26 Oct 2007 mirwin   » (Master)

Wikiversity Little change other than a few people are back who had gone missing. Steady trickle of activity on the recent changes log.

Citizendium LOL As one of Larry Sanger's designated TROLL!s there is zero chance that I will be participating there even if it managed to improve on Wikimedia Foundation leadership/governance challenges. However, I was curious to see what improvements they would choose to make to encourage success in building a viable community to create and maintain their fork of Wikipedia. I was amazed to find that there was very little change other than an up front charter/manifesto that Larry and Larry's appointed buddies with P'hds would be in charge and that people were expected to use their realife identities and give some basic background information. Sanger recently advertised that he would be a fixed term dictator, not god king for life, as if this would be a major improvement. After some consideration I still conclude that:

1.) Sanger is still an obnoxious self righteous idiot who probably still does not understand that an ad hominem attack from an alleged P'hd in Philosophy periodically makes him look like a ten year old who has managed to get his game system attached to the internet.

2.) Sanger either brilliantly took my advice regarding mana and engineering and decided on slow increments of "improvements" or independently figured out that duplicating the early Wikipedia environment was a best way to rapidly recreate Wikipedia's success. Time enough to fix the flaws after a viable project emerges. By which time he will have filtered the community to people willing and eager to put up with him in charge. Exactly the Bomis strategy with Wikipedia. Brilliant! Again!

So a measure of success is inevitably how long Citizendium takes to achieve its first viable community fork.

Wikipedia I have been browsing around Wikipedia doing some research and considering ways to possibly attract attention to some Wikiversity projects in which I have interest. I find the current predictions of doom scattered around the web from various knee jerk opponents (note: not critics, their constructive observations serve a critical feedback function) who still claim it is a temporary phenomenon which cannot possibly succeed longterm amazingly tenacious.

Regarding finances. Yes they still beg for money and allow the servers to slow down response when inadequate rather than doing some basic long term business planning and getting the foundation on a solid continuously funded footing. However, a rapid response emergency plan has been floated as a balloon on the mailing list. Display google ads only to unregistered users and only on select pages. Revenue potential is high and immediate. Jimmy Wales has previously promised no advertising but I suspect in an emergency someone could throw him reluctantly into the briar patch.

Regarding qualityThere have been multiple technical fixes to quality control and improvement identified/explained/proposed on the mailing lists, around the Wikipedia site, in gas station rest rooms, and even the occasional review by a constructive critic. Most of these are certainly viable and easy to implement ranging from based upon stable/unstable release versions of articles, to threshhold support voting, to etc. etc. Meanwhile is it simple enough for anyone to revert vandalism or article rot or click back a few versions in the history log. Most detractors ignore this through ignorance or malevolence. Personally I think it is idiotic that a stable version is not shown until a threshold of trusted users (can anyone say advogato style trust metric) have clicked on an [advance to stable version delivered to users] button as myself and many others have proposed on various mailing lists.

Many other viable quality control approaches have been proposed and ignored. Why? I think it is an artifact of early days when growth was paramount. That growth recruited people who love the immediacy of the "Wiki Way!" Whenever the subject of rationalizing version control and presenting the best available approved version to the public as default rather than whatever was edited a few seconds ago the Wiki Way people get hysterical. Every one else agrees that Wikipedia is not finished yet and drops the subject to calm the hysterical vanity tweakers, my tweaks must show immediately to public, cannot be restricted just to active editors for mere eternities much less days or weeks or months until ready to satisfaction of trusted community members.

Of course, Wikipedia will never be finished. Just like a massive engineering project the redlines will never be done. Revision control is necessary to have the correct version of drawings depicting best known information regarding the facility or equipment. If used for critical engineering decisions the information must still be checked against the physical plant, equipment or software release versions and any redlines hanging in the drafting queue or in technicians work order files. Granted it is the responsibility of the Wikipedia user to use the information responsibly ... why force them to start from random crap rather than the currently available best version ..... (but if we say it is good they may not check it, but does this make us responsible if they use the information to modify a nuclear reactor?, but they should know that after they check all the information they can fix the mistakes, and on and on) ....

This is an area where Citizendium may do us a huge favor. See? Even successful forks can be invigorating? By their promise to provide easy access to certified best information available for use at your own risk, they may force the Wikipedia community to stop foot dragging and make it easy for casual users to read the best source text/code available by default rather than whatever tweaks, vandalism, mistakes, or propaganda happen to be sitting in the priveleged top queue position pointed at when John Q. User (not a Wikipediac) requests some reference information. (but we are a Wikipedia not an encyclopedia or a body of reference information, editing here is only a fun way to waste time). This vast improvement of Wikipedia achieved merely by delivering the stable code vs. the unstable code to the user by default will not hurt Citizendium. Their market niche is clearly defined by a committment that P'hds will have reviewed the material presented to the random public. Even should the vast pool of talent at Wikipedia (which does indeed include quite a few actual P'hds along with some pond scum alleging they have credentials) match Citizendium's article quality. Their credentialism should provide a large niche market secure from Wikipedia in the short term.

Regarding project status Overwhelming success. There may never be another GFDL general purpose English language encyclopedia in the history of mankind that is not derived at least in part from Wikipedia. I cannot speak for other languages but they appear to be progressing as well. In a few more years that statement will be probably be true for most specialized encyclopedias. Even though "Deletionists" keep claiming much human knowledge is not "notable" others keep putting it and more back. Why would anyone start from scratch when such an excellent body of reusable work available in machine transparent form exists?

Regarding long term survivalThe English Wikipedia community seems to have its ups and downs and it would not surprise me to see the Wikimedia Foundation implode. I suspect there are backups around the web that could be acquired by legal action if necessary if someone wanted to restore the database and start a new community from scratch. Personally I suspect the Wikimedia Foundation will eventually get its act together and that the overall community and database is now rather robust and indestructible in a Phoenix kind of way.

Most importantly.Should Wikimedia Foundation implode tommorrow and Wikipedia databases be corrupted and unsalvageable .... the project (which I view as a prototype for open engineering activities) is still a resounding success. It has proven that gnu/linux is not a fluke or a fad. People scattered worldwide interested in improving the human condition via a gift economy can accomplish anything that a government or a wealthy capitalist could accomplish in the past and possibly a few things they could not. Can an operating system be developed and given away for free? Yes. But that is software whisper/shout the critics. Can an encyclopedia be developed and given away for free? The superset Wikipedia (Wikipedia is more than an encyclopedia.) was. But that is merely information. Can a Manhattan style nuclear bomb or an Apollo style ICBM be developed over the internet by thousands of volunteers scattered worldwide. Based upon this google search for open source car I suspect the answer is yes. But it would be expensive and why would you want to? I am curious as to whether open engineering could help make Lunar and Martian Tourism an "affordable" reality within a few decades. Perhaps open source fabbers will make hardware prototyping a bit cheaper in the near future. Notice the interesting open source media wiki the site is using to facilitate coordination. An open source CAE suite beginning with 3D models and spitting out engineering drawings and machine code for milling machines is probably a bit much for engineers to ask of the open source community before committing to open source software in a big way. The problem as always is preexisting proprietary data formats and art. How one would go about setting up a user friendly collaborative site to talk to computer illiterate engineers about what software tool features would be useful in open design of various components and/or systems for various purposes is currently a bit beyond me. Fortunately sites such as Wikiversity and Fabber at Home seem to be springing up in time to begin supporting the soon to be arriving thousands of open source engineering teams that may be required to open the new frontier for the benefit all people everywhere.

Latest blog entries     Older blog entries

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!