Wikipedia:Articles for deletion/Comparison of XMPP server software

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was delete. While I can see there is some disagreement, I see a consensus to Delete based on Dylnuge's argument. Liz ^{Read! Talk!} 01:23, 4 November 2022 (UTC)[reply]

Comparison of XMPP server software[edit]

[Hide this box] New to Articles for deletion (AfD)? Read these primers!

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

Masses of primary-sourced tech cruft and original research * Pppery * _{it has begun...} 00:25, 28 October 2022 (UTC)[reply]

delete as blatant transgression of WP:NOTGUIDE. PC Magazine and its ilk exist for a reason. Mangoe (talk) 02:17, 28 October 2022 (UTC)[reply]
Note: This discussion has been included in the deletion sorting lists for the following topics: Computing, Software, and Lists. – dudhhr^talk _contribs (he/they) 04:34, 28 October 2022 (UTC)[reply]
Delete Clearly original research. Agletarang (talk) 17:57, 28 October 2022 (UTC)[reply]
Delete, but I think it's a bit more complex than some of the other nominations like the Excel add-ons AfD. In those cases we had comparisons of fully non-notable software. Here we're getting into comparisons of software that have their own articles, so they may be notable (I think notability is questionable for some of these, to say the least, but examining the notability of every list item seems a bit out of the scope of the discussion, and I don't need to do it for my deletion argument). Per WP:LISTCRIT, when a list can't be exhaustive it should have selection criteria, and a common one is Every entry meets the notability criteria for its own article in the English Wikipedia. My previous argument used entry notability as a criteria because: 1) these comparisons are, in part, a list of software (some are even categorized as such) 2) the lists cannot be exhaustive and 3) none of the entries are notable, and there's not another good WP:CSC. Comparisons of non-notable software are a trivial delete vote, and we don't have that here.
I do think there's a broader argument against having these kinds of software comparison matrices at all, which I brought up in my reasoning on the DNA melting prediction software AfD. However, this argument feels a bit more complex; we have 300 total pages in Category:Software comparisons and its subcategories (as of my counting; several are in AfD right now but this should stay order-of-magnitude correct). WP:OSE, I know, but I think it merits some notice that the wide variety and presence of these suggests that there may be prior existing consensus around having them. I disagree, but it'd be a bit disingenuous for me to say WP:NOTDIRECTORY and call it a day when these pages are so prevalent.
My main problem with these articles is that they seem essentially impossible to maintain, in the spirit of WP:NOTDIRECTORY. I can see an argument for the value in general of having community sourced Consumer Reports like comparisons for notable products, but given the maintenance and review that would be required to make them trustworthy, they feel like a bad fit for Wikipedia. I'm not a fan of any page where someone who visits it is unlikely to trust that it is accurate, neutral, and useful. I see a lot of problems here: the selection of what to compare on is WP:SYNTH (and selecting things to make a specific product look good is easy, companies do it all the time in marketing material; neutral subject matter experts would be needed to ensure things don't fall into WP:PROMO), keeping things complete and accurate would require regular review as well (most edits to these pages update information for a specific product, and not the entire chart at once), and most of these things don't have good sources besides marketing material from the developers (and company claims about a product are not a good source for something like this). Overall there are plenty of ways for me to construct a policy-based "this is a bad page" vote, but I want to be clear that my main reason for deleting is that I think a page like this is unmaintainable and we shouldn't have pages that can't generally be trusted.
@Pppery: As a procedural note, have you considered making a broader WP:MULTIAFD (or for that matter, even a more general "Software comparisons on the wiki" RfC) instead of nominating these individually? It seems like you've got several of these up for nomination right now and while my reasoning applies to all of them, a more all-encompassing discussion might be merited. Dylnuge ^{(Talk • Edits)} 18:06, 28 October 2022 (UTC)[reply]
This is a WP:MULTIAFD of two articles already. While I would probably support deleting most of Category:Software comparisons and its subcategories for the same reasons, I've generally been nominating comparisons of only non-notable entries. What happened here was I noticed while reviewing the entire tree for other non-notable comparisons that this article was so ridiculously laden with fancruft that I ended up removing 5/6 of it, and then noticed the rest was all primary-sourced and warranted a AfD. IMO each of these is different enough for a separate discussion, although I intend to stop working on this task for a week to allow for these AfDs to close and confirm I'm not overstepping. * Pppery * _{it has begun...} 18:19, 28 October 2022 (UTC)[reply]
Yeah, by "broader" I meant encompassing more/most of these; didn't mean to imply this one wasn't already covering two articles. FWIW I don't think you're overstepping by nominating these (I guess obviously I agree with the nominations, so that's not an un-biased opinion or anything). Dylnuge ^{(Talk • Edits)} 18:51, 28 October 2022 (UTC)[reply]
Before I started editing the article about XMPP servers, there was fully funcraft, there was no criterion for selecting XEP - in the end, I didn't know what to do with it, so I just left it, although I also thought about removing it. In any case, it was a significant improvement. Article about XMPP clients it does not contain it, although the whole structure of the article is basically based on one source (XMPP compilance), so it will probably be removed precisely because of the difficulty of finding secondary sources - unfortunately, this is no longer a popular technology and finding links to media references is almost a miracle, despite my best efforts. About W:OR, the easiest way would be to simply remove clients that do not have their own wikipedia articles. Twomithe (talk) 07:49, 29 October 2022 (UTC)[reply]
To clarify a bit, I don't think the OR issue here is selection criteria; I entirely agree that "notable XMPP clients/servers only" (i.e. those that merit their own Wikipedia article; technically whether or not they have one but generally determined by the existence of an article) is a totally fine criteria (and a pretty standard one, per WP:CSC). The WP:SYNTH comes with what things are compared on, and how those criteria are established and maintained. Are most readers going to care whether XEP-0401 (Ad-hoc Account Invitation Generation) or XEP-0455 (Service Outage Status) are implemented in these clients? Probably not—I picked two that, per the article, aren't known to be implemented anywhere—but how do we decide what matters? If we list every single RFC (or in this case, XEP) for an open protocol, the table becomes disorganized and hard to use. If we list only some, deciding which matter and don't is almost certainly a form of synthesis. There's also likely some OR involved in filling it out for each program—and where sources do exist, they're still likely non-independent sources like marketing copy which might be misleading. This isn't just something that applies to XMPP, either; even the collection at Comparison of programming languages suffers from this.
Honestly I think looking at Comparison of XMPP clients makes this point pretty clearly. The "?" character appears 1549 times on this page; that is far more than the word "Yes" (608 times) or "No" (pretty much confined to the Operating system support table; exact count was trickier since it's a substring of "not," "note," and "notable"). I can't see how these pages would be useful for people who actually need a comparison of these programs. Dylnuge ^{(Talk • Edits)} 15:43, 29 October 2022 (UTC)[reply]
There is a non-SYNTH (and not ridiculously crufty) inclusion criteria possible for the server list if you blank the entire "List of other RFCs/XEPs Supported" section (and the "Summary" section which summarizes it). The result is that the bases to compare are those specified by XEP 0459, which seems to be a reasonably standard feature set. I originally did that, but realized the result was still reliant exclusively on original research and primary sources (There's also likely some OR involved in filling it out for each program—and where sources do exist, they're still likely non-independent sources like marketing copy which might be misleading) so the entire thing needed to go.
I have no plans to systematically nominate other comparisons of bluelinks; this one caught my attention because of its ridiculous overdetail, and the only other ones I seem to have nominated are Wikipedia:Articles for deletion/SQL compliance (again overdetailled in the same way) and Wikipedia:Articles for deletion/Comparison of layout engines (Scalable Vector Graphics) (the problems are different). Finally, I suggest you write up the argument you are making here as an essay at Wikipedia:Software comparisons * Pppery * _{it has begun...} 15:59, 29 October 2022 (UTC)[reply]

So summarized, does exists the following options:

Insert all RFC/XEP - aside from the aforementioned clutter, which would require a logical arrangement to avoid it, it would make little sense since there are tons of unused, discontinued XEPs out there. Are you sure we would like to have them in the article? If not, you can't avoid OR 100% probably. For example, in the current shape of the article about XMPP clients, you have 2 XEPs that are marked as deprecated.
"If we list only some, deciding which matter and don't is almost certainly a form of synthesis."
1. many software comparison articles are often based on completely arbitrarily selected functions for no reason at all so it's 100% OR.
2. Other option, it use selection of a criterion based on the source. This is what the article about XMPP clients says. The choice is then not arbitrary and you avoid redundancy. This option seems to be the most optimal in this case, although not everyone likes it (digression) - For example, this edition has been classified as spam, even though I did exactly what is shown in this article, that is, I based the criterion on the original source. So what does matter?
3. Option, which represent article about XMPP servers it's synthesis 1 option and 2.2. which I guess it would be the closest to consensus. An article that selects criteria based on the sources is always better than without them. And that's probably the best thing than can be achieved. Any other option will be arbitrary.
"they're still likely non-independent sources like marketing copy which might be misleading."
1. the main problem of the current state of the article is the fact that it is based on primary sources - they are of better quality, but it also breaks WP:N. Although I have added some media sources, mostly from non-English language sources, I don't know if this is enough. Twomithe (talk) 12:34, 30 October 2022 (UTC)[reply]

Strong Keep Please see my previous comment on this visigothic attempt to a delete a notable article. XMPP has massive and active community and one of the core messaging protocols of the last 20 years. This is reflected in the article getting an average of 50 people a day looking it over the last 10 years, yet you want to delete. Those number indicate that its actively used for comparison and useful as valid encyclopeadic article. scope_creep^Talk 16:35, 2 November 2022 (UTC)[reply]
Comment It is not clearly WP:OR. Each entry has been added by the expert, so you don't know what your talking about. scope_creep^Talk 16:36, 2 November 2022 (UTC)[reply]
Can you clarify what exactly you mean by that? OR doesn't mean that the person adding the content is wrong; plenty of OR is, in fact, correct. The people writing the article might be experts! I have no reason personally to suspect the information is wrong (mostly, the information is missing, but where it isn't, I suspect it was at least accurate at the time it was added—though it's not clear to me if it's dated or not). Wikipedia pages need to be verifiable and maintainable by any editor—clear sources (mostly secondary) allow for this. Dylnuge ^{(Talk • Edits)} 23:59, 2 November 2022 (UTC)[reply]
Delete, as with essentially all of these "comparison" articles, this is a mess of original research that violates WP:NOTGUIDE and does not pass GNG in any case. Devonian Wombat (talk) 00:45, 4 November 2022 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.