Talk:Eigenclass model

The first restoration
I seem to have missed the discussion regarding deleting this page and/or redirecting it solely to Ruby. Please don't just delete the page and point it to your favorite programming language. In reviewing deletions and redirects, it would seem appropriate to review the number of inbound links to a page in order to ascertain the contexts in which the page is relevant. I have un-deleted this page. If it is felt that the term 'eigenclass' is particular to the Ruby community (as the term 'eigenclass' was used previously in a mathematical context), a better solution may be to rename this page to something like Metaclass_model or to merge with the Metaclass page.

Wes Turner (talk) 12:44, 7 October 2013 (UTC)

In particular, the relation between code class models and RDF ontological forms cannot be found on any other page on wikipedia, AFAICT. What harm is there to the existence of this page? What sort of quorum was achieved prior to the deletion of this article?

Wes Turner (talk) 12:48, 7 October 2013 (UTC)


 * https://en.wikipedia.org/wiki/Comparison_of_programming_languages#See_also
 * https://en.wikipedia.org/wiki/Rosetta_Stone
 * https://en.wikipedia.org/w/index.php?title=Special%3AWhatLinksHere&target=Eigenclass_model&namespace=

Wes Turner (talk) 13:07, 7 October 2013 (UTC)

The second restoration
I undeleted the page since it seems to me that no reasonable consensus has been reached in the deletion discussion. The default result of a deletion discussion is to keep the article. Redirecting the page to Ruby_(programming_language) does not make sense. The section tells nothing about the eigenclass model, it even does not contain the word "eigenclass" or "model". Moreover, as pointed out by Wes Turner, the links pointing to the article have not been taken into account.

The subject of the article is the common core part of object models of programming languages (or even ontology languages) that stick, at least conceptually, to the everything is an object pattern.

The uniformity is achieved by "eigenclass completion" &mdash; a common refinement by introducing auxiliary objects called eigenclasses. For some languages, this refinement is purely fictitious, for some partially. In Ruby, every eigenclass (except those of immediate values) can be made an actual object to which properties or methods can be bound.

The problem is that there is no established term (in the literature) for the "core part of object model via eigenclasses". There is just "object model" and Ruby's "eigenclass". The term "eigenclass model" is made up. (I agree that the article might give false impression that "eigenclass model" is an established term in computer science.)

The article provides a uniform view of the core part of object oriented programming. This requires a lot of original research, since such a view is not present in the "reliable" literature.

Consider the value obtained by putting various terms into connection.

  OO terms: object, class, metaclass, inheritance, object model, name resolution, linearization.

 Set theory terms: set membership, subset, powerset, non-well-founded, urelement, superstructure.

 Order theory terms: partial order, upset, order embedding, closure operator, monotone map.

 Programming languages: Ruby, Python, Scala, Java, Smalltalk, Objective-C, CLOS, Perl.

 Ontology languages: RDF Schema, OWL.



There is no other page on Wikipedia that shows how these terms are interrelated.

My solution is to preserve the page with a clear original research mark, emphasized by a red line on the left side of the page.

If you still cannot stand the existence of the page, please consider &mdash; instead of any redirection &mdash; writing a stub page with an epitaph that would point to some old revision of the page.

Hundblue (talk) 20:58, 9 October 2013 (UTC)




 * This requires a lot of original research, since such a view is not present in the "reliable" literature.
 * My solution is to preserve the page with a clear original research mark, emphasized by a red line on the left side of the page.


 * This is not how Wikipedia works, as that article clearly states:


 * This page in a nutshell: Wikipedia does not publish original thought: all material in Wikipedia must be attributable to a reliable, published source. Articles may not contain any new analysis or synthesis of published material that serves to advance a position not clearly advanced by the sources themselves.


 * If you still cannot stand the existence of the page, please consider &mdash; instead of any redirection &mdash; writing a stub page with an epitaph that would point to some old revision of the page.


 * Again, this seems to be based on a fundamental misunderstanding of the purpose of Wikipedia. If you're still unclear on this, start with these:


 * NOT
 * But it's true!
 * Conflict of interest


 * chocolateboy (talk) 18:26, 10 October 2013 (UTC)

A proposal: Redirect to Metaclass
I agree that the article violates the NOR principle but I doubt that erasing the page contents makes Wikipedia better. The following things should be considered:

  The deletion discussion was closed based on a single contribution. No real discussion took place.

 What is the field of interest of the Wikipedian who   opened the AfD discussion: Chocolateboy,  closed the discussion: Black Kite?  Are these editors qualified enough to make the right decision as to what should be done to this particular article?

 What are the arguments for discarding the articles that violate the NOR principle?



I still think that redirection to Ruby_(programming_language) does not make sense. A far more appropriate solution would be redirect to Metaclass. This section contains basic description of the eigenclass structure of the Ruby object model. Furthemore, some of the present links to the article would retain some meaning if such a redirection is established. (Some not. In particular, any redirection invalidates the link in RDF_Schema.)

Hundblue (talk) 12:49, 11 October 2013 (UTC)


 * No real discussion took place.


 * What's to discuss? You already "agree that the article violates the NOR principle". If you disagree with the policy itself, neither the deletion discussion nor this talk page are the place to challenge or overthrow it.


 * What is the field of interest of the Wikipedian who...


 * Our interest is in improving Wikipedia. Rather than wasting your time fighting or impugning us, I suggest (again) that you take the time to read the linked articles, which make it abundantly clear that content is included in Wikipedia on the basis of verifiability (by reliable, third-party sources), not on the basis of its (alleged) "truth" or the (alleged) credentials of its contributors.


 * A far more appropriate solution would be redirect to Metaclass.


 * The Metaclass article was almost as bad as this article. I've cleaned up parts of it, but it's still almost entirely unsourced.


 * If you want to redirect the article to that section, feel free, but please note that Wikipedia is not a platform for the promotion of self-published research.


 * chocolateboy (talk) 16:49, 11 October 2013 (UTC)


 * Hundblue, can you link to some examples of academic literature or similar that use Eigenclass outside the context of Ruby? I couldnt quickly find anything substantial regarding OWL and RDFS using the term Eigenclass.
 * The redirect to Ruby is not ideal because that section doesnt mention 'Eigenclass'; the closest it comes is mentioning 'metaprogramming'.
 * I can see it would be good for Eigenclass to redirect to Metaclass, as the reader is interested in a specific concept, and they are far beyond wanting to read about the type of information found in the Ruby article. The page Metaclass needs to explain the Ruby concepts better. John Vandenberg (chat) 02:06, 16 November 2013 (UTC)


 * The OO term 'Eigenclass' is presently used only in the context of Ruby so that I cannot provide the requested links. I made a rewrite of the version from 20 September of the Metaclass section. The current version is presumably more conformant to the NOR principle. Please feel free to redirect the Eigenclass model page to Metaclass. (I don't feel free enough.) Hundblue (talk) 09:03, 21 November 2013 (UTC)
 * The OO term 'Eigenclass' is presently used only in the context of Ruby so that I cannot provide the requested links. I made a rewrite of the version from 20 September of the Metaclass section. The current version is presumably more conformant to the NOR principle. Please feel free to redirect the Eigenclass model page to Metaclass. (I don't feel free enough.) Hundblue (talk) 09:03, 21 November 2013 (UTC)

AGF?
Chocolateboy, I must admit that I have difficulties with assuming good faith in your recent edits. You just reduced all my contributions to Wikipedia to two sentences (plus a few links) and you well know it. Are you feeling better now? Deleting pages or their parts seems to be your favorite activity on Wikipedia. You are not much concerned about the consistency (and the value in general) of what remains after your edits.

You write that I should feel free to redirect the article, but presently, I do not feel free to make any contributions to Wikipedia. It appears to be a waste of time.

Thank you for your generosity (maybe it is rather a discipline) of not erasing the term "Eigenclass model" from Wikipedia so that old versions of the page are available.

Hundblue (talk) 18:48, 11 October 2013 (UTC)

Third Restoration
To be clear: I am very -1 on removing this page.

There are common roots to these data/language modeling patterns which I feel are being disrespected by redirecting this page solely to Ruby (a language which is generally supported by a very open and accepting community). Again, please stop vandalizing this page by redirecting to Ruby.

The term 'eigenclass' does not occur anywhere within the page which you keep redirecting to. Eigenclass does seem to be a common term in the Ruby Community. Again, please stop vandalizing this page by redirecting to Ruby.


 * https://en.wiktionary.org/wiki/eigenclass
 * http://books.google.com/books?id=irt7nOFaR3sC&lpg=PA32&ots=LaA7GbL7SH&dq=eigenclass%20-ruby&pg=PA32#v=onepage&q=eigenclass%20-ruby&f=false
 * search: "eigenclass" -ruby
 * search: "eigenclass model" -ruby

Wes Turner (talk) 11:08, 28 January 2014 (UTC)


 * Reverted again for the reasons stated above. This has been discussed before and the consensus is that it's "a WP:COATRACK for a massive expansion of the concept beyond what is found in reliable sources". You're welcome to add content that complies with Wikipedia policy. This orgy of original research obviously doesn't, as has already been acknowledged ("This requires a lot of original research", "I agree that the article violates the NOR principle" &c.).


 * chocolateboy (talk) 16:17, 28 January 2014 (UTC)


 * If there is rework to do, there must be a startingpoint ... but how to give this topic a chance if it is reverted each time someone tries to reuse the old content. The admissibility question is does this topic is admissible ? I think yes. Would you accept the restauration if some of the original work was removed ? TomT0m (talk) 14:32, 17 April 2015 (UTC)

How to request an unmerge ?
Moved from some other place Hi, I'm a little lost on the forum jungle here, so pardon me if I'm not at the right place to ask :)

I'd want to request the unmerge the article Eigenclass model, whose content seems simply to have been lost in the fusion process (the term does not appears at all in the target article, which is what personally annoys me the most in fusions on Wikipedia.

The rationale behind the fusion was not about the content of the article, but more on his title. His name seemed to be attached to a particular implementation of the concept, but the concept is real and has an history. Plus there is a lot of book in which the term is used. So this is not a source problem, nor a notoriety one. I don't really understand why the article was merged, especially because this implied a loss of content, and the redirect does not really make sense as is. The terms power types Powertype (UML) and singleton class seems to exists in literature in the metaclass context. So there if it's just a question of rewording and neutrality, this is totally doable.

I just see the issue has been controversial and a suubject of edit warring, so I want to do this especially carefully and right.

I want to assure I was not part of the previous controversy, and that I discovered the subject on a connex article I'm working on Metaclass (semantic web). This article's content seems very interesting, even if some other more generic like metamodeling are as well. TomT0m (talk) 12:25, 17 April 2015 (UTC)
 * You would first have to establish that the concerns raised Articles for deletion/Eigenclass model have been addressed. -- TRPoD aka The Red Pen of Doom  13:08, 17 April 2015 (UTC)
 * Where would I do that ? I think I did argument a little bit here already :) . I also think the article, if merged, should better be merged into metaclass or something. It's akward to merge it in ruby as the main subject is metaclasses, or "singleton class" or something. I notify the other involved in the merge.

Addition : This paper (not really a published one it seems, but seems quite complete) cites "the eigenclass model", outside of the ruby community, at ms research lab http://research.microsoft.com/pubs/226237/Luca-Cardelli-Fest-MSR-TR-2014-104.pdf and cites (Perrotta 2010) as a source : P. Perrotta. Metaprogramming Ruby. Pragmatic Bookshelf, 2010.