==Attempts to remove MaraDNS from the Wikipedia==
I always feel hurt and upset whenever some “deletionist” at the Wikipedia tries to delete my MaraDNS article. A now-banned Wiki editor tried to delete the article over three years ago and another editor put it on the Wikipedia deletion page just a couple of weeks ago.
==MaraDNS will stay on the Wikipedia==
I am pleased that the MaraDNS article was able to survive this “Articles for deletion” nomination with a strong “Keep” consensus because of the significant coverage in third party sources establishing notability. Because of how Wikipedia works, now that the MaraDNS article has clearly been shown to be notable enough to keep on the Wikipedia in this deletion discussion, it is very difficult for another editor to try and delete it in the future. Indeed, Wikipedia has a policy called “Notability is not temporary” which means MaraDNS is now permanently notable enough to stay on the Wiki for the foreseeable future.
While I have finished up actively developing MaraDNS in 2010, and am slowly winding the project down, MaraDNS will always be something important to me. Indeed, MaraDNS may very well become be my legacy, the thing I will show my grandkids when they ask me what I accomplished in my life, and it’s good to know it will have a Wikipedia article.
==Down the Wikipedia rabbit-hole==
Wikipedia, being a large community, has created their own terms for describing Wikipedia-related things.
In Wikipedia-speak, for an article to remain on the Wikipedia, it needs to be “notable”. Notability has a very specific meaning in the Wikipedia world: It means that the subject in question has “received significant coverage in reliable sources that are independent of the subject”.
“Significant coverage” also has a specific meaning: It means that the subject in question has more than a vaguely-defined “trivial mention”. For the purposes of this blog posting, I will consider “significant coverage” more than 100 words discussing MaraDNS.
“Reliable” additionally has a specific meaning: It generally means something published by an established journalistic entity and/or publishing company, or a peer-reviewed academic paper.
Here are three sources which are both reliable and devote well over 100 words to discussing MaraDNS:
- Jan-Piet Mars, in 2008, wrote Alternative DNS Servers: Choice and Deployment, and Optional SQL/LDAP Back-Ends, a book published by UIT Cambridge Ltd. and now available as a free PDF download. In this book, there are over 7,000 words (an entire chapter and mentions elsewhere in the book) that discuss MaraDNS.
- Carla Schroder wrote a two-part article about MaraDNS that is published in Enterprise Networking Planet: Buck DNS Monoculture with BIND Alternatives (part 1) and Buck DNS Monoculture with BIND Alternatives (part 2). This two-part article talks about pretty much nothing but MaraDNS; that’s over 2400 words from a reliable third party source.
- A peer-reviewed article by Sooel Son and Vitaly Shmatikov, The Hitchhiker’s Guide to DNS cache poisoning (PDF download) devotes well over 300 words discussing MaraDNS’ recursive resolution algorithm. This article was published in the book “Security and Privacy in Communication Networks: 6th International ICST Conference, SecureComm 2010.” (edited by Sushil Jajodia, Jianying Zhou) printed by Springer.
There are a number of other mentions in other sources; click “show” on the green box in the MaraDNS deletion article for them.
==Why did they try to delete the MaraDNS article?==
Considering that MaraDNS easily meets even the most strict Wikipedia definition of “notable”, how did the article get nominated for deletion?
While I can not enter the mind of another person and determine what exact feelings they had when adding the deletion discussion for MaraDNS, we can make some reasonable speculations.
Wikipedia is a crowd-sourced web site; like open-source software, the majority of contributors are volunteers. Since no one is getting paid, it is hard to motivate people to do the tedious work which an encyclopedia sometimes needs.
The request for deletion appeared about a week after I put up a request for a neutral third-party editor to update the MaraDNS article. That request was ignored for over a week; nine days after my request, MaraDNS was nominated for deletion.
It took an editor only seven minutes to say, in so many words, “MaraDNS is not notable, delete it”. I got the seven-minute figure because seven minutes elapsed between this editor’s last edit before marking MaraDNS for deletion and his final MaraDNS-deletion related edit.
It took me, on the other hand, about three hours to look through all of the links provided by the relevant Google searches, find notable mentions of MaraDNS, and catalog them (who wrote it, when did they write it, who published it) to add them to the deletion discussion. It probably took me another hour or two to write my own version of the MaraDNS article, to help the Wikipedia no longer have an outdated article about MaraDNS.
In summary: Seven minutes work to say “let’s delete MaraDNS”. Five hours work to establish MaraDNS’ notability and update the article.
I don’t blame unpaid editors for sometimes taking the easy way out; with the number of editors per reader decreasing, and with Wikipedia’s increasing popularity to the point it’s an Alexa top 10 site and almost always one of the first hits when one does an Internet search, I am sure a lot of editors have become very jaded with the number of “POV-pushers” (people who edit the Wiki not to make it a better encyclopedia, but to have it promote their company or product) out there.
==The history of MaraDNS’ Wikipedia article==
MaraDNS’ Wikipedia article is almost as old as the Wikipedia itself; it was created in December of 2001, before even the stable MaraDNS 1.0 release came out (that happened in June of 2002).
The article pretty much languished until 2005, when, after getting permission, I started expanding the article. Before I started editing it, it was a 4-sentence stub; after making a few edits over a couple of months, it had become a much more extensive article.
All of this was done during the “growth” period of the early Wikipedia, when the priority was to grow the Wikipedia and get as many articles as possible written. Sometime between 2005, when I was able to edit the MaraDNS article without controversy, and 2007, things changed.
The Wikipedia had become more popular; as a result, a number of parties otherwise uninterested in the Wikipedia started coming to the Wikipedia to promote their products and services.
This kind of editing violates a guideline called “Conflict of Interest” (COI in Wikipedia-speak). With many Wiki newcomers attempting to use the Wikipedia as a promotion vehicle, the consensus changed from one where people could edit an article they were personally involved in as long as they tried to stay neutral to one where it was strongly discouraged to edit an article one was personally involved with.
By 2007, an administrator asked me to stop editing the MaraDNS article. After that, all of my MaraDNS-related edits, with one exception, have been the kind of “non-controversial” edits that an editor with a “conflict of interest” is permitted to make.
The only “controversial” edit I made has been an edit where, after having the MaraDNS release number on the Wikipedia page be a year out-of-date (it claimed that 2.0.04 was the current version even though 2.0.05 was released a year before my edit), I finally took it upon myself to update the MaraDNS version number to 2.0.07.
(It’s an open question whether I would have gotten consensus to edit the MaraDNS article in 2005 if more editors had been part of the discussion back then. I think I would have; the change in Wikipedia’s culture towards editors with a conflict of interest [COI] happened between 2005 and 2007. To wit, a COI notice on another article was added in October of 2006.)
==How the Wikipedia has changed==
Back in 2001, when the MaraDNS article was first created, MaraDNS was a computer program which would not pass today’s arbitrary “notability” guidelines. It’s an open question whether MaraDNS ended up getting enough attention from journalists and academic researchers to become a subject that passes Wikipedia’s “notability” tests because it has had a Wikipedia article which the press and academia noticed.
I think it is very unfortunate that it’s a lot harder for a new open source project to get a Wikipedia article than it used to be. Open source developers are people who work very hard but usually do not get financially compensated for their effort. It’s very important that today’s hard-working open source developers get compensated in other ways; letting even new open-source projects have a Wikipedia article would be one way for the Internet to show appreciation for their hard work. I really wish that Wikipedia’s current crop of admins did set the bar for new articles to be so high.
Indeed, while I used to be a regular Wikipedia editor—I contributed a lot more to the Wikipedia than editing the MaraDNS article, back in the day—I have mostly stopped editing the Wikipedia after seeing a number of useful open source projects get deleted from the Wiki. My Wiki user page used to have this on it:
I find the deletionist movement here very disturbing and very much against this hacker ethic. I think there is something fundamentally broken with the way the Wikipedia allows articles to be deleted because the subject they cover is not “notable” enough; to me the deliberate destruction of information goes against the hacker ethic of making information freely available. It’s sad that the Wikipedia allows a single person using TW [Note for blog: TW is a tool used by some editors to help automate the editing of Wikipedia articles] to mark dozens of useful articles to be deleted in just a few minutes, and has a culture that encourages this what I see as destructive behavior.
Then again, I should really just let go of the Wikipedia and let the deletionists go on their rampage here; if they continue unabated, Wikipedia will ultimately become a less useful place for readers, but that’s Wikipedia’s problem, not my problem.I can not speculate why other editors have quit editing, but undoubtedly many of them also quit because they did not like how deletionists became so powerful.
Since I devoted so much time dealing with Wikipedia’s MaraDNS article, I did not get a chance to do any actual MaraDNS development this month. Maybe next month, but, then again, I am in the process of transitioning from working on MaraDNS once a month to once every other month. I probably will not devote time to MaraDNS again until June, unless a critical security bug with a CVE number is found.
To post a comment about an entry, send me an email and I may or may not post your comment (with or without editing)