5 ms·
Ask HN: What could be done to build a competitor to Wikipedia?
Wikipedia is electing new Arbitration Committee members just now, which prompts this question to the Hacker News community. How could someone set up a competing free, online encyclopedia to show users an alternative to Wikipedia's approach to administration and set up a friendly competition that might improve free, online encyclopedias in general?
I've seen a lot of complaints about Wikipedia here on Hacker News over the years. Yet I also see a lot of comments and even submissions of new stories that come straight from Wikipedia. This love-hate relationship suggests that Wikipedia has plenty of strengths along with its weaknesses.
For a while I've entertained the "thought experiment" of what it would take to build a new free, online encyclopedia and do it right. But how to start? I used to think of how I would run a free, online encyclopedia if I were project funder and founder and desired to be a project leader as an occupation thereafter, just to consider criteria for the hypothetical project. Reading recently about the DARPA self-driving car race reminded me that maybe a DARPA prize or X Prize approach might be even better than designing a whole project from scratch. Why not just announce a prize for a new free, online encyclopedia project, with some criteria, and let a variety of motivated teams think through how to build a new encyclopedia? Just set a few goals, and let the people who enjoy that kind of thing figure out the details. That has worked amazingly well for some projects that previously seemed impossible.
So what goals do you suggest for a Free Online Encyclopedia Challenge Prize? What could a new project do to show it is building a good encyclopedia, and that it deserves a cash prize to keep its growth going as it wins user acceptance? What specific goals would show that the project passed a reasonable "finish line" in a race to build a viable competitor to Wikipedia?
- Jemaclus 13y agoWell, first of all, you'd run into the chicken and egg problem. How do you fill a competitor with data without blatantly copying Wikipedia's data? How do you fill in data without editors? How do you get editors without data? It's not unsolvable, but a very tricky problem. Second, you'd have to overcome Wikipedia's massive marketing efforts, particularly with the word "wiki". People often use the term "wiki" as a verb in the same way they use "google" as a verb. How do you get people to think "I'll look that up on Clonopedia" instead of "I'll look that up on Wikipedia"? It's not insurmountable, but potentially a very, very expensive proposition. Third, despite Wikipedia's flaws, it is really one of the best collections of knowledge on the internet. It's so vast that the editing has slowed down because everything goes on there. This is like winning an architecture competition where they submit the Taj Mahal and you submit a sandcastle. It's not impossible, but... well... good luck. That said, I would enjoy some serious competition for Wikipedia and similarly placed entities (Google, etc).
- jamesjguthrie 13y agoI don't know about everybody else but I don't use wiki as a verb, it's a noun in my vocabulary. I'll say "Google it" but I won't say "wiki it" or "Wikipedia it." I'll say "Check the wiki" as if every other wiki on the internet doesn't exist and Wikipedia is the wiki.
- J_Darnley 13y agoI don't use wiki as a verb either. There's more than one wiki and if you want to find the right information you need to check the right one. If you want encylopedic knowledge about Star Trek you will check Memory Alpha. Star wars, Wookiepedia. I don't quite remember the "origin story" behind other wikis but thought it was because Wikipedia didn't want the content which now fills these others. Anyway, because these other wikis exist with, I presume, far more detail than Wikipedia contains is the reason Wikipedia won't become /The Wiki/.
- tokenadult 13y agoIs the licensing of Wikimedia Foundation project content consistent with using that content to form a new project, as long as there is attribution? I think so, but I would have to check. That could lead to the odd situation of New Project linking to Old Project from most articles, but that would also allow a head-to-head comparison of article quality, which just might heavily favor New Project.
- uptownhr 13y agoWhy can't you just start off with copying wikipedia? Create a wiki branch and link it to your branched site for every article. Make a chrome/firefox extension that allows anyone to easily do that. What is something you'd change/fix?
- Houshalter 13y agoIt's pretty easy to clone wikipedia. As a random example, here's some site that maintains a copy of wikipedia as it was in 2007: http://www.bazpedia.com/en/index.html http://www.bazpedia.com/en/index.html Now set up whatever policies for content or editing you want. Then set up some kind of system for automatically updating it with new content from the main wikipedia without overwriting anything your own users have changed. This way you get the benefit of both worlds.
- bonemachine 13y agoEven in the unlikely event that you'd prove successful at replicating the uncounted millions of hours of volunteer labor that have gone into replicating the, generally speaking, stupendously successful accomplishments of Wikipedia thus far: What makes you think you wouldn't run into pretty much the same bottlenecks, dilemmas & systemic frustrations that WP is currently finding itself mired in at this juncture, also? TL;DR don't spin your wheels trying to "compete" with WP. Better instead to find ways (technical / non-technical) to synergize with it + chart new ways in which it can evolve, keeping the momentum and positive energy base it already has.