Wikipedia:Articles for deletion/Spectral Hash

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 no consensus. –MuZemike 23:52, 24 June 2010 (UTC)[reply]

Spectral Hash

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

Non-notable cryptographic hash function with limited actual usage. KTC (talk) 23:28, 10 June 2010 (UTC)[reply]

Delete, not notable. --Nuujinn (talk) 16:37, 15 June 2010 (UTC)[reply]

Relisted to generate a more thorough discussion so consensus may be reached.
Please add new comments below this notice. Thanks, Ron Ritzman (talk) 00:01, 17 June 2010 (UTC)[reply]

Note: This debate has been included in the list of Computing-related deletion discussions. -- N/A 0 04:34, 17 June 2010 (UTC)[reply]

Keep Notability is based on mention by sources (which this has, from NIST), not on the amount of use an algorithm sees. Nor is WP the right body to judge its quality. We should quite rightly have an article on bubble sort too. Andy Dingley (talk) 09:30, 17 June 2010 (UTC)[reply]

Comment, the two references linking to NIST are primary sources, and do not constitute significant coverage in reliable sources. We're not here to judge the algorithm, but rather it's notability. --Nuujinn (talk) 12:19, 17 June 2010 (UTC)[reply]

Then are you going to delete NIST hash function competition too? I cannot see any way to separate the notability of the competition, and that of its lesser-known entrants (some, presumably the winner, might well go on to further notable things). Andy Dingley (talk) 12:45, 17 June 2010 (UTC)[reply]

Comment, the notability of the conference isn't relevant here, per WP:OSE. What the lesser-know entrants may do in the future is also irrelevant, per WP:CRYSTAL. --Nuujinn (talk) 13:04, 17 June 2010 (UTC)[reply]

You've failed to comprehend my point. You wish to reject the two NIST refs for Spectral Hash. Are there still refs for NIST hash function competition that wouldn't have to be rejected by the same criteria? Andy Dingley (talk) 13:15, 17 June 2010 (UTC)[reply]

Comment, I don't know, I glanced at the NIST hash function competition article, but haven't evaluated the sources, since that's not relevant to the discussion here. It may well be that the NIST hash function competition is notable, perhaps it is not--but in either case, the notability of a competition does not confer notability to all of it's participants. The Boston Marathon is notable, but most of the runners aren't. Also, I do not "wish to reject the two NIST refs", I'm just pointing out that since those references are the presentation of the algorithm, they do not establish notability per WP:GNG. --Nuujinn (talk) 13:26, 17 June 2010 (UTC)[reply]

Now I'm failing to understand your point. I'm happy to agree that NIST hash function competition is notable, but the majority of the refs in that article (I don't have time to dig) are no more (and perhaps less) secondary sources for that article than the two NIST-published refs for Spectral Hash. Now IMHO, these are all adequate refs for independence, quality of authorship and avoiding the issues of primary sources - so both articles stand. Whilst the first NIST-published ref for Spectral Hash is authored by the algorithm's authors, the second is little more than a conference schedule and list of entrants, authored and published by NIST. That isn't a great ref for detail, but it is IMHO a strong ref that a contest happened, and that Spectral Hash took part in it. That is as much, IMHO, as we need to demonstrate to show notability of it, sufficient to pass AfD. Andy Dingley (talk) 14:26, 17 June 2010 (UTC)[reply]

Comment, I'm not saying that the NIST hash function competition is notable, so I'm not sure what you mean by agree. If the references for that are no better than the ones in this article, notability is probably in question for that article as well. But that's not relevant to the discussion here. The fact that Spectral Hash was part of a competition does not establish it's notability. Do you really believe that the references constitute significant coverage? --Nuujinn (talk) 14:40, 17 June 2010 (UTC)[reply]

I see at least one independent RS that attests to its existence (and its entry in the competition). I consider that enough. Now strict policy calls for "multiple", so in the absence of time to look for further ones (and this encyclopedia would be better if editors spent more time in the Google: namespace and less arguing in the WP: namespace), I have no real answer to that. However I'd regard that as a churlish reason to seek deletion of an extant article on a useful topic (hubris in algorithm design) and I'm sure that other refs are out there, should anyone have time and effort to look for them (I expect there's more discussion of the flaw and its demise than of its initial release). Whilst a mass run like the Boston marathon might not convey notability on its competitors, some more selective contests do. Andy Dingley (talk) 15:14, 17 June 2010 (UTC)[reply]

Comment, I still don't see significant coverage in any of the references. We disagree, it happens. As a side note I point out that "more" doesn't list all participants, and many listed including winners are redlinked, "selective" lists winners (many of whom are redlinked), and contests lists all participants but there are a lot of redlinks there, too. I'm not sure what point you're trying to make, but again, WP:OSE. --Nuujinn (talk) 15:27, 17 June 2010 (UTC)[reply]

The difference with bubble sort is that it is notable from where it is taught / mentioned by lots of courses, books, tutorial etc. The only notes with this algorithm is its entry into an open competition. Just because the competition is notable doesn't mean one of its entry (which didn't get very far at that) is. The notability requirement requires multiple independent sources with significant coverage of the subject. The sources for Spectral Hash does not fulfil this requirement: 1) Not independent as it's the presentation of the algorithm by its designers; 2) Not significant coverage as it's merely the timetable listing of the 1st conference; 3) A paper co-authored by one of the co-designer talking about hardware implementation so not really independent; 4) Less about the algorithm and more about an attack on it. KTC (talk) 18:55, 17 June 2010 (UTC)[reply]

Delete. It seems this article is being used as a resume-builder for Gokay Saldamlı, Cevahir Demirkıran, Megan Maguire, Carl Minden, Jacob Topper, Alex Troesch, Cody Walker, Çetin Kaya Koç. It's a "new family of hash functions" (their website) with really no third-party significance. — Timneu22 · talk 11:56, 17 June 2010 (UTC)[reply]

"Resume building" would seem a little odd when this has function was so quickly rejected as insecure! Andy Dingley (talk) 12:04, 17 June 2010 (UTC)[reply]

Keep. This has notability due to being selected by NIST for the first round of the competition, and independent sources in the form of discussions of the competition, despite the fact that it's not a very good hash function. Compare FEAL, a completely insecure block cipher that is also unquestionably notable. Yes, it could use more direct sourcing in the article itself, but those sources exist. — Gavia immer (talk) 22:35, 18 June 2010 (UTC)[reply]

Question What "independent sources in the form of discussions of the competition"? If you have sources, please bring them to the discussion. Sources are always welcome! --Nuujinn (talk) 23:47, 18 June 2010 (UTC)[reply]

I mean discussions like [1] and [2] (warning, PDFs). I'm sure there are others. — Gavia immer (talk) 00:42, 19 June 2010 (UTC)[reply]

PDFs are fine. 2 is not linked. 1 is a good source, but not significant coverage, just a passing mention. But please, add any additional references you can: if notability can be established, that's a good thing. --Nuujinn (talk) 01:30, 19 June 2010 (UTC)[reply]

Fixed link 2; I dunno how I mangled it in the first place, so thanks for letting me know. — Gavia immer (talk) 01:56, 19 June 2010 (UTC)[reply]

No problem, but I still must be missing something, there's a footnote to an article with spectral hash in the title, do you consider that significant coverage? I do not think there has been reference to this algorithm put forth here that goes into any depth whatsoever about the algorithm itself that is not the work of the authors of the algorithm, and as such, I cannot regard it as notable. --Nuujinn (talk) 13:57, 20 June 2010 (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.