Talk:List of important publications in computer science
|This page was nominated for deletion on 2006 December 1. The result of the discussion was no consensus.|
|List of important publications in computer science received a peer review by Wikipedia editors, which is now archived. It may contain ideas you can use to improve this article.|
|This article is of interest to the following WikiProjects:|
|Material from the associated project or article page was split to other pages. The page histories of the associated projects or article pages now serve as the attribution history for part of the contents of those pages. Please see the events in question for more details:
- 1 Programming Languages?
- 2 Primality Testing?
- 3 Page too long
- 4 The need for SOME references...
- 5 Missing: Operating_Systems:_Design_and_Implementation
- 6 Cryptography
- 7 Human Computer Interaction (HCI)?
- 8 Collaborative Networks?
- 9 Don't alphabetize
- 10 Proposed selection criteria
- 11 Why is Further reading tagged?
- 12 The problem with this format
- 13 Lists importance
- 14 That's not Programming Languages
- You'll have to be more specific. Which one? The Sieve of Eratosthenes?—greenrd (talk) 16:01, 6 June 2008 (UTC)
Page too long
- I agree this page is too large in terms of both size in kilobytes and overall length. I'm not entirely sure breaking each section is the best approach though. Some of these sections would find themselves on the AfD list almost immediately, not only for the subjective nature (that's a whole other topic...), but their size and relevance. I do think, however, that some of these sections can be consolidated or removed; this would create a more organised article and facilitate possible splits. At the very least, "Algorithmic information theory" should be merged with "Algorithms" or "Information theory". Anyone who specialises in one of these fields will probably argue why that section should be separate but, in the end, Wikipedia is intended for a general audience and should be formatted as such. Some possibilities for broader categories (some topics are listed in more than one category because I'm creating a list of possibilities...):
- Theoretical computation—Computability, Computational complexity theory, Algorithms, Information theory, Algorithmic information theory, Formal verification
- Software systems—Operating systems, Databases, Software engineering
- Applied computing—Cryptography, Compilers
- Intelligent systems—AI, Machine learning, Computer vision
- Interdisciplinary applications—Computational linguistics, AI, Computer vision
- Parallel computing, Distributed computing
- Networking—Computer networks, Internet
- Programming languages, Compilers
- Computer architecture, Computer networks (with the "Internet" category being eliminated and its publications added here)
- History of computation
- Computer graphics, Computer vision
- Computer science humour (actually, this should be removed all together. All 3 publications can be placed elsewhere easily, and the element of humour can be noted as a reason why the publication is "important")
- In terms of article/list length, the titles of each application should be smaller! It may only seem like a little thing, but when there are this many sections, it adds up. This might also allow for a teeny (almost negligent) change in size. As for that tag...alphabetizing might not be the best solution (perhaps by date?), but this is certainly in need of organization!
- Me too. This page is in an urgent need of clean-up, but every time I plan to do something I am overwhelmed by the sheer length of this page – and clean-up in one section would only lead to internal inconsistency. Splitting the page would help tremendously: then we could try to tackle the areas one by one. Editors who are interested in specific areas could focus on those articles. Whoever has time and energy, please go ahead and do the splitting!
- If some parts end up being AfD-listed, I think it is only a good thing. It encourages people to find references and to clean-up unsourced claims. We must have some references – I am actually surprised that this page hasn't been deleted yet.
- Regarding the specific categories: We could perhaps try to follow Outline of computer science to identify the categories. But I agree with the above suggestion that we should perhaps have just one category for Theoretical computer science instead of splitting it into Theory of computation vs. Algorithms + data structures.
- One more thing: After splitting, it is also possible to do some merging. For example, "List of important publications in theoretical computer science" could be just a section in Theoretical computer science. The page List of important publications in computer science would contain pointers to all such lists, whether they are separate articles or sections. — Miym (talk) 13:22, 24 August 2009 (UTC)
- I agree with your interpretation. This page can be a sort of parent page for all the subpages, with appropriate links and text. I'll get the splitting process in motion by splitting the Theoretical CS stuff to a new article, and leaving a Main article: ... redirect on this page. I'm not confident in my understanding of the other CS fields to perform the split into a coherent page. But once I start, perhaps people will follow. --Robin (talk) 14:08, 24 August 2009 (UTC)
- Thanks, I have now created List of important publications in concurrent, parallel, and distributed computing. — Miym (talk) 16:28, 24 August 2009 (UTC)
Performed one more split: List of important publications in networks and security. I guess another logical category would be Artificial Intelligence, which can cover all the publications in AI, Machine learning and Computer vision. --Robin (talk) 18:36, 3 October 2009 (UTC)
- As splits have been done I am removing the split tag request. 17:40, 4 November 2010 (UTC)
The need for SOME references...
Without getting into a debate about the inclusion of these "important publications" lists in Wikipedia, we do need to make this article more reliable. Most of the time, a publication with any degree of impact will be deemed that somewhere. Number of citations with respect to years since publication is one good assessor, and there are plenty of sites around that will rank the publications this way for you. When a publication is deemed a breakthrough, the Bible of whatever field, etc., it is mentioned in the first chapter or introductory material of other publications. For most of these older publications, they have been reprinted with a preface explaining the history and notability of the book (or whatever). Citations DO exist, and they are not hard to find. This list will always be subjective to some extent, but at least providing a reference for SOME of these will improve objectivity. —Skittleys (talk) 01:32, 12 July 2008 (UTC)
- Yes, absolutely, some objective standard of sourcing "breakthrough" is needed. I have no idea what this criteria should be. Something along the lines of being mentioned in standard textbooks as the breakthrough paper would work. Merzul (talk) 20:24, 12 July 2008 (UTC)
This list is missing Operating_Systems:_Design_and_Implementation by Andrew S. Tanenbaum. That's THE textbook on OS design. AM I missing something? -- Duesentrieb-formerly-Gearloose (?!) 09:07, 12 March 2009 (UTC)
- Yes, I should think it's a good idea to add it to the page. --SeyedKevin (talk) 03:23, 31 December 2009 (UTC)
Hi, I think "New Directions in cryptography" (W. Diffie and M. E. Hellman) is a quite important paper in cryptography, can someone include this? —Preceding unsigned comment added by 18.104.22.168 (talk) 17:07, 20 January 2010 (UTC)
- See List of important publications in networks and security. — Miym (talk) 17:15, 20 January 2010 (UTC)
Human Computer Interaction (HCI)?
While this was "the rage" in 2005/2006, it is not considered a subfield of computer science in the way that Graphics or Compilers or any of the other headings is. Also, the papers listed are not the most cited, missing the 2005 paper: Singh: Collaborative Networks as Determinants of Knowledge Diffusion Patterns Management Science 51(5), pp. 756–770, 2005. Robert.pless (talk) 04:26, 18 August 2010 (UTC)
- I'm guessing that the point was to alphabetize the sections. I have done that and removed the tag. RockMagnetist (talk) 23:52, 15 November 2011 (UTC)
Proposed selection criteria
A perennial issue on this and similar pages is how to set criteria for the list. I have carefully considered existing guidelines and tried to craft a broad set of policies that satisfy them. I have posted it on the Science pearls talk page. I would welcome your comments. Of course, these guidelines are not intended to be binding for any particular page, but might help you choose your own selection criteria. RockMagnetist (talk) 00:11, 8 November 2011 (UTC)
Why is Further reading tagged?
The references in /* Further reading */ are general references with the purpose of establishing the notability of the list. They seem like a good start. Cybercobra, why did you tag this section? RockMagnetist (talk) 19:18, 15 November 2011 (UTC)
- I'm guessing that the problem is the lack of a References section or citations. I have renamed the section, added citations and dropped the tag. RockMagnetist (talk) 20:46, 15 November 2011 (UTC)
The problem with this format
I have just done an experiment that illustrates the biggest problem with the current preferred format. I moved Machine learning inside Artificial intelligence, which is consistent with both Outline of computer science and Artificial intelligence. However, now the section head for Machine learning is on the same level as the other articles in Artificial intelligence because they get a subsection each. A system that is incompatible with a hierachical topic structure is seriously flawed! For an alternative approach, see Bibliography of biology and List of important publications in physics (the latter is not fully converted to the new format). If you would rather stick with the current format, feel free to undo my latest edit. RockMagnetist (talk) 20:19, 15 November 2011 (UTC)
In a hierarchical format, Computer vision would go in Artificial intelligence; Information retrieval would go in Databases; and Operating systems would go in Computer architecture.RockMagnetist (talk) 20:28, 15 November 2011 (UTC)
There is no requirement that a WikiProject conform to the policies of any other WikiProject. Therefore the statement that other WikiProjects rate the importance of lists is not relevant. Ideogram (talk) 00:07, 28 February 2012 (UTC)
- Look at Article statistics in Wikipedia:WikiProject Computer science. Perhaps you should discuss this with the rest of your Wikiproject first. RockMagnetist (talk) 00:28, 28 February 2012 (UTC)
That's not Programming Languages
I don't think any PL researcher would consider most of the papers under "Programming Languages" notable. Only two of them are on Benjamin Pierce's list of "Great Works on Programming Languages" ( http://www.cis.upenn.edu/~bcpierce/courses/670Fall04/GreatWorksInPL.shtml ). A lot of them are language specs, which is to PL as the Mathematica manual is to math. One of them (SICP) is not on PL at all. I'll be replacing these in a week unless there's protest. 22.214.171.124 AurakDraconian (talk) 01:16, 3 December 2012 (UTC)
- Benjamin Pierce's list seems to be the result of a poll. How authoritative is it? I would be happier with another source or two for the great works on PL. RockMagnetist (talk) 01:45, 3 December 2012 (UTC)
- The poll was held on the TYPES forum, to which mostly programming language theorist are subscribe. The papers listed there also have insanely high citation counts. Also http://www.cs.cmu.edu/~crary/819-f09/ and http://www.cs.uu.nl/wiki/Techno/ProgrammingLanguageTheoryTextsOnline.
- I agree that the list should focus on programming language theory and not on programming language references, although I would consider SICP part of the former.
- Perhaps several of the publications listed there currently are better moved to List of important publications on computer programming. —Ruud 10:48, 3 December 2012 (UTC)