Opening Statement by Mitchell Szczepanczyk
[Copied from OpenZuka's defunct phpBB forum - BM]
The idea behind openzuka was probably best articulated by Benjamin:
"I'm in favor of a campaign to keep Google on the side of the forces of light, but I also think that with the IPO it's time to recognize that Google is going to increasingly favor corporations and the establishment in everything it does. We need an open source alternative to Google, whether it survives its coming war with Microsoft or not. Hard to imagine now, but maybe a technological innovation like distributed processing will make possible a replacement (that and killer search algorithms), the same way cheap computers in clusters, programming genius, and near-ideal site design made Google's rise to dominance possible."
To which I responded:
"Benjamin.
History may remember this post as the start of something big. You've just given some of us an amazing homework assignment.
I work in software development for a living. I don't know if you have any computer skills (it sounds like you do), but would you be interested in getting the ball rolling on developing some open-source Google equivalent or near-equivalent?
That goes for anyone else reading this post. Do you have any computer skills? Are you a fan of open source? Interested in taking on Google? Email mitchell@chicagomediaaction.org and let's get something started. "
So this is that something that we're getting started, which we've named "openzuka".
An explanation about the story behind the word "openzuka": In trying to figure out possible names for the project, I and others came up with such ideas as "OpenGoogle" [my coinage], "Moogle" [my friend Mario on WHPK], "SearchStorm" [inspired by the usual Chicago weather], "Droogle" [don't remember], and, quasi-heartedly, "FuckYoogle" [a friend of mine at CAN TV]. I was also toying with the idea of incorporating the word "szukaj" [pronounced SHOO-kye], the Polish word for "search". (Dammit, English needs more Polish loanwords.)
I ran "szukaj" past my (non-Polish speaking) ex-girlfriend [and now friend] Laura, who said that the spelling would trip up most people. She suggested the alternative "zuka" instead.
"zuka", it so happens, is already a word/name being used (and guess what search engine I used to find this out). E.g., there's a nature reserve in South Africa named "zuka".
It was Benjamin who, in St. Louis, at the Media Democracy Showcase during the National Conference for Media Reform, came up with the idea of combining "open" with "zuka". Paydirt. Another search (guess where?) revealed that the sole instance of "openzuka". Plus, no one reserved the domains "openzuka.com", "dot net", or "dot org".
We then ran this idea past an impromptu focus group -- our friends Jessica and Brian at The NewStandard. They liked the name (heck, it was Brian's rant on his blog about Google News that got this project started), so we decided on the spot to name it that. It so happened that there was a camera crew on-hand recording this entire conversation.
And now, here we are. Openzuka.
My credentials in a nutshell: I work in software development for a living, and I hold a master's degree in computer science from the University of Chicago. I can code in PHP, MySQL, Java, Javascript, C, C Shell, Perl, and a bit of Python. I have built some fairly large-sized applications, so obviously I'm warmed up to take on Google, AskJeeves, Altavista, MSN Search, and that new Bittorrent search.
I've been thinking about how we build openzuka. I don't have the answers, but I do have two questions to ask:
(1) Do we copy the internet (insofar as possible) into some gigantic distributed cache, just as Google and other search engines have done? If so, where do we put it? (On a distributed basis, I suppose; I don't have cash to buy and maintain a farm of some 10,000 servers.) If not, how do we build the index the internet?
(2) What is the criteria for a given search? I suggested (as Ben mentioned) as a starting point to use the diversity of incoming links: The more links a site or page gets from other sites, the higher it ranks is in the search.
One possible answer to question (1) would be to somehow leverage the power of Apache servers. After all, Apache is open-source (as is Openzuka), and an impressive percentage of servers around the world, maybe 60% or so, use Apache. How might we tie in all those servers into Openzuka?
By the way, I've asked a professional graphic designer friend of mine to design a logo for Openzuka. [Which is now featured on this site! - BM]
