Changeset: 27888534
revert un-discussed mass edits by Test360
Closed by woodpeck_repair
Tags
bot | yes |
---|
Discussion
-
Comment from Test360
Setting "landuse=wood" ( http://wiki.openstreetmap.org/wiki/Tag:landuse%3Dwood_%28Don%27t_use%29 ) instead of "landuse=forest" seems a mistake according to the OSM wiki.
-
Comment from woodpeck
I agree. This does not, however, mean that you can simply chnage all landuse=wood to landuse=forst (or natural=wood) automatically. Your changesets have not been reverted because the content was wrong, but because the process was wrong.
-
Comment from Test360
I would like to know why the 27640972 changeset ( http://www.openstreetmap.org/changeset/27640972 ) which is clearly not a mass edit as been reverted by this changeset.
-
Comment from Test360
Same request for the 27641137 changeset ( http://www.openstreetmap.org/changeset/27641137 ), which was a local edit that fixed an obvious mistake.
-
Comment from Test360
Same request for the 27662643 changeset ( http://www.openstreetmap.org/changeset/27662643 ) which can't be smaller because it change the way the river is tagged.
-
Comment from Test360
Same request for the 27662544 changeset ( http://www.openstreetmap.org/changeset/27662544 ) which can't be smaller because it deals with a river.
-
Comment from Test360
Same request for the 27662224 changesert ( http://www.openstreetmap.org/changeset/27662224 ) which can't be smaller because it change the way the river is tagged.
-
Comment from Test360
Same request for the 27661940 changesert ( http://www.openstreetmap.org/changeset/27661940 ) which can't be smaller because it change the way the river is tagged.
-
Comment from Test360
I would like to know why the 27630561 changeset ( http://www.openstreetmap.org/changeset/27661940 ) has been reverted. What is the problem? The changes in this changeset were not done automatically. And nobody complain in the discussion of the changeset.
-
Comment from Test360
I would like to know why the 27631195 changeset ( http://www.openstreetmap.org/changeset/27631195 ) has been reverted. What is the problem? The changes in this changeset were not done automatically. And nobody complain in the discussion of the changeset.
-
Comment from Test360
Same request for the 27631546 changeset ( http://www.openstreetmap.org/changeset/27631546 ).
-
Comment from Test360
Same request for the 27631744 changeset ( http://www.openstreetmap.org/changeset/27631744 ). Again: not an automatic edit, no complaint.
-
Comment from Test360
Same request for the 27631975 changeset ( http://www.openstreetmap.org/changeset/27631975 ). Not an automatic edit, no complaint.
-
Comment from Test360
Same request for the 27632063 changeset ( http://www.openstreetmap.org/changeset/27632063 ). Not an automatic edit, no complaint.
-
Comment from RM87
http://www.openstreetmap.org/changeset/27631744 - highway=services is ment to describe areas and makes no sense on ways. It would be ok to assume that this tag implies area=yes automatically. You should have made this change in the wiki not in the data. In case of wiki change a drop of note to tagging list would be also.
-
Comment from RM87
http://www.openstreetmap.org/changeset/27631195 , http://www.openstreetmap.org/changeset/27632063 , http://www.openstreetmap.org/changeset/27632063 , http://www.openstreetmap.org/changeset/27631975 and http://www.openstreetmap.org/changeset/27632063 same as http://www.openstreetmap.org/changeset/27631744
-
Comment from RM87
http://www.openstreetmap.org/changeset/27661940 , http://www.openstreetmap.org/changeset/27662224 , http://www.openstreetmap.org/changeset/27662544 and http://www.openstreetmap.org/changeset/27662643
You should not make mega relations. These are hard to edit and even harder to fix as they tend to broke extremly easily.
You should instead make a waterway relation of relations and ways to group the separate pieces of river parts together. See: http://wiki.openstreetmap.org/wiki/Relation:waterway -
Comment from RM87
http://www.openstreetmap.org/changeset/27641137 seems ok. You may redo it. This time with a little bit more descriptive comment. It is hard to notice that a space was removed from the word shared _lane . If a good changeset is in the middle of not so good changesets it might be not noticed.
-
Comment from RM87
http://www.openstreetmap.org/changeset/27640972 5 of 6 changes were wrong (should have been surface=asphalt) and the sixth might have been wrong (felt like compacted on some parts). 5 are fixed with changeset http://www.openstreetmap.org/changeset/28018519
6.th one I did not touch.
Centerlines are clearly seeable on bing and the road is fairly long so the surface must be asphalt on those 5 roads. -
Comment from Test360
About "highway=services": the OSM wiki clearly state: "Make sure you add the area=yes tag in case you tag a way, because it is not considered the default for closed ways."
-
Comment from Test360
The size of the relation is irrelevant. The changsets on river has been reverted because of « un-discussed mass edits », wich they are not. Where is the discussion about the size of the relation of these changesets before this revert?
-
Comment from Test360
About http://www.openstreetmap.org/changeset/27641137 , you just state that it seems ok, but you don't explain why it has been reverted. I would like to know.
"good changeset is in the middle of not so good changesets it might be not noticed": are you implying that all these changesets have been reverted without understanding them?
-
Comment from Test360
About http://www.openstreetmap.org/changeset/27640972 , you discuss the correctness of the changeset, but you miss the point: why a non-mass edit changeset have been reverted in this changeset?
The correctness should have been discussed **before** the revert.
-
Comment from SomeoneElse
@Test360 - although it's perhaps easy to get lost in a conversation entirely conducted in changeset discussion comments, perhaps it would help to read from the top again, including the comment that says "Your changesets have not been reverted because the content was wrong, but because the process was wrong".
-
Comment from Test360
@SomeoneElse
Before my first comment, I didn't understand that this changeset reverts 80 of my changesets. I thougth that it was the revert of only one changeset. That's why I discuss the correctness. Then I understand."Your changesets have not been reverted because the content was wrong, but because the process was wrong": that exactly why I asked my questions. I don't understand what is wrong with the process of several reverted changesets. I would like to understand to avoid that in the future.
-
Comment from Test360
@RM87
"You should have made this change in the wiki not in the data": I don't understand at all this sentence. Can you explain in details? Thanks. -
Comment from RM87
About "highway=services" do you really believe everything that is written or not written in the wiki?
I do agree that the changeset comment « un-discussed mass edits » does not fit for those river changeset reverts. Nevertheless there is a waterway relation model for the thing you did with a multipolygon relation.
I do not know why http://www.openstreetmap.org/changeset/27641137 was reverted, because I'm not the person who reverted it. Frederik probably can explain why he did add this changeset for reverting.
-
Comment from SomeoneElse
@Test360, re the process: There's a mechanical edit policy described in the wiki here:
https://wiki.openstreetmap.org/wiki/Mechanical_Edit_Policy
The very first requirement is "Discuss". It's there for a number of reasons. OpenStreetMap is a shared database, but it's actually more than that - it's the results of people actually getting out of the house walking around for a while and writing down what they see about the physical world. That takes time and effort, and for the results of that time and effort to be lost in some tag homogenisation would be very sad indeed, so one of the reasons why it's important to discuss mechanical edits before performing them is simply just human politeness. Another reason (already mentioned elsewhere) is that sometimes new mappers use the wrong tag for things, but they obviously don't know that they are doing so. It's important to discuss with them how something may be tagged better in the future. A third reason why it's important to discuss mechanical edits before performing them is that sometimes the basis on which the mechanical edit is to be performed is incorrect (the "all deciduous trees and broadleaved" assumption was an example of this sort of incorrect assumption). You simply cannot believe everything you read in the wiki. A fourth reason why it's important to discuss beforehand is that, even if the change is a good idea, it's important that people who use the data know that it is going to change. Someone might be rendering "wood=deciduous" on a map; they need to be told to change the way that they render trees before the change is made to the database.
Please, sit down and read https://wiki.openstreetmap.org/wiki/Mechanical_Edit_Policy and follow what it says.
-
Comment from RM87
About the wiki changing: the wiki page of highway=services should have say'd that this tag automatically implies area=yes on ways instead of insisting of adding area=yes to every single way. This can be done because the other option of closed way tagged an highway=services in the meaning of a non area makes no sense.
A change like this on a wiki requires going through the tagging list. That in turn requires a person with quite 'thick skin' and you seem to fit. -
Comment from Test360
@SomeoneElse
Thank you for this detailed response, I get your point. I don't know if you had intended to answer my questions with this message, but they are left unanswered.I have read the mechanical edit policy. Right now, I do *not* complain about the revert of my mechanical edits, I rather try to understand why some of my changeset *that are not mechanical edits* have been reverted as mechanical edit.
-
Comment from Test360
@RM87
Thank you for this clarification. I now understand.I see several reasons to not change the current lack of implication between highway=services and area=yes. So I won't try to change that by discussing it on the tagging list.
So do you agree that right now, a service area which is tagged with highway=services and without area=yes is an obvious error?
Nevertheless, the important point is that the changesets I mentionned earlier that concerns this issue were not mechanical edits. I had checked each service area with Bing. And I'm pretty sure that grouping several edited ways into one changeset is not a reason to revert.
-
Comment from Test360
I would like to know why the http://www.openstreetmap.org/changeset/27643030 has been reverted: not a mechanical edit, just a group of two very local changes.
-
Comment from Test360
I would like to know why the http://www.openstreetmap.org/changeset/27643284 has been reverted: not a mechanical edit, just a group of modifications that have been review one by one thanks to Bing.
-
Comment from Test360
I would like to know why the http://www.openstreetmap.org/changeset/27643485 has been reverted: not a mechanical edit, just a group of modifications that have been review one by one thanks to Bing and based of the nearby streets in OSM.
-
Comment from woodpeck
You had made a large amount of mechanical edits interspersed by a few manual edits. The changeset comments were not suitable for determining whether something was a manual edit or a mechanical edit. I would normally have reverted *everything* done by you in that time span but instead I made an effort to look into the changesets and leave those in place that seemed to be manual edits. It is quite possible that even though I invested that time to understand your insufficiently described edits, *some* manual edits were still in the list and accidentally reverted as collateral damage.
-
Comment from Test360
Why haven't you asked me what changeset are mechanical? If you had notified me of the revert of my 80 changesets before, I would gladly have provided you the ones that were mechanical.
"I would normally have reverted *everything* done by you in that time span". That is a very wrong way to deal with the changes I have done. The correct way is to carefully analyse each changeset. You should take a look on this page: http://wiki.openstreetmap.org/wiki/Vandalism . "If the edits are dubious but it can't be proved to be incorrect then we should contact the person and ask for some additional information"
If you find an insufficiently described edit, the right way is to ask the author for explanations.
You seems to underestimate the scope of the "collateral damage". So far, I have found 14 changesets that appears to have been wrongly reverted. Several of them represents hours of work. Moreover, I will spend several hours to redo them.
-
Comment from gileri
@woodpeck I'm interested (and I believe other mappers too) in seeing a resolution of this discussion and changesets.
Could you answer the points raised ?
-
Comment from woodpeck
Could you point me to the forum or mailing list where this changeset is being discussed? Only a few days ago I received inquiries from user "tuxayo" about this and I can hardly believe it is accidental that several people stumble across a year-old changeset?
-
Comment from gileri
I'm not sure how it's relevant to the discussion of this changeset, but for full disclosure :
You're right, I stumbled upon this changeset by accident, and pointed @tuxayo to it, as we've had private discussions about this kind of changesets and their reversal by DWG.
I think we can now return on the original topic at hand.
- Comment from woodpeck
- 295274114, v3
- Brissy-Hamégicourt (183795635), v13
- 234469538, v5
- 187047103, v4
- 273024706, v3
- 300061260, v3
- Cimetière (211203538), v4
- 49204344, v3
- 39493898, v4
- Campo Europa (171552274), v5
- 47188162, v3
- Cimetière (51451782), v9
- 315903837, v3
- 295637503, v4
- 199724487, v4
- 222376765, v8
- 41377717, v13
- 50820387, v8
- 318043957, v3
- Orchard Drive (22691097), v3
- 3304966, v4
- 4250622, v3
- 398785, v16
- 1702670, v5
- 1358993, v3
- 1660941, v3
- 402459, v3
- 3156599, v4
- 402289, v3
- 413554, v3
- 2175750, v3
- 1159366, v3
- Loch Druidibeg (3108649), v4
- 3931478, v3
- 3156601, v8
- 1387517, v3
- 414034, v3
- L'Isère (278498), v67
- Bus 3: Champs de Mars => St Blaise (3161077), v13
- 1387527, v3
Welcome to OpenStreetMap!
OpenStreetMap is a map of the world, created by people like you and free to use under an open license.
Hosting is supported by Fastly, OSMF corporate members, and other partners.
https://openstreetmap.org/copyright | https://openstreetmap.org |
Copyright OpenStreetMap and contributors, under an open license |