a sorted, duplicate free (aka “DISTINCT”) list of national park names for the UK you need to use something like https://overpass-turbo.eu/s/11In

With today’s (19 November 2020) build, the Windows Task Manager “not responding” status issue seems to be gone. Also the linux wine environment crashes less frequently. I hope this is expected behaviour and not some coincidence?

I’m seeing some infrequent issues when P3 fails to show parts of the map. I couldn’t figure out, if the loading process itself isn’t working properly at times, or some parts of the map are simply not drawn. In any case, I don’t see any error message.

By the way, could you include some more detailed build information in the “Help” menu, maybe? That would be really helpful when reporting issues.

I tested both versions in the same small townish area, that would still fit well in the 50k node limit for an export on z16. Although working on z16 feels a bit sluggish, it’s still feasible on P2. I experienced issues in particular when panning on z17+z18. Maybe there’s something special in my set up, that impacts P3. Let’s see what other users report back.

I also tested Potlatch 3 on Windows 10 now. While the Flash based Potlatch 2 works quite smoothly on that machine, I’m seeing frequent lockups w/ Potlatch 3, in particular when panning the map. Windows Task Manager reports the app as “not responding”, and an impatient users gets prompted to kill the process. It usually takes about 5-30 seconds until it all clears all up again. Maybe there’s some processing going on that blocks the main thread, and Windows thinks the app is just unresponsive and dead.

Oh well, this seems to be somewhat unstable and tends to crash when panning the map.

those should be two separate commands after all:

cp Potlatch.air ~/.wine/drive_c/users/USER
wine ‘C:\Program Files (x86)\Adobe\Flash Player\AddIns\airappinstaller\airappinstaller.exe’ C:\users\USER\Potlatch.air

At a first glance, it seems to be working using wine:

https://get2.adobe.com/air/ –> download windows version

wine AdobeAIRInstaller.exe

download & unzip potlatch

cp Potlatch.air ~/.wine/drive_c/users/USER wine ‘C:\Program Files (x86)\Adobe\Flash Player\AddIns\airappinstaller\airappinstaller.exe’ C:\users\USER\Potlatch.air

wine ‘C:\Program Files (x86)\Potlatch\Potlatch.exe’

(replace USER by your login user name in Linux)

Documenting things that don’t work is a good way to save others lots of time (and frustration). In general, people trying this out should be aware that installing some really old software could potentially screw up their system (end of mandatory disclaimer).

Tried a Docker image on Ubuntu 20.04 b/c I didn’t want to install some ancient libs on my box:

Dockerfile:

FROM nucleardreamer/air-docker
COPY Potlatch.air .
COPY Potlatch_CPU.air .

docker build . -t my_image

sudo docker run -ti –rm -e DISPLAY=$DISPLAY –net=host -v /tmp/.X11-unix:/tmp/.X11-unix my_image /bin/bash

Unfortunately, installing the app using the installer crashed with a segfault. I gave up at this point.

root@ubuntu:/air# air-install -silent -eulaAccepted  `pwd`/Potlatch_CPU.air 
The program 'air-install' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadValue (integer parameter out of range for operation)'.
  (Details: serial 196 error_code 2 request_code 130 minor_code 3)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the --sync command line
   option to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.)
Application crashed with an unhandled SIGSEGV
Crashlog has been dumped in /tmp/airCrashLogs/1118_1641_67pN6X

-> https://github.com/simon04/whodidit/issues/49

Die Karte zeigt auch ein Bauunternehmen als Baustelle: osm.org/node/2134184963

Die Overpass query scheint hier auf den ersten Blick noch etwas ausbaufähig zu sein.

https://lists.openstreetmap.org/pipermail/osmf-talk/2020-July/006973.html mentioned “We’re looking for a site reliability engineer who can help the OpenStreetMap Foundation’s sysadmin team manage the growth in the demand for its services.[…]”. Typically you would expect an open selection process following such an announcemnt. Having a preferred candiate in kind kind of defeats the whole idea of consulting the membership on any details about this position.

BTW: No, thank you, I’m not a sysadmin kind of person :)

to hire a sysadmin/Site Reliability Engineer. We have someone in mind, but for the benefit of the candidate, we cannot name names yet.

I was somehow expecting a public announcement (not just some random tweet like https://twitter.com/OSM_Tech/status/1287395222847139846), followed by a more formal candidate selection process that includes interviews, etc. Maybe I missed this, or this was all done behind the scenes.

The positive numbers would still fit perfectly well into uint64, so that wouldn’t be much of an issue here. osm2pgsql uses an internal node cache that derives some block numbers from object ids. If those ids happen to be too large as in this case, block numbers exceed internal limits and eventually the whole processing aborts. Renumbering makes sure there’s no conflicting ids while at the same time keeping object ids in a reasonable value range.