For an Apple fan, Macbook displays cannot be mentioned without speaking of "the crispness of". They've spent so much time gobbling that stuff up that they unconsciously repeat the marketing mantras. Marketingspeak on the brain.
Way back when the Amiga was still a barely viable platform, there used to be fanaticism... but it was absolutely nothing like what Apple fans perpetuate now. This stuff is more like pod people from the film.
What the fuck are you smoking, and where can I find some of that shit?
BitTorrent has had integrity verification from the very start. In like, 2002. Per-chunk SHA1 hashing. "Very poor data integrity", my eye.
The signing bit happens like this: you sign the commit object that's at the "tip". This references the log object and the tree object, which in turn references other tree objects and file objects and so forth. All referencing happens via SHA1 hashes, which guarantees as good a level of security as the initial OpenPGP-compatible hash of the tip commit. It's not hard to understand. All hashes can be checked, and the tip commit's signature can be verified using an OpenPGP-compatible tool such as GnuPG.
This is incorrect. While it is true that at any given moment a torrent client has connections to a few tens of peers, which peers among the swarm these are changes constantly. The sample size may be small, and typically is by design, but the sampling frequency is quite high.
Over time (typically a hour or two), combined with the "random unchoke" feature of the tit-for-tat algorithm, this tends to optimize to prefer peers who have demonstrated a high upload bandwidth.
Introducing a complication such as geolocation data to this can only lead to swarm bisection and other nasty misfeatures that aren't generally foreseen by people enamoured with the Latest, Hottest Thing... Such as the Azureus developers, by how their Java code is structured.
It's not explicit, but the tit-for-tat algorithm which is at the core of the self-organization mechanism of BitTorrent already favours fast connections. These generally translate to geographical closeness, but they also end up preferring links that are otherwise quick: for instance, in Finland I get quite a bit of stuff from Sweden thanks to the interconnections and the swedish Bredbandsbolaget.
Adding a GeoIP style thing to BitTorrent can only make the algorithm perform worse, as it would prefer addresses by geographical locality rather than locality as defined by network topology.
No, resizable implies that the size can be changed. It does not imply both making it larger and smaller; in fact, a filesystem that can only be shrunk would qualify as resizable.
Yeah, what with all the tens of releases of RSDL that Con put out and everything, and the multi-year development.
Then again, we are talking about the Linus who picked BitKeeper due to a friend of his whispering in his ear, nearly causing a fork of the Linux project. Even today, Linus doesn't see any fault in his reasoning: it was all Andrew Tridgell's fault that Larry McVoy got pissed off and withdrew (by his conscious action) the free BitKeeper license.
When the hell is Alan Cox coming back to Linux development? Long ago he was the one person who could whack Linus over the head and get him to not stick with his original knee-jerk decision. I guess in that way, Linus has finally become RMS.
Come out of it man. You're just wanking over ZFS because it's a big black monolith from a supposedly successful, inscrutable (i.e. unhelpful) company. You have no idea how it works, and it doesn't have a track record of stability and reliability that was in any way comparable to JFS, XFS, Reiser3 or even ext2.
Besides, this stuff about "unified file systems" is just a huge crock. There's no need to have an One True Filesystem: going with the one that is good for the task at hand (such as ext2 for database tablespaces, XFS for "lots of large files" and reiserfs for your newsspool or squid cache) is proven to work already. Problem solved, and not in a filesystem jealously guarded by a single company.
It's about a silly ideal of "code going with data". To anyone reasonable, on the Intarbutts that's a bloody horrible idea just because of the security implications of running code downloaded from J. Random Webpage on the client computer.
Yet these OOP wankers call structs and arrays "plain old data", as though being time-tested and proven over some fifty years was suddenly a bad thing because it isn't the Latest New Thing. Apparently some people will do anything to have more mechanisms, particularly ones that dictate policy, in the name of hypothetical "flexibility" that never appears or is required.
They estimated that it'd last for 10 years. It took the Doom9 forum people 5 _weeks_ to hack it. That's like, less than a hundredth of the estimate (i.e. 5/520).
I wonder. They must not have heard that architectures with an obfuscated instruction set are also reverse-engineerable? I distinctly remember reading an article on the Transmeta VLIW machine's opcode and instruction packet format... and that one has never been officially released at all.
Well, that's the market for you. The government represents the people. The people should be an equal actor in the marketplace, on the same line as a huge ugly telco.
I don't see anything wrong with that. It's called democracy and a free market (long as the telco isn't charged, you know, customs fees).
Or would you rather have a stagnant one-pony market, where the best available Internet connection is something like a 128 kbit/s ISDN that you pay per-minute _and_ per-megabyte charges for? Because that's what you get in a regional monopoly. Competition from the local government is still competition, and oftentimes it works quite well. Of course the telcos will scream red menace, but then again who wouldn't -- no incumbent likes having to do better and bilk their customers less for better service.
3) Intel/VIA/TMSC/IBM Fabs sticks to contractual obligations it has with AMD that carried over from when it was still AMD Fabs. 4) AMD keeps dominating the x86-64 server market.
These corporate types aren't stupid, you know. This is the obvious fear factor, and the stockholders would never go along with the plan if this fear were not addressed.
The difference is that Intel is chiefly a semiconductor manufacturing company, whereas AMD is chiefly an R&D company. As evidenced by Intel's chips being based on a silicon process that's about half a generation ahead all the time, and AMD's being smart, high-bandwidth designs hampered by slower silicon process development.
Thing is, creationism is objectively wrong. The Earth was not created 7000 years ago. There should not be any teaching of objectively wrong things in science class aside from as an example of nonscience in the later, philosophy-of-science type courses.
And as for conclusive proof, you should be aware that in natural science there is no such thing. And evolution and origin of mankind and stuff like that are most definitely matters of natural science.
Of course your "god" will explain everything purely and perfectly and conclusively, so how the heck could some mere humans even hope to compete -- except for those bits where, you know, heliocentrism turned out to be objectively wrong and such. Having learned to rely on a preacher's version of god, you expect everything to be nice and platonically pure and conclusive and final. Carved on slabs of marble, so to say.
Unfortunately, the world doesn't quite work that way.
That's interesting -- I've found the opposite, being a regular potsmoking programmer. Having a few tokes and a couple of cups of tea or something often makes me less perfectionist and more inclined to just keep it real so to say. This is very conducive to higher-level design than doing legwork for a O(h) data structure that'll see perhaps twenty elements in its lifetime, instead of just using a linked list.
I guess for me, weed limits my mild OCD symptoms. It also boosts creativity, and I've found many documentation tasks easier under the influence. Things become easier to consider from a "what if... the exact opposite applied?" point of view, which is an absolute boon for design or prototyping work.
Reviewing a stoned night's output is of course a good idea, but then code and design reviews are a good idea for many many reasons having nothing to do with the herb.
(...) Heroin doesn't affect the motor reflexes and lasts a long time (and destroys the user, and is permanently and physically addictive the first time it's used).
You don't actually believe that first time shit, do you? Every heroin addict will tell you that they only became really reliant on the horse after a couple of weeks of shooting up constantly time and again. And to get there, you need either serious pain or serious stress (e.g. from touring).
The giggle fits, the uncontrollable munchies, the feeling you have to constantly go take a wee, and most of the "wow I'm high lol guys" type effects tend to go away with experience. That is to say, smoke moar.
I've broken personal gaming records several times (mostly driving games, go figure) while in the "whittling" part of a good spacecake high. You know, where you start e.g. rolling joints or playing RTS games just because you want to do something small and fascinating.
And that's not talking about all the C code I've written and designed while high. Besides some weird comments in places, I don't see any problems with this sort of thing. "PERHAPS BONGHITS WILL FIX MY MAKEFILE", you know. I'd be surprised if California's software industry hasn't been running on weed for the last couple of decades.
Slashdot Mirror
For an Apple fan, Macbook displays cannot be mentioned without speaking of "the crispness of". They've spent so much time gobbling that stuff up that they unconsciously repeat the marketing mantras. Marketingspeak on the brain.
Way back when the Amiga was still a barely viable platform, there used to be fanaticism... but it was absolutely nothing like what Apple fans perpetuate now. This stuff is more like pod people from the film.
What the fuck are you smoking, and where can I find some of that shit?
BitTorrent has had integrity verification from the very start. In like, 2002. Per-chunk SHA1 hashing. "Very poor data integrity", my eye.
The signing bit happens like this: you sign the commit object that's at the "tip". This references the log object and the tree object, which in turn references other tree objects and file objects and so forth. All referencing happens via SHA1 hashes, which guarantees as good a level of security as the initial OpenPGP-compatible hash of the tip commit. It's not hard to understand. All hashes can be checked, and the tip commit's signature can be verified using an OpenPGP-compatible tool such as GnuPG.
This is incorrect. While it is true that at any given moment a torrent client has connections to a few tens of peers, which peers among the swarm these are changes constantly. The sample size may be small, and typically is by design, but the sampling frequency is quite high.
Over time (typically a hour or two), combined with the "random unchoke" feature of the tit-for-tat algorithm, this tends to optimize to prefer peers who have demonstrated a high upload bandwidth.
Introducing a complication such as geolocation data to this can only lead to swarm bisection and other nasty misfeatures that aren't generally foreseen by people enamoured with the Latest, Hottest Thing... Such as the Azureus developers, by how their Java code is structured.
It's not explicit, but the tit-for-tat algorithm which is at the core of the self-organization mechanism of BitTorrent already favours fast connections. These generally translate to geographical closeness, but they also end up preferring links that are otherwise quick: for instance, in Finland I get quite a bit of stuff from Sweden thanks to the interconnections and the swedish Bredbandsbolaget.
Adding a GeoIP style thing to BitTorrent can only make the algorithm perform worse, as it would prefer addresses by geographical locality rather than locality as defined by network topology.
No, resizable implies that the size can be changed. It does not imply both making it larger and smaller; in fact, a filesystem that can only be shrunk would qualify as resizable.
You'll have to find another expression.
Yeah, what with all the tens of releases of RSDL that Con put out and everything, and the multi-year development.
Then again, we are talking about the Linus who picked BitKeeper due to a friend of his whispering in his ear, nearly causing a fork of the Linux project. Even today, Linus doesn't see any fault in his reasoning: it was all Andrew Tridgell's fault that Larry McVoy got pissed off and withdrew (by his conscious action) the free BitKeeper license.
When the hell is Alan Cox coming back to Linux development? Long ago he was the one person who could whack Linus over the head and get him to not stick with his original knee-jerk decision. I guess in that way, Linus has finally become RMS.
What, all 4 of you?
Come out of it man. You're just wanking over ZFS because it's a big black monolith from a supposedly successful, inscrutable (i.e. unhelpful) company. You have no idea how it works, and it doesn't have a track record of stability and reliability that was in any way comparable to JFS, XFS, Reiser3 or even ext2.
Besides, this stuff about "unified file systems" is just a huge crock. There's no need to have an One True Filesystem: going with the one that is good for the task at hand (such as ext2 for database tablespaces, XFS for "lots of large files" and reiserfs for your newsspool or squid cache) is proven to work already. Problem solved, and not in a filesystem jealously guarded by a single company.
It's about a silly ideal of "code going with data". To anyone reasonable, on the Intarbutts that's a bloody horrible idea just because of the security implications of running code downloaded from J. Random Webpage on the client computer.
Yet these OOP wankers call structs and arrays "plain old data", as though being time-tested and proven over some fifty years was suddenly a bad thing because it isn't the Latest New Thing. Apparently some people will do anything to have more mechanisms, particularly ones that dictate policy, in the name of hypothetical "flexibility" that never appears or is required.
I suggest everyone tag this article "whore", "whoring", "blogwhore" and "blogwhoring". Because that is what it is.
I seek whatever the fuck I want, and you can shut the hell up about it.
By that reasoning, the retard is you. What do you produce? What gives you the "right" to express an opinion? Nothing, that's what.
Yeah go tell that to 4chan
They estimated that it'd last for 10 years. It took the Doom9 forum people 5 _weeks_ to hack it. That's like, less than a hundredth of the estimate (i.e. 5/520).
I wonder. They must not have heard that architectures with an obfuscated instruction set are also reverse-engineerable? I distinctly remember reading an article on the Transmeta VLIW machine's opcode and instruction packet format... and that one has never been officially released at all.
Well, that's the market for you. The government represents the people. The people should be an equal actor in the marketplace, on the same line as a huge ugly telco.
I don't see anything wrong with that. It's called democracy and a free market (long as the telco isn't charged, you know, customs fees).
Or would you rather have a stagnant one-pony market, where the best available Internet connection is something like a 128 kbit/s ISDN that you pay per-minute _and_ per-megabyte charges for? Because that's what you get in a regional monopoly. Competition from the local government is still competition, and oftentimes it works quite well. Of course the telcos will scream red menace, but then again who wouldn't -- no incumbent likes having to do better and bilk their customers less for better service.
Or...
3) Intel/VIA/TMSC/IBM Fabs sticks to contractual obligations it has with AMD that carried over from when it was still AMD Fabs.
4) AMD keeps dominating the x86-64 server market.
These corporate types aren't stupid, you know. This is the obvious fear factor, and the stockholders would never go along with the plan if this fear were not addressed.
The difference is that Intel is chiefly a semiconductor manufacturing company, whereas AMD is chiefly an R&D company. As evidenced by Intel's chips being based on a silicon process that's about half a generation ahead all the time, and AMD's being smart, high-bandwidth designs hampered by slower silicon process development.
A similar move would hurt, not help, Intel.
Well it says profiling. I'm sure it won't hurt unless they mean it to.
Purple friendrank -- I'ma grip and sip.
Might that be because infinite data structures don't often exist in mainstream and/or commercial software applications?
Might that be because mainstream programming languages don't support infinite data structures?
Might that be because infinite data structures don't often exist in mainstream and/or commercial software applications?
Thing is, creationism is objectively wrong. The Earth was not created 7000 years ago. There should not be any teaching of objectively wrong things in science class aside from as an example of nonscience in the later, philosophy-of-science type courses.
And as for conclusive proof, you should be aware that in natural science there is no such thing. And evolution and origin of mankind and stuff like that are most definitely matters of natural science.
Of course your "god" will explain everything purely and perfectly and conclusively, so how the heck could some mere humans even hope to compete -- except for those bits where, you know, heliocentrism turned out to be objectively wrong and such. Having learned to rely on a preacher's version of god, you expect everything to be nice and platonically pure and conclusive and final. Carved on slabs of marble, so to say.
Unfortunately, the world doesn't quite work that way.
Go to school high. Get high scores.
That's interesting -- I've found the opposite, being a regular potsmoking programmer. Having a few tokes and a couple of cups of tea or something often makes me less perfectionist and more inclined to just keep it real so to say. This is very conducive to higher-level design than doing legwork for a O(h) data structure that'll see perhaps twenty elements in its lifetime, instead of just using a linked list.
I guess for me, weed limits my mild OCD symptoms. It also boosts creativity, and I've found many documentation tasks easier under the influence. Things become easier to consider from a "what if... the exact opposite applied?" point of view, which is an absolute boon for design or prototyping work.
Reviewing a stoned night's output is of course a good idea, but then code and design reviews are a good idea for many many reasons having nothing to do with the herb.
(...) Heroin doesn't affect the motor reflexes and lasts a long time (and destroys the user, and is permanently and physically addictive the first time it's used).
You don't actually believe that first time shit, do you? Every heroin addict will tell you that they only became really reliant on the horse after a couple of weeks of shooting up constantly time and again. And to get there, you need either serious pain or serious stress (e.g. from touring).
The giggle fits, the uncontrollable munchies, the feeling you have to constantly go take a wee, and most of the "wow I'm high lol guys" type effects tend to go away with experience. That is to say, smoke moar.
I've broken personal gaming records several times (mostly driving games, go figure) while in the "whittling" part of a good spacecake high. You know, where you start e.g. rolling joints or playing RTS games just because you want to do something small and fascinating.
And that's not talking about all the C code I've written and designed while high. Besides some weird comments in places, I don't see any problems with this sort of thing. "PERHAPS BONGHITS WILL FIX MY MAKEFILE", you know. I'd be surprised if California's software industry hasn't been running on weed for the last couple of decades.
(this comment was written while high, also.)
Whot, your previous employers never heard of a non-disclosure agreement?
Silly you for not negotiating harder!