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. NW (Talk) 17:16, 25 September 2009 (UTC)[reply]

MurmurHash[edit]

MurmurHash (edit | talk | history | protect | delete | links | watch | logs | views) – (View log)
(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

Article is unsupported by any reliable sources. The article topic is not notable enough for reliable sources right now. It does not seem that this topic should be in the wikipedia at present- multiplicative hash functions are two a penny.- (User) Wolfkeeper (Talk) 22:36, 11 September 2009 (UTC)[reply]

Probable delete - I originally PRODded the article, but it was removed with the comment "remove prod - not a clear cut case", presumably because of the use in projects like Memcached. However, I'm not sure that this notability is inherited, so instead we need reliable sources. As the nom says, there are currently no reliable independent sources that discuss this subject (if some can be found, it's likely that I will change my !vote). Oli Filth(talk|contribs) 22:43, 11 September 2009 (UTC)[reply]
STRONG KEEP - Anyone who looks through the 16,000 hits on Google [1] and claims there are just no reliable sources is kidding themselves, and us. The project's own homepage on GooglePages [2] is a reliable source for the existence of the algorithm and for the verifiable specifics of it. The links to other notable projects that have chosen to include the algorithm, including memcached, maatkit and hadoop [3], are reliable sources that show that the algorithm is itself notable. Despite being relatively recent and lacking a strong connection to academia, it is already being referenced in academic papers [4]. Why is it being adapted by these projects? Because, according to their benchmarks, it gives good results while being much faster [5]. In short, this is a significant new algorithm that has broad appeal in the open-source coding community. There is absolutely no question that MurmurHash is notable; the only question is why Filth has tried twice to delete any mention of it from Wikipedia. 208.80.104.2 (talk) 22:24, 14 September 2009 (UTC)[reply]
None of the 16,000 hits are, to my knowledge, reliable sources and the [6] is just a self published website. Anybody could write anything in there; it doesn't even prove that murmurhash exists or that it is any good.- (User) Wolfkeeper (Talk) 22:38, 14 September 2009 (UTC)[reply]
I would be fascinated by your explanation for how 16,000 web pages mention an algorithm that "doesn't exist". It would also be great if you could explain how you can personally dismiss them all despite the fact that academic papers and open-source project sites are accepted as reliable sources in the other articles on hashing algorithms. It would be just as interesting to find out why anyone would even imagine that MurmurHash doesn't exist when in fact they can see it for themselves right here on Wikipedia. It's an algorithm, not a mountain range: proof of existence is a rather different question. Alternately, perhaps could consider that you are overstating your case to the point of self-parody and should instead focus on a credible argument for exclusion. 208.80.104.2 (talk) 14:36, 15 September 2009 (UTC)[reply]
The kind of level I would want for this would be a mention in communications of the ACM or something similar.- (User) Wolfkeeper (Talk) 22:48, 14 September 2009 (UTC)[reply]
Oh, you mean like Pearson hashing, an obsolete 8-bit algorithm that the ACM covered 20 years ago? Thank you for sharing your personal standard, but I fail to see why anyone else should accept it. 208.80.104.2 (talk) 14:36, 15 September 2009 (UTC)[reply]
The standard is here: WP:RS. Please limit your discussion to how the article does or does not meet this standard, and completely cut out the personal attacks, it reflects badly on both you and the article.- (User) Wolfkeeper (Talk) 15:47, 15 September 2009 (UTC)[reply]
Very weak keep The MurmurHash algorithm may be notable, just stroll through the web pages. It is used by different notable projects, mentioned by the NIST and so on. The problem is the article, which does not talk about, how the algorithm works what makes it different from others and so on. We only find some prose about the version history spiced with a little bit of advertising. --Kgfleischmann (talk) 08:28, 16 September 2009 (UTC)[reply]


Hello, I am the author of MurmurHash (Austin Appleby). MurmurHash is "brand new" as far as hash functions go - people have been coming up with them for decades - and has seen only a few references in academic material so far. That said, it is in fact a significant improvement over previous algorithms and has been embraced by a number of open- and closed-source projects - both the ones mentioned in the article, and in internal projects at Google, Microsoft, Amazon, Yahoo, and no doubt others. A number of these projects discovered MurmurHash through browsing Wikipedia.

I did not create the Wikipedia article and I've avoided making any significant edits to it due to being unsure whether that was acceptable behavior here, but if the primary complaints here are that the article is insufficiently meaty then I'd be more than happy to elaborate on why MurmurHash happens to be considerably faster and more effective than previous hashes - the topic would touch on aspects of modern processor pipelines as well as statistical tests derived from cryptanalysis.

Given another year or so I'd expect that enough people might encounter MurmurHash to earn it an article in Dr. Dobbs or Linux Journal or whichever sources are considered sufficiently reliable. In the meantime the algorithm exists in a state much like a mathematical proof - it exists, it has been published, and its properties are easily and objectively verifiable. I am a professional and experienced software developer and not a dedicated researcher, and since MurmurHash was a product of necessity and not formal research it was not presented first via a technical journal nor submitted to peer review before publication. I published it using the most expedient means necessary and assumed that users would eventually either refute or confirm my claims - most all have been confirmed, with a few caveats regarding performance on older chip architectures.

If the above qualities are insufficient to qualify MurmurHash for a Wikipedia article, then I will be rather disappointed by its deletion - the article has proven to be a useful point of reference for software engineers and researchers, and removing it would seem to me to be a step backwards. Aappleby (talk) 22:32, 17 September 2009 (UTC)[reply]

Well, it fails WP:NOTABILITY, and the purpose of the wikipedia is to contain WP:verifiable knowledge ("verifiability over truth"). Right now, it's not verifiable. If it became verifiable then it very probably would become notable and you could get the deletion overturned.- (User) Wolfkeeper (Talk) 23:33, 17 September 2009 (UTC)[reply]
Relisted to generate a more thorough discussion so consensus may be reached.
Please add new comments below this notice. Thanks, \ Backslash Forwardslash / (talk) 23:51, 18 September 2009 (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.
More