include changesets_count with current edit count in changeset tags, like iD

[mkoniecz]

I noticed that iD started since ​https://github.com/openstreetmap/iD/commit/25d8a8aa08f9353e01c1d1e893a9cd1025ebd806 adding changeset tag with information how many edits user made before making this one.

It helps to distinguish clueless initial edits from complex things that may be intentional on reviewing suspicious changesets.

It is also good reminder that somebody may be just starting and to avoid complaining too much.

It provides information unaccessible without unreasonable detective work.

I propose that JOSM would also start doing this ( see ​https://www.openstreetmap.org/changeset/52169327 for example of edit made with iD ).

URL:http://josm.openstreetmap.de/svn/trunk
Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b
Last:Changed Date: 2017-12-31 03:09:43 +0100 (Sun, 31 Dec 2017)
Build-Date:2017-12-31 02:33:46
Revision:13265
Relative:URL: ^/trunk

Identification: JOSM/1.5 (13265 en) Linux Ubuntu 16.04.3 LTS
Memory Usage: 408 MB / 871 MB (123 MB allocated, but free)
Java version: 1.8.0_151-b12, Oracle Corporation, Java HotSpot(TM) 64-Bit Server VM
Screen: :0.0 1920x1080
Maximum Screen Size: 1920x1080
Dataset consistency test: No problems found

Plugins:
+ OpeningHoursEditor (33876)
+ buildings_tools (33735)
+ continuosDownload (68)
+ imagery_offset_db (33774)
+ reverter (33865)
+ todo (30303)

Last errors/warnings:
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet
- W: java.io.IOException: Attribution is not loaded yet

[Don-vip]

I'm not sure about this, it sounds like user tracking and I'm not sure JOSM users would be happy about that.

It could be useful for mapping parties organized by people planning to use JOSM but it should probably be disabled by default as most of our users are not beginners.

[mkoniecz]

it sounds like user tracking

Usually I am on side of exterminating tracking (I use Duck Duck Go instead of Google, taped camera in laptop etc etc) but it does not feel like tracking at all. Users are already tracked by their username, it is not a private information, it is not exploitable for anything nefarious, anybody wishing to do something evil may anyway easily obtain this information.

most of our users are not beginners

Still, some are - and it is useful to tell the difference easily. Recently I made mistake by assuming that somebody is expert due to using JOSM and I send message that is probably confusing for somebody who made about 50 edits.

There is also significant difference between somebody who made 500 edits, 5 000 and 50 000 and I can easily imagine situation where it would change my approach.

[stoecker]

I also don't like that approach.

An alternative suggestion would be to mark a changeset as beginner:

• when less than 100 changesets
• first active not before 6 months ago

This way starters would be marked without tracking information.

[Klumbumbus]

I don't think that tracking is such a big problem with that single changeset count number as other tools reveal much more information about your tagging behavior. e.g. ​hdyc or ​osm-suspicious. Or even worse uploaded gps tracks of the ​type public or identifiable (maybe starting or ending at your house...)

[stoecker]

A little bit there a litle bit there...
That other places reveal same information is no reason to violate privacy unecessary.

[bastiK]

The current number of edits is already publicly visible on the user page ​https://www.openstreetmap.org/user/<username>.

I don't see what the big deal is with privacy in this case. It is fairly useful information for fellow mappers. And already now, it is can be easily obtained by anyone with technical know-how or with enough patience to click through the changeset list.

[anonymous]

can be easily obtained by anyone

And at the same time it is easy enough that saving it in the changeset is not making situation worse and tedious enough that during typical looking at changeset it is not useful to do that.

[pnorman]

I'm against this. Not because of privacy information, but because it duplicates information easily accessible through other means.

If it's a user looking into what someone else has done, the information is one click away on the user's page. If it's an analysis tool like osmcha, they have a database of changesets, and it's a trivial operation.

If the OSM community decides they want to add to the information displayed on a changeset page by adding the number of changesets, it should do so with a website change, not editors downloading information from the user API and re-uploading it as a changeset tag.

There are no privacy implications, because the information is already available.

[Klumbumbus]

Replying to pnorman:

If it's a user looking into what someone else has done, the information is one click away on the user's page.

Not exactly as the user page shows the current edit count not the edit count when the user created the changeset of interest.

If the OSM community decides they want to add to the information displayed on a changeset page by adding the number of changesets, it should do so with a website change, not editors downloading information from the user API and re-uploading it as a changeset tag.

Sounds reasonable.

[Don-vip]

Too many users object this change or are not at ease with it.