choffstein

Members
  • Content count

    2754
  • Joined

  • Last visited

Community Reputation

1090 Excellent

About choffstein

  • Rank
    Contributor
  1. Minecraft servers crashed. Now free to play for everyone!

    So ... Minecraft is the Dwarf Fortress of 2010?
  2. I haven't done any reading of the academic literature, but my guess is a PCA analysis on image scan. Basically, compile a huge collection of finger-print scans and find the 'eigen-features'. That way you can identify a specific scan by the linear combination of eigen-features. You can then measure the distance between two scans as the distance between their factor loadings. I believe this technique is used in some facial recognition software as well.
  3. Diddle, a DRY Data Definition Language

    Quote:Original post by capn_midnight Actually, Migrations looks like it's solving a completely different issue. It's not solving the stuttering-like-Elmer-Fudd problem, it's simplifying the deployment problem. Deployment is still a problem with my database designs, but it's not really the purpose of this project. I typically use something like Red Gate SQL Compare to do my deployment environment migrations. It works really well, runs pretty fast, and can do automatic backups prior to changes. Then I don't think I quite understand the purpose of your project. Is the goal to simply minimize typing? I would say the Ruby migration library is pretty expressive for what it does. It is deployment and definition all rolled into one.
  4. Diddle, a DRY Data Definition Language

    Might be worth checking out Ruby on Rail's Migrations. They are pretty expressive, well tested, and already do just about everything I think you are trying to do. You can just use Ruby as your scripting language to manage your database. The ruby gem (library) standalone_migrations does all the heavy lifting of tearing the relevant work out of Rails for you! Nevertheless, very cool!
  5. Quote:Original post by deathtrap "Health" and "Health food" threads scare the crap out of me, because of all the 'wisdom' usually spouted there is usually very few sources to back up said claims. I'm not saying people are saying incorrect things, I'm just saying a few more sources would be great to read through so we can educate ourselves more. To get the best advice go to a registered nutritionist though after getting a medical check up by a real doctor. So called 'registered nutritionists' also used a food-pyramid for 20 years that had you having 6-11 servings of carbohydrates a DAY. Whoops ... turns out that was a bad idea. So far, I see very down-to-earth suggestions here: eat natural foods and avoid crap. Can't ask for much more than that.
  6. A few simple rules have never steered me wrong 1) Shop on the outside of grocery stores. This is where the fresh fruits, vegetables, meats, and dairy are normally found. 2) If you can't understand the ingredients on the back (or, a five year old cannot pronounce them) -- don't eat it. 3) If your great-grandmother wouldn't recognize what the food is -- don't eat it. 4) If it doesn't perish, don't eat it. FOOD SHOULD GO BAD! These simple rules will force you to eat fresh, healthy foods.
  7. Pixelcraft: The Colors of Frogatto

    I'm no artist, but that was a cool entry to read!
  8. Introduction to CUDA

    Check out the thrust project if you want to use CUDA but with a C++ STL feel to it. The project is fairly mature at this point, and a lot of tests I have seen show that you don't end up losing a lot of speed using thrust vs going to CUDA.
  9. Anyone know any good stocks to invest in?

    Quote:Original post by way2lazy2care I always wanted to try watching CNBC or some other business news site and investing in their recommended stocks without doing research just to see how well it would work. I have to imagine they average a hefty profit on their predictions or they wouldn't have a job. Either way, watching business news consistently will give you a better idea of what to invest in than this forum most likely. After a week or month of watching you'll have a really good idea for where the market is shifting. One would hope -- but research seems to point to the exact opposite. Check out Cramer's results. The most generous of studies find that he adds zero stock selection alpha (i.e. his portfolio outperformance, when risk adjusted, is diminished to mere market performance) -- and the more vicious ones find that he is absolutely atrocious in his recommendations. In my opinion, CNBC is an awful source of investment advice. At the end of the day, it is all for entertainment (hence why they keep putting on more attractive, female anchors).
  10. Anyone know any good stocks to invest in?

    Yeah, this is a great way to lose your money. But a serious response: For someone who cannot afford a financial advisor and wants to invest on their own, I would highly recommend dollar cost averaging into low-cost index ETFs. You would probably do well to split your money across US, International, and Emerging markets, as well as put a portion into US Bonds, High Yield Bonds, and Treasuries. If you don't have enough money for the latter, there are ETFs that can give you access for lower cost. Look up "Lazy ETF Portfolios" in google (e.g. SeekingAlpha article) for an idea of how to construct your portfolio to get good all around market exposure within your risk tolerance (i.e. how do you want to allocate between equities, bonds, currencies, commodities, real-estate, cash [never underestimate the value of liquidity], et cetera). Dollar-cost averaging is a technique that has you investing a fixed amount in the market over equal time-periods, allowing you to purchase more as index prices go down, and less when they go up. This works extremely well over the LONG RUN in index products. Remember: RETURN IMPLIES RISK! If you get a high return, it is because you took on a considerable amount of risk. When you invest in a company, you are taking on a considerable amount of risk. Some that can be diversified away with other holdings (specific, company risks), and some that can't (economic risks). There is NO FREE LUNCH IN INVESTING.
  11. It was fun. The gameplay didn't seem very deep. I just bought a medium sized ship, loaded it with lots of cannons & ammo, then searched for ships w/ 70+ damage. Once I killed them, I would repair at port. Rinse, wash, repeat. Other than that ... there didn't seem like much to do...
  12. Well, all I can say is that I am very, very, very impressed, and I can really see something like this being applicable technology to something like a netbook or other smaller portable devices. It would be interesting to hear more about the back-end side of it. Where does the server run? Can I run an instance of the server on Amazon EC2 and run 10 web interfaces to that server? What sort of technology do I have to dedicate to it? All in all, a very cool project -- with which, I can't imagine you would have too much trouble finding employment...
  13. Like Quake? You'll love this (almost as much as pie)!

    Quote:Original post by owl Yeah. I meant that soon we won't need to install the games in our machines. They will be served to us as we play them. I can even envision that someday, games won't even run in our PCs, we will be served with the video while our input gets transmited to the server. People won't even need good video cards to play games with complex graphics. There was a really slick demo from a start-up that is doing exactly this. They were doing Far-Cry (I think), served remotely from their game-centers (located at major isp-hubs in the country), to dumb terminals attached to your TV. The trick was in their video compression software. But basically, the dumb box just displayed video and sent your controller signals over the net. It was really smooth, and really impressive. I can't find the video, unfortunately. But it looked like a really, really cool system.
  14. Convex Bounding Hull for N-Dimensional Point Cloud

    Ended up just piping stdin to qhull -- which works for now. Thanks.