Slashdot Mirror


User: sketerpot

sketerpot's activity in the archive.

Stories
0
Comments
1,473
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 1,473

  1. Re:No real spoilers in the clip. on Footage From Star Wars: Episode III · · Score: 1

    I thought that Jar Jar did much better in episode 2, since he only got a few minutes of screen time, and then only as a weak-minded fool who sold out the entire republic.

  2. Re:Good. on Apple and Pepsi Ad Sports RIAA Targets · · Score: 1

    True, although the RIAA did send goon squads around to various stores to confiscate pirated CDs. I hope the police take a dim view of the impostors.

  3. Re:WOO HOO! on Google Social Network: Orkut · · Score: 1

    Actually, this is a good thing. Have a popular site to allow people to continue their vacuous clique discrimination, and then stay away from them. The more they're circle jerking, the less they're bugging you.

  4. Re:bah on Is Your Silver-based Thermal Paste Really Silver? · · Score: 1

    I thought that it had a very heat-conductive outer layer and a very insulative inner layer. That way the heat is dispersed better.

  5. Re:bah on Is Your Silver-based Thermal Paste Really Silver? · · Score: 1

    You had teeth? Talk about luxury! I had to brush my teeth with a stick, which is thoght to have anti-bacterial properties, and I didn't even have teeth to brush, but I still brushed them!

  6. Re:Forshortened, and three heretics ago... on IBM Supporting Linux On Power Processors · · Score: 1

    We'd need a tree structure, too. Something like, ((KDE, GNOME)/XFree86/), bash)/GNU/Linux, only much bigger and thornier. I wonder if the Ents are into Lisp....

  7. Re:Is lying to Congress illegal? on SCO Lobbying Congress Against Open Code · · Score: 1
    No, they would worm out of it. Come to think of it, it may be illegal. It wouldn't surprise me.

    Think about it: having another rule can only be to the benefit of these guys. There are so many rules out there that nobody knows all of them, and they could just pull this one out of the book to thwack someone they don't like. Plus, if there's one thing these guys are good at, it's squirming and writhing in slime.

  8. Re:most humans have 2 hands moron on A Glance At 24 Keyboards & Mice · · Score: 1
    Apparently you don't have an FPU.

    Really, though, the debate seems nearly moot to me, since it isn't that hard to type one-handed on a standard keyboard, and it beefs up your arm muscles. That said, however, some optimization would be welcome on the off-chance that I lose a hand.

    This comment was typed with one hand.

  9. Re:Best Keyboard... on A Glance At 24 Keyboards & Mice · · Score: 1
    As for the PS/2/USB debate (yeah, not sure how else to write that)

    We need some kind of grouping symbol. Parentheses and brackets are used regularly in ordinary text, but those little curly braces languish unused except by, say, TeX. S, shout it proudly: {PS/2}/USB debate!

  10. Re:Question of gender... on A Glance At 24 Keyboards & Mice · · Score: 1

    Perhaps you should read that again. The thing here is that a man submitted a story, and the editor responsible thought that Robyn was a woman. The editor was acting as you would expect an editor who is used to having story submissions from both men and women to act.

  11. Re:Better security? on The State of IPv6 · · Score: 1

    OTOH, it could be a good thing to have complexity moved from userland programs over into a stable, actively maintained IPv6 imlementation.

  12. Re:and by the way on NASA Cancels Hubble Mission, and Other Space Bits · · Score: 1
    If the Chinese, with a monopoly on the moon, decide they want to start lobbing missles at us from there, they'd be pretty well fucked. You see, there are many countries with missiles who would take a very dim view of china's moon missle thing. If china used conventional missiles, I don't see how it would be that much different from ICBMs except that we get more time to respond and they bypass our northern missile-detection systems. They'd be screwed. Or didn't you realize that the US has the most powerful military in the world?

    If China sends nukes at us, we would be screwed, whether the nikes were from China or from the moon. The normal response is to nuke China, resulting in Mutual Assured Destruction, which is where we stand today.

    Not everything revolves around your paranoid militarism, or that of the moderator who thinks you're "interesting". What part of that don't you understand, troll?

  13. Re:Revisit Sojourner! on Spirit Rolls on Mars · · Score: 1
    They could take the material home and put it on their mantle as a souvenir and it wouldn't hurt anyone.

    Actually, that would be cool. I wonder if NASA would consider putting bits of old RTGs (test models that never left earth), encasing them in glass or plastic or something, and selling them as desk toys. I'd buy one.

  14. Re:Revisit Sojourner! on Spirit Rolls on Mars · · Score: 1

    Alright, since you seem so fixated on unlikely events, let's look at the consequences that have *already been discussed*. Suppose that an RTG broke open for some spectacularly hypothetical reason, spewing its contents into the atmosphere over, say, a populated region. IIRC, you'd have a fairly safe form of plutonium dispersed over a wide area. You might have traffic jams of hysterical people fleeing from an imagined threat, but the radiation increase in any spot would be negligible. The plutonium would be in a fairly safe form, and eventually people would get over the shock.

  15. Re:They did produce 60,000 lines of code .... on SCO Fails to Produce Evidence · · Score: 2

    They are not only unimportant, but in the US you can't copyright, say, a phone directory because it's just a list of facts without the "requisite originality" to be copyrighted. I don't see why errno.h should be any different.

  16. Re:Gentoo! on SCO Fails to Produce Evidence · · Score: 3, Funny

    Once it was compiled, it was able to realize that it didn't have the ethernet card driver installed---faster than the others! Back to kernel configuration! (Whee! Fun! Seriously, almost.)

  17. Re:and bush says... on USA To Return To Moon By 2015, Then Mars · · Score: 1

    It would be much cheaper for the porn industry to get their hands on a rented KC-135-type plane and do the zero gravity thing in the atmosphere.

  18. Re:Wow! Metrics! on Mars Rover Rolls And Turns · · Score: 1

    Oops. Thanks for clearing that up. Darn newspapers not adequately differentiating between NASA and JPL....

  19. Re:Wow! Metrics! on Mars Rover Rolls And Turns · · Score: 1
    They wouldn't, but their subcontractors certainly would, and (as you pointed out) did.

    Down with Imperial units! Metric now!

  20. Re:Spamkiller doesn't care on Filter-foiling Gibberish Becoming A Spam Staple · · Score: 1
    Yes, that looks like a better solution than the current naive method. Interestingly, what my filter does it treat @ as a token separator. Naturally, tokens like "xan" and "gra" get rather high spam probabilities.

    I wonder if you could also try matching tokens longer than, say, four letters, to your recorded spam words using some spellchecking algorithm. Something like apache's mod_speling might help.

  21. Re:Spamkiller doesn't care on Filter-foiling Gibberish Becoming A Spam Staple · · Score: 1

    I'm not talking about anything that hard and fast, oh sarcastic one. Just taking them into account may improve spam filters, although of course you'd have to do tests to make sure that the overall effect was positive. If y0u h4v3 a message w|th lots of |nterupt|0nz, you usually don't want to see it, in my experience.

  22. Re:An attempt to make Bayesian analysis a pain? on Filter-foiling Gibberish Becoming A Spam Staple · · Score: 1

    I guess you could periodically perform a spring cleaning on your spam corpus by removing all but the n most common words, where n is some manageable size.

  23. Re:The problem with this technique on Filter-foiling Gibberish Becoming A Spam Staple · · Score: 3, Informative

    In most adaptive filters, only words that have been used a certain number of times are taken into consideration. For example, the original Plan for Spam algorithm ignores any word that doesn't appear over 5 times in the corpus.

  24. Re:filtering on Filter-foiling Gibberish Becoming A Spam Staple · · Score: 1

    A bayesian filter should do the same thing, but with more ability to tune it. Plus, if one of your regular correspondents often misspells, say, "teh", then you can train your filter to ignore it.

  25. Re:Why? on Filter-foiling Gibberish Becoming A Spam Staple · · Score: 1
    By the way, why are Bayesian filters mostly run at home, rather than by ISPs? I can think of two possible reasons. First, one of the strengths of Bayesian filters is that everyone's email is different. One person's good words may not be another person's. However, you could maintain some of that diversity by having each ISP do further training on a basic corpus.

    The other reason is that email servers are loaded enough as it is, and Bayesian filters are somewhat processor-intensive. If so, then we really do have a use for the faster and faster processors that are coming out all the time even if we don't do video editing. Whee.