I’m glad to see this diary, I’m the author of another project Keqing (although the repository is currently owned by the organisation, I’m the only one who mainly maintains it), although it started very early but I’m only coding in my free time, because There are still many features waiting to be implemented, so I haven’t dared to introduce it to others. osm_easy_api looks very well developed, so I’m afriaid that I’ll be “reinventing the wheel” in my Keqing package. I actually noticed that we may have already worked towards the same goal, for example on parse data and quick extraction of data that have specific tags. Of course, I don’t have confidence in my own code, my implementation is also not concise enough. As the author of an OSM-related library also based on python, do you have any suggestions on how to reduce duplication of work I’m doing and calling between those 2 package each other?

(Besides, also looking forward to your comments on my humble work)

(Reply to @Adamant1 ‘s comment above, diary/400909#comment54304) Indeed, I may have had unrealistic expectations of Bing’s set of tools, I should realize that it is a commercial company after all.

(to @Adamant1 ‘s comment /diary/400909#comment54306) Maybe it is possible to review the edited mapper’s account? Similar to previous organizational edits?

The only other way to feedback the edits into OpenStreetMap would be if Microsoft did mass imports of their users’ contributions every couple of months or something, but doing it that way is clearly a non-starter.

Hope that doesn’t cause a lot of conflicts with editing in between these months. So I hope that Microsoft will do more in terms of open data. I’ll continue to see if the content I’ve previously edited with Map Builder in a number of different countries

I don’t know what it is, but you’re right - it looks like the edits have stopped appearing.

If this is because Bing is organizing their mappers to gradually review all contributions from Map Builder, then I think this is acceptable and hope this is confirmed, then I will retract my previous allegations and really apologize

Hope to find such evidence and possibilities as explanation.

(Continuing from the above response to @Adamant1)

And in my opinion, what we are discussing is different from issue between OHM and OSM, because OHM stores a lot of ancient administrative division boundaries, demolished objects, and they don’t meet the requirements of on the ground rule in OSM. That’s the difference.

@woowoowoo @Adamant1

There may be some misunderstandings here, but I must point out that my question against Bing exists on a moral level rather than a legal level. Yes, Bing can do this from point of license view, including all of us here cannot prevent its use data licensed under the ODBL - if Microsoft correctly signs the OpenStreetMap contributor attribution correctly.

Why I still have to say they are hijacking OSM is because If these edits are not contributed back to OSM and just stay in Bing’s own database, but still see data from OSM was shown during the Map Builder editing process, it will make people mistakenly believe that their edits are actually being contributed to OSM, that is why I think Microsoft Bing should be criticized - unless Bing can get this part of the edits back upstream in some form.

If they can do it, I apologize for it

The only attractive thing about it for me is its anonymity, it is difficult for the government to trace the identity of OSM editors

I changed my idea, those many edits I‘ve made the other day through Bing’s Map Builder have never entered the OSM database in any form yet, so Bing is hijacking OSM

The only attractive thing about it for me is its anonymity, it is difficult for the government to trace the identity of OSM editors, especially in countries where OSM editors receive government interference

e.g., Where I’m living in.

I just tested editing in Bing Map Builder and they don’t show up on OSM in real time, so I do have concerns with Bing’s toolchain, are they just intercepting edits, or are they really hiring skilled mappers validate it? But if it really appears in OSM in real time, will it be abused as a tool to bypass the block?

It’s really worth thinking about

Aha, this project let me remember the “EqualStreetNames” made by OSMBE community: https://equalstreetnames.org/

That project use name:etymology:wikidata to specify the name from which person.

似乎依然留有大量的残留内容没有被清理

这位在编辑的时候过分依赖卫星图像，对DEM模型不正确的卫星图像过拟合，因此很多桥梁隧道，明明是笔直的，却被错误的视为弯曲。

因为中国的山区建设非常快，地形也发生了较大改变，而部分影像中使用旧的高程模型就会造成这种弯曲。很显然这些问题仍需中国mapper的手动修正。

FYI：https://t.me/osmChinaPT/3690 ~ https://t.me/osmChinaPT/3699

We don’t know how Bing is doing in other parts of the world before, but as far as the Mainland of China community concerned, we found that Bing Imagery was very outdated at least two years ago. We usually use the Esri or Maxar Imagery in the editor as the background for Armchair Mapping of unmapped area.

However, the iD editor selects Bing Aerial Imagery by default, so many newcomers often change the roads/buildings to the state of many years ago when they first come into OSM (some areas are only 1-2 years behind, and the images in many economically undeveloped areas are at least behind 5 years), so a big challenge for the OSMChina community is how to discover that newcomers have inadvertently drawn historical elements that no longer exist, and explain to them on time.

Thanks to other mapper friends that also found this problem.

我们不知道Bing在世界上其他地区的表现如何，但就中国大陆社区而言，至少在两年前我们便发现Bing影像已非常过时。我们通常采用编辑器内的Esri或Maxar影像作为对欠绘制区域Armchair Mapping时的依据。

但iD编辑器默认选择Bing航空影像，因此很多新人第一次接触OSM时仍经常会将道路或建筑改至多年前的状态（部分地区仅落后1-2年，而很多经济不发达地区的影像至少落后5年），因此中国大陆社区面临的一大挑战是如何发现新人不慎绘制了已不存在的历史要素，并及时向其说明。

很感谢有其他编辑者朋友亦发现了这一问题。

“铁路”还是“线”的问题

我比较担忧的是，我们如何面对维基百科命名习惯里面的“京沪铁路”风格的编辑者进行的善意修改。（尤其是绑定wikipedia以后，同时带着name=京沪线+wikipedia=zh:京沪铁路）

这个问题不少国内铁路老编辑者都非常有意见，还打过架，我们得尽可能取得更多人的意见统一。

统一ref

说实话包括我在内大部分人应该不知道有这个ref，才会“善意”推测的写上ref=京沪这种字眼

监控与防破坏

国内铁路里程非常多，分摊到每一个active mapper头上的里程很多，可能什么时候出来一个新编辑者在不知道规范的情况下就把周围很多给改了，还不好fully revert。只是写进osmwiki，新人看不到的话就不了解我们的规范，还是有点难办的

英文翻译格式

我不发表意见了，先看看大家能提出什么命名格式，不过希望不会变成之前公交命名投票的那么乱的情况。

我的想法是做到尽可能多的兼容 因此我赞成直接ref和railway:ref双管齐下，保证不管从哪个角度都能通。

然后关于name的问题，道岔如果要有name我觉得总不能叫“X号道岔”吧……不然就只能照抄ref？

偏个题，轨道的话name可以是“13道”但不建议是“北京站13道”，因为它已经在北京站的区域内。而ref就可以是纯数字的13。（这个是特意查过OSM对于大包小问题的态度的，子一层不用重复说明）

而hanchao先生所言的“无现场调查”，我倒是认为这次并不适用。因为这并非新增对象，而属于对众所周知的品牌的规范化，虽然是否应移除这个alt_name仍有争议，但Thregren移除alt_name的这个理由，“NSI数据（name-suggestion-index）中，中国大陆及港澳的版本，已不存在alt_name”，我想是站得住脚的，因为NSI确实不能对过往编辑做出回溯性修改。且若是有争议，这份争议应当被送到ideditor/NSI社区，而非中国编辑者社区，因为这是一个技术问题。

举个例子（按照时间顺序）：

• A品牌在2020年的NSI数据中不包含alt_name，因此2020年编辑者在用ideditor时就不会向改品牌的POI上添加alt_name。
• 而在2021年NSI数据中包含了alt_name，那么在2021年使用ideditor完善过的A品牌就会包含alt_name。
• 但若2022年，NSI数据中这个alt_name被移除了，之前2021年的编辑者编辑过的POI中包含的alt_name并不会随着NSI数据的变化而移除，而需要编辑者手动维护OSM数据，删掉POI上的alt_name。

–[zh-Hans ↑][en ↓]–

As for hanchao’s reference to “no survey” accusation, I don’t think it applies to this issue. Although it is still debatable whether the [alt_name] should be removed. Thregren’s reason for removing it, “The alt_name for KFC in Mainland of China, Hong Kong and Macau no longer exists in NSI(name-suggestion-index) data”, is a valid one, at least can be discussed. Since the NSI really cannot handle old version edits. And if there is a dispute, this dispute should be sent to the ideditor/NSI community, not the Chinese mapper community, because it is essentially a technical issue

As an example (in chronological order).

• Brand “A” does not contain [alt_name] in the 2020 NSI data, so mapper will not add [alt_name] to the POI when they use ideditor in 2020.
• And the NSI data added [alt_name] in 2021, then Brand “A”, when been refined using ideditor in 2021, will contain the [alt_name].
• But if this [alt_name] is removed from the NSI data in 2022, the [alt_name] stored in the POI previously edited by mapper in 2021, will not be removed with the changing NSI data. It require the mapper to manually maintain the OSM data and delete the [alt_name] on the POI in 2022.

就同一个问题的修改，用大范围的变更集并非没有先例，如（ osm.org/changeset/115447422 ），且在评论中进行了细致的说明进行如何的变动。Thregren特别备注“没有编辑台湾地区”，我想就是因为知道bbox可能包含台湾西部而特别声明的。若是想如您所说，专注于绘图而不必为修改集评论区困扰，那最好的方法显然是信任其他编辑者的comment。

甚至连Overpass Turbo代码都留下来了，在代码中明确点明了”{{geocodeArea:China}}->.searchArea;”

–[zh-Hans ↑][en ↓]–

It is not unprecedented to use large changesets for the same particular issue, such as (osm.org/changeset/115447422), and there is a detailed comment of what changes made in this changeset. The user, Thregren, specifically notes “No edits to Taiwan area”, I think because he knew that the bbox might include western Taiwan and to avoid arguing between 2 community. If you want to “focus on mapping, not changeset comment discussion” as you said, then the best way is obviously to trust the comments of other editors.

Even the Overpass Turbo code was given, with a clear “{{geocodeArea:China}}->.searchArea;” in the code.

我可以直连 不过，如果真的为了利益，商业公司游说政府封禁也不是没有先例，参见百度学术挤兑谷歌学术的前例。

然后这个wheelmap网站唯一被诟病的就是 1. 匿名编辑是否违反原则 —— 但是西方社区都没表态，我们没啥可说的，看来修改并没有作妖 2. 修改集经常太大 —— 似乎我们也左右不了什么

但是思路和出发动机是好的，无障碍也是应该做的

喜欢标军区的就只有某个特定的用户而已

我的建议是，即使要记录贡献，作为作业得分的依据，也不应当把真实姓名放出来，而是用其他代号来代替姓名，比如学号之类。 并且也不应该直接在note或source字段中填写

也不是完全无人修正，他们完全依靠社区来对学生进行监督修正。 如果社区没有发现这个问题，错误数据就会长久的存在于OSM中，包括学生的姓名。

看起来第二份不带地形的更整洁一些，但文字或许太突出了

无论如何，感谢您在多语言地图上做出的贡献！

首先，这里没有出版方，这个地图并非又某个具体的出版社出版。 （如果真的有这么一家出版社敢这么出版地图，那么不光中华人民共和国会希望暴捶它，很多国家都会因为领土问题想暴捶它的） 这个地方的所有数据都是全世界各地的人绘制的，这些绘制者可能持有不同的政治观点和立场，因此存在争议是必然的