Wikipedia:Articles for deletion/Flora-2

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 keep. (non-admin closure) —cyberpower ^Chat_Offline 16:06, 23 March 2014 (UTC)

Flora-2

Flora-2 (edit | talk | history | protect | delete | links | watch | logs | views) – (View log · Stats)

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

Academic software which doesn't appear notable per WP:GNG. All publications are primary sources by the same group of researchers at UBuffalo and Stony Brooks. There are about 200 citations to the papers combined, perhaps that's enough for some, but given the bloom of OWL publications, this is actually a rather low citation count. A similar system called Pellet has 1654 citation to one paper, so that might be notable, but not this. Someone not using his real name (talk) 11:38, 29 January 2014 (UTC)

Note: This debate has been included in the list of Software-related deletion discussions. Someone not using his real name (talk) 11:39, 29 January 2014 (UTC)

Note: This debate has been included in the list of Academics and educators-related deletion discussions. Someone not using his real name (talk) 11:43, 29 January 2014 (UTC)

• Keep - The system is not an OWL reasoner like Pellet as the above implies. The comparison of citation counts is therefore irrelevant. Casonj (talk • contribs) 04:53, 30 January 2014 (UTC)
  • Someone built F-OWL using flora-2, so perhaps it's not that unwarranted to compare flora-2 with another [semantic] reasoner. You are probably right that Pellet being an OWL reasoner from the get-go has affected its (academic) popularity relative to non-OWL reasoners. But F-OWL only has some ~80 citations, so not every reasoner that's OWL'd flies high. Someone not using his real name (talk) 21:57, 5 February 2014 (UTC)
    • Your insistence on putting your ignorance on display it truly remarkable. Michaelkifer (talk) 05:35, 6 March 2014 (UTC)
  • Also, I haven't done anything like a formally tally, but a lot of the citations for flora-2 are in OWL/semweb publications. So you can't say it hasn't been (positively) affected by the boom in OWL/semweb research/publications. Maybe it's not 100% fair to compare it with Pellet for citations, but it's not 100% irrelevant either. If you think another class of reasoners is a more appropriate comparison citation-wise, please suggest which. Someone not using his real name (talk) 22:10, 5 February 2014 (UTC)
  • Ok, but why did you even start this process without really knowing the difference? And who defines the threshold anyway? Would 30 citations be too few? 40? There are a sufficient number, in my estimation, of relevant publications as well as open source projects using Flora-2. It's true that many references are authored by the developers, so what? That means the ideas have been vetted over and over by academics independent of the development team through peer review. That's noteworthy. As someone mentions below, one paper was recognized as a "Most Influential Paper" at a relevant academic conference (not semantic web - logic programming). Isn't that noteworthy? Lastly, I don't appreciate being accused of being a sock puppet. I have had as yet nothing to do with the development of Flora-2. I just happened to be someone who finds Flora-2 noteworthy and feel it's ridiculous that it's being nominated for deletion. I have edited Wikipedia before, but the fact that I don't spend all my time editing it or have all the policies memorized does not make me a sock puppet. Casonj (talk) 23:16, 5 February 2014 (UTC)
    • There is unfortunately not a guideline for this because the number of citations to consider enough varies by field. My personal rule of thumb is that less than 1,000 citations for a compsci paper is questionable notability for Wikipedia. And academic biographies are judged by similar citation-based rules of thumb using the h-index; see WP:PROF and WP:AFD/PROF for the latter. Someone not using his real name (talk) 21:54, 7 February 2014 (UTC)
• Keep - The anonymous protagonist for the deletion betrays only a specious understanding of the subject. Flora-2 is a system in the Logic programming (LP) family of knowledge representation languages. The expressive power, the capabilities, and the application domains of these systems are completely different from the OWL reasoners. The intersection between LP and OWL is rather small and uninteresting from the logic programming point of view. The Flora-2 page has also been updated by the users and more bibliography (with citation count in the thousands) was added. MichaelKifer (talk • contribs) 19:56, 1 February 2014 (UTC)
• Keep - This is a unique system that extends classical logic programming with FLogic, Transaction Logic, HiLog and defesible logic. There is no other system like it in the world for knowledge representation. Moreover, it is used by hundreds of users. Therefore, it would be a mistake to remove the Wikipedia page. PIFodor (talk • contribs) 20:28, 30 January 2014 (UTC)
• Keep In addition to what the above comments have discussed, much of the research upon which Flora-2 is based has proven influential. For instance, Transaction Logic, implemented in Flora-2, received a retrospective "Most Influential Paper" award at the 2013 International Conference on Logic Programming . Other underlying logics of Flora-2 have also received awards. So in addition to being an open-source system used by various research and commercial users, Flora-2 implements a unique combination of logics, which have been specially recognized by others in the fields of logic programming and deductive databases. — Preceding unsigned comment added by Writtenfarces (talk • contribs) 21:56, 1 February 2014 (UTC)
• Keep - I can only support what has already been written and add that Flora-2 is being taught at least on two universities I know about, I my self am teaching three courses in which Flora-2 is used. Mschatten (talk • contribs) 00:46, 2 February 2014 (CET)

• Delete (see additional commentary below) - so many !votes, so few policy-based arguments. Needs to meet WP:GNG. Does it? Doesn't look like it. Stalwart111 08:33, 5 February 2014 (UTC)

A decade of papers focussed on Flora at serious international conferences. No, of course that doesn't meet GNG – where's the MTV or Nickelodeon coverage?Andy Dingley (talk) 18:36, 5 February 2014 (UTC)

You're right, Andy, because what I called for was coverage on the Real Housewives of LA! Hmm. There are 9 references attached to the article and every single one was authored or co-authored by the creator of the software. A similar reference list for commercial software or a short film or a company or a piece of music or really anything else not "academic" in focus would be shot down instantly. Yesterday I contributed to an AFD about a college salsa dancing group where every reference was authored by the coach, the university or a single local newspaper. Great people doing great things for their college/field-of-study on the world stage but nothing by way of independent sources. Good causes (including academia) don't get a free pass. Stalwart111 22:17, 5 February 2014 (UTC)

So what's the issue here? If you claim that sourcing is too single-sourced, that's a valid concern. We have ways to fix that, by editing the content of the article. If you claim that there's a COI, then we have review policies for dealing with that as well.

What did WP actually do? It cited GNG, on the grounds that the topic is insignificant and non-notable. Easy way to completely remove an article one dislikes, but it's a quite inappropriate response to your concerns as stated here.

What did WP do next? A sockpuppet investigation! Alleging that Michaelkifer is the invention of another editor. Whilst simultaneously being the academic who's using this article for evil self-promotion. It doesn't really work both ways. Andy Dingley (talk) 22:40, 5 February 2014 (UTC)

Actually, I had no idea about the SPI until after I commented (and haven't looked at it), nor did I trawl the history of the article looking for possible COI and I don't think I've alleged one here. I generally expect to see some connected contributors when it comes to niche academic topics and I don't think these guys are in it to make millions from their software so it's probably less of an issue than areas where I usually work like WP:SKATE (I have plenty of experience with MTV references!) where companies pop up to promote their useless crap all the time. My issue is that we have a small academic community that created something and now talks about it extensively (and covers it) among themselves. The difficulty here is that the one person most likely to be considered an expert on this software is the fellow who helped create it. He'd otherwise be considered a very reliable source on related topics but that's a stretch here because of his lack of independence from the subject. An MTV analogy? Joe Jackson would be considered a reliable source with regard to Motown generally, but a less-than-independent source with regard to the Jackson 5. Put up a couple of sources from people other than those in the dev team and its a different story. The ones we have at the moment, given they don't meet WP:GNG requirements with regard to independence, probably won't suffice. Stalwart111 23:28, 5 February 2014 (UTC)

• Note to closing admin: there's a good chance that meatpuppetry is occurring here, see Wikipedia:Sockpuppet investigations/Casonj for details. Mark Arsten (talk) 16:42, 5 February 2014 (UTC)
  • Didn't we go through this last time WP went after SemWeb topics? Proof of innocence from sockpuppetry is not the same thing as proof of meatpuppetry. Obscure topics are obscure: people who know nothing of such an obscure subject will be ignorant each in their own special way. People who work in that obscure field are likely to be knowledgeable about a topic, and for their knowledge to be very similar to each other. This is not evidence of a bad-faith action and per strong WP policies it is wrong for WP editors to imply or to act as if it is, in the absence of convincing evidence to the contrary.

I am thoroughly tired of WP's inbuilt bias against SemWeb topics (WP is out to delete semantic reasoner itself at present for having "too many" links!) I am particularly tired of having to meet SemWeb people and then have to defend WP to them and explain that, despite appearances, it's not just a childish project where nothing other than K-pop and Marvel comics are taken seriously. Sure, this isn't List of Disney Princesses or anything important like that, but academic speciality is not proof of unimportance. Andy Dingley (talk) 18:36, 5 February 2014 (UTC)

You should read WP:GREATWRONGS. It's not Wikipedia's fault that the academic research/publication boom in SemWeb apparently did not have much real-world relevance/applications and thus has not received much outside coverage. Someone not using his real name (talk) 21:20, 5 February 2014 (UTC)

<sigh> You miss the point completely. The SemWeb community doesn't give a damn whether semantic reasoner or even semantic web are deleted (yes, go ahead and AfD that too, it's the sort of petty bitchiness that WP increasingly practices). It's only Wikipedia, it doesn't matter. You make the classic Wikipedian mistake of thinking that when an article is deleted from here, it makes it vanish out in the real world too.

WP does (allegedly) have a policy of AGF. Yet when SemWeb articles are slated for deletion by the ignorant, the reaction to a bunch of academics saying that they're significant and sourced is instead "you're all just sockpuppets". Andy Dingley (talk) 21:34, 5 February 2014 (UTC)

I don't see how WP:VANISPAM is better when it's written by an academic. This page (flora-2) was written by someone with an obvious WP:COI. Nobody else cared about it to cover it on Wikipedia. Someone not using his real name (talk) 22:02, 5 February 2014 (UTC)

Um, Would you say that FB is not noteworthy if Mark Zuckerburg started the article? I think you jumped the gun at first and now are trying to save face. I don't see it working. The software is noteworthy; you just didn't do your homework. Casonj (talk) 23:30, 5 February 2014 (UTC)

It's funny that you mention WP:GREATWRONGS where it says "wait until it's picked up in mainstream journals", and this idea/software has been. Casonj (talk) 23:51, 5 February 2014 (UTC)

I never said there was proof of meatpuppetry, so I don't know where you got that from. But when a group of long-dormant and brand new accounts all become active at the same time to vote the same way in the same discussion then I think it's fair to say there's a possibility that meatpuppetry is involved, whatever the merits of the article in question. Mark Arsten (talk) 23:07, 5 February 2014 (UTC)

---

Relisted to generate a more thorough discussion so a clearer consensus may be reached.

Please add new comments below this notice. Thanks, Michaelzeng7 (talk) 20:08, 5 February 2014 (UTC)

---

• Keep A quick Google search brought up several journal entries and academic websites regarding Flora-2. There seems to be many valid sources for Flora-2.Spirit of Eagle (talk) 23:06, 5 February 2014 (UTC)
• Comment Stalwart: References to publications unrelated to any of the developers have been added. I am sure that you will now remove all your objections, right? Also, FYI: WP articles have no value in the academia whatsoever. Michaelkifer (talk) 23:48, 6 February 2014 (UTC)

• Keep - Michael, that's exactly what I was looking for. There are a couple of very strong sources there including international publications that provide coverage of the subject's capabilities, features and importance in detail. You might consider linking directly to the PDF of the Schatten/ƒCubrilo/S’eva paper and some of the links are to subscription-only papers (which should be noted) but those aren't required steps for them to be considered. The added references are more than enough for me. Objections withdrawn. Stalwart111 00:05, 7 February 2014 (UTC)

Comment The reason for the subscription-only links is because these are the publisher sites that published the corresponding works. They are sure to stay around for a long time, while personal web sites tend to undergo all kinds of changes. So, these links are better for WP. People can always Google to find free copies, if available. Regarding Schatten/ƒCubrilo/S’eva, this is a link to the conference site and there is a link to the paper itself on that page. This is better because the referenced page provides information for people who need to cite the paper in question. Michaelkifer (talk) 08:41, 7 February 2014 (UTC)

All perfectly fine - there's a little template you can use to tag sources as subscription only, that's all. It's not required, just helpful for readers. We would generally cite the source itself but your explanation makes sense and it makes no real difference to the substance or validity of the source. I'm still comfortable with both. Stalwart111 09:55, 7 February 2014 (UTC)

F-Logic (which has ~2K citations in GS) has its own article. I'm not sure that other implementations of F-logic (and papers thereof) justify a separate page for Flora-2. Schatten's application papers have "F-logic" in title, although it's quite possible they've just used the Flora-2 implementation "as is", although that's not immediately apparent from their abstracts. One other academic research group using Flora-2 doesn't raise the notability a whole lot more than the aggregate (academic) citations. (And yes, I do realize F-logic and Flora-2 and [1] etc. have the same people behind it. Someone should create a bio for Michael Kifer as he clearly passes WP:PROF with a GS h-index of 45 [2]. But I'm not sure we need a WP:COI article for every paper of his.) Someone not using his real name (talk) 22:07, 7 February 2014 (UTC)

Flora-2 is not an implementation of just F-logic, but of a combo of different logics, which doesn't occur in any other system, as someone already mentioned. All the newly referenced papers both use and refer to Flora-2, as you can easily check yourself. If you are the same anonymous user as elsewhere in this discussion page then you appear to be quite adept at shifting arguments. You have traveled quite a long way from your original post, so maybe one day you will become a user yourself :-). It is free both as in "free beer" and in "freedom." Finally, it is not just "one other research group." Check again. There are also thousands of downloads each year. But thanks for raising the issues: it certainly has helped to improve the page. Michaelkifer (talk) 05:49, 8 February 2014 (UTC)

Technically correct, but most independent sources cited in the article see F-logic as its primary/defining characteristic (whether that's fair or not). Someone not using his real name (talk) 06:26, 11 February 2014 (UTC)

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.