I’ve gone with CSV, quoting the string field. On the technical site, it’s using Python’s csv.writer with unix_dialect and QUOTE_NONNUMERIC.

sqlite would be overkill for what is 15kb of data in text form in a single table.

If I were truly reporting website domains, I wouldn’t worry about quoting, since domain names have a restricted set of characters. Instead, what I’m reporting is extracted domains from the referer header, and there’s nothing that stops an app from sending https://www.foo,bar.com/ as a referer header. They’d eventually get blocked for sending a fake header, but it doesn’t rule out them appearing in the logs.

That is related to the origin of “” in the hosts log. They are from referers that don’t parse as a valid URI.

In practice, for the type of log processing I’m doing, I’m likely to ignore that. If I run a command like fgrep '"umap.openstreetmap.fr"' hosts* | sed 's/hosts-//' | sed 's/.csv:"umap.openstreetmap.fr"//' > umap.csv I get a CSV I can open in a spreadsheet and do further stuff with.

The first file is up at https://planet.openstreetmap.org/tile_logs/hosts-2022-06-26.csv. I’m preparing the files to backfill into 2021

Actually, I think the best approach to properly get to current (non legislative but) de-facto speed limits is to not look at OSM data but have a fleet of cars constantly on the roads and sourcing the data for the current traffic situation for that. E.g. like Google does it (with data sourced by users of Google Maps). But whoever else may be collecting this kind of data (big taxi companies?, navigation system manufacturers?) , it is proprietary data and there is no good reason for them to give that away for free other than maybe to combine forces against a common market leader (Google).

You need this information for good routing anyways, which is the main use-case for speed limit data. Typically you want to develop routing cost functions based on actual speed, not max speed.

No, the tile logs are from the CDN and are an accurate count of successful requests for tiles. Prior to 2021-04-13 the logs were only from the second layer of the old CDN.

The logs include successful requests on tiles where there were at least 10 requests, and the requests came from at least 3 distinct IPs. Most of them represent real views, but there are some artifacts.

Hi Alessandro, Immaculate Mwanja has recently posted about the project in the hot mailing list: https://lists.openstreetmap.org/pipermail/hot/2022-February/015767.html

Unfortunately I tried contacting them back in January without reply, asking for information required by the OEG.

Immaculate Mwanja, do you have a link to the documentation on the wiki the OEG call for? In particular, I’m interested in the “plans for a ‘post-event clean up’ to validate edits”

With the large number of people blocking you, have you considered modifying your behavior so that they no longer feel the need to do so?

OpenStreetMap Carto has its own schema which is currently fairly similar to the osm2pgsql defaults. These are completely different than the API DB schema.

How will OSMF ensure progress is made on infrastructure things that are stuck? See examples below. - Rob (OSMUK) Translations on diary posts (this was an action from last year’s OSMF Board & LC meeting). How to move the issue on? Mailing lists migration to Discourse (GH issue)

There seems to be a misunderstanding about Discourse. We’re looking at testing it out, and the first step to that is adding OAuth 2.0 support to the website. This work is in progress, not stuck.

For the translations on diary posts, I don’t know the status because it is not an infrastructure thing.

I wouldn’t assume it’s due to the lack of a Hungarian translation. Japan, Indonesia, and Russia all had similar response rates as Hungary and Japanese, Indonesian, and Russian were all provided languages.

I have access to some other by-country usage stats and want to see the relationship between them and survey responses, but that analysis will take more time and I won’t get to it this week.

Not quite - it’s a linear regression forced through the origin, weighted by number of responses.

What is your monitor, and have you zoomed your browser in or out? High DPI monitors and zooming can both cause blurriness as the image tiles need to be scaled up or down.

I did a quick check on the short-circuiting with buildings which are more likely to be low node-count polygons. point2 took 715s and point3 took 565s. This comes at the cost of point3 potentially returning points outside the polygon for very weirdly shaped polygons with four corners, e.g. those in a v shape.

Large parts of the Canadian map have been imported from CanVec, a vector map provided by the Canadian Natural Resources organization. In the area of Beresford, the import happened back in 2012. That’s already eight years ago. And the quality of the data wasn’t particularly great. Roads are inaccurate, forests cover half the urban areas, bridges aren’t actually placed over the water, you name it.

What makes it worse is in some parts of the country, the water and forest data in CanVec were 30 years old already.

Different imagery sources are better in different places. All imagery sources can have offsets, so it’s not unique to Bing.

What’s needed is someone systematically evaluating the quality of Bing, Esri, and Maxar where mapping is done and deciding which is best overall.

You’ve indicated a connection between your OSMF candidacy and your paid work. Do you intend to ask the board permission to do OSMF work on paid time as a conflict of interest under appropriate legislation. I have always taken time off for board duties (e.g. take vacation time for in-person meetings, rearrange my work schedule for monthly meetings).

It’d be good to see this work on osm.org via submitting bubble wrap to be a featured tile layer. Are you interested in submitting it?

It’s good to see the improvements. The calendar on the wiki is underused by some parts of the community and this should help bring up usage. With the other tools built around its data like WeeklyOSM’s, adding your event to it gets a lot of coverage.

It might help for potential volunteers for the MWG to know the following: -average time to on board for membership waiver review or other MWG tasks -approximate time that would be expected 1 hour a week etc. - types of skills needed

It’s hard for me to judge that, since I’ve been in the MWG for some time and set up some of the new tools. Being experienced makes it difficult to know how much time it will take for someone else. Perhaps you, as a newer MWG member, could provide an estimate based on your experience learning the new stuff.

I believe everyone in the MWG wants to be transparent about waivers, reporting overall numbers at a minimum. There are limits to this, of course - country breakdowns are tricky. It’s not as much of an issue on money transfer waivers, because anyone who requests the associate member list can get the list of members from Iran, and the inability to transfer money from there makes it obvious they’re all fee waiver members, but the financial need members could come from anywhere.

Right now we’re moving to OTRS to track inbound requests, which should make reporting easier. I had to go in and count emails to get these numbers, which isn’t sustainable.

CiviCRM has country and OSM username information, so it’s technically easy to get a breakdown of fee waiver members’ countries. We need to add fee-waiver to https://wiki.osmfoundation.org/wiki/Membership/Statistics but it hit the first of the month today so we haven’t yet done it.

The most surreal and memorable OSMF board meeting yet

Ah, these youngsters.

+1