Wikipedia:Articles for deletion/Google distance

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 Merge. Rlevse 21:52, 3 March 2007 (UTC)

[edit] Google distance

Google distance (edit|talk|history|links|watch|logs) – (View log)

The paper that defined this wasn't published in a peer-reviewed journal, nor, as far as I can tell (from citeseer/citebase) were any of the three article which cite it (and two of those are by the author of the original paper). It has a reference from New Scientist (here, which is offline, see archived version) and one from Baseline (here, which I've never heard of, and which may have been written by the author of the original article as well). I believe that the lack of references from reliable published sources independent from the author of the article which originates the term indicates that this isn't a notable term. Sopoforic 03:19, 24 February 2007 (UTC)

• delete It seems more a definition which would be better in wikitionary or whatever it's called. I can't see any particular information in the article so I can't see any reason to keep it. Mike 15:41, 24 February 2007 (UTC)
• Delete- irrelevant and fails to meet Wikipedia policies —The preceding unsigned comment was added by Astrotrain (talk • contribs) 16:40, 24 February 2007 (UTC).

• Weak delete. Vitanyi is an established scholar; I'm not concerned about the lack of peer review yet. But though his paper has attracted around 20 citations in Google scholar, it's not clear how many if any of them are about this "Google distance". A better source would seem to be The Google Similarity Distance by the same authors, but that's also a preprint with no citations yet. It seems premature to consider this encyclopedic. —David Eppstein 03:16, 25 February 2007 (UTC)

• Agree with Bejnar: Merge into Semantic relatedness, Semantic similarity, or both (after those two are also merged). —David Eppstein 22:50, 1 March 2007 (UTC)

• very Weak keep The present evidence for N is that New Scientist thought this worth an article. Many items in this field are now published only in arXiv, even by the very best scholars. But since this isn't peer-reviewed publication, they would have no be established by citations of the articles from unquestionable sources. All the alternatives to conventional peer review make use of subsequent citation as the test for quality--which is reasonable because such citation is from the same group from whom the reviewers would have come. I would consider one additional one to New Scientist enough. I think if necessary this might prove to be a good test case for how to deal with these non-conventional quality control systems, but it would not do so until it has been widely cited in convention publications. I suggest re-creating it at a later date with the references that will have accumulated. DGG 07:55, 25 February 2007 (UTC)
  • Comment: I agree that enough citations, even from non-peer-reviewed sources, would make this worth keeping. I'm only just on the side of deleting this, which is why I mentioned everything I could find in the nomination; a few extra (reasonable) sources would make me think this ought to be kept. It is written (somewhere, probably, I think) that things published by known scholars on e.g. their own webspace should be considered also as probably reliable. I just think that this probably isn't notable enough, yet. --Sopoforic 12:38, 25 February 2007 (UTC)
  • Comment I found this announcement for a refereed paper that's supposed to appear next month. I also found this paper, which defines a different quantity as the "google distance". With two substantially different definitions wandering around, it may be too soon to write about this concept for Wikipedia. DavidCBryant 17:42, 25 February 2007 (UTC)
• Keep - With all due respect the arguments for deletion strike me as weak.

1. The article provides more than a definition by providing the algorithm used for measuring Google Distance and this is of some interest. What's provided may be insufficient but this merely justifies a Stub tag and not deletion.
2. I'm just flummoxed by the suggestion that this concept is irrelevant since the way Google measures website popularity and related notions has been in the news and the subject of lawsuits.
3. If there is confusion between the concepts of "Google Distance" and "Google Similarity Distance" then that should be cleared up. But I don't see how that in itself is grounds for deletion.
4. The article mentioned by David Eppstein is in the IEEE Transactions of Knowledge and Data Engineering , The Google Similiarity Distnce

Also, four articles came up on the ACM digital library in regards to "google distance", and, what can I say, found over 800 hits when I googled "google distance". There is no question that the article needs to be improved, but I'm not seeing how any of these are grounds for deletion. Could something cite the specific Wiki policy that they think is relevent here? A B Carter (talk) 17:24, 25 February 2007 (UTC)

The policy I was concerned about when I nominated it was WP:N. Specifically, the part about being the subject of multiple independent non-trivial works. Granted that a google search for "google distance" yields a few hundred results, but a fair number of those are unrelated, and many of the related ones are from blogs and such. I agree that the case for deletion is pretty weak--I just also think that the case for keeping the article is pretty weak. Basically, especially given how happy the media is to report on google, I'd expect that if something like this could be called notable, a couple of newspapers/magazines (in addition to new scientist) would have mentioned it. That may have happened, and, if so, we ought to keep the article; however, I couldn't find such mentions.

Incidentally, as far as I can tell, this Google distance isn't Google's celbrated sorting algorithm that they use (i.e. pagerank), but some other thing which happens to use google as a tool. So, I don't think that the notability of google necessarily implies that this is notable. --Sopoforic 01:50, 26 February 2007 (UTC)

Agreed, I was initially confused on this point. I would also agree that as a consequence notability is an issue, though I'm still for a weak keep. A B Carter (talk) 02:34, 26 February 2007 (UTC)

Comment "… some other thing which happens to use google as a tool." It's a measure of lexicographic similarity. It's one of the first attempts to use the actions of a large number of individuals (their choice of words for web pages) as the basis of artificial intelligence. If you haven't looked at the article recently, you might want to read it again. DavidCBryant 13:00, 27 February 2007 (UTC)

• Delete lacks notability. SakotGrimshine 20:27, 26 February 2007 (UTC)
• Keep Merge per Bejnar. Many references to "google distance" in google scholar. This is a rather similar case to Wikipedia:Articles for deletion/Eurolinguistics. John Vandenberg 13:35, 27 February 2007 (UTC)
  • Not all that convincing. I only got 19 references of which 6 referred to articles by Vitanyi, 3 to "google distance learning", 2 to Poesi which appear to offer a different notion of "google distance" and 5 of unknown origin. I'm still a weak keep but this offers no additional support, in fact, it kinda weakens it for me. A B Carter (talk) 19:52, 27 February 2007 (UTC)
• Merge into Semantic relatedness. This is not a new idea, "semantic distance" has been measured for years using a variety of different algorithms. See for example [1]. This does not deserve its own article. --Bejnar 22:36, 1 March 2007 (UTC)
  • I think we have a winner! Merge is an excellent option. The entry is clearly on the cusp: I wouldn't have given it an entry but now that there is one I'm reluctant to simply delete it. A merge allows us to keep the contribution in some form; if later research supports it it can always be split off into its own entry. A B Carter (talk) 23:06, 1 March 2007 (UTC)
• Merge seems an excellent choice. There is not enough here for a separate article.NBeale 10:52, 2 March 2007 (UTC)
• merge into something and leave a redirect to that page. Reason: reading the article I actually found what is the formula. It is more appropriate as an example in an article about semantic connectedness or something like that. Having an article by itself probably is not justified, but erasing the whole info?..:Dc76 00:13, 3 March 2007 (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.