scale.metaoptimize - A StackOverflow-like site about high scalability and NoSQL
scale.metaoptimize.com
scale.metaoptimize.com
For years before stackoverflow, everyone copied the PHP-BB UI. Stackoverflow changed it.
New sites can come up with a better and different design. The ones that do, stand a better chance of making it big than ending up as many of those stackexchange1.0 sites (that died).
I'm the person that launched the site.
I'm running it on a shared host. I apologize for any attendant problems with quality of service. (and yes, I do see the irony :)
There is a pretty obvious fix. I'd need it switch it to a virtualized host or its own machine. Sorry I haven't done this yet. I launched the site as a hobby project.
My previous site: http://metaoptimize.com/qa/ got a large amount of visitors last week and no one complained about the load. Load randomly spiked to 60 on my shared host because of another user's job.
OSQA devs tell me that running OSQA on its own machine gives very good performance. Even the largest OSQA site, like LockerGnome (http://lockergnome.net/) supposedly only has a single digit CPU percent usage when run on its own machine.
I recently launched an OSQA (the Django+Python software running my site) for ML+NLP: http://metaoptimize.com/qa/
That last site was incredibly successful. I have a few hundred signups, there is activity on the site every ten minutes or so, and everyone on the site is very grateful. I'll post some quotes at the end of my comment.
I decided to try this again with another niche topic that I find very important: scalability. There is a lot of undocumented folk wisdom, as well as little known tips and tricks, that I believe should be discussed in a Q+A format with voting.
I am sorry you consider my site distateful. I took the default OSQA skin and made some small customizations.
I hope in reading the following comments about the ML+NLP site, you'll see that people genuinely value what I put up. I am hoping similarly to create value on another topic that a lot of people wrestle with daily.
Things people are saying about MetaOptimize Q+A:
Alexandre Passos (Unicamp): "Really thank you for that. As a machine learning phd student from somewhere far from most good research centers (I'm in brazil, and how many brazillian ML papers have you seen in NIPS/ICML recently?), I struggle a lot with this folk wisdom. Most professors around here haven't really interacted enough with the international ML community to be up to date" (http://news.ycombinator.com/item?id=1476247)
Philip Resnick (UMD): "Looking at the questions being asked, the people responding, and the quality of the discussion, I can already see this becoming the go-to place for those 'under the hood' details you rarely see in the textbooks or conference papers. This site is going to save a lot of people an awful lot of time and frustration."
Aria Haghighi (Berkeley): "Both NLP and ML have a lot of folk wisdom about what works and what doesn't. A site like this is crucial for facilitating the sharing and validation of this collective knowledge."
Ryan McDonald (Google): "A tool like this will help disseminate and archive the tricks and best practices that are common in NLP/ML, but are rarely written about at length in papers."
esoom on Reddit: "This is awesome. I'm really impressed by the quality of some of the answers, too. Within five minutes of skimming the site, I learned a neat trick that isn't widely discussed in the literature." (http://www.reddit.com/r/MachineLearning/comments/ckw5k/stack...)
Digging deeper, I can see that OSQA is based on the same tech as CNProg, which has been discussed on HN before. If I recall correctly, most people shared my displeasure with the clone. But now because OSQA is open sourced, people think it is OK?
When people spend a lot of time and money to design something through blood, sweat, and tears, like I'm sure the StackOverflow team did, it is not cool for someone else to come by and just clone it blindly. This also applies to free and open source software. Derivative, inspired work is one thing. But this is a step away from copy-paste.
There is so much more to software than what your CSS layout is, and we should all look to our fellow designers and programmers and steal the best ideas and use them ourselves. That's how software evolves.
From a legal standpoint, didn't we settle the whole look and feel issue way back when with Apple vs. MS?
really? You think so?
I am glad that there is an open source option, but I hate the fact that its all most 95% copy paste job of SO.
There is a difference between design inspired from something and blatant copying something.
I had not heard of OSQA. I'm about to apologize to the person who started this particular site (other thread), as it's not his fault. But being open source doesn't make it OK to copy someone else's hard work.
In fact, I'd say it's pretty much the same idea, just with a different community.
Oh wait.
Let good ideas/designs prosper and spread. Humanity is better off that way.
This only has seven questions, but it is living and breathing. Isn't that is what is important about a website?
Until they fix it I guess the solution is to use external alternatives. I'm not sure I like the stackoverflow copying done by OSQA, but it's open source and easy to set up. I'm not the creator of this site, btw, just a user who thought it would be interesting to the HN community.
The point in the democratic process we're pushing is so the site doesn't have seven questions on day one and someone checks it out and goes "Gee neat..." and never comes back again. The 400 people committed to the GIS proposal have committed "to visit at least three times per week, to ask at least three questions during the beta phase, and to answer as many questions as I can for at least three months." That's a pretty good head start for a site where you want expert answers to your questions and you want them fast.
We're still tweaking the specifics of how much commitment a site needs, but we're certain it's more than a link on a few websites the first day. The design of the software goes a long way, but in order for these sites to not fall flat they have to give answers fast, and they have to attract experts. You can't make that happen by just launching a URL; we're working full time on making sure we figure out how to do that.