File talk:NEA by survey (1995-2017).svg

File:Neo-chart.svg requires updating

Hi NEA by survey (1995-2017).svg,
Thank you for your contributions to Commons. I noticed File:Neo-chart.svg is out of date. statistics through the end of 2023 are now available. Could you update it? Thanks again. Ellehan (talk) 21:10, 18 January 2024 (UTC)

Renaming

A bit of confusion here I'm afraid! I requested an image rename from:

File:Neo-chart.svg → File:NEA by survey.svg

However, Mateus2019 moved it to this name instead (File:NEA by survey from 1995 to 2024.svg). The problem is that this graph is going to be updated past 2024 while the name suggests it won't be changed.

I'm not sure how best to proceed but ideally the move would be undone and then pointed to the correct target, sorry for the bother Mattes! Aluxosm (talk) 14:18, 4 April 2024 (UTC)

Hello, no worries. Your renaming proposal was not very descriptive (could include a lot of years), so I added the timespan. (file descr. quote "Chart of NEO's detected by sky surveys between 1995 and 1 April 2024"). So my suggestion is to leave it like that and other updated graphs in the future become new files with a new name and time spans in the file name. Or, if you want, I can move it to File:NEA by survey.svg (dunno if that is the idea of the original uploder or the rest of the community) though. Best wishes, --Mateus2019 (talk) 15:53, 4 April 2024 (UTC)

@Mateus2019: It was a little vague for a reason, just as with the other images in Category:CNEOS discovery statistics, they've been regularly updated for the last 10+ years so time ranges didn't really make sense (descriptions can be easily changed). Specifically though, this would align it with the PNG version of the same image (File:NEA by survey.png).

I think it'd be better having names that allow for updating rather than having to upload a new file every year, so renaming it would definitely be preferable.

This would leave the issue of this image (File:NEA by survey from 1995 to 2024.svg) presumably being redirected to an updatable version but it could be worse. Whatever you think works the best. Cheers, Aluxosm (talk) 21:24, 4 April 2024 (UTC)

I am sorry for the fuzz (I have not seen a file which is updated over years keeping its name constantly ... but ok). I'll let File:NEA by survey from 1995 to 2024.svg being deleted w/o redir and the file will be named File:NEA by survey.svg after all. Greez, --Mateus2019 (talk) 04:02, 5 April 2024 (UTC)

Please upload separate files, and redirect to newest version

Please note that we do not have dynamic files per Commons:Overwriting existing files. So best to upload your new versions, and then change the base redirect to point at the newest version of this file. We will obviously need to update the file targets at each wiki to reflect the base form.  — billinghurst sDrewth 06:04, 5 April 2024 (UTC)

@Billinghurst and Mateus2019: Apologies, I'm a bit new here so please excuse any misunderstandings; this just feels like a huge amount of work for a simple graph. I read through Commons:Overwriting existing files before proposing the change and it seemed to me that it was fine as long as Template:current was in place. Could you please point to the specific advice? I can't find any mention of "dynamic files" on that help page.

• Commons:Overwriting existing files#Files with current data: files may be identified to reusers as ones where information may be updated ... A file named "File:Germany location map.svg" is expected to indicate the current boundaries, and is expected to be updated.
• Commons:Overwriting existing files#Files with current data_2: Gives an example image of File:IAEA members.svg that has been updated since 2008 and does not redirect to a version with a range of dates.
• Commons:Overwriting existing files#Link to a variable content: Talks about this specific issue and does mention this redirect strategy, it is however the second option after: Label the image page with the {{Current}} template (see § Files with current data).
• File:NEA by size.png had been renamed by an admin and it the name didn't appear to present any problems.
• The same admin, User:Abzeronow (sorry to drag you in) recently marked two new SVG versions of the images as updatable and added a current template to one, still seemingly okay with the names.

I'll always go with consensus but to me it still seems as though I was. I'm mostly just struggling to see how having dozens of out of date versions of a graph like this would be useful for anyone, even ignoring the effort it'd take to maintain them all.

If we do end up going with the redirect option, could you suggest some tools for "fixing" the other images in Category:CNEOS discovery statistics; having to manually do this for every image at least once year doesn't sound like a lot of fun 😅. Cheers, Aluxosm (talk) 08:56, 5 April 2024 (UTC)

@Aluxosm: The best way to handle this is to have

• ... (1995-2017).svg
• ... (1995-2019).svg
• ... (1995-2020).svg
• ... (1995-2024).svg
• ... and so on for any future versions

then update for "NEA by survey.svg" to be updated to point to latest version of the file. This allows ongoing no maintenance. I would argue that for the set that you are looking at the progression itself has ongoing relevance for each data point. It is typically how many many progressions have been managed (anecdotal comment, nothing measured). I would also say that the graphic has relevance to have the series cited at a corresponding Wikidata item. [It could argued that the preferred file could be managed at WD to display the most updated item.] Overwriting should ideally be the last chosen option.

I feel that prior to something being updated that such a conversation would be considered at the beginning, have a consensus if there is to be a change in the process, rather than just occur as someone's decision.  — billinghurst sDrewth 10:30, 5 April 2024 (UTC)

<sigh> I see that there is also a PNG version, and that just makes it all ugly and uglier to manage, especially with the irregular updates. Makes me stand by my encouragement about separate and uniquely named files. Can I recommend people to actually look to actually managing the dataset per Help:Tabular Data and that readily allows the automated graphing, and labels in the language of the wiki, and most definitely for all other related files. Then it is solely the dataset and the expectation of an update.  — billinghurst sDrewth 10:39, 5 April 2024 (UTC)

I think requiring separate uploads for each year is going to discourage future updates. Yes, I can see a niche historical interest in a 1995-2020 chart, but it's more important for our reusers to keep the information as up to date as possible. The charts were designed to be updatable, and by requiring a new file for each year, we are adding a new burden on future uploaders and are essentially chaining ourselves to the past because we fear all destruction even when creative destruction would be better. Abzeronow (talk) 16:09, 5 April 2024 (UTC)

I'm an editor who has updated the png version of the files multiple times over the years, and I am the main editor of the article using the files. I am completely against the separate uploads idea, for the following reasons.

• I have seen dynamic updates as standard practice for diagrams of running stats in many different fields.
• Having read Commons:Overwriting existing files thoroughly, I find you still can have dynamic files: by using {{Current}}.
• These files aren't own work, they are the public domain product of NASA JPL, and are dynamic files at the source.
• The original files are updated every few days, not annually. Over the years, we Wikipedia editors tried to limit updates on Wikipedia to once a year, best near January, and update the data in the article to match the diagrams. (This wasn't done now, and I won't do it now before going to bed, so I may need to upload a new version of each file if the original files are updated anyway by the time this is settled so I did it, as fortunately the source hasn't been updated in the meantime).
• On all of the diagrams, the stats for past years can change somewhat (as the determined characteristics of the asteroids can change as additional observations improve precision), but not by much, so updated diagrams basically include the significant info in previous versions, and it makes little sense to keep them as separate files (past versions won't be used in any Wikipedia article for sure).

So my vote is on keeping a single file with no date span and using {{Current}} with the parameter "|interval=annually", but I accept another solution if I am in the minority or my interpretation of the policy is shown to be wrong.

Independently of that, I propose to copycopied additional info (like caption) & more precise language (not NEO but NEA, no apostroph etc.) from the png versions of the files, and maybe add a reminder in the description about the preferable update frequency. Rontombontom (talk) 00:05, 6 April 2024 (UTC)