User:Pcap/On software notability

Work in progress, feel free to make suggestion on the talk page, but do not edit this page.

There are several competing essays and proposed guidelines on the notability standards a piece of software must meet before it's included in Wikipedia. Too often WP:GNG has become a word-counting game, ignoring real-world notability, so here are my ideas about this:

End-user software
I consider some software notable if any of the following is satisfied, (but preferably more than one of these):


 * Has dedicated editorial reviews or appears in several round-up reviews in reputable publications, on-line or paper. I give more weight to editorial reviews which aren't hosted on download-focused pages. In general, the editorial reviews on the download-focused sites (e.g. Softonic, Softpedia, CNet download pages) are rather superficial.


 * Has tutorials dedicated to it in reputable on-line publications or books (the latter do not usually contain product reviews).


 * Has several mentions, preferably recommendations, in the mainstream, general interest press (usually in their tech columns). These do not have to be lengthy. It's unreasonable to expect most software to be covered at length in the mainstream press, but software noticed there is usually worthy of notice on Wikipedia too. Normally one finds more detailed coverage of software like this specialist venues, but there may be exceptions, usually if the technical function of the software is rather trivial.


 * Has an independent measure of high popularity. For instance, it ranked high in a download top for a sufficiently long period of time, or was ranked high in a reader survey (these need to be statistically significant, with thousands of respondents). Some examples of tops like this are the end-of-year download charts of Chip.de and the reader surveys of Linux Journal. This is a similar argument to that used for WP:NSONG ("ranked on national or significant music charts").

Back-end software
This include libraries, database engines, and file formats that are not exposed directly to the user. In general, one cannot expect reviews for this kind of software.


 * Has tutorials dedicated to it in reputable on-line publications or books. Conference tutorials and talks should be taken into account if they are independent, but also if t


 * Is central to the operation of multiple notable end-user software. It's silly to have many red links for software of this kind when a technical description can be obtained from the primary sources, and its real world importance is easily documented by use in notable products. Some significant AfDs in this area are Parchive and GT.M.


 * Documented large user base. This needs to be approached with care as it's easy to synthesize the appearance of significant market share from disparate examples.

Academic software
Most academic software is written to test or illustrate some research ideas rather than be used in production. It's generally difficult to find extensive third-party coverage of software like this. Software like this needs to be considered like other academic work. Care must be take not to overvalue citations in related work which are merely a matter of due diligence in the absence of which a paper would likely get rejected.

License
Although Wikipedia prides on free content, the topics are discussed in neutral manner, so no preference for inclusion should be given to software based on how it is licensed.