Wikipedia:Articles for deletion/SLYNUX
From Wikipedia, the free encyclopedia
This page is an archive of the proposed deletion of the article below. Further comments should be made on the appropriate discussion page (such as the article's talk page or on a Votes for Undeletion nomination). No further edits should be made to this page.
The result of the debate was KEEP--Doc (?) 23:19, 27 September 2005 (UTC)
[edit] SLYNUX
Advertising (and poorly-written advertising at that...) - Delete. CLW 10:24, 20 September 2005 (UTC)
- Weak Keep Seems to be a valid linux distro, garners 136,000 (ok, 134 actual) google hits. Just needs some serious cleaning. Anetode 10:51, 20 September 2005 (UTC)
- Comment: There are only 134 uniques in Google. The "136,000" number means nothing (even though Google does show it). --rob 11:03, 20 September 2005 (UTC)
- Well, technically it means that somewhere out there are about 135866 pages that replicate those 134, which means something in and of itself. Not arguing about its notability, though, just the 'Google not showing nothing' thing. :) Anetode 11:23, 20 September 2005 (UTC)
- Well, you're sort-of right, and sort-of wrong. The real problem, is some web servers have an infinite namespace. Meaning, that every time Google requests a path from a domain, it gets a "valid" response, and content is returned (but it's the same page, over and over). There's no duplication whatsoever on the original web site, but Google "counts" each and every url, which produces the same result. Depending on how a site makes it's links this can produce absurd hit counts. It's like a teacher taking attendance, and every time the teacher calls a name, the same kid yells "present", no matter what name is called. As well, Google will even count urls it never even *attempted* to fetch, simply because there was a link to them, and the linking text had the magical keyword. These snippet-free url's literally don't exist, but do get counted. In fact, here's what's funny, domains with non-existent/invalid TLDs get counted (probably not in this case) --rob 12:37, 20 September 2005 (UTC)
- Once again, Google never displays more than 1000 results, and the 134 unique ones are sorted out of the 1000. For example, a search for sjgames (Steve Jackson Games) finds 134 unique hits (of 450.000), and a search for mistress finds 885 (out of 11.200.000). If you don't believe me, find me a keyword which receives more than 1000 unique hits. - Mike Rosoft 06:00, 21 September 2005 (UTC)
- Actually, your twisting things. The first thousand, contains URLs from the same web sites, as those past the first thousand. The sites near the end tend to be the non-unique ones (redundant with earlier one). So, if the hit count is well below one thousand, it's likely those past the 1000-point were just redundant; and your distinction is a meaningless point, not to be worried about. Now, if it gets close to the 1000-mark (say 900+), then sure, there could well be more than a 1000 real hits. Also, you ignore, that Google counts as hits, URLs that are invalid, and that it never even *tried* to fetch. The snippet-free listings, tend to be at the end, and would presumabley be in the post-1000 group. Anything past the last verified, unique hit in Google's results, is pure speculation. No approach will find all web pages, since Google doesn't even get all web pages. The purpose of a Google test is to get a *verified* count. The fact the actual count could be higher, is always true, and pointless to mention. I've seen a case, where over 1,000,000 hits came from a *single* domain, which shows how worthless the non-unique hit count is. --rob 06:49, 21 September 2005 (UTC)
- Once again, Google never displays more than 1000 results, and the 134 unique ones are sorted out of the 1000. For example, a search for sjgames (Steve Jackson Games) finds 134 unique hits (of 450.000), and a search for mistress finds 885 (out of 11.200.000). If you don't believe me, find me a keyword which receives more than 1000 unique hits. - Mike Rosoft 06:00, 21 September 2005 (UTC)
- Well, you're sort-of right, and sort-of wrong. The real problem, is some web servers have an infinite namespace. Meaning, that every time Google requests a path from a domain, it gets a "valid" response, and content is returned (but it's the same page, over and over). There's no duplication whatsoever on the original web site, but Google "counts" each and every url, which produces the same result. Depending on how a site makes it's links this can produce absurd hit counts. It's like a teacher taking attendance, and every time the teacher calls a name, the same kid yells "present", no matter what name is called. As well, Google will even count urls it never even *attempted* to fetch, simply because there was a link to them, and the linking text had the magical keyword. These snippet-free url's literally don't exist, but do get counted. In fact, here's what's funny, domains with non-existent/invalid TLDs get counted (probably not in this case) --rob 12:37, 20 September 2005 (UTC)
- Well, technically it means that somewhere out there are about 135866 pages that replicate those 134, which means something in and of itself. Not arguing about its notability, though, just the 'Google not showing nothing' thing. :) Anetode 11:23, 20 September 2005 (UTC)
- Delete, per comment by Rob. Usrnme h8er 11:13, 20 September 2005 (UTC)
- Delete I didn't nominate it because I wasn't sure if linux distros automatically warranted an article, but that not being the case, ditch it...I got 126 "real" google results (makes it much easier when you go 100 results per page)...the 100,000+ result is just repetition it seems Paul 13:08, 20 September 2005 (UTC)
- Just counting search result totals is not enough. It is important to read the pages that the search finds. Moreover, whilst Linux distributions do not automatically warrant an article, the WP:CORP guidelines for commercial products can be reasonably employed to determine whether a distribution meets the notability criteria as if it were a commercial product. Furthermore, the notability litmus test for software that I outlined in Wikipedia:Articles for deletion/ExamDiff (second nomination) is whether people, independent of the author, have written non-trivial works of their own that focus upon the software in question.
Actually reading the pages that the search finds, I find independently written news coverage of this distribution in three separate news publications, including one of India's national newspapers. This satisfies both the WP:CORP criteria for a product and the software notability criterion that I outlined. Keep. Uncle G 18:17, 20 September 2005 (UTC)
- Yes, obviously, one has to do more than count results. Of the stories linked to, just 2 are English. Both of these are "independent" but, in reading them, it's fairly clear no analysis was done, and the author merely reported what they were told on technical matters. So, these articles were trivial. Given that the size of the enterprise isn't what garners notability, I think what we need to find, is if this distro is a substiantially different and unique product. If SLYNUX is one of group, of barely differnt distros, it's not worthy of a topic. If it's Windows-like interface is special, and noted, that may make it unique (though it's hardly the first to make such a claim). The issue with GNU license is, is one can take another's GNU Linux distro, make small changes, and call it your own. Not all of these warrant articles, since not all are truly unique entitities. So, I'm still on the fence. --rob 18:47, 20 September 2005 (UTC)
- To counter the bogus argument from unique Google hits, weak keep. - Mike Rosoft 06:00, 21 September 2005 (UTC)
- Keep - Obviously, the original hit count in Google was wildly silly and absurd. But, I'm going to make my decision based on something different from that, and different from the technical signifance of the product (which isn't supported). I think there's a human interest story, which was covered, and is probably notable, at least in India. If the same thing was made by a 30-year old from Silicon Valley, it probably wouldn't be worth mentioning. Also, I think as articles related to this are created and put up on AFD, they can quickly be re-directed to this one point (e.g. Kishore Prakash (AFD), MarileSoft (speedied), and Sarath Lakshman (no attempt yet)). --rob 07:33, 21 September 2005 (UTC)
- Keep, it's verifiable and wiki isn't paper. It doesn't come under any point at What Wikipedia is not. ··gracefool |☺ 18:32, 25 September 2005 (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 an undeletion request). No further edits should be made to this page.
