Speaking of illigitimate Bebop, can anyone recommend a DivX player that doesn't crash and burn on certain episodes, but is nicer than Windows Media Player? I'm running DivX Player 2.1 on WinME (circumstances beyond my control). It handles most things better than WMP, but it crashes in the middle of Session 24.
It's just a wild guess, but, well, I've seen radio interference from weather drastically amplified in the vicinity of iron pipes, for instance. Improperly insulated power cords, malfunctioning power supplies... that sort of thing I would look at.
Do you have any links to good web sites about that sort of thing? I have wireless internet, but the ISP is a bit... amateur. It's very cheap, but my connection craps out frequently. It's still good enough to use the web and download Futurama episodes, but it could be better. A lot better.
If nothing else, it'll artificially boost China's aerospace industry and wean them away from dependence on American collaborators like Boeing.
And what happens after China's moon frenzy dies off? The Chinese aerospace industry will probably have greatly overextended itself, and once the money stops flowing in like it used to, they're going to have big problems.
I predict bad things in the future for the Chinese aerospace industry. Perhaps this is a job for foresight exchange...
Uhh... Evil Islamic terrorists woke the US up on Sept. 11. Have you forgotten already?
Sep. 11, 2001, to be precise. How could anyone forget that with people like you always waving it in our faces all the time? Hopefully one day it will be the sole province of history books and old geezers.
I would prefer that the US stay alert and defend itself!
The funny thing about the US department of defense is that it seems to devote most of its money to attacking other countries. Don't get me wrong, Saddam Hussein and friends deserved what they got, and the US pulled off the invasion nigh perfectly. I still wish that the US could try subtler tactics in Syria or whoever is next in line, like asassinating (or capturing and bringing to an international war crimes trial; I won't be picky, and that's probably more humane) the worst of the leaders, which would stir things up a bit while being less invasive than... well, an invasion.
Hell, lots of things are better than just declaring "Guarded" status and attacking people. Oh well, I've mostly stopped caring and resigned myself to being powerless to affect anything political.
Python has advantages and disadvantages. It is a great language that should have killed VB years ago, very complex code has a tendency to work well the first time, it is in general a very aesthetic language. OTOH, VB has a very quick way to make quick and dirty GUIs, whereas in Python you generally have to take more time, although there are some very good GUI toolkits for Python, like Qt, GTK, wxWindows, and even Tk or MFC.
Python is good at connecting with C/C++/Fortran, but it requires a little bit of extra effort to turn DLLs into modules. There are, of course, tools to help you: SWIG, SIP, pyfort, and others. Once you have python modules, though, they're very pleasant to use.
I don't know what sort of stuff VB has for, say, fetching things from the internet, but I know that Python has excellent stuff there. urllib is great for lightweight fetching, and you can get heavyweight if you need to. There is a simple HTTP server class, and in fact there are classes for a lot of things that fit beautifully with the rest of the language.
There are some modules fot python that may be particularly useful for research like Numeric and SciPy, and I seem to recall some biology module for python. I don't know what VB has in their place, but I'd bet that it wasn't as pleasant to use.
Different languages, different things. But given a choice, I'd pick Python.
Which is why you use forward error correction. Have you ever scratched a CD? It can still play, thanks to FEC. (Cross Interleave Reed-Solomon Code, to be specific---good at correcting fairly small numbers of errors, like somebody drilling a 1 mm hole in a CD).
Sweet. That would be great for multimedia stuff (sound files would be simple), and I suppose you could make some sort of homebrew TiVo thing. At least you could make a DivX/DVD player. The firewire sounds nice for that too, as does the ethernet.
Re:Probably it will always stay...
on
BitTorrent Guide
·
· Score: 1
If that becomes the sort of common sense thing that everybody who knows about BT knows, then I think we can get past the percieved legal issues (it's a lot like one of those P2P filesharing piracy things!). BTW, it's sort of funny to see slashdot running cheerful stories on a partial cure to the slashdot effect.
BT can be very cool, but people have to realize that it is just as legitimate as mirror sites.
Then the story was written by people who didn't bother to think first. You're right; any plan to sell the salt for a profit will fail. Even if the ideas of using this as a salt production method are used and fail, somebody will probably use this as a way of reclaiming soil and selling the salt to defray the costs. Therefore, it's nice to think about. And who knows, there may be something to the "health salt" ploy. With enough advertising, perhaps people would be dumb enough to fall for it....
Wouldn't a motorcycle be a bit cold in the winter? The site rejects bikes as a reasonable major part of transportation for that reason since there aren't all that many places with climates temperate enough.
That reminds me of the way I use to reduce the crime caused by arcologies (in 2000). Parks actually worked better than police stations. That seems to be the sort of thing the plans on the web site are planning (green space in every block, that sort of thing).
There are plenty of cops and legislators and judges who are annoying pains in the butts of the masses. A law banning making virtual child pornography (no actual children involved) was struck down, so I don't see what legal ground they would have, even if they think that Lain is child pornography.
I'm not going to hold in my opinions on net filtering software and address your post's parent: do you have any idea how easy it is to get around filtering software? Do you honestly think that filtering software actually works? You may think that filtering software is some magical technology that blocks out the filth on the internet while letting most legitimate pages through, but IT'S NOT! If more people understood this rather than turning to a magical "solution" that doesn't work, we'd all be a lot happier!
How the hell do you know that this guy didn't run, say, Freenet? There are ways. How do you know he didn't use a CGIproxy over HTTPS? There are plenty available. Really, installing web filtering software is like seeing a bunch of barrels of apples, noticing that there is one rotten apple, tossing out the whole barrel, then declaring that you've gotten rid of all the rotten apples.
Do you just delete spam? I find it unpleasant to even see the spam sitting there and to have to press delete. So I use POPfile to delete my spam for me. One problem down---for me. Now if only such good spam filtering could be available for the masses....
If Stallman insisted on getting GNU in there anywhere, I'm pretty sure it would be the more logical (and less easy to mindlessly troll about) NASA/GNU[/Linux/whatever]. Anyway, RMS would, I'm sure, leave NASA alone on this unless they tried making, say, a GUI on top of GNU and calling it an operating system. Then he would be on them like a dog.;-)
Re:The first /.-proof website?
on
Gentoo Games
·
· Score: 1
If only part of the picture needed to change, I'd use static positioning with CSS to put down first the unchanging part of the image (just a PNG file in a directory somewhere) and then to overlay the dynamic part in just the right place. And yes, you could store quite a bit of that in a database....
Who cares? I don't think the SubGenius people would have any problems with calling a browser "Bob". After all, if they won't cut you a little slack, who will?
Ah, but multiply the time it takes to perform, say, a cryptographic hash function, by the number of emails spammers send out, and you get some big delay indeed, no matter what the hardware. If you then combine this with widespread bayesian filtering and incremental filtering tarpits, you get an internet that is a very hostile place for spammers to be. It would be like a less deadly version of the Darwin Award winner who stuck up a gun store in texas and tried to rob them.
It wasn't so bad because it was nearly dead, rather than a huge thing that gets packaged with most office software and goes wandering around like a zombie of a language that just won't die. Why can't it be replaced by a good language like Python? (I also hear good things about Ruby, and the example programs look pretty clean. I haven't gotten around to learning it yet.) I mean, who wants to keep writing "dim" statements until the end of their days?
Let me guess: you were forced to read The Crucible in High School and, instead of properly forgetting it, your brain actually remembered what the Puritans did to Giles (Giles Corey?).
Indeed. It would also be nice if they would actually link to Spamhaus.org rather than just misspelling it. There was one bit near the end where it said that the spammer's business wasn't sending out spam because, for some unknown reason, people tended to associate spam with scum. I wonder why....
2. Even if you don't assign a cut-off point as in #1, you just make "characteristics" out of the number of new tokens. For example, if you have a message that contains 50-60% new tokens, that itself becomes a new Bayesian token. Perhaps, over time, Bayesian will find that "messages with 50-60% new tokens have an 80% chance of being spam." So the fact that they munge becomes a damning factor even if the computer can't identify the actual munging.
3. You add new characteristics as in #2. Perhaps another characteristic is "Messages that contain no body except for a URL." Perhaps 85% of those messages are spam, and Bayesian can count that as a damning characteristic. Or, perhaps, messages where over 50% of the body are devoted to URLs have a 90% chance of being spam. All these add new "characteristics" that can be used to calculate a spam probability for Bayesian.
These are great ideas. If you're going to introduce that sort of thing into a Bayesian filter, it's best if you have the characteristics being processed the same way as the rest of the tokens: with Bayes' Law. I have some more ideas for characteristics. Something determined by looking in a seperate Bayesian sweep of the subject line in which the tokens are characters, so that, say, "*____-__" would have a very high spam probability. Take the probability from this and make it just another characteristic. Do something similar on the host names in the headers, splitting the tokens into things like "sketerpot", "chase3000", and "com", and then letting your corpus be a blacklist. Can anybody think of some more?
Yes, it is amusing to watch big companies pay people to make lists of "words" like V!^gra and still just raise the bar into the spam world a bit. My Bayesian corpus is still pretty young, yet it's only let a single spam through this week, and none for several weeks before that. Still, it would have a good deal of satisfaction if we could put the spammers out of business. Perhaps if all email programs got their act together, standardized a format for corpus exchange (I love that phrase;-)), and made Bayesian filtering with a fairly decent corpus that would automatically put spam in a place you don't normally look an integral part of the program, and made the spam folder empty after 24 hours, then we could finally finish these spamming bastards, rather than wringing our hands about it and getting politicians to pass laws about it to increase their popularity. Cheers.
Slashdot Mirror
Just wondering....
Do you have any links to good web sites about that sort of thing? I have wireless internet, but the ISP is a bit... amateur. It's very cheap, but my connection craps out frequently. It's still good enough to use the web and download Futurama episodes, but it could be better. A lot better.
And what happens after China's moon frenzy dies off? The Chinese aerospace industry will probably have greatly overextended itself, and once the money stops flowing in like it used to, they're going to have big problems.
I predict bad things in the future for the Chinese aerospace industry. Perhaps this is a job for foresight exchange...
Sep. 11, 2001, to be precise. How could anyone forget that with people like you always waving it in our faces all the time? Hopefully one day it will be the sole province of history books and old geezers.
I would prefer that the US stay alert and defend itself!
The funny thing about the US department of defense is that it seems to devote most of its money to attacking other countries. Don't get me wrong, Saddam Hussein and friends deserved what they got, and the US pulled off the invasion nigh perfectly. I still wish that the US could try subtler tactics in Syria or whoever is next in line, like asassinating (or capturing and bringing to an international war crimes trial; I won't be picky, and that's probably more humane) the worst of the leaders, which would stir things up a bit while being less invasive than... well, an invasion.
Hell, lots of things are better than just declaring "Guarded" status and attacking people. Oh well, I've mostly stopped caring and resigned myself to being powerless to affect anything political.
While we're at it, I'll just say this: Israel and Palestine both stink!
Python is good at connecting with C/C++/Fortran, but it requires a little bit of extra effort to turn DLLs into modules. There are, of course, tools to help you: SWIG, SIP, pyfort, and others. Once you have python modules, though, they're very pleasant to use.
I don't know what sort of stuff VB has for, say, fetching things from the internet, but I know that Python has excellent stuff there. urllib is great for lightweight fetching, and you can get heavyweight if you need to. There is a simple HTTP server class, and in fact there are classes for a lot of things that fit beautifully with the rest of the language.
There are some modules fot python that may be particularly useful for research like Numeric and SciPy, and I seem to recall some biology module for python. I don't know what VB has in their place, but I'd bet that it wasn't as pleasant to use.
Different languages, different things. But given a choice, I'd pick Python.
Which is why you use forward error correction. Have you ever scratched a CD? It can still play, thanks to FEC. (Cross Interleave Reed-Solomon Code, to be specific---good at correcting fairly small numbers of errors, like somebody drilling a 1 mm hole in a CD).
Sweet. That would be great for multimedia stuff (sound files would be simple), and I suppose you could make some sort of homebrew TiVo thing. At least you could make a DivX/DVD player. The firewire sounds nice for that too, as does the ethernet.
BT can be very cool, but people have to realize that it is just as legitimate as mirror sites.
Then the story was written by people who didn't bother to think first. You're right; any plan to sell the salt for a profit will fail. Even if the ideas of using this as a salt production method are used and fail, somebody will probably use this as a way of reclaiming soil and selling the salt to defray the costs. Therefore, it's nice to think about. And who knows, there may be something to the "health salt" ploy. With enough advertising, perhaps people would be dumb enough to fall for it....
Wouldn't a motorcycle be a bit cold in the winter? The site rejects bikes as a reasonable major part of transportation for that reason since there aren't all that many places with climates temperate enough.
That reminds me of the way I use to reduce the crime caused by arcologies (in 2000). Parks actually worked better than police stations. That seems to be the sort of thing the plans on the web site are planning (green space in every block, that sort of thing).
It may get split by slashdot, but it's all one word...
There are plenty of cops and legislators and judges who are annoying pains in the butts of the masses. A law banning making virtual child pornography (no actual children involved) was struck down, so I don't see what legal ground they would have, even if they think that Lain is child pornography.
How the hell do you know that this guy didn't run, say, Freenet? There are ways. How do you know he didn't use a CGIproxy over HTTPS? There are plenty available. Really, installing web filtering software is like seeing a bunch of barrels of apples, noticing that there is one rotten apple, tossing out the whole barrel, then declaring that you've gotten rid of all the rotten apples.
Do you just delete spam? I find it unpleasant to even see the spam sitting there and to have to press delete. So I use POPfile to delete my spam for me. One problem down---for me. Now if only such good spam filtering could be available for the masses....
If Stallman insisted on getting GNU in there anywhere, I'm pretty sure it would be the more logical (and less easy to mindlessly troll about) NASA/GNU[/Linux/whatever]. Anyway, RMS would, I'm sure, leave NASA alone on this unless they tried making, say, a GUI on top of GNU and calling it an operating system. Then he would be on them like a dog. ;-)
If only part of the picture needed to change, I'd use static positioning with CSS to put down first the unchanging part of the image (just a PNG file in a directory somewhere) and then to overlay the dynamic part in just the right place. And yes, you could store quite a bit of that in a database....
Who cares? I don't think the SubGenius people would have any problems with calling a browser "Bob". After all, if they won't cut you a little slack, who will?
By the way, anyone caught sending me spam will be forced to work at home stuffing Viagra into inkjet cartriges. Be afraid!
Ah, but multiply the time it takes to perform, say, a cryptographic hash function, by the number of emails spammers send out, and you get some big delay indeed, no matter what the hardware. If you then combine this with widespread bayesian filtering and incremental filtering tarpits, you get an internet that is a very hostile place for spammers to be. It would be like a less deadly version of the Darwin Award winner who stuck up a gun store in texas and tried to rob them.
Curse you, Microsoft, for resurrecting Basic.
What part of the midwest are we talking about here?
Or something else....
Indeed. It would also be nice if they would actually link to Spamhaus.org rather than just misspelling it. There was one bit near the end where it said that the spammer's business wasn't sending out spam because, for some unknown reason, people tended to associate spam with scum. I wonder why....
3. You add new characteristics as in #2. Perhaps another characteristic is "Messages that contain no body except for a URL." Perhaps 85% of those messages are spam, and Bayesian can count that as a damning characteristic. Or, perhaps, messages where over 50% of the body are devoted to URLs have a 90% chance of being spam. All these add new "characteristics" that can be used to calculate a spam probability for Bayesian.
These are great ideas. If you're going to introduce that sort of thing into a Bayesian filter, it's best if you have the characteristics being processed the same way as the rest of the tokens: with Bayes' Law. I have some more ideas for characteristics. Something determined by looking in a seperate Bayesian sweep of the subject line in which the tokens are characters, so that, say, "*____-__" would have a very high spam probability. Take the probability from this and make it just another characteristic. Do something similar on the host names in the headers, splitting the tokens into things like "sketerpot", "chase3000", and "com", and then letting your corpus be a blacklist. Can anybody think of some more?
Yes, it is amusing to watch big companies pay people to make lists of "words" like V!^gra and still just raise the bar into the spam world a bit. My Bayesian corpus is still pretty young, yet it's only let a single spam through this week, and none for several weeks before that. Still, it would have a good deal of satisfaction if we could put the spammers out of business. Perhaps if all email programs got their act together, standardized a format for corpus exchange (I love that phrase ;-)), and made Bayesian filtering with a fairly decent corpus that would automatically put spam in a place you don't normally look an integral part of the program, and made the spam folder empty after 24 hours, then we could finally finish these spamming bastards, rather than wringing our hands about it and getting politicians to pass laws about it to increase their popularity. Cheers.