Wikipedia:Articles for deletion/ERROL

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. Spartaz ^Humbug! 06:32, 3 December 2011 (UTC)

ERROL

ERROL (edit | talk | history | protect | delete | links | watch | logs | views) – (View log)

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

Every reference on this page is by Yoav Raz or someone who has coauthored with him. Google scholar has 5 hits (three of them linked) for "Entity Relationship Role Oriented Language" and nothing in the first five pages (50 docs) for "ERROL" appears to be related. It does not appear to have been used in any notable systems. In short there appears to be no evidence of notability for this language. Stuartyeates (talk) 07:30, 25 November 2011 (UTC)

Note: This debate has been included in the list of Computing-related deletion discussions. —Tom Morris (talk) 08:58, 25 November 2011 (UTC)

• Weak keepDelete - it's too long and quite repetitive but I can't see any clear reason to delete it. It's more than just an experimental language to illustrate a single point. Admittedly it's unlikely to feature on anyone's list of languages they've heard of (SQL is extremely dominant in the field), but the references suggest it is notable enough to have appeared in a few refereed journal articles, though the number of distinct authors is small. --Northernhenge (talk) 19:59, 25 November 2011 (UTC)
• Comment: Google scholar for the narrow search expression [errol "entity relationship" language markowitz] [1] gives 29 hits. 9 of them are related to the authors of ERROL, Victor M. Markowitz and Yoav Raz. 20 of them are references to ERROL by other authors. The Google scholar search indicated at the page top as a reason for deletion (5 hits) is far from providing a complete citation picture of ERROL; as well it is far from being an indication to the level of interest in ERROL by database query language professionals and many other people interested in convenient access to database information. ERfan111 (talk) 21:49, 26 November 2011 (UTC)

This number is also supported by Arnetminer which includes only notable conferences and journals: The single conference ERROL article has a 26 count. Without RRA (2) and other Dr. Markowitz's citations (3) the number 20+ for (conferences and journals) citations is supported.ERfan111 (talk) 09:04, 29 November 2011 (UTC) — ERfan111 (talk • contribs) has made few or no other edits outside this topic.

• See also Wikipedia:Articles for deletion/Reshaped relational algebra --Northernhenge (talk) 22:31, 26 November 2011 (UTC)
• Comment: I'd also include Talk:Entity_Relationship_Role_Oriented_Language in this AfD, which is an orphaned talk page (which I believe can technically be speedied but should probably be included here). Stuartyeates (talk) 22:38, 26 November 2011 (UTC)
• Keep - The question here is the notability of ERROL. ERROL and its related articles have been published in notable refereed journal and conferences (see references in ERROL). ERROL has won the national computer science award by ILA in Israel. ERROL has been referenced in at least 20 academic publications by authors different from, and unrelated to the ERROL authors (see my comment above). This establishes sufficient notability. ERfan111 (talk) 23:40, 27 November 2011 (UTC)

Is there a reference for ERROL receiving this award? Neither the ILA nor the award appear to have wikipedia pages, so it seems unlikely that winning the award automatically grants notability to the recipients. Stuartyeates (talk) 00:45, 28 November 2011 (UTC)

The only source I know is Dr. Yoav Raz's personal site (resume and ERROL page). Dr. Raz may have/know other sources. Also Dr. Victor M. Markowitz who is a co-recipient should be able to acknowledge this (email?). Both were at the Technion - Israel Institute of Technology (CS) at that time and they should have a record. Also ILA themselves, of course. For me the testimony of Dr. Raz is more than satisfactory, and I see no reason to further inquire. Regarding "unlikely that winning the award automatically grants notability to the recipients" I already have seen this idea when you actively participated in the delition of the article Yoav Raz. But now you seem to be confusing award recipient with the reason for award: They got the award because they created ERROL! I find this argument here very strange. BTW, the notability of Dr. Raz is primarily because of Commitment ordering which you have also tried to delete and rightfully failed. ERROL (together with Reshaped relational algebra which you have deleted as well) is just one of his achievements. ERfan111 (talk) 09:55, 28 November 2011 (UTC)

"For me the testimony of Dr. Raz is more than satisfactory, and I see no reason to further inquire." - if implemented, that would be an uncommonly clear violation of NOR. --Northernhenge (talk) 23:00, 29 November 2011 (UTC)

More time is needed to try to get other evidence for the award. ERfan111 (talk) 07:45, 30 November 2011 (UTC)

Publishing articles in peer-reviewed and citing each other's work is what scientists do for a living. This alone doesn't each of the approximately 2 million computer science papers that have been published notable enough for an article in Wikipedia. Clearly, winning a Turing Award would make both the author and their work instantly notable. Winning a departmental award given only to local faculty probably not. Do your have 1) some background information on the ILA award and 2) can you cite any papers that discuss ERROL in more depth than merely citing it in the bibliography? —Ruud 10:42, 28 November 2011 (UTC)

As a person involved with Academia you should know that honest unbiased scientists cite articles only when relevant, and not "citing each other's work is what scientists do for a living". I hope that as a young scientist yourself you do not adopt such ugly cynical practice, which no academic institute would endorse. Thus accusing 20+ articles (of authors different from the ERROL authors, and unrelated) is indeed an unjustified insult to all these authors.

Of course ERROL is not among the most notable (Turing award - a very high tree), but you well know that most scientific articles here are not backed by Turing award or alike. The ILA award is a National Israeli award. Also not the most notable, but notable among all Israeli IT professionals and academics at least. ERROL's notability is not confined to Israel alone, as the 20+ citations show.

About ILA award: I do not know much about it, except that computer science is among few categories in which it is given. You can google ["ILA award" Israel] and see some related entries. Hebrew search probably will provide more entries.

"can you cite any papers that discuss ERROL in more depth than merely citing it in the bibliography?": In other deletion discusion of WP article connected to Yoav Raz, where like User:Stuartyeatesyou have been active in, it already was said that most citations (in respected articles) do not elaborate on the cited articles, and only text books do this. I have not seen an elaboration on ERROL, but only on its foundation Reshaped reletional algebra.

ERfan111 (talk) 12:12, 28 November 2011 (UTC)

I did not intend my comment to be interpreted in such a cynical way. I meant to say that citing is so common that it does not automatically imply scientific work is notable if it has received a few citations, not that citing is in any way a morally deplorable action. Pretty much all published work receives some citations in other work, but these vary in weight from acknowledging prior work ("a colleague at a different university did related work in [x]") to "[x] is the seminal paper this text book is based on" followed by an extensive discussion of the original paper n follow-ups by others. The context of citations is very important here (also see a few of the common criticisms of the h-index), even though the absolute citation count here doesn't seem to be very high in the first place.

For example, all the references I've included at Generalized algebraic data type have received an adequate number of citations, but likely none of those works (including the theory and algorithms developed therein) is notable enough to warrant an individual article. It's only the subject take as a whole - comprising of several papers by several authors at several universities - that is notable. Even though, and contrary to your claim, most of these articles are discussed at more depth than only one or two sentences in other articles. Encyclopedias are supposed to summarize scientific knowledge, not to republish everything that has already been published in academic journals. CCS is an example of work that obviously deserves a separate article, CO is a borderline case and RRA, in my opinion, did not receive enough attention in academia to warrant a separate encyclopedic entry. I'm not yet convinced where on this scale ERROL might fall, clearly somewhere between CO and RRA, but that could still become either a weak keep or a delete. —Ruud 14:04, 28 November 2011 (UTC)

"Pretty much all published work receives some citations in other work" - I disagree: An article is cited only if of interest to other work (assuming no abuse and cynicism). Quite many get no citations at all, or very small numbers. Randomly pls check in arnetminer.

[ERROL] "did not receive enough attention in academia to warrant a separate encyclopedic entry" - I disagree. What is the "right" number and type of citations? ERROL and RRA are unique and special and provide a new, user-friendly approach to data management languages. They are described in several articles cited in ERROL, which are cited in the relevant articles in their area. Not always notability can be measured just by the number and type of citations, as common in Academia. Though not known to be implemented beyond its prototype it has received a national award and it is a "heavy enough" subject to warrant a WP article(s). Deleting it will be a disservice to WP users and closing for them an important aspect of data management. ERfan111 (talk) 14:47, 28 November 2011 (UTC)

• Comment: Most academic articles are incremental in their contribution. The ERROL approach is a jump, a new concept. A new Idea needs to be sold, also in Academia, and the authors did not seem to make much effort here (Dr. Raz did not submit his accepted to DKE major ERROL article "since was too busy"; see his ERROL page). Thus ERROL needs to be judged by objective experts in Data management and Database languages who can properly evaluate it (WP should have access to such), rather than by a citation count and arbitrary threshold. ERfan111 (talk) 18:17, 29 November 2011 (UTC)

Judging the merit of ERROL need to happen in academic journals, not here on Wikipedia. It seems ERROL has had little impact in the former (unless you can point me to a few papers that I've overlooked). —Ruud 20:28, 29 November 2011 (UTC)

20+2(+) independent citations is not "little" though not "huge". ERfan111 (talk) 07:45, 30 November 2011 (UTC)

• Delete Borderline notable academic work. Does not pass WP:GNG ("topic has received significant coverage in reliable sources that are independent of the subject"). A complete lack of inline references, peacock language and excessive zealotry surrounding these article does not help either. —Ruud 20:28, 29 November 2011 (UTC)

The above comment is a total distortion of facts (a continuation by Ruud on a journey of deletion and tagging with no explanation (by "Neutrality disputed") of at least 13 articles related to Commitment ordering and Dr. Yoav Raz):

1. "Does not pass WP:GNG ("topic has received significant coverage in reliable sources that are independent of the subject")" - No. 20+ direct independent citations (some more indirect by RRA). National award.
2. "A complete lack of inline references" - No. The article uses the legitimate Inline parenthetical referencing. See Wikipedia:Citing sources#Parenthetical referencing.
3. "peacock language" - No. Simple language. Facts and only facts.
4. "excessive zealotry surrounding these article" - No. You may find some fanaticism from me here, but not in the article. Very restrained language, with understatement. Facts and only facts.

ERfan111 (talk) 21:25, 29 November 2011 (UTC)

• Comment - I previously said weak keep. The above discussion has not made me feel any stronger about it. It's a concern that no other significant sources have come to light. --Northernhenge (talk) 22:54, 29 November 2011 (UTC)

The sources known to me are summarized in my comment above. Additional possible one can be the ILA CS award citation and its academic committee, if found. The discussion has not added any new sources, but rather took place to dispute additional claims against the article. ERfan111 (talk) 23:42, 29 November 2011 (UTC)

That's a shame. It would be good to have more articles in this area but the nomination said: "there appears to be no evidence of notability for this language". I now agree. I've changed my position to delete. --Northernhenge (talk) 23:50, 29 November 2011 (UTC)

How many more independent citations beyond 20+ you need to be "sufficient for notability"? ERfan111 (talk) 11:34, 30 November 2011 (UTC)

• Comment: User:Ruud Koot, who is not an expert in the related areas (of ERROL and Commitment ordering), needs to explain why he has tagged with "neutrality disputed" related to Commitment ordering 10 (!) articles at least (he probably followed all links in CO to its utilization). He was asked to explain it in the respective article's discussions pages, but no reply. Such tagging requires a good understanding of the subject. If he cannot explain the tagging with a reasonable explanation, he should disqualify himself here due to bias. ERfan111 (talk) 00:34, 30 November 2011 (UTC)

Ruud also wrongly downgraded the importance ranking of Commitment ordering from "High" to "Low" though he is not an expert in database concurrency control. There is no doubt about the high importance of CO. Changing importance should be done only by a person with a very good knowledge and understanding in the subject. ERfan111 (talk) 13:21, 30 November 2011 (UTC)

I'm not sure that User:Ruud Koot has to do anything, unless there's a policy I'm overlooking. I'm not sure that I've seen all ten, but I've seen several and they seem fine to me. (List them if you like and I'll go through them.) "neutrality disputed" is a perfectly reasonable tag to use when a WP:SPA is active on the article. Stuartyeates (talk) 06:47, 30 November 2011 (UTC)

Again you are confusing the facts: I, ERfan111, am a WP:SPA. Ruud has put the 10+ tags for Commitment ordering done by User:Comps. He is not a WP:SPA. I see he has been involved with multiple articles, especially his work in Database, reorganizing it and increasing its size from about 26k to >100K. Thus Ruud's tagging is for another reason, not WP:SPA, and I want to know why. I suspect a bias of Ruud against Dr. Yoav Raz who is the author of Commitment ordering and a co-author of ERROL and RRA. ERfan111 (talk) 07:29, 30 November 2011 (UTC)

Please don't say what I may or may not be doing without direct knowledge. If you are going to make claims against User:Ruud Koot I suggest that you give diffs or point to direct evidence. Stuartyeates (talk) 07:45, 30 November 2011 (UTC)

You said the reason is WP:SPA, and this is clearly incorrect regardless of any knowledge (?) you may have, as I show above. I do not know what diffs you are referring to, but anybody can see Ruud's "neutrality disputed" tag for Commitment ordering (seen at edit mode) in Concurrency control, Serializability, Global serializability, Distributed concurrency control, Software transactional memory, Snapshot isolation, and more. The most interesting is his tag in Two-phase commit protocol#Tree two-phase commit protocol: Commitment ordering is not mentioned there and irrelevant, but the word "commit" appears all over, and the section is referenced to Yoav Raz's article and the textbook Weikum and Vossen 2001. Weikum and Vossen 2001 explicitly say their respective book section is based on that Yoav Raz reference. So, just a "naive" mistake by Ruud to put there irrelevant tag? It also clearly shows he does not understand the subject, confusing "Commit protocol" with "Commit ordering. ERfan111 (talk) 08:19, 30 November 2011 (UTC)

I did not say the reason User:Ruud Koot tagged those articles was because of an SPA. What I know of User:Ruud Koot's motivations is solely what can be deduced from public actions on WP and to the best of my knowledge they have not disclosed motivations explicitly. I put forward a possible motivation which seemed reasonable to me. Stuartyeates (talk) 08:41, 30 November 2011 (UTC)

Ok. It is clearly not SPA. So, I would like to hear from him his reasons.ERfan111 (talk) 08:59, 30 November 2011 (UTC)

User:Ruud Koot also took an active part in the deletion of the articles Yoav Raz, History of commitment ordering, Reshaped relational algebra, related to Yoav Raz, as well as the attempt to delete Commitment ordering, as the discussion records show. All these "achievements" of Ruud here and above took place during the last three weeks. ERfan111 (talk) 08:27, 30 November 2011 (UTC)

The deletion of Yoav Raz should be appealed: His Commitment ordering, B-tree concurrency control, and Dynamic two-phase commit protocol works are detailed and referenced in the textbook Weikum and Vossen 2001, the latest text on transaction concurrency control (ERROL is another area). This fact which validates his notability, I see has been overlooked in the Yoav Raz deletion discussion. ERfan111 (talk) 11:06, 30 November 2011 (UTC)

The process for this is laid out at Wikipedia:Deletion review. Stuartyeates (talk) 17:32, 30 November 2011 (UTC)

• Comment on citation: Some academic areas are more popular (often temporarily) than other. This means that they enjoy a high number of publications. If they are "crowded", works get closer in ideas, and thus multiple citations and comparisons exist to show distinction for publication acceptance. If a cited subject is not close or similar, only a reference is sufficient, with no elaboration. A very hot subject today is, for example, Quantum computing with an explosion in publications. Completely different and modest is the area of database languages (used to be hotter in the 1960s-70s, at early databases, before SQL and ERROL). Thus much less total citations. From an independent author point of view, ERROL, being different and apart from almost all database languages, typically needs no elaboration and comparison to the author's (different) work when cited. ERfan111 (talk) 13:44, 30 November 2011 (UTC)

The situation is different in textbooks. They intend to elaborate. Unfortunately I do not know of a book on database languages in the last few decades. I'm sure that such a book will dedicate to ERROL more than an adequate coverage, but such a book is quite unlikely due to lack of sufficient interest beyond few researchers, as far as I can see. Regular modern texts cover few languages in use like SQL and possibly languages for the implemented data models Object model and possibly XML. ERfan111 (talk) 06:00, 1 December 2011 (UTC)

If you're not aware of any database language books published recently, I suggest that you start here. Stuartyeates (talk) 06:18, 1 December 2011 (UTC)

• Delete. Preceding discussion pretty much nails it. To establish notability under WP:GNG, you need reliable independent secondary sources. There aren't any. Notability here means other people not connected with the product thought it was worth writing about, not just the author thinks it's pretty cool and wrote a lot about it. Msnicki (talk) 01:08, 1 December 2011 (UTC)

You ignore what I wrote in the comment above. Articles cannot be too long and do not detail beyond citation if unnecessary to establish their own uniqueness relatively to the citation. 20+ citations in a quite dormant area is sufficient for establishing notability. ERfan111 (talk) 06:00, 1 December 2011 (UTC)

What part of the need for independent sources to establish notability seems confusing? You don't have them. Stick a fork in it. It's done. Msnicki (talk) 06:14, 1 December 2011 (UTC)

People deserve to know about languages beyond SQL and some other implemented languages, especially if a language presents a new idea with potential, and is not some superficial idea, but rather covered with thorough cited research work. People delete the article here in this discussion by far too simplistic criteria, beating the purpose of WP. Kind of automatic. ERfan111 (talk) 06:27, 1 December 2011 (UTC)

People deserve to know about languages beyond SQL and some other implemented languages [...] I invite you to look over Query language, which lists a great many languages other than SQL. Stuartyeates (talk) 07:02, 1 December 2011 (UTC)

Yes, either historic, from the past when database languages were dealt with extensively, or dedicated languages in use. I wonder how many will survive the scrutiny here. Also ERROL is there in blue, and I hope it does not turn red... ERfan111 (talk) 07:20, 1 December 2011 (UTC)

Independent sources are the selective committees that selected the ERROL articles for publication: The committees of the Entity Relationship 1983 conference and the Journal of Systems and Software. ERfan111 (talk) 08:47, 1 December 2011 (UTC)

An independent source is the 1984 ILA computer science award citation text and the names on the committee that decided this. I do not have it yet. ERfan111 (talk) 06:38, 1 December 2011 (UTC)

• Delete. Much ado about nothing. Just do it and move on. History2007 (talk) 18:18, 1 December 2011 (UTC)

Right. Just one article among all the many WP is nothing, so why all the ado about its deletion? One article more or less will change nothing. Thus, when you are absolutely board, and find nothing else to do but spending several months writing a WP article, do not be stupid (everybody, not only History2007): Be very careful about notability. Otherwise your article will eventually be deleted. Stay on safe ground. Write only about a very notable subject rather than, for example, about an esoteric database language. Take for example a notable subject like Supercomputing. But an article already exists! So take another absolutely notable subject like China, and write a new article, as History2007 did, about Supercomputing in China. Notability squared! (notability*notability). When using this formula you are completely safe, with a very large number of possible new articles, and a huge contribution to WP. All that you have to do later is to update your article from time to time as the map changes, which will continue to take care of your boredom, while enjoying looking from time to time how other articles are being deleted massively (you may even add your own Delete for the fun, and also feel fulfilled when "protecting" WP from unnecessary, noisy articles). ERfan111 (talk) 06:48, 2 December 2011 (UTC)

• Delete. This article gives WP:UNDUE coverage of one specific entity-relationship language. What would be more appropriate would be a Wikipedia article on such languages more generally, giving appropriate coverage to the several other ones that are more notable than ERROL. As for the 29 hits in an early comment from ERfan (whose name does not exactly convey an image of neutrality): most of them appear to be primarily about other e-r languages and tools and only cite ERROL as part of their competition, rather than having nontrivial coverage of ERROL specifically. For instance, the first one, "A graphical data manipulation language", does not even mention ERROL in the main text of the paper, only includes a reference to it at the end in a collection of "nongraphical ER languages". The second one, on SQL/EER, only has one sentence citing ERROL and four other languages. The third one (skipping the ones by the ERROL authors themselves) is the one on QBD*, again includes it in passing as one of five citations for the phrase "fourth generation languages". So I'm also not convinced that there's really adequate depth of third-party coverage. —David Eppstein (talk) 23:01, 2 December 2011 (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.