A study showed that playing Go has positive effects on young brain development similar to studying music. It's a tremendously deep game with simple rules.
We haven't had much problem with linux nis in a crossplatform environment, except one of linux nis server's ypserv dies sometimes.
It's direct maps that autofs doesn't support.
We haven't seen the ypbind problems you describe, but I'm pretty sure you're not alone on this.
I believe the default recv and xmit has been bumped up. In fact, I believe linux (or was it just redhat?) cranked this so high that it confused netapps - there was a patch to lower it again, IINM.
> # There's a database of devices. Probe order on > the bus has NOTHING to do with what device > matches with what entry in/dev. If I have 2 > SCSI cards on my PCI bus and then I add a third > SCSI card inbetween the two existing ones, every > single device on the original SCSI cards stays > with the same entry in/dev. The two original > controllers are "c0" and "c1", the third (new) > one, though it is probed before c1 and after c0, > is called "c2", because it is important to keep > everything the same for c0 and c1. This is > important on servers that have multiple SCSI > cards!! (And, this database of devices is stored > in ASCII in case you need to fix it manually, > unlike the equivalent thing in AIX...)
I like Solaris ok, but I'm not convinced this is an advantage. The choice isn't between "renumbering controls" and "not renumbering controllers", but rather "renumbering controllers when you install a controller" and "renumbering controllers next time you reinstall the system".
Personally, I prefer "renumbering controllers when you install a controller", because then it's more obvious what you should attribute the sudden problems to. If you put it off until install time, you may not even have the same admin working on the machine anymore, in which case it can be really confusing.
...about reading this article. Give the sensationalistic journalism a hit? Nah, rather not. But I'd like to have the knowledge about the warts of such a nice system.
In the end I clicked the link. And it didn't render correctly in galeon. So I decided to ignore the article afterall.
1) Sun sponsored a free implementation of NFSv4 for linux.
2) Solaris' NFS was perfectly usable for us in 2.4, probably earlier.
3) Sounds to me like you were doing something wrong. Our suns have been highly reliable.
4) If you think that, you don't have much design experience. Some things need to be on the same box, unless you go with something like GFS on RAID, which isn't exactly inexpensive.
Actually, sparc isn't proprietary, and neither is sbus. Both are open standards. x86 proprietary, despite being more common and cheaper. Not sure about PCI.
It's really the industry that needs to rethink things - to get with the program and support open standards. Sun's been making the right moves, and getting slapped around for it.
1) Itanic runs 64 bit apps fast, and 32 bit (x86) apps slowly. Would a 64 bit, itanic-native x86 emulator like that from connectix, run x86 apps faster than the "native" x86 support in itanic? Could the support be made seamless, so the enduser doesn't care if they're running an x86 binary or an itanic binary using something based on connectix's product?
2) This kills off the market for a major away-from-microsoft-only migration tool - virtualization products. Sure, microsoft benefits from selling windows licenses in the VM products, but how many of the people buying a virtualization product wouldn't have bought a windows computer instead if they couldn't get the virtualization? And how many of those folks wouldn't then have had the money for a mac, or a sun, or a linux-only x86 computer in addition to the "required" windows box? How many of the people running virtualization products are running microsoft oses only as long as they have to, until they can find workable app alternatives in their OS of choice? In a not insignificant number of cases I bet it wouldn't even -be- their OS of choice, anew or still, if it weren't for the virtualization product. Virtualization products are a large anti-lockin force. Microsoft is anti-anti-lockin.
Personally, I think that itanium faces an uphill battle, and that if x86-64 wins, AMD will be sitting pretty - more than the article indicates. Compatibility is king, especially compatibility with all those windows apps that are currently out there, and if a chip vendor's servers can run the same binaries as their desktops, you have a winner because desktops are where the volume is.
Intel built its lead through what seemed like excessive compatibility during the 80x86 vs 68k wars - they now ignore compatibility at their peril. I don't think they've built up so much brandname recognition that they can suddenly ignore this.
Sure, Redhat has every right to try to turn a buck however they see fit. But they have to reckon with the possibility of losing market share or even profit if they make a wrong turn (or do the same poor thing too long, too).
I think the writing is on the wall. This is a minimal-backlash way of making what used to be the real redhat distribution less attractive, and the AS product more attractive, because most of their profit last quarter came from AS.
It's minimal backlash, because a lot of people don't get that it's not practical for professional support people to upgrade all thei boxes once a year. Sure, I upgrade my home box every time they make a new release, but at work, we do all we can to keep a stable computing base for our clients.
A long time ago, back around redhat 4.2, we converted from debian to redhat. I made that switch whole-heartedly. I became a redhat advocate. But I'm having serious second thoughts about that now. Like I said, I don't blame them for trying to make a profit, but their goals may no longer be compatible with my personal goals or the goals of the team I work in at the university here.
OpenBSD dug around in community software, came up with tons of fixes, and didn't tell anyone about them. If you say to them "You should share your results", they just command you to grovel around in their source tree to find out what they did.
Frankly, it's poetic justic that Sun is telling OpenBSD that they have to grovel around in a Linux source tree to find what they need.
People, Microsoft loves it when you tweak their box this way. Publically, they're against it, but you're helping their bottom line, and as long as your tweaks are unofficial, they're marginal enough to be ignored. In fact, even if microsoft itself sold the modchip, it'd still be helping their bottom line to purchase it.
Do you think they care what weird software you put on their box? No. What they care about is their sales numbers, which is the only way your tweak impacts them. Their sales numbers, when high enough, mean more titles for xbox (and less for cooler platforms). Their sales numbers mean microsoft proprietary hardware, which will probably eventually run windows.
Imagine, microsoft with a wholly controlled (no, your silly modchip doesn't matter to them, it's still wholly controlled) hardware platform, to match their wholly controlled OS. It's a dream come true for their stock holders, and a very bad thing for FOSS, and a very bad thing for competition in the console market - because you can bet that if they get more than a toe hold on that market, they're going to run with it and never let another competitor see the light of day ever again.
So please, if you love high prices and lack of software freedom, keep snapping up those xboxes! Tweak to your heart's delight! Microsoft will be your friend (at least until they can make a buck by selling your soul)!
Smarteiffel (formerly smalleiffel) can compile to jvm.
I believe jvm isn't a good fit for some languages, but it does appear to work out for higher level stuff like eiffel and (the already mentioned) python (jython).
You're confused because we're exceptionally diligent here in applying our patches and following security practices. But with hundreds of systems, over an indefinite number of years, there's -no- way we're going to avoid 100% of breakins without pulling our stuff off the net. Bear in mind that the defenders have to take care of -every- vulnerability, while the attackers (dolts) have to just find one vulnerability - basically, it's worlds harder to defend than to attack.
You're also confused in saying I'm relying on people respecting our machines. It's 100% reasonable to encourage that respect technically -and- politically -and- legally, but I have 0 delusions that my efforts will make all idiots out to prove they're "hackers" stop being idiots.
You're also confused in saying I have no influence. Believe it or not, people, including you, read slashdot. Some will disagree, others will agree. That's influence. And when we talk to the police about how we'd like things handled, that's influence. And when I write my congresscritters about how I'd like to see things handled, that's influence.
(BTW, some hackers really are hackers. By that, I mean the ones who code up useful software systems, including -writing- exploits. But the script bunnies who use the exploits without the foggiest clue how they work are beneath contempt, especially when they start bragging about how cool they are)
Actually, if I were a betting man, I'd bet you were a script bunny who was in denial that someday you might face serious consequences for what you do in your leisure time. But if I just up and claimed that about you, I'd be making the same kind of (rather obvious, really) mistake you've made in your claims about me - so I won't claim it.
Obviously the usual method of law should be used, but they shouldn't be merely slapped on wrist, nor are they anything like graffiti artists.
Everytime some idiot joyrider breaks into one of our systems, we end up reinstalling, which takes a lot of time - and even that isn't really enough to definitely prevent reentry. You may think they just log in and start up an irc program, but how are the white hats supposed to know that? There's no way of telling if there was more to it or not, so you have to treat it seriously - reinstall, maybe even change all the passwords, inspect all the.forwards, all the cron jobs, and so on.
Personally, I believe they should be put on trial, and the potential punishments should be severe. Nothing like graffiti, this is more like stealing a car for some joyriding and leaving a note saying you've planted explosives and drugs somewhere in the car, and "just try to find it, nyah, nyah", and also that they've had a duplicate key to the car made (they may not have, but the note says they did, maybe just to freak out the owner, or maybe they really did...).
Because all of these things can potentially be done to a system, once it's been broken into.
If the media hadn't decided to misuse the word hacker so egregiously, we probably wouldn't have nearly the problems we do with stupid kids trying to prove "how smart they are". Face it, breaking into computers is stone stupid. But somehow the media likes to call people who do it "too smart for their own good" which is a horribly positively spun way of saying "kind of a little bit tech smart, but incredibly lacking in overall perspective".
Jesus, if he existed, wasn't born in the winter. Christmas was scheduled in December to steal the thunder of "pagan" winter celebrations like Saturnalia, which was about giving presents to children.
I've come to the conclusion that slashdot movie commentary is worse than useless. It's annoying.
Please, stick to tech. I can only pay attention to your movie reviews for the sake of doing the opposite of what you recommend.
I almost didn't see Nemesis because of the stupid comments here about it. That would've been a mistake. There are numerous other examples.
And what's with this fascination with Anime? Does it really have Anything to do with techies? I'm sure there are some techies who like Anime, but I'd guess it's far from the majority. In fact, if it weren't for slashdot, I doubt most of us would know what it is (I would, but I'm weird:)
Why would you -want- to run a modified xbox, when you can get a playstation? Why would you put up with being marginalized by microsoft, when you can be encouraged by Sony?
Sell your silly xbox (and its games) and get a real console (and its games) from a vendor with a clue.
I'm sure I'm starting to sound like a broken...
on
Behind Deep Blue
·
· Score: 1
...record, but give up on chess already. If the computer doesn't beat the human expert this time, it won't be long.
But go players won't be bested by computers for some time to come. I personally find it a bit depressing playing a game that a computer can beat me at, but with go, I can beat the best programs I've tried, which not long ago were the best in the world. Humans are far more challenging opponents to someone who's studied go part time for a few years.
The ones who use Linux tend to lack the jobs and income to be able to buy nice things like this to begin with.
I totally disagree. There are the young, I-can't-afford-anything-else folks running linux, but I believe they're the minority among linux users.
Besides moron, you could wipe the thing and install Linux on it. That should be enough to get you frothing at the bits.
But if you did, would you still have PPC in rom? I'd like to see this answered by someone who's actually investigated the issue.
Fault Palm for dropping the ball for taking so long to come out with a new OS (Palm OS 6 with Be technology). You can't slack when MS is on your tail.
Actually, isn't it pretty common for a market leader to avoid making big changes? If something's working, don't fix it. Personally, given how long it takes to get a Very-Different OS ready, I'd say Palm was on the ball here.
Why do so many geeks fail to realize that the ultimate geek challenge, go/baduk/weichi, hasn't really been dented by mathematicians, or even computer scientists?
Yes, there's been some productive analysis of the endgame, but that's So Vastly Simpler than the opening or middle game.
I'd contact the FSF to see if they're interested in developing a pool of patents they can cross license.
I have no faith that I could feed a dog for a year on proprietary gpg extensions.
But if my patented idea, held by the FSF, made it into a lot of key software, I'd be pretty much assured some big points in interviews when I say "Oh, you've used xyztech? Yeah, my algorithm is in that."
No guarantees the FSF is interested in doing this. But I hope so. If they aren't, I'd contact the OSI.
Slashdot Mirror
Check out http://kgs.kiseido.com/
A study showed that playing Go has positive effects on young brain development similar to studying music. It's a tremendously deep game with simple rules.
Give me a break. Microsoft isn't going to give up their lockin.
Using XML is:
1) An attempt to gain buzzword compliance
2) An attempt to popularize their dialect of an open standard
A document in XML can be every bit as proprietary as a binary document.
This is just going to make people wonder why non-MS-Word wordprocessors do such a "crappy" job of "reading/writing XML".
What's up with the D-H support in OpenSSL?
If I'm not too worried about MITM attacks, is it possible to make apache and other applications work without all that certificate stuff using D-H?
We haven't had much problem with linux nis in a crossplatform environment, except one of linux nis server's ypserv dies sometimes.
It's direct maps that autofs doesn't support.
We haven't seen the ypbind problems you describe, but I'm pretty sure you're not alone on this.
I believe the default recv and xmit has been bumped up. In fact, I believe linux (or was it just redhat?) cranked this so high that it confused netapps - there was a patch to lower it again, IINM.
> # There's a database of devices. Probe order on /dev. If I have 2 /dev. The two original
> the bus has NOTHING to do with what device
> matches with what entry in
> SCSI cards on my PCI bus and then I add a third
> SCSI card inbetween the two existing ones, every
> single device on the original SCSI cards stays
> with the same entry in
> controllers are "c0" and "c1", the third (new)
> one, though it is probed before c1 and after c0,
> is called "c2", because it is important to keep
> everything the same for c0 and c1. This is
> important on servers that have multiple SCSI
> cards!! (And, this database of devices is stored
> in ASCII in case you need to fix it manually,
> unlike the equivalent thing in AIX...)
I like Solaris ok, but I'm not convinced this is an advantage. The choice isn't between "renumbering controls" and "not renumbering controllers", but rather "renumbering controllers when you install a controller" and "renumbering controllers next time you reinstall the system".
Personally, I prefer "renumbering controllers when you install a controller", because then it's more obvious what you should attribute the sudden problems to. If you put it off until install time, you may not even have the same admin working on the machine anymore, in which case it can be really confusing.
...about reading this article. Give the sensationalistic journalism a hit? Nah, rather not. But I'd like to have the knowledge about the warts of such a nice system.
In the end I clicked the link. And it didn't render correctly in galeon. So I decided to ignore the article afterall.
1) Sun sponsored a free implementation of NFSv4 for linux.
2) Solaris' NFS was perfectly usable for us in 2.4, probably earlier.
3) Sounds to me like you were doing something wrong. Our suns have been highly reliable.
4) If you think that, you don't have much design experience. Some things need to be on the same box, unless you go with something like GFS on RAID, which isn't exactly inexpensive.
Actually, sparc isn't proprietary, and neither is sbus. Both are open standards. x86 proprietary, despite being more common and cheaper. Not sure about PCI.
It's really the industry that needs to rethink things - to get with the program and support open standards. Sun's been making the right moves, and getting slapped around for it.
1) Itanic runs 64 bit apps fast, and 32 bit (x86) apps slowly. Would a 64 bit, itanic-native x86 emulator like that from connectix, run x86 apps faster than the "native" x86 support in itanic? Could the support be made seamless, so the enduser doesn't care if they're running an x86 binary or an itanic binary using something based on connectix's product?
2) This kills off the market for a major away-from-microsoft-only migration tool - virtualization products. Sure, microsoft benefits from selling windows licenses in the VM products, but how many of the people buying a virtualization product wouldn't have bought a windows computer instead if they couldn't get the virtualization? And how many of those folks wouldn't then have had the money for a mac, or a sun, or a linux-only x86 computer in addition to the "required" windows box? How many of the people running virtualization products are running microsoft oses only as long as they have to, until they can find workable app alternatives in their OS of choice? In a not insignificant number of cases I bet it wouldn't even -be- their OS of choice, anew or still, if it weren't for the virtualization product. Virtualization products are a large anti-lockin force. Microsoft is anti-anti-lockin.
I thought x86's ram limit was 4 Gig?
Personally, I think that itanium faces an uphill battle, and that if x86-64 wins, AMD will be sitting pretty - more than the article indicates. Compatibility is king, especially compatibility with all those windows apps that are currently out there, and if a chip vendor's servers can run the same binaries as their desktops, you have a winner because desktops are where the volume is.
Intel built its lead through what seemed like excessive compatibility during the 80x86 vs 68k wars - they now ignore compatibility at their peril. I don't think they've built up so much brandname recognition that they can suddenly ignore this.
If that's true, then it's a HUGE change from how they used to handle security fixes.
Did you forget the huge number of times Theo posted saying "Yeah, you just found this, but it was fixed in OpenBSD years ago. Heh."
Face it, OpenBSD made its name this way. If it's been discontinued, that doesn't mean it's not still reaping the fruit of its ill-gotten gains.
Sure, Redhat has every right to try to turn a buck however they see fit. But they have to reckon with the possibility of losing market share or even profit if they make a wrong turn (or do the same poor thing too long, too).
I think the writing is on the wall. This is a minimal-backlash way of making what used to be the real redhat distribution less attractive, and the AS product more attractive, because most of their profit last quarter came from AS.
It's minimal backlash, because a lot of people don't get that it's not practical for professional support people to upgrade all thei boxes once a year. Sure, I upgrade my home box every time they make a new release, but at work, we do all we can to keep a stable computing base for our clients.
A long time ago, back around redhat 4.2, we converted from debian to redhat. I made that switch whole-heartedly. I became a redhat advocate. But I'm having serious second thoughts about that now. Like I said, I don't blame them for trying to make a profit, but their goals may no longer be compatible with my personal goals or the goals of the team I work in at the university here.
OpenBSD dug around in community software, came up with tons of fixes, and didn't tell anyone about them. If you say to them "You should share your results", they just command you to grovel around in their source tree to find out what they did.
Frankly, it's poetic justic that Sun is telling OpenBSD that they have to grovel around in a Linux source tree to find what they need.
People, Microsoft loves it when you tweak their box this way. Publically, they're against it, but you're helping their bottom line, and as long as your tweaks are unofficial, they're marginal enough to be ignored. In fact, even if microsoft itself sold the modchip, it'd still be helping their bottom line to purchase it.
Do you think they care what weird software you put on their box? No. What they care about is their sales numbers, which is the only way your tweak impacts them. Their sales numbers, when high enough, mean more titles for xbox (and less for cooler platforms). Their sales numbers mean microsoft proprietary hardware, which will probably eventually run windows.
Imagine, microsoft with a wholly controlled (no, your silly modchip doesn't matter to them, it's still wholly controlled) hardware platform, to match their wholly controlled OS. It's a dream come true for their stock holders, and a very bad thing for FOSS, and a very bad thing for competition in the console market - because you can bet that if they get more than a toe hold on that market, they're going to run with it and never let another competitor see the light of day ever again.
So please, if you love high prices and lack of software freedom, keep snapping up those xboxes! Tweak to your heart's delight! Microsoft will be your friend (at least until they can make a buck by selling your soul)!
I believe jvm isn't a good fit for some languages, but it does appear to work out for higher level stuff like eiffel and (the already mentioned) python (jython).
You're sadly confused.
You're confused because we're exceptionally diligent here in applying our patches and following security practices. But with hundreds of systems, over an indefinite number of years, there's -no- way we're going to avoid 100% of breakins without pulling our stuff off the net. Bear in mind that the defenders have to take care of -every- vulnerability, while the attackers (dolts) have to just find one vulnerability - basically, it's worlds harder to defend than to attack.
You're also confused in saying I'm relying on people respecting our machines. It's 100% reasonable to encourage that respect technically -and- politically -and- legally, but I have 0 delusions that my efforts will make all idiots out to prove they're "hackers" stop being idiots.
You're also confused in saying I have no influence. Believe it or not, people, including you, read slashdot. Some will disagree, others will agree. That's influence. And when we talk to the police about how we'd like things handled, that's influence. And when I write my congresscritters about how I'd like to see things handled, that's influence.
(BTW, some hackers really are hackers. By that, I mean the ones who code up useful software systems, including -writing- exploits. But the script bunnies who use the exploits without the foggiest clue how they work are beneath contempt, especially when they start bragging about how cool they are)
Actually, if I were a betting man, I'd bet you were a script bunny who was in denial that someday you might face serious consequences for what you do in your leisure time. But if I just up and claimed that about you, I'd be making the same kind of (rather obvious, really) mistake you've made in your claims about me - so I won't claim it.
Have a nice day.
Obviously the usual method of law should be used, but they shouldn't be merely slapped on wrist, nor are they anything like graffiti artists.
Everytime some idiot joyrider breaks into one of our systems, we end up reinstalling, which takes a lot of time - and even that isn't really enough to definitely prevent reentry. You may think they just log in and start up an irc program, but how are the white hats supposed to know that? There's no way of telling if there was more to it or not, so you have to treat it seriously - reinstall, maybe even change all the passwords, inspect all the
Personally, I believe they should be put on trial, and the potential punishments should be severe. Nothing like graffiti, this is more like stealing a car for some joyriding and leaving a note saying you've planted explosives and drugs somewhere in the car, and "just try to find it, nyah, nyah", and also that they've had a duplicate key to the car made (they may not have, but the note says they did, maybe just to freak out the owner, or maybe they really did...).
Because all of these things can potentially be done to a system, once it's been broken into.
If the media hadn't decided to misuse the word hacker so egregiously, we probably wouldn't have nearly the problems we do with stupid kids trying to prove "how smart they are". Face it, breaking into computers is stone stupid. But somehow the media likes to call people who do it "too smart for their own good" which is a horribly positively spun way of saying "kind of a little bit tech smart, but incredibly lacking in overall perspective".
Christmas wasn't hijacked by consumption fever.
Consumption fever was hijacked by christmas.
Jesus, if he existed, wasn't born in the winter. Christmas was scheduled in December to steal the thunder of "pagan" winter celebrations like Saturnalia, which was about giving presents to children.
I've come to the conclusion that slashdot movie commentary is worse than useless. It's annoying.
Please, stick to tech. I can only pay attention to your movie reviews for the sake of doing the opposite of what you recommend.
I almost didn't see Nemesis because of the stupid comments here about it. That would've been a mistake. There are numerous other examples.
And what's with this fascination with Anime? Does it really have Anything to do with techies? I'm sure there are some techies who like Anime, but I'd guess it's far from the majority. In fact, if it weren't for slashdot, I doubt most of us would know what it is (I would, but I'm weird
Why would you -want- to run a modified xbox, when you can get a playstation? Why would you put up with being marginalized by microsoft, when you can be encouraged by Sony?
Sell your silly xbox (and its games) and get a real console (and its games) from a vendor with a clue.
But go players won't be bested by computers for some time to come. I personally find it a bit depressing playing a game that a computer can beat me at, but with go, I can beat the best programs I've tried, which not long ago were the best in the world. Humans are far more challenging opponents to someone who's studied go part time for a few years.
I totally disagree. There are the young, I-can't-afford-anything-else folks running linux, but I believe they're the minority among linux users.
Besides moron, you could wipe the thing and install Linux on it. That should be enough to get you frothing at the bits.
But if you did, would you still have PPC in rom? I'd like to see this answered by someone who's actually investigated the issue.
Fault Palm for dropping the ball for taking so long to come out with a new OS (Palm OS 6 with Be technology). You can't slack when MS is on your tail.
Actually, isn't it pretty common for a market leader to avoid making big changes? If something's working, don't fix it. Personally, given how long it takes to get a Very-Different OS ready, I'd say Palm was on the ball here.
Yes, there's been some productive analysis of the endgame, but that's So Vastly Simpler than the opening or middle game.
More here. .
I understand go is EXPTIME-complete or EXPSPACE-complete depending on the ko rule (which isn't that central to the flow of the game).
I'd contact the FSF to see if they're interested in developing a pool of patents they can cross license.
I have no faith that I could feed a dog for a year on proprietary gpg extensions.
But if my patented idea, held by the FSF, made it into a lot of key software, I'd be pretty much assured some big points in interviews when I say "Oh, you've used xyztech? Yeah, my algorithm is in that."
No guarantees the FSF is interested in doing this. But I hope so. If they aren't, I'd contact the OSI.