Wikipedia:Featured list candidates/List of software patents

List of software patents

The following is an archived discussion of a featured list nomination. Please do not modify it. Subsequent comments should be made on the article's talk page. No further edits should be made to this page. The closing editor's comments were: 14 days, 0 support, 2 oppose. Fail. Juhachi 00:57, 6 May 2007 (UTC)[reply]

Summary:This article was nominated for deletion way back in 2004. Since then, I have reorganised it and have brought in some strict inclusion criteria to respond to the original objections that the list was extremely POV. It is now heavily referenced, and includes links to main articles on the topics raised as well as highlighting notable software patents that do not, of themselves, warrant a complete article. Comments on the talk page from previously concerned editors have been positive and have helped to push the article through several revisions into its current strucutre.

Details:

Useful The article brings together a group of existing articles related by well-defined entry criteria and lists other important events on this notable topic, the inclusion of which are objectively sourced.
Comprehensive The article is a dynamic list, since new software patents and new notable examples of software patents are appearing all the time. However, I am not aware of any major omissions - the big ones like the GIF and MP3 patents are there, for example.
Factually accurate Heavily referenced article and I'm a patent attorney so know what I'm talking about, as are several other contributors.
Uncontroversial The topic is controversial, as is the definition of "software patent". However, the article recognises this and explains the content of the list in the lead section such that the list itself is not controversial.
Stable New patents are often added but there appear to be no outstanding conflicts over content or presentation.
Well-constructed This was the toughest aspect in a topic which is so broad. The reaction on the talk page to my categorisation of patents, however, has been positive.

Looking forward to comments! GDallimore (Talk) 15:02, 21 April 2007 (UTC)[reply]

Oppose. ~~Abstain.~~
1. The introduction reads like a tendentious justification of the list and patents on software: "All modern industry depends upon programmed computers. As a consequence, a large number of patent applications filed, and patents granted, relate to programmed computers to some extent." (emphasis added) The causal link is dubious. All modern industry depends on good management. However, as far as I know, very few patent applications filed, and patents granted, relate to management. The introduction therefore lacks neutrality.
2. The criterion for inclusion implies a grey area. The condition includes "[patents] which are notable due to their acknowledgment by the press, notoriety within the open source community or by virtue of high-profile litigation." I would suggest that it may omit major components of the subject. Example: EP0001640 (Koch & Sterzel - major opposition and appeal at the EPO).
3. The criterion for inclusion is based on the Bessen/Hunt technique. However, this is just one possible technique proposed for identifying software patents, based on Bessen and Hunt own definition of a software patent ("We construct our own definition of a software patent (there is no official definition) and assemble a comprehensive database of all such patents." [1]). The selection of this particular technique and the omission of any other possible definition and technique for identifying software patents may indicate that the article is not comprehensive.
4. The definition of a software patent is controversial. There is no official definition. It appears that the European Patent Office (EPO) does not use the term "software patent". The reason why the EPO does not use it but instead prefers the expression "computer-implemented invention" is not entirely clear to me, but this is a sign that there is a case for disputing the title itself. Why not using List of software patents and patented computer-implemented inventions?
5. The article does not present a world-wide view. There is no Japanese patent and it seems there is a US bias.

--Edcolins 21:10, 21 April 2007 (UTC)[reply]

I have taken the liberty to number Edcolins objections for ease of reference and respond below:

Introduction changed to better reflect a statement made two paragraphs earlier in the originally cited High Court judgment.
I agree that cases like Koch and Sterzel are of interest to patent attorneys and are important when studying the history of law in this area, but such history is discussed in other articles at length (see Software patents under United Kingdom patent law). Rather than simply listing cases of historical interest that are already listed in these other articles, this article's criteria for inclusion means that the focus is on patent/applications that would be relevant and/or of interest to a wide range of readers, not just specialists per Wikipedia:Make technical articles accessible. Under the Wikipedia-friendly entry criteria chosen, no major "components" are omitted per WP:FL? 1(b). Moreover, the entry criteria ensure that the article is a unique resource on the subject.
The fact that Bessen/Hunt is only one way of identifying software patents is not a reason for opposing the nomination as far as I can tell. The article makes clear what the criteria for inclusion are and the Bessen/Hunt technique is an especially practical and verifiable method for checking that a particular patent meets those entry criteria. The Bessen/Hunt technique is also known to provide a broad definition of "software patent" [2] so the resulting list is certainly not going to lack comprehensiveness on that basis.
I have already addressed this point in my original nomination. The general topic is undoubtedly controversial, but the article itself is completely open about what it is doing and how "software patents" are being defined. The article is therefore not controversial in itself. The particular definition of software patent chosen has been taken from a paper prepared by a reliable source. The fact that you (and I!) diagree with the definition and terminology does not alter that and our personal views on the topic are not reasons to oppose the nomination of the article. I've already tried reducing usage of the term "software patent" on Wikipedia, as you well know, but Wikpedians in general want articles about software patents it seems, not computer-implemented inventions. While I may not be happy about this, I can at least make sure that everyone understands what the preferred term implies and this article is an important part of that.
The lack of Japanese patents appears to be a reflection of the lack of litigation in Japan, not a flaw in the list. If you are aware of any notable Japanese patents, please add them! The US bias appears to be representative of the large amount of litigation in the US, and the fact that more software patents are granted there than anywhere else in the world. There is therefore no systematic bias within the article itself. I would point out that I am from the UK and therefore have no reason to promote a US bias. GDallimore (Talk) 16:13, 22 April 2007 (UTC)[reply]

The new introduction "Patents and patent applications involving computer programs cover a wide range of topics, demonstrating the importance of the question of whether these patents, often labelled software patents, should be granted." fails verification: Paragraph 185 reads "If nothing else, the wide range of topics (...) demonstrate the importance of the exclusions." The legal importance of the exclusions of Art. 52 EPC is said to be demonstrated, not "the importance of the question of whether these patents, often labelled software patents, should be granted." IMHO the article provides an interpretation, which fails verification. The introduction is better than the previous one (more neutral) but may still qualify as original research.
What about U.S. patent 5,794,210? Cited here and here, and here for instance... How can we verify that the list is comprehensive?
...
You wrote "The general topic is undoubtedly controversial, but the article itself is completely open about what it is doing and how "software patents" are being defined.". I am just wondering how the article can be "completely open" while at the same time comprehensive and stable (""Comprehensive" means that the list covers the defined scope by including every member of a set, or, in the case of dynamic lists, by not omitting any major component of the subject.").
...

I will fill the "..." later. --Edcolins 21:07, 24 April 2007 (UTC)[reply]

Edited again, although previous version was synonymous but avoiding UK-style language relating to "exclusions".
It's a dynamic list. If you think something is notable, add it. The fact that you can find brief mentions of a patent in a couple of papers and a press-release style article doesn't make it massively notable and doesn't mean that "major components are omitted". It's funny that I had that particular patent cited against me once as prior art...
...
You're objecting to the entry criteria, here, not the content of the article. The entry criteria are clearly stated so there is no controversy over what they are and are clearly defined so there is no controversy over what should be included. Ergo, no controversy.
...

GDallimore (Talk) 21:24, 24 April 2007 (UTC)[reply]

1. Solved.

2. to 5. The list seems to be open and the notability criterion seems not well-defined. Also, it seems to me that this is a list of important software patents, like a list of important operas (look at the entry criterion), and a more objective criterion (for the notability aspect) may need to be added... On the other hand, some featured lists appear more "open" than others, take Cultural depictions of Joan of Arc: "The entries represent portrayals that a reader has a reasonable chance of encountering rather than a complete catalog. Lesser known works, particularly from early periods, are not included." (and the title is not Important cultural depictions of Joan of Arc..).

So I don't pretend to be an expert of featured list nominations... and I don't want to be the only active opponent down here! So, anybody out there willing to provide a third opinion? Thanks! Until then, I abstain. --Edcolins 20:20, 26 April 2007 (UTC)[reply]

Oppose per the second reasn given by Edcolins. The current criterion does not strictly enough define what should be included. -Phoenix 22:16, 21 April 2007 (UTC)[reply]
- As discussed above, the second reason for the opposition is not that there is no strict definition of what should be included, but that the criteria for inclusion are narrower than he would like. The criteria for inclusion are actually well defined and I have responded to Edcolin's actual reason opposing the nomination above. GDallimore (Talk) 16:13, 22 April 2007 (UTC)[reply]
  - As mentioned above, I do not agree that the criterion for inclusion is well-defined. "[Software patents] which are notable due to their acknowledgment by the press, notoriety within the open source community or by virtue of high-profile litigation" is probably not sufficiently well-defined... --Edcolins 20:27, 26 April 2007 (UTC)[reply]

This is a summary response to all of the objections above. I have been going over various Wikipedia essays and guidelines, most importantly Wikipedia:Notability and Wikipedia:Lists in Wikipedia which I think contain some very useful guidance about creating well-defined entry criteria for a controversial topic such as software patents. Having gone over these, I have made the following changes:

I have changed the way the entry criteria are expressed to make clear that, by "notable" software patents, we mean software patents that are notable under Wikipedia's general notability guidelines and includes a couple of guiding examples. I think this solves the problem of whether the entry criteria are well-defined on that front.
I have added the {{dynamic list}} template to make clear that this list will always be subject to change.

The only remaining issue, I think, is whether this list can ever be non-controversial by virtue of the controversial question of how you actually define a software patent. Without doubt, the method presented by the article can be verified against every patent in the list, so there is no controversy there. The only controversy is in the choice of this particular definition as the entry criteria. I offer the following:

The definition is presented in an paper by well-established academics in the field. The paper was also apparently quite important in the CII debate since there was so little reasearch at the time (and there still is). Finally, Public Patent Foundation used this criteria themselves to identify software patents (as the article used to point out, but the link has now died). I think it therefore complies with Wikipedia:Reliable sources.
There is only controversy in the particular definition of "software patent" if, along with that definition, you apply the hardline view that everything that is a software patent is bad. The paper doesn't say this (although it is, sadly, generally critical of software patents) and the article itself highlights important developments as well as the more notorious patents. Therefore there is no systematic bias within the entry criteria for or against software patents and, as a result, I think WP:NPOV is complied with.

Are there any other criteria for objecting to this definition of software patent in an article titled "list of software patents"? GDallimore (Talk) 21:51, 26 April 2007 (UTC)[reply]

Comment I suggest the list should include the 'Freeny' patent (assuming it counts as a software patent), which caused many software & web companies trouble worldwide; see e.g. [3] Ben Finn 19:24, 27 April 2007 (UTC)[reply]

Feel free to add it yourself (as it appears to meet the software patent definition) but I know nothing about it and the article you cite doesn't give much info.GDallimore (Talk) 19:46, 27 April 2007 (UTC)[reply]

I don't really know enough about it. Ben Finn 14:48, 29 April 2007 (UTC)[reply]

Further comment. The new notability criterion violates WP:SELF. In addition, don't assume the Wikipedia notability criterion to be stable and clear. The entry criterion is still unclear IMHO. I oppose.

What about creating a criterion, which would not contain any self-reference, such that:

For a software patent to be included in the list, its number or a clear and unambigous reference to the patent, must have been cited in at least five (ten?) of the following sources:

New York Times - Washington Post - BBC News - The Times - Guardian - FT - Die Welt - Le Monde - Le Figaro - Wall Street Journal - International Herald Tribune - epi Information - Intellectual Asset Management - Journal of Intellectual Property Law & Practice - Managing Intellectual Property - Patent World - IEEE Spectrum - Nature - Scientific American - Slashdot? The Register? Heise News Ticker? .. what else? Japan, Chinese newspapers?

Sources would need to be provided (of course). The criterion for entering the list would be explained clearly at the bottom of the article. The list should preferably be based for instance on a published list of ten most influencial newspapers in the world, and the like for the scientific, economic and legal communities. The list of reliable sources should be justified to ensure stability of the criterion. Just a suggestion... --Edcolins 21:03, 27 April 2007 (UTC)[reply]

Sorry, but that's just silly. Even though I don't think that it what WP:SELF is talking about, I've changed the article to simply say "notable" and give some examples of how notability might be achieved. "Notability" is a clear concept and the suggestion that it isn't clear when opposing this nomination is, frankly, stunning in light of the huge amount of effort that has gone into the WP:Notability guidelines.

You suggestion of requiring 5 citations in a list of particular sources is massively onerous on any editor - I doubt that I could find five such references for a single one of the patents currently cited in the list and wouldn' expect an editor to go that far. It's also rule creep (or whatever the term is) when such precision in identifying something that is notable is simply not necessary. Compare the definition of "software patent". That is extremely easy to apply to any patent you may want to check against the entry criteria. Same for "notability". If you can find a couple of reliable sources, then you're fine - not only do you not need to insist that the sources be from particular places, but you don't want to because such a list might bring with it systematic bias.

So, I think you opposition on this point is flawed and groundless as well as not in keeping with Wikipedia guidlines. Your suggestion for a different entry criteria seems equally flawed. GDallimore (Talk) 15:18, 28 April 2007 (UTC)[reply]

Looking back over this last comment, I think I need to explain better why I think the current reason for the opposition is untenable: If notability is an adequate requirement for whether or not an article can be on Wikipedia, then it must also be an adequate requirement for whether a particular patent can be included in this list. Therefore the entry requirements are clear. GDallimore (Talk) 09:28, 30 April 2007 (UTC)[reply]

Thanks for your response. Regarding your edit summary "you could have fixed this youself".. Well, I have not fixed it myself, because I could not see any satisfactory way to fix it. Since your fix is a come back to the previous version [4] which I already commented, I have nothing to add. I don't think my suggestion is massively onerous to contributors. Google can help and some research might be necessary indeed.

My suggestion might not be enough anyway, it still leaves untouched the issue of the verifiability of the comprehensiveness (Wikipedia:Verifiability, ""Comprehensive" means that the list covers the defined scope ... in the case of dynamic lists, by not omitting any major component of the subject." - 1. (b) WP:FL?). How can you help us to be convinced that no major component is omitted?

I am not an expert in this kind of nomination. Third opinions would be welcome on the following suggested question:

Can the notability of an entry in a list constitute a criterion for inclusion which is sufficiently clear and precise for the list to become a featured list? If yes, in which circumstances and under which conditions?

Thanks. --Edcolins 22:16, 30 April 2007 (UTC)[reply]

A comparable example can be found in this discussion: Wikipedia:Featured list candidates/List of HIV-positive people. That article had a discussion of clear entry requirements (notability was considered a perfectly adequate entry requirement and was, in fact, an assumed entry requirement) and comprehensiveness ({{Dynamic list}} addressed the concerns). GDallimore (Talk) 23:27, 30 April 2007 (UTC)[reply]

Thanks for pointing out this featured list of HIV-positive people. I didn't know this one. The inclusion entry seems to be: if there is a Wikipedia article, then an entry can be added in the list. This ensures that a proper debate is possible for each entry. The present list is different in this respect. What about using the same criterion: if there is a Wikipedia article for the software patent, then it can enter the list. (Does this violate WP:SELF? I dunno)

Other suggestions for improvement:

(i) What do you think of the new layout? It improves readability I think. The current layout is a bit confusing IMHO.

(ii) We need an automatic tool for checking whether the patent/patent application description meets the Bessen/Hunt technique. Otherwise I don't see how to easily check the Bessen/Hunt criterion against a European patent. A macro parsing the esp@cenet description page would be needed.

(iii) The entries in the "miscellaneous" section do not fit with the remaining categorization. The patents in this section should be merged into the other categories (otherwise they could appear in two categories - since there two classifications). Only one categorization criterion, the technological field, is needed.

--Edcolins 16:59, 1 May 2007 (UTC)[reply]