Does the script provide reasonable feedback to the user about how far through the process it is, and in the event of an error does it provide reasonable feedback to the user about what has gone wrong and how to fix it?

The reason that I’m asking is that I (or someone) will need to update Manually building a tile server (Ubuntu 22.04 LTS) et al and it’d be good to know whether the same caveats are needs as are around get-external-data.py, which offers no user feedback and does not “fail safe”, hence the warnings such as “not much will appear on the screen…” on the switch2osm pages.

“get-external-data.py” caught another user out yesterday (you may have seen the IRC discussion).

I’d have expected that the “flatpak” option above is the easiest route, but If it’s an older one that doesn’t support Crostini then “Crouton” should be an option: https://github.com/dnschneid/crouton . That project is now maintenance-only since Crostini has essentially superseded it, but it’s still a useful option for older, cheap but still capable devices.

If you want something self-contained on a Chromebook, I don’t think it has anything like the resources to run even just a rendering website. You can run something like JOSM on it (via Linux, via Crostini / Crouton). This stackoverflow question suggests that QGIS might be an option?

Can you explain in a bit more detail why this is a problem?

I’m a fairly regular user of taginfo (I’ve even updated project files so tag consumption information will appear there) but I’ve never been bothered by what is displayed in the description field on e.g https://taginfo.openstreetmap.org/keys/naptan%3ANotes#wiki - you’re going to need to click on the link anyway to see the page.

I’m mentioning this because someone’s suggested that this diary entry be included in the next WeeklyOSM issue, and the editors there (including me) are scratching their heads about what to say about it.

Below we have an unpaved footpath joining onto a paved footpath.

For completeness, the paved one is here and the unpaved one is here. There’s actually no difference in the style between paved or unpaved footpaths and paths - if one of them had a lower trail_visibility, or had informal=yes then they would appear like the “informal” example above.

Roads are slightly different - unpaved ones such as this do appear as tracks that everyone can access.

This is just a quick update to address all the CCI who believe it is essential to have comprehensive and descriptive comments.

You were asked to use descriptive comments on future changes at osm.org/changeset/120723949 because your earlier lack of descriptive comments caused a lot of confusion.

It’s not some personal vendetta against you - OpenStreetMap is something that we create together, and we all work together better if there’s effective communication and we can all understand what each other is doing.

Anything that is a barrier to that means that together we’re less effective than we could be (and I’d argue that comments like “Anyone possessing even a minimal IQ…” don’t help, either).

Best Regards,

Andy

@matheusgomesms

So what’s the point to having the repository open anyway? It is time to OSMF run things more professionally, having them more controlled for core stuff, just like the main render on OSM. I don’t know, hire some people exclusive for that, get more money to fund that, whatever it takes.

I can pretty much guarantee that there will be at least one candidate standing for the board at the next elections with that as a platform. If you’re not an OSMF member already I’d definitely join so that you can influence the direction the foundation takes.

landuse=education or highway=busway, proposals that were discussed heavily for YEARS through the community, and then finally got accepted, but never made to Carto because God only knows

Why do you think that a “professional OSMF standard style” would accept all changes that have been “discussed heavily for years … and then finally got accepted”? Is it not possible that these professionals might say “actually, we have a perfectly good way of tagging that already, so we don’t want to support another way of tagging the same thing”?

To take a specific example, with a “person who looks after a map style” hat on I looked at the usage of highway=busway, and just treating it as an alias of highway=service made the most sense. I’m sure that that wasn’t what the proponents of the tag had in mind, but it does make most sense in the area I render.

There can’t be any one map style that does everything - it would be impossible to understand. I try to be extremely inclusive of what to render (sort https://taginfo.openstreetmap.org/projects by keys and the style I look after is second only to MapComplete) but some of those are “really the same as [some other tag]” or “only rendered from zoom 21” - you can’t assume that any new addition (parcel lockers, busways, whatever) will be treated by renderers as a “new and different thing”, because it probably isn’t.

Responding to a couple of Mateusz’ points:

Basically, anyone who has map style and is willing to obtain/donate rendering can have their own map style and osm.org website.

That’s simply not true - it’s flatly contradicted by the “Global scope and coverage” section of osm.wiki/Featured_tile_layers/Guidelines_for_new_tile_layers posted above. As an example, I maintain a map style that is focused on features present in England and Wales, but is of less and less interest the further you go from there. It would make no sense to suggest that people in Iran use a map style that shows different sorts of “Enlgand and Wales public rights of way” and 300 different combinations of pub features. If I tried to suggest it as a global style on osm.org people would (rightly) object on the grounds that it is not interesting to most people around the world.

It used to be true that “anyone could have their own map style” - before the https enforcers had their way, it was trivial to swap an OSM tile layer for a different one using browser plugins (and the switch2osm guides did actually suggest that).

and is it cost of 50 euro/month or 15 000 euro/month for non-default layer.

That is a good question - elsewhere today a starting price of €150 per month was suggested for a global server. That sounds plausible to me, but clearly it does put this sort of thing out of the reach of many or even most people. For comparison, a map server serving GB and IE only costs me about €20 per month.

There’s another reason why “another global map style” isn’t a good idea - global map styles are just boring, lowest-common-denominator affairs, the McDonalds or Burger Kings of the map world. OpenStreetMap should be “everyone’s local map”, and that should also mean “everyone’s local map style” too. The things that are “interesting” to show on a map in one place are irrelevent elsewhere; for example highway shields are important in the US but almost no-one else understands the problem.

There’s lots wrong with OSM Carto as a map style and as a project, but I think that the far bigger problem is the lack of flexibility of osm.org itself to be able to support other styles.

Hello melo_mo,

The current mapping in this area is the result of a community discussion, which resulted in https://forum.openstreetmap.org/viewtopic.php?pid=606038#p606038 back in 2016..

If you think that the current mapping is incorrect or the situation have changed, then please do discuss with the rest of the OSM community - at the moment, the main talk list https://lists.openstreetmap.org/listinfo/talk is probably the best place to do that. Although a few people have have commented on this diary entry, not everyone will see it.

Best Regards,

Andy Townsend, on behalf of OSM’s Data Working Group.

DWG shouldn’t stand by and do nothing with this user’s vandalism! He deserve a ban!

Andy from the DWG here. We’ve had exactly 2 reports about this user over the years, once in 2019 (about something completely different) and one 2 days ago, which didn’t link to any actual data but just said “This user left many strange lines on the map”. It was only by seeing someone reply here immediately after I’d dealt with the other ticket that I was able to make the connection. Earlier I’d already asked the user to reply about the odd circles: osm.org/user_blocks/5984 .

If you want the DWG to deal with something please read “I’ve seen a problem; what should I do?” at https://wiki.osmfoundation.org/wiki/Data_Working_Group .

(likely inaccurate and possibly hilarious machine translation follows)

(随后可能是不准确且可能很搞笑的机器翻译）

来自 DWG 的 Andy。 这些年来，我们已经收到了 2 份关于该用户的报告，一次是在 2019 年（关于完全不同的事情），一次是 2 天前，该报告没有链接到任何实际数据，只是说“该用户在 地图”。 只有在我处理完另一张票后立即看到有人在这里回复，我才能建立联系。 早些时候我已经要求用户回复奇怪的圈子：osm.org/user_blocks/5984。

如果您希望 DWG 处理某些事情，请阅读“我发现了一个问题；我应该怎么做？” 在 https://wiki.osmfoundation.org/wiki/Data_Working_Group 。

Well it did have some content previously, which was possibly removed following my comment above…

Thanks - as long as that happens before you start editing, that’s great.

Hello,

If you’re a member of an organised editing team can you please read and follow https://osmfoundation.org/wiki/Organised_Editing_Guidelines and make sure that your activities are listed at osm.wiki/Organised_Editing/Activities .

Best Regards,

Andy

Hello PopeyePopcord,

I’ve asked on talk-ca about the suitability for the licences that you are using for use in OSM - see the thread here. I’d suggest that you dicuss it in more detail with the Canadian community.

Thanks for the information provided so far on e.g. osm.org/changeset/119761959 about the suitability of that source.

To answer the question on osm.org/changeset/119755674 (“how is using open data not allowed”), the problem is that not all “open” data licences are compatible with each other (or with OSM). There have been examples of government datasets (e.g. in the UK) that were initially believed to be licence-compatible with OSM but later found to contain IP from a non-open source. It may be that other people have done much of the work of licence checking for you - but we need to be sure that the data that you are using really is available under a licence compatible with OSM.

Best Regards,

Andy

I can hide it if you want (I’m a moderator)

Hello,

I would strongly recommend that you do not put too much effort into this project as your responses at https://help.openstreetmap.org/questions/83528/can-different-named-areas-of-sea-overlap suggest that you’re ignoring the way that most other mappers in OSM map things.

If that’s the case, and your changes do not reflect realiy, it’s likely that we (OSM’s Data Working Group) will need to revert some of these changes.

In the real world it’s simply not true that every one piece of sea has “only one name”, just like it’s not true that any one piece of ground has. For example, I live in a city with a name, but that is also in England, and also in the United Kingdom. It can get more complicated of course - see e.g. https://web.archive.org/web/20190225133138/http://ma3t.co.uk/euanmills/euanmills/tifd.html and http://mappinglondon.co.uk/2013/londons-localities/ .

Best Regards,

Andy Townsend, on behalf of OSM’s Data Working Group

Beyond the technical part described here, I’d also try and understand the history - Who added it and why? What were they actually trying to do? Do they need help? In many cases of course it’ll be pretty obvious that things like this simply no longer add value, and have long been remapped as something else.

Have you tried communicating with the people who are organising the HOT projects or with the mappers themselves?

Indeed, I had attempted to reply in a humorous manner that the meaning of this diary entry really wasn’t very clear. I’m sorry that didn’t work, and I’ve messaged bo_hot privately to apologise about that and suggest what might have helped better communication in the first place.

I understand most of the words; it’s just that when they’re joined together they don’t make sense (for example - what exactly is a “future community product owner”?)

I can guess some, since I’ve dealt with salespeople fluent in corporate BS for many years. I presume that:

“… will be sitting down to identify and triage the most pressing issues expressed by the Tasking Manager community”

means “… will fix some bugs” and

“If you have issues, features or enhancements that are either new, or outstanding, I highly encourage you to start getting them shaed and consolidated in the TM github repo”

means “Please log any bugs you know about”.

However, many of the people that I’d hope you’d want to read whatever you’ve written may not be, especially if they haven’t previously been exposed to this particular American flavour of corporate-speak.

As an aside, documents hosted on Google and Slack channels may well not be accessible to everyone. OSMF has a committent to open communications channels; shouldn’t Humanitarian OpenStreetMap Team have something similar? Don’t you want what you write to be visible to everyone?

Andy

(for the avoidance of any doubt, writing in an entirely personal capacity)