Some references (which may or may not be reliable under your definition - I am not an expert at such things), include several issues of Mac Addict Magazine, several issues of Mac World magazine, and dozens of other hard-print magazine articles. I can cite references if they are worthwhile, but I will wait to see your response as to whether or not they are reliable. Its integration into Emacs was an event of note, but does not define the game. I originally posted its source code in the '80s to USENET. It was originally written in MACLisp. Not that it means anything, but I have been getting fan mail and e-mail asking questions about the game every day since the early '90s. I am too humble to state it as OMPIRE did, but I've been told by many others that it is a piece of history. As a side effect of its (non-exclusive) integration into Emacs, it's been claimed that it is the game installed on more computer than any game in the world other than Solitaire. This is due to the fact that most web servers and all Mac computers install it by default. My signature (with the tildes) doesn't seem to work right (maybe you can tell me how to fix it?) I'm known on here as both "Ron Schnell" and "aviators99". I will now type the tildes: Ron Schnell 00:26, 8 July 2015 (UTC)

Also, doing a Google Books search with the term "dunnet text adventure" (without quotes) yields quite a few references; not all of which are meaningless (although some are indeed very minor). In particular, the ones that reference how to run it from the command line (as opposed to from Emacs) I would consider to have standalone importance. Ron Schnell 00:33, 8 July 2015 (UTC)

A few things: (1) If indeed you are the creator, our conflict of interest guidelines strongly encourage you to not edit the article directly, but through suggested edits. (2) Yes, Mac World and Mac Addict articles should be helpful and would count as coverage in mainstream, reliable publications. That said, the Google Books search you mention brings up all passing mentions—single sentences that use the word "dunnet" (not enough to constitute significant coverage). Article topics must have significant coverage in multiple reliable, independent sources. ( ?) If the Mac magazine sources cover the game, review it, talk about any of its historical impact as you conceive it, then we have actual material to write an article. Otherwise, we're left with an unsourced article, or worse, one that relies on unreliable blogs and hearsay. We delete those. (3) Lots of cult classics don't have articles on Wikipedia. It is an encyclopedia of the verifiability and not of fan mail volume. We leave it to reliable, secondary sources to determine whether something is indeed "a piece of history" rather than leaving it to your or my opinion. In other words, we don't just trust OMPIRE's statement on the game's significance to ARPANET—we require sources of repute. (Type four tildes in a row and they will magically turn into your signature when you click "save".) –  czar 02:17, 8 July 2015 (UTC) reply

Yes, I try not to edit the article directly. Her's a full Macworld article: http://www.macworld.com/article/1047210/oldschooladventure.html, Here is a Mac Addict article: http://www.maclife.com/article/columns/terminal_101_4_emacs_easter_eggs. They cover it around once per year, and always act as though it's a Mac-specific thing, causing misinformation. This has previously been helped by Wikipedia, as it sets them straight that 1) It is not an easter egg, and 2) It is not a Mac game; it is a UNIX game (and now not only a UNIX game since there is a web version, as demonstrated above). I'll also remind you that it has "outlasted" Zork and Adventure as text adventures that are currently shipping on new computers from the factory today. You can download rewrites of Zork and Adventure from various pages, but Dunnet comes automatically on 20+ million Apple computers per year plus an unknown number of new web servers per year (probably more than Apple's 20+ million), written in its native language. I do put the 4 tildes, but "sinebot" usually replaces it for some reason. Ron Schnell 14:20, 8 July 2015 (UTC) — Preceding unsigned comment added by Aviators99 ( talk • contribs)

((collapsetop | (( sinebot troubleshooting advice -- not related to the AfD. )) ))

Ron, methinks the problem with SineBot is that your 'Ron Schnell' sig is non-hyperlinked to any user-talkpage , see the helpdocs at User_talk:SineBot#FAQ. Or if you haven't reconfigured your sig-config, maybe you are pasting something, rather than literally clicking edit then typing your content then literally typing tildeTildeTildeTilde at the end of your comment then clicking save? If you have two usernames, and are logged into Aviator99, you cannot sign as RonSchnellUid1234 ... similarly, if you aren't logged into any wikipedia UID, then putting the quadTilde into your posting will 'sign' the posting you just saved with your IP addr, rather than your wikipedia-username. Hope this info helps you get it worked out, if not, there are a few places you can seek technical advice -- see WP:Q for the list, you probably want either "Live chat help" via IRC or the "Teahouse" for fastest response, if those don't get you fixed up then try some of the other places mentioned at WP:Q. 75.108.94.227 ( talk) 03:50, 12 July 2015 (UTC) reply

((collapsebottom))

Should clarify even further that not only is it still written in its native language, it is the same code. I even made some changes last year that are now in the code base. If there is any reason you would like to confirm that I am the author, run the game. My e-mail address is in the "help" command, and I would be happy to respond to you. Ron Schnell 14:24, 8 July 2015 (UTC) — Preceding unsigned comment added by Aviators99 ( talk • contribs)

Right, so OMPIRE added those sources to the article already. The issue is that, as you noted, they aren't enough to confirm basic aspects of the game, such that we'd have to rely on original research rather than reliable secondary sources. The first source has some info but the second is simply a blurb—there isn't enough to write an encyclopedia article on this game. I would suggest hosting this content on a Wikia or another wiki if you want it preserved. –  czar 15:30, 8 July 2015 (UTC) reply

I actually believe there are enough secondary sources to keep the article. Ron Schnell 20:10, 10 July 2015 (UTC)

Hi Ron, thanks for your help with wikipedia, and thanks for writing Dunnet; I agree that yes, probably there are enough secondary sources with in-depth coverage to keep the article... but only barely, from what sources we have seen in this discussion. Here's a good one, [1] and here's another good one. [2] [3] (See WP:NSOFT-essay for why the second pair only count as 'one' source in the wiki-verse... but note that is an essay rather than a guideline or a policy or a legal mandate.) For the purposes of this AfD discussion, only cites with in-depth-coverage -- aka several paragraphs and preferably the article-title about Dunnet specifically -- count towards satisfying wiki-notability.

   The other stuff (longevity / fanmailVolume / briefMentions / shipmentCounts / etc) are no help here at AfD, even though some of those things may be helpful later on. The single-sentence mentions are WP:NOTEWORTHY and thus the material they mention can belong *somewhere* in wikipedia, but for justifying the dedicated article Dunnet (video game), most AfD folks like to see at least three multi-paragraph in-depth wiki-reliable sources. Fan-reviews like these ones [4] [5] are NOT helpful towards proving wiki-notability, because they are not editorially-controlled wiki-reliable-sources, although like WP:NOTEWORTHY mentions the fan-reviews are sometimes helpful later (see my longer reply below). For the moment, though, the best way to get out of AfD with a keep-result is to help us find additional multi-paragraph wiki-reliable-sources. "...dozens of other hard-print magazine articles. I can cite references if they are worthwhile..." Yes, those are what will help -- but note that, it needs to be more than a name-drop ("emacs also includes the game Dunnet") and more than just a repetitive blurb ("to play Dunnet type M-x dunnet and hit return then type help"). There needs to be some *meat* to the citation, some in-depth discussion of the gameplay, or of the codebase, or of the history of Dunnet, or something like that. We have two of those already, which is technically enough albeit barely; so, finding more will help cement the keep-vote, if that makes sense.

   The in-depth sources don't have to be online at the moment, [6] or even in English, as long as they specifically talk about Dunnet with a reasonable amount of depth. Hardcopy-only refs are fine, as long as the publication is 'wiki-reliable' aka either magazine/newspaper/academicJournal/governmentAgency/similar publisher with some sort of editor. MIT A.I. Lab memos are probably also acceptable as 'academia' sources, an IETF RFC would also suffice as 'governmental' agency, and many blogs/ezines are also wiki-reliable for specific fields. FLOSS apps like Dunnet sometimes get a bit of a break, I will submit; wikipedia is FLOSS, and although WP:NOTPROMOTION is a pretty firm policy, the strict application of WP:RS standards to FLOSS apps is often a bit looser -- so for instance, if you have an email-chain from RMS that covers the history of Dunnet, that might be considered 'WP:RS' in some loose sense (especially since we already have a couple good WP:RS citations from MacWorld and CultOfMac/LifeHacker/MacAddict). In any case, please don't be insulted by this AfD, which is supposed to be a discussion of wiki-notability, as diametrically opposed to notability-in-real-life; plenty of reasonably-famous apps get their articles marked for deletion, at least once, for instance MantisBT was almost deleted in 2010, when it was probably in the top three bug-trackers on earth. The keep-vote was mainly because of the comment Pcap made as of 20:10, 2 March 2010, which linked to four in-depth cites (or three if you want to be strict and count the two from linux.com as 'one' source for AfD purposes). 75.108.94.227 ( talk) 03:50, 12 July 2015 (UTC) reply

Dunnet'83 was originally in MacLisp, Dunnet'92 was a port to eLisp which is still actively [7] [8] maintained by multiple programmers, and as of Dunnet'15 has been ported to Javascript [9] as well by a third party. Besides GNU Emacs, the videogame is also found in XEmacs [10] and apparently (from the article-text) in the Scheme variant called Guile. So, although the most widespread port of Dunnet is included within Emacs, the game is NOT subsidiary to Emacs, and does not belong (except as a WP:NOTEWORTHY mention) solely within the article on Emacs. Dunnet began as a standalone-videogame before Emacs existed... not counting the TECO precursor... and as of 2015, Dunnet is still being ported (as a standalone-videogame) to places outside of Emacs and eLisp (e.g. to Javascript and Guile). If we want to justify a dedicated article, we need multiple in-depth sources:

  WP:RS#1 -- blurbed in 2005, as part of a group [11] and then a few months later given more depth in a dedicated review [12].

  WP:RS#2 -- reviewed in 2013, as part of a group [13] then as two dedicated articles on two different tech-sites [14] [15] both of which qualify as 'online magazines' per wikipedia tradition, if memory serves.

  There are also various WP:NOTEWORTHY mentions, in 2009 [16] [17] and 2014 [18] that I ran across.

  Not-quite-WP:RS methinks, but worth noting as evidence of continued interest amongst retro-gamers, two fullsize reviews [19] [20] by bloggers in 2008, one a graphics designer who works at some kind of interactive applications firm, the other by a university reference-librarian. See also my final paragraph below, suggesting that these not-quite-WP:RS can still be useful.

  If we want to get WP:RS#3, the usual rule-of-thumb for justifying a dedicated article, we have to switch gears, and stop thinking of Dunnet as a computer game which is reviewed in consumer-electronics magazines -- instead we must see Dunnet as a teaching-tool-slash-GPL'd-text-adventure-reference-implementation for the edification of budding programmers, and thus as something used in computer science academia, and the computer programming industry.

((collapsetop| Dunnet as a computer-science-tool rather than as a videogame ... analysis of 3 scholar.google.com mentions, 3 university-coursework mentions, and 5 programming-industry-press mentions ))

  There are three scholar.google.com published books/papers that make WP:NOTEWORTHY mention of Dunnet. The first is 283 cites for GNU Emacs Manual by Stallman et al, ISBN  1-882114-05-1, which has mentioned Dunnet at least since 1994 [21] and August 1996 [22] as well as continues to mention it in the 2015 edition of the book [23] on epage#429 aka printedPage#407. True, it is just a one-sentence blurb... but this is the printed documentation, and the Emacs distribution-blob also contains not just the Emacs manual, but also the entire source-code of the game, and the online-helpdocs-for-the-game, which the game itself includes. The people in charge of the Emacs application are the FSF, which is independent from Schnell; they are including his codebase, and his helpdocs, in the primary Emacs app-distro itself -- and documenting this inclusion via a short pointer in their official helpdocs, since the game (as a game) is self-documenting. Emacs is a programmable programmer's text-editor; the inclusion of several videogames is thus a sort of easter egg, to show off the power of the embedded scripting language eLisp, and prove that Emacs is more than just a mere text-editor.

  The other two scholarly sources which specifically mention Dunnet are the 1998 paper [24] Developing software with GNU: An introduction to the GNU development tools which has 3 cites on scholar.google.com, as well as in the 2001 book [25] pdfPage#29 of Advanced Linux Programming ISBN  0-7357-1043-0 which has 137 cites on scholar.google.com.

  Beyond the inclusion of Dunnet as a stock component in the Emacs application-distro, there are a bunch of Linux-distros and Unix-distros which have included Emacs (and by incorporation Dunnet) inside their operating-system-distribution-blobs; as the OSX-specific articles above indicate, pretty much every Apple desktop operating system since 2001 has included Dunnet. Also, since every ^{citation needed} Linux distro offers Emacs in their package-repo, Dunnet is also available in all those operating system variants. Of course, the history of Dunnet goes back further: it was available, as a stock component of Emacs, on most flavors of commercial UNIX and most variants of BSD before the dominance of Linux [26] and OSX [27] eclipsed those older platforms. Since Dunnet was written back-in-the-day in 1983 for PDP-10 mainframes, Dunnet in actuality precedes the desktop and the microcomputer; it is a big-iron videogame, like Spacewar, albeit less venerable and thus correspondingly less famous.

  Delving into WP:OR for a brief moment... much like SpaceWar, the players of Dunnet are mostly computer hackers, hence the sci-fi cyberpunk theme (e.g. you have to be comfy with the CLI of both DOS and UNIX-like systems to successfully play Dunnet). This is by stark contrast with 'interactive fiction' games like Zork, which are text-adventure-precursors to the RPG. The mechanics of Dunnet are similar to Zork, but the type of player is quite distinct: most anybody can play and enjoy Zork, in the same way that most anybody can read and enjoy LOTR books ... but to play and enjoy ^{citation needed} Dunnet, you need to be a computer programmer, or at least, a sysadmin. Hence, Zork became a commercial product, and converted from loosely-open-source over to a proprietary source code model, selling hundreds of thousands of copies to consumers that owned 8-bit microcomputers in the early 1980s, whereas "open source" Dunnet was for the PDP-10 (the environment where the *developers* of Zork did their programming... as opposed to the microcomputers where *consumers* of Zork played the game). Later, Dunnet began to be distributed with Emacs, the programmer's programmable editor, whereas Zork was not -- both Dunnet and Zork were written in in LISP variants, but Dunnet was not encumbered by a proprietary codebase. I suspect, given the copyright-1992-by-FSF message of dunnet.el nowadays, that Dunnet'92 might be the first GPL'd text adventure. A pointer to the earliest release of the source-code on USENET might help; I was only able to find this [28] from 1992, which suggests the initial Dunnet'92 codebase was *not* GPL at first (noncommercial-only semi-copyleft license). End WP:OR.

  The point of this history-lesson is that Dunnet was available on a bunch of operating systems over a bunch of decades, and in particular, was included as a stock easter-egg hidden inside the programmer's programmable editor Emacs: it is no coincidence that Dunnet is mentioned not just in the manual for that programmer's editor, but also in a book like Advanced Linux Programming, which presumes you will be using Emacs for your programming work, and mentions the easter-egg in passing, as something programmers would find cool. By contrast, Zork was interactive fiction, something meant for gamers to enjoy; from a non-scientific 2002 ranking, [29] over a dozen retro-gamers -- since 2002 was over a decade after text adventures had become 'obsolete' in some sense -- had played and voted on the games Anchorhead (game), Trinity_(video_game), Spider and Web, A Mind Forever Voyaging, Unnkulian, Christminster (interactive fiction), Jigsaw_(video_game), Excalibur_(video_game), Curses (video game), I-0_-_Jailbait_on_Interstate_Zero, Wishbringer, Photopia, The_Hitchhiker's_Guide_to_the_Galaxy_(video_game), Planetfall, Lurking Horror, Infidel_(video_game), and Moonmist... as well as the variants of Adventure_(disambiguation)#Games and the Zork series. Dunnet wasn't even on the list... because although it is a text-adventure, it isn't really interactive-fiction in the normal 8-bit-microcomputer commercial-software sense. That it was *excluded* from the list of the usual interactive fiction games enjoyed by retrogamers circa 2002, is actually evidence that Dunnet is wiki-notable: it ain't your everyday run-of-the-mill text adventure, which might not deserve a dedicated article, being like almost every other text-adventure and thus failing the WP:ROTM essay-criteria... instead, it seems clear that Dunnet is Something Completely Different from run-of-the-mill. Being atypical doesn't guarantee that Dunnet is wiki-notable, but having multiple WP:RS and a couple dozen WP:NOTEWORTHY mentions across two decades, *plus* being atypical, convinces me Dunnet should be a keep.

  As evidence for the hypothesis that Dunnet is a computer-science-related open source text-adventure reference implementation of sorts, not just another videogame (though it is also a videogame), we see that the game itself was specifically utilized in several college programming-courses: VLSI in 1999/2003, [30] [31] Java in 2003/2004, [32] Haskell in 2007 [33] (looks like lecture-slides but I don't speak german), and presumably others. Dunnet is still considered cool in computer-science academia as of 2010, [34] and *also* cool amongst retro-gamers as of 2005ish-thru-2014 per sources above. Background info: during the 1990s and early 2000s, a large percentage ^{citation needed} of university-level intro-to-programming courses used Emacs, and furthermore, included some variation of a homework-lab which required you to program your own videogame, often a text-adventure like Dunnet (and sometimes using Dunnet specifically during class as the links above prove). This is pretty typical of LISP programming textbooks, even in 2010, [35] although I cannot tell if chapter#5 building-a-text-game-engine of this recent intro-to-LISP-programming-book specifically mentions Dunnet or not -- there is no online version of the book.

  Outside of computer-science-related academia, Dunnet has a couple mentions in the computer software industry press. 2005 at a publisher of books about programming, [36] 2009 article on the use of easter-eggs as a way to improve user-engagement within software-applications, [37] , and a pre-2004 mention [38] by videogame designer Eri Izawa (who worked [39] on Asheron's Call, Gods and Heroes, and Hellgate:_London among other videogames). There are also at least three computer-industry-non-academia books with passing mention of Dunnet, specifically in Xemacs on page 445 of Teach Yourself Suse Linux In 24 Hours, [40] and specifically for OSX on page 147 of Mac Hacks: Tips & Tools for Unlocking the Power of OS X Mountain Lion, [41] as well as Rule the Web: How to Do Anything and Everything on the Internet Better. [42] The latter book is a bit more than passing mention, since it has some reasonably-in-depth content about text adventures, which includes mention of Dunnet -- most of the other books are programming and sysadmin related tomes, which mention dunnet as a cool easter-egg, whereas the RuleTheWeb book is specifically talking about Dunnet-the-videogame. Not all the OSX books are like that; Mac OS X Power Hound Panther Edition is definitely treating Dunnet like an easter-egg; it gives four sentences or so to the topic, which is slightly more than passing mention, but not quite in-depth coverage methinks. [43] There is something about Dunnet on page#588 of The UNIX Companion from 1995 by Harley Hahn, but google's search-OCR is buggy so I was unable to see if it was a brief mention or more substantial. There is also a passing mention in Learning GNU Emacs by ESR et al on page 466, which specifies that dunnet implements a major mode. [44]

  As an aside, although it must not be allowed to impact our deliberations here, I will note that there is a fanatical religious sect [45] which may wreck havoc upon wikipedia should Dunnet (video game) be deleted from mainspace; however, we must not allow zealots to strike fear in our hearts! If the evidence of WP:RS given above is found insufficient to satisfy WP:N for a dedicated article Dunnet (video game), then I suggest we move the material on Dunnet to a subsection on text_adventure#Dunnet, or failing that, to a subsection of Ron Schnell#Dunnet (which would require writing that BLP).

((collapsebottom))

  So where is WP:RS#3, then? At the end of the day, there is no singular and obvious WP:RS#3 that I can point to, showing how Dunnet played a decent-sized role in computer-programming-academia-and-industry from the mid-1990s through circa 2010 or thereabouts, but in aggregate, methinks that all the links given in the green box are pretty strong evidence. It is also plausible that WP:RS#3 may already exist, perhaps in offline-form as a chapter from some CS textbook of the mid-1990s before the web really took off.

  If in fact WP:RS#3 doesn't yet exist, it seems plausible to predict that it most likely *will* be written in the future: Dunnet was written in 1983, gained WP:NOTEWORTHY status in 1994 as a computer-science-easter-egg-slash-teaching-tool, and by 2005 had fullsize WP:RS#1 as a videogame-proper (for retrogamers on newly-more-popular UNIX-like OSes), followed in 2013 by another burst of coverage (again for retrogamers) in the form of fullsize WP:RS#2, with several fullsize not-quite-WP:RS gamer reviews in 2008. The timespans involved suggest that coverage of Dunnet in wiki-reliable sources has not ceased, and although wikipedia articles ought not predict the future, wikipedians can use common sense when determining whether WP:N is satisfied. In my judgement, the article on Dunnet does already satisfy wiki-notability criteria ( "significant coverage in multiple reliable sources" even though we only know about two rock-solid sources at present rather than the traditional-by-convention minimum of three) and thus Dunnet (video game) should remain as a dedicated article in mainspace; merging it into text_adventure#Dunnet could also be done (and is a good idea regardless of whether or not the dedicated article is retained).

  There was a specific objection expressed by user:czar, that we don't have enough depth-of-coverage to write the wikipedia article. However, I think that worry is misplaced. Once we have determined WP:N is satisified, by showing multiple-wiki-reliable-sources-with-reasonably-deep-coverage, we can then go ahead and use the non- WP:SPIP-independent albeit-non-editorially-controlled blog-reviews [46] [47] to fill in the plot-subsection of the wikipedia article (and other such non-contentious facts), so long as we are careful to only pull material from them which is 'unlikely to be challenged'. Along the same lines, we can also use Schnell himself as a source -- preferably via his published writings on his 'official' blog rather than via his wikipedia username of course -- see WP:ABOUTSELF, although of course once again (as with the 2008 blog-reviews) we have to be careful to stick just to uncontroversial facts. For an example of this, see Colossal_Cave_Adventure#References which uses a mixture of judiciously-selected WP:ABOUTSELF as well as traditional WP:RS citations... although unlike the intermixing found at the Colossal_Cave_Adventure article, I strongly suggest it is much better to put all the WP:ABOUTSELF stuff (and by the same token info from the not-really-quite-WP:RS blogger reviews as well) underneath Dunnet (video game)#Notes rather than under Dunnet (video game)#References, per the wiki-helpdocs at Help:Footnotes#Footnotes:_groups. (( Later update, I have just updated Jigsaw (video game) in this fashion, for example. ))

  Disclosure: I played Dunnet, sometime in the previous millenium; at some point prior to that, I took one of the programming-classes where writing an adventure-videogame in Scheme was one of the lab-homeworks, and Emacs was the required text-editor for the course (Dunnet wasn't mentioned in the course-materials that I remember however -- I just ran across dunnet.el later, on my own). That said, I have no WP:COI in the wikipedia sense of that phrase; I've never heard of Schnell until today, and didn't even remember the game was called Dunnet -- but after I read the wikipedia-article today, I recalled the gameplay experience quite distinctly. It's a fun game, ^{citation needed} if you are into that sort of thing. My keep-vote is based on having 2 very solid WP:RS videogame reviews, plus a fairly large number of WP:NOTEWORTHY mentions in academic-and-commercial settings, over a timespan from 1994 through 2014 at minimum; that says wiki-notable, to my ears. 75.108.94.227 ( talk) 03:50, 12 July 2015 (UTC) reply

...which goes to show just how much original research from primary sources is needed to state anything of substance about this software/game (esp. its supposed impact). If it is important, let reliable, secondary sources say so. As it remains, all we have is what we already discussed: a blurb, a short review, and a slew of passing mentions in non-notable media and blogs. To then hedge that someone might write about it in the future is exactly what we do not. All the gathered information is better off hosted on another wiki until more journalists take up the mantle. –  czar 05:34, 12 July 2015 (UTC) reply

Hi czar, you are setting up a false choice. The discussion here is not about whether to delete all mentions of Dunnet from wikipedia, purging it en masse, to be banished to wikia or some other non-wikimedia location until it has 'more' journalists giving videogame reviews. See WP:NOTEWORTHY, and also please remember that deletion ought to be the last resort. What we are discussing, is whether Dunnet (video game) should be a dedicated article, or rather, merged-and-then-redirected to become a subsection of the text_adventure article. I lean towards the former, since the latter article is currently a redirect to interactive_fiction, and although Dunnet is known to the interactive-fiction gamers, per these non-WP:RS blogs by such gamers in 2008, [48] [49] it is really more-than-slash-different-than the usual 'interactive fiction' videogame covered in that article. The evidence that Dunnet had an impact on computer science education, is at present all WP:NOTEWORTHY ... which is not, of course, grounds for outright deletion of that WP:NOTEWORTHY info, as I keep taking pains to point out. Dunnet also is WP:NOTEWORTHY as an easter_egg, and as a stock component of Emacs the programmer's editor.

  I expect those aspects of Dunnet won't take up much real estate: a sentence about it being mentioned in the computer industry press with cites to where, a couple sentences about the integration into Emacs and the port to eLisp, and a couple of sentences explaining how Dunnet is often perceived as an easter egg, but unlike typical easter eggs is also a standalone program. The bulk of the article (or the article-section iff merge-n-redirect is the decision of the AfD closer) will concentrate on our major secondary sources which have some reasonable depth: 800 words in 2005 by MacWorld, [50] [51] and then a series of reviews in 2013 by MacLife [52] / LifeHacker [53] / CultOfMac [54] with another 600 words. It won't take long to summarize those, especially since portions of them are redundant ("Dunnet is a text-based game ... type emacs -batch -l dunnet to play..."), but even after eliding the repetitions, the end result will be a good wikipedia article per WP:SIZERULE.

  As to your point about WP:CRYSTAL and predicting the future, I can only repeat what I already said, emphasis added: The timespans involved [1994-2014] suggest that coverage of Dunnet in wiki-reliable sources has not ceased, and although wikipedia articles ought not predict the future, wikipedians can use common sense when determining whether WP:N is satisfied. My point was that, even if the 'final' cleaned-up properly-sourced version of the Dunnet-article is relatively short as of 2015, it is unlikely to be a WP:PERMASTUB, since it is still generating press thirty years after it was first written for ITS; hence, my lean towards keeping the dedicated article Dunnet (video game), rather than merging into a subsection of text_adventure#Dunnet. By contrast, look at the original version [55] of the article on Jigsaw (video game) compared to our version now ... there is only one in-depth source for that, most of the other info is WP:ABOUTSELF and WP:BLOG, so it makes sense that Jigsaw (video game) could be reasonably merged-and-redirected to a new subsection Graham Nelson#Jigsaw... but it also doesn't really hurt anything to have Jigsaw (video game) as a dedicated article, so long as we stick to what the sources say. The article on Dunnet can be written quite well, no WP:OR needed, with the sources we already know about; if Schnell finds additional hardcopy sources with some depth that will only improve the situation, though I still don't think additional sources are necessary for a keep-result here at AfD. 75.108.94.227 ( talk) 23:14, 13 July 2015 (UTC) reply

Jigsaw (video game) ... there is only one in-depth source for that, most of the other info is WP:ABOUTSELF and WP:BLOG, ... Jigsaw (video game) could be reasonably merged-and-redirected to a new subsection Graham Nelson#Jigsaw... but it also doesn't really hurt anything to have Jigsaw (video game) as a dedicated article, so long as we stick to what the sources say

Much of what you said is spot on, so I'm not sure how you ended up with this conclusion. By your logic, anything that has ever had a secondary source can be bolstered into its own article with enough self-published and primary sources, but that's exactly what we don't do at AfD. An article is notable only if it has significant coverage in multiple reliable, independent sources. ( ?) If you freely admit that only one or two reliable sources cover a topic, then there is patently no significant coverage! It doesn't matter that we can cobble together a WP page with a ton of primary sources—WP draws the line when games, people, ideas do not receive any formal reviews, do not receive any coverage more than a single (or two) blurbs describing its basic function, as there's no way we can write authoritatively as an encyclopedia with such a dearth of coverage. "Non-WP:RS" fan sites are still, by definition, unreliable, and unsuitable sources for an encyclopedia. Hence, I suggest copying the material to Wikia. I'd entertain a redirect but I think the links to the related topics are weak. Alternatively, it could go into a List of interactive fiction games but I think that would be its own (mostly unwarranted) mess. –  czar 01:27, 14 July 2015 (UTC) reply

Hi czar, I've replied to your points in the green box below, but since I feel we're close enough, I've gone ahead and collapsed it. Here is my 'final' listing of sources, ordered roughly by importance to AfD.

* 1400 words of coverage (or thereabouts) in MacWorld, MacLife, LifeHacker, and CultOfMac.

* WP:N: 2 or 3 dedicated news articles. [56] [57] [58] [59]

* Kinda- WP:N: 2 subsections of news articles. [60] [61]

* WP:NOTEWORTHY 2 books by RMS and ESR on Emacs and eLisp, which application-and-programming-language-distribution includes the both full source & in-game helpdocs for Dunnet, to show off the power of Emacs/eLisp (plus to provide a fun game -- tetris and a dozen lesser games are also stock). [62] [63] [64]

* WP:NOTEWORTHY 2+ computer science academia-books (plus book above by RMS also qualifies here) with scholar.google.com cites. [65] [66]

* Likely- WP:NOTEWORTHY use as an 'example-app' in homework given by 2+ college profs (besides Schnell himself). [67] [68] [69]

* WP:NOTEWORTHY 4 or 5 computer-user-press as useful for retrogaming. [70] [71] [72] [73] [74]

* WP:NOTEWORTHY 4 or 5 computer-programming-industry-press as an easter egg. [75] [76] [77] [78] [79]

* WP:ABOUTSELF: 3+ with depth (the author is a wikipedian so more may be forthcoming). [80] [81] [82]

* WP:BLOGS: 3 with depth, using their real names and independent of the author&game, can be used iff necessary for gap-filling with non-contentious unlikely-to-be-challenged factoids. [83] [84] [85]

* WP:EL: short introductory walkthru, plus explains distinction between a TUI videogame like Nethack& Rogue-likes versus an interactive-fiction-text-adventure videogame like Dunnet& ADVENT, the latter pair relying on pure text with no ascii art. [86]

* Also worth pointing out: FLOSS codebase since 1983 for ITS mainframes, at the peak of the commercialization of text-adventures on early home computers; codebase still active in 2015 with multiple contributors. [87] [88]

* Maybe worth pointing out: ships with Emacs for the past two decades, and thus comes stock on OSX and most Linux distros today, [89] [90] [91] aka millions of new machines per year.

My calculations (below) indicate that using these sources (above), Dunnet (video game) will have roughly 300 words of body-text, which is compatible with WP:SIZERULE. Thanks for your time and efforts, folks. 75.108.94.227 ( talk) 06:32, 15 July 2015 (UTC) reply

((collapsetop| point out List of IF, more on Jigsaw as the UBER borderline, wordcounts of uber-borderline Jigsaw vs cleaned-up-Dunnet vs quasi-borderline GooglePhotos ))

((On your last point, there is no List of interactive fiction games, but the equivalent exists at text adventure#Notable works, and Dunnet belongs on that list in my judgement although it is not yet there because the article is officially called 'interactive fiction' whereas Dunnet is more of a 'text adventure' in the traditional nerdy sense... the fit is not perfect, since Dunnet differs from interactive fiction like Jigsaw (for all intelletual gamers of all genders), specifically in terms of who Dunnet appeals to (for all command-line-savvy computer nerds and cyberpunk-sci-fi fans). Still, the only place that it makes sense to put Dunnet is with text adventures, either as a subsection thereof with no dedicated article, or as a notable-works-pointer therein which links to the dedicated article. Until and unless we generalize the article on interactive fiction into the more generic text adventure, rather than vice versa squeezing the more generic text adventure topic into the somewhat stricter confines of the interactive fiction genre, this problem will remain for Dunnet and all other text adventures that are only quasi-IF-like. ))

   Yes, I freely admit Jigsaw only has one single in-depth source, the 2011 videogame-review in village-voice. That said, Jigsaw is furthermore WP:NOTEWORTHY in that the NYT called it "acclaimed", and that some PhD english prof called it "epic...notable". Does that satisfy the strict reading of WP:42, or even a pretty loose reading? Obviously not. Does it improve the encyclopedia, to contain the paragraph about Jigsaw, fully sourced? Yup. It doesn't really matter if we put the Jigsaw paragraph under Graham Nelson#Jigsaw, or leave it as Jigsaw (video game), but it would be foolish to move it to wikia. The logic here is simple: see WP:IAR and WP:Don't_cite_WP42_at_AfD, which tell us that if what we are doing improves the encyclopedia qua encyclopedia, ignore all rules to achieve the primary goal, and in particular, suggests to us that strict application of WP:42 is the wrong way to go about arguing at AfD. Jigsaw has only one source, and yet, the article is just fine. Specifically, you had a problem with this perceived extrapolation from my Jigsaw example: "[by that logic] anything that has ever had a secondary source can be bolstered into its own article with enough self-published and primary sources" This isn't what I'm saying, because I'm not making a rule I'm making an exception to the rules, for a specific article (aka not "anything" only some relatively-rare things qualify). Your assertion is that, since Jigsaw violates the rule about multiple-sources, it therefore MUST be completely deleted and moved to wikia. You are wrong on two counts: first, the multiple-source violation is obviously correctly identified, but wikipedia has no firm rules, and if other evidence of significance exists (NYT article on the game-author and PhD english prof reviewing games of that author), then we might very well be quite justified in overriding the WP:42 rule, yellow highlighting and all. My *actual* logic is straightforward: look at material on a case-by-case basis, and if it improves the encyclopedia, then deleting it (aka move-to-wikia) is always wrong, whereas rearranging it (aka move-n-redirect) can sometimes be an improvement. This is the pretty much the same logic that WP:FAILN says to apply; it's not just me. Furthermore, though it is true that the Jigsaw article is 'cobbled' together and 'bolstered' with WP:ABOUTSELF and such in a few places, there is exactly zero 'original research' therein, and no WP:SYNTH either; doing legwork to dig up published sources is the opposite of WP:OR, not the twin of it. Since this is an AfD about Dunnet, however, I suggest we continue discussion about jigsaw on usertalkpages, if necessary.

   What about Dunnet then? By the strictest of counting, it has two sources: 800 words in Macworld of 2005, 600 words in MacLife/ LifeHacker/ CultOfMac of 2013. Is that ' multiple' sources? Yes. Is that ' several' sources? Maybe, depending on how you count them; 2 sources glomming bursts together, 3 sources if we only count articles entirely dedicated to Dunnet, 4 sources if we only count publishers with dedicated wikipedia pages, and 5 sources if we count them all individually. But as always, AfD is about qualitative significance, not about counting sources (though of course it pretty much always boils down to countage). Thus, whether we keep the article Dunnet (video game) as a dedicated article, or merge-and-redirect to text_adventure#Dunnet, is a judgement call. In your judgement, 'two' sources is patently not enough. In my judgement, looking deeply at the specifics of the other WP:NOTEWORTHY coverage of Dunnet (several WP:NOTEWORTHY hits in books by RMS and ESR plus computer-press coverage of the easter-egg aspects ... neither of which any run-of-the-mill videogames have), the two-to-five in-depth sources we already have, plus the couple-dozen noteworthy-mentions in completely distinct fields, are plenty to show significance. To prove my point, I fixed up the Jigsaw article, which has one in-depth WP:RS, a few WP:NOTEWORTHY sentences, and generates about a paragraph of wikipedia-article-text.

   The body-text of the cleaned-up article on Dunnet will also be rather brief, probably two or three paragraphs and thus slightly longer than Jigsaw but not by much. But as with the Jigsaw article, there will be plenty of meat in the refs section (four times as much meat as Jigsaw roughly speaking), and where there are missing tidbits of info that most wikipedia articles contain (given the specific tidbits are uncontentious), we can pull them from secondary independent blog-reviews that happen to be non-WP:RS, or even from primary sources using WP:ABOUTSELF, which is perfectly within wiki-policy. Dunnet has roughly the same number of in-depth refs as are currently being used in Google Photos, [92] which currently has a positive-coverage-burst from May/June with 4500 words, [93] [94] [95] [96] and a negative-coverage-burst from June/July with 1750 words, [97] [98] [99] leading to the wikipedia-article containing 642 poz-words aka compression-ratio 7:1 and 177 neg-words aka compression ratio 10:1 which is pretty decent (72% in sources and 78% in wikipedia). Dunnet only has poz-coverage, so assuming the same 7:1 ratio holds, I expect the ' finished' article will have something in the neighborhood of 200 words summarizing the videogame-reviews, plus a few sentences covering the non-videogame aspects, total roughly 300 words. Jigsaw currently has fewer sources, and is currently 163 words, so the relative sizes of Google Photos versus to-be-cleaned-up- Dunnet (video game) versus already-cleaned-up- Jigsaw (video game) makes sense to me. Google Photos is an obvious keep, despite having only two bursts of coverage; there will be more we can confidently predict (in talkspace as opposed to mainspace), and even if not, the article is fine like it is. Jigsaw would usually be moved-n-redirected to an equivalent 163-word-paragraph inside Graham Nelson#Jigsaw, or under text adventure#Jigsaw, and such may very well occur, despite brief NYT and PhD quotations to the contrary. Dunnet is the borderline case, but it is certainly closer to Google Photos than to Jigsaw on the spectrum... not only does Dunnet have two bursts of coverage containing a handful of in-depth sources like Google Photo does (albeit with lesser word-counts), Dunnet has similarly-diverse WP:NOTEWORTHY mentions to Jigsaw (albeit without the newspaper).

   Anyways, I've enjoyed our discussion here, but I went ahead and collapsed this reply, since I feel we are coming close to the end of what I can productively contribute to this AfD. If you'd like to switch to usertalk, I am happy to explain my interpretation of 'significant coverage' further, or talk about WP:SIZERULE, or predict the future of wikipedia, or whatever. I have placed a backup-copy of Dunnet (and Jigsaw pending the AfD that I foresee may be happening there soon) into draftspace, and will try to add a few sentences I've worked up for Draft:Dunnet there and/or at Talk:Dunnet. 75.108.94.227 ( talk) 06:32, 15 July 2015 (UTC) reply

((collapsebottom))

Folded away. 75.108.94.227 ( talk) 14:09, 25 July 2015 (UTC) reply

• comment, additional international sources, with some mild depth. User:Aviators99 found two more cites in the 2013 burst, picked up by the German-language computer press (looks WP:RS to me -- [107] [108]), which wrote a German-language listicle, [109] [110] noting the game was English-only (not yet internationalized) in approximately 97 words of Dunnet-specific text, plus a Dunnet-specific screenshot. The other was in the Australian-English computer press, Dunnet-dedicated article of ~150 words, [111] looks WP:RS, [112] [113] brief but interesting as the screenshot shows part of the plot, midway through the game. Plus of course, these indicate press-interest on three continents. Four continents, if you count the localization into Japanese by a FLOSS activist on their blog, [114] though of course that doesn't qualify as "press" interest. 75.108.94.227 ( talk) 03:21, 20 July 2015 (UTC) reply
• comment, additional source, User:Aviators99 ran across this 2007 listicle, ~78 words but quote "extremely addictive", looks WP:RS, [115] [116] [117] [118] publisher is redlink AppleMatters but has been used as WP:RS before per Special:Search/AppleMatters. 75.108.94.227 ( talk) 19:11, 22 July 2015 (UTC) reply

Summarizing as of now, in total we have three bursts of English-language WP:N coverage with some depth in 2005 (listicle+dedicated), 2007 (listicle), and 2013 (listicle+dedicated), plus international press with some depth in Australia/Germany/etc, as well as WP:NOTEWORTHY mentions in scholarly & programming-industry WP:RS from 1994 through 2014+ (unusual for a videogame to appear outside videogame-specific-media-channels). I believe these recently-added sources push us over the unclear-notability-borderline and well into wiki-notable territory. User:czar, you asked to be pinged if non-English sources were discovered, please see these. [119] [120] [121] [122] (mostly mentioned above already). 75.108.94.227 ( talk) 19:11, 22 July 2015 (UTC) reply

75.108, the verbosity on this page is an impediment to anyone actually trying to participate in the AfD. This is absurdity. There is no reason to post links from unreliable or what you call "noteworthy" sources. We're here to discuss the topic's notability—whether or not we can write an article on the subject based on its coverage in reliable sources. The other stuff that would fill out the remaining details (the self-published sources, the "noteworthy" but not "notable" stuff you mention) are all not useful right now. Furthermore, all of your recent links are passing mentions! The fundamental issue is that those single sentence mentions together do not constitute significant coverage and then do not pass the general notability guideline. This should be obvious from the basic idea that we can't write an article without reliable source coverage (doesn't matter how many unreliable blogs mention the item or however many self-published sources may be available) and that we would have unreliable articles written for everything ever mentioned if that were the case. We can't write an article based on links from low-level blogs that only mention that the game exists. Our standards for notability require depth of coverage. In all of the articles you've linked, has there been one review? Has there been one discussion of its broader impact? Or are all of them listicles—not actual write-ups, but clickbait side-articles—that only mention the game in passing as an Easter egg? This AfD has descended into absurdity. –  czar 19:47, 22 July 2015 (UTC) reply

And all of this belies the original point—the same point to which we return after boatloads of links: that the coverage only discusses Dunnet as a minor feature within Emacs. It should only be mentioned in context at the Emacs article and if anything, this Dunnet page should redirect there. –  czar 19:54, 22 July 2015 (UTC) reply

How is this, this, and this a listicle? Or the several hundred word discussion here? These are not at all passing mentions. Calling articles "clickbait" isn't really helpful; it's just a pejorative opinion. These are websites (or books) which generate revenue and have significant traffic and readership. II | ( t - c) 01:14, 23 July 2015 (UTC) reply

Since it has come to this, here is a full analysis of the links that purport to together constitute significant coverage:

1. Macworld 2005: explains basic gameplay, fine, but you probably didn't notice that your fourth link, Total Snow Leopard is a paraphrased copy, which shows how useful the original was: it explains that it's a text-based adventure and here are some commands.
2. MacLife 2014: says nothing about Dunnet other than mentioning its existence (and incorrectly, at that, as a MUD!) So then Lifehacker, Cult of Mac (your other two links), and Kotaku post exact copies of the same material with catchy headlines and but fluff it up with insubstantial information that doesn't say anything other than that this Easter egg exists. We can't add anything to an article based on these sources. They count as the "same link" for our intents, but they are still, together, worthless. The standard definition of " clickbait" fully encompasses the usefulness of these articles as reliable sources in an encyclopedia.

Any article written on this topic—without any reviews, without any development information, based on small blurbs in extremely low-quality articles—will not be encyclopedic. The coverage about this game is enough to warrant a mention on a relevant page or list—that's all. What's disappointing is that all of the text above scares away any of the AfD regulars (who aren't coming with an outside interest) who would be able to tell you the same thing I've repeated since the original nom. –  czar 01:38, 23 July 2015 (UTC) reply