My problem is that Apple broke Bluetooth in a MAJOR way with 10.2.8, and with Panther right around the corner, it looks like it'll never get fixed. That's practically illegal- "we broke it, so just buy the update." Um, no- and as a result, I think I'll be downloading Panther, not buying it.
Apple broke a lot of things in a MAJOR way in 10.2.8, and they've fixed bost of them in 10.2.8.1. I bet they fix bluetooth if it's broken so bad.
And if not, you still aren't forced to upgrade. You can reinstall from 10.2.6, which is sort of a bitch to do, but it's still a way out. You can also compile your own SSH fixes (i do stuff like that sometimes).
It worked (and did the obvious thing) on whatever shell I was using when I made that sig, because I pasted that from the shell. However, you're right, it doesn't seem to work now. Maybe it was a really old shell (from my operating systems project) or there was an instance of/bin/rm in the history.
Given the number of "That doesn't work if . is at the end, dumbass!" flames I get with this.sig, it's time to change it anyway.
IIRC, which I may not, you could set up handlers in HyperCard which were preserved across stacks. Which is a problem if you open a trusted stack, and a malevolent handler is still lying around from an untrusted one.
And yes, in this case you've sandboxed it wrong. But sandboxing for scripted apps like that can be much harder than it sounds, especially if they have complex features. That was just an example of how easily you could screw it up and allow CSS.
I would call it USB 2.0 compliant. But I would not talk about how USB 2.0 is capable of 480Mbps (in huge red letters) on my package, when my device does not go anywhere near that fast. Such branding creates in the minds of consumers the following equation:
If it complies with the USB2.0 protocols, but isn't any faster than 12Mbps ("full-speed"), it's still USB 2.0 and can be branded as such. And therein lies the problem, because companies claim that their slow drives, cameras or whatever are "full-speed" USB 2.0, and extol the virtues and great speed of USB 2.0. They probably can't get sued for this bullshit though, because they didn't technically lie.
It is not technically false, but it is intentionally misleading, which is pretty much the same thing everywhere outside of the courts.
I've seen products marked "Full-Speed USB 2.0! USB 2.0 can transfer data at up to 480Mbps..." etc, etc. And of course if you look at the fine print in the specs on the other side, it says their max data transfer rate is 12Mbps. I've bought a few of said products anyway, because they were cheap and shoddy and I was looking for something cheap, but always with a little guilt for contributing to such a bullshit company.
On a side note, does anyone know if this drive actually supports hi-speed USB (ie, anything more than 12Mbps data transfer)? It uses evasive language and descriptions, has a restocking fee for returns, and no tech specs available...
Imagine if you could easily pull up cards from stacks on other computers across an AppleTalk network -- it would have very much resembled an early version of HTML -- only more powerful.
Yeah, but remember that just like Word, hypercard has them crazy macroviruses... it would be a bitch to get a trust-model worked out that would protect against macrovirus and cross-site-scripting vulnerabilities. Even under OSX, where you could chroot / su it into a very small sandbox, you have to worry about CSS: if it could redefine a procedure in memory, used by more trusted stacks, you could end up screwed anyway.
This feature shouldn't really need hardware support. It should be able to make a memory image in a file on the hard disk, and when it reboots, if the image exists, to reload it.
But it's easy to do that if you don't mind your app running slow. Just write it in Java, or some other language with no buffer overflows; that's n fewer things to check.
This is simply not true. If you add in some other commonly requested attributes, like "full of overly complex GUI iCandy," "every feature under the sun," or the like, then you might have to decide.
The key is simplicity. A simple, well-designed, and carefully coded solution can be cheap, fast, and secure; the simplicity of the design reinforces all three of these.
What?! What exactly wasn't true about what was said?
These guys are all famous security researchers, and what they say in this report is mostly if not entirely true, but how they say it is intentionally inflammatory. They bash Microsoft left and right, and regard them throughout the paper as an evil empire. While Microsoft is in many ways an evil empire, the analysts sacrificed objectivity and even-handedness for additional force to drive their point. Very rarely did they point out mitigating circumstances. Furthermore, they did not mention the dangers of heterogeneity in computer systems, such as: More training required to use the software. More training required to securely administer it, meaning more incompetent admins. More ports of software, which may be rushed or otherwise buggy. (This is especially problematic with patches). More difficulty in creating and applying patches. More difficulty in researching bugs, as ports may be substantially different. More divided development efforts, with increasingly divergent platforms. Fewer security appications per platform, for above reasons.
Furthermore, heterogeneity does not retard the spread of a Flash worm, only a worm of Warhol speed or below (although it does limit the damage caused). Granted, we haven't seen any worms even of the Warhol type, but the paper does warn of future attacks...
Disclaimer: I am Christian, but I believe that the creation account in the bible is metaphorical.
This orchestration of life is almost certainly bullshit. Even if a life-form could evolve from his bubbles, it would not share many of the features of life on earth. These things are pretty much miniature ball lightning.
However, many of the experiments into the origin of life are quite reasonable. Scientists have a pretty good idea of the environment about the time that life arose (at least, the time it arose if you trust fossil evidence). So they try to simulate things like lightning strikes or tidal pools in a similar environment, and they find that it creates many of the prerequisites for life "as we know it," including amino acids, nucleic acids, and microscopic spheres bounded by structures siimilar to prokaryotic membranes (no, I'm not talking about the plasma experiments).
Such experiments do not have humans "designing" life, but merely trying to recreate conditions which could have started it.
In any case, development of life this way can still be consistent with a God that created the universe, and possibly guided the development of life.
It's not just EXT2, though, unless you're speaking of some particular use for that reservation (the journalling perhaps?).
Ext2 doesn't have journalling. That's Ext3. But no, I'm not talking about fragmentation (internal in the case of Ext2), or data structures, or journalling. However, Ext*, FFS, UFS, etc all reserve 5% of the disk space to sit free, to make it easier to find free contiguous blocks when they allocate things; this is why defragging rarely helps a Unix partition. Root can use the extra 5%, so as to make disks report that they are over 100% full. I think that this is a silly choice for personal machines; the system should tell the user that it'll be slow, but still let them use it.
I wish I could find an FS which could flexibly handle differing file sizes and so the File Size vs. Space Used were always within a small amount, able to handle 10-byte files and 100+MB files without leaving lots of unused HD space. That would make the FS work harder though, right?, plus make it more difficult if you ever tried defragging.
http://www.namesys.com/v4/v4.htmlReiser4 is supposed to be released this summer. ReiserFS (at the same site) accomplishes this pretty well. They are log-structured and therefore have one heck of a time defragging, but they have greatly increased performance between defrags. So if your load is bursty (say, during the day), then these might be a good choice.
You have been warned: Reiser4 is beta! Don't put a beta filesystem on a production machine! Although it is designed with crash-protection in mind (easier in an LFS), it might corrupt everything!
What I want to know is, is there a filesystem designed for USB Flash drives? They have very different characteristics from spinning media: large blocks must be erased at the same time, but nonlocal accesses are perfectly OK.
Of course they're lying! Advertisers always lie. But the thing is, they won't get busted in court, because what they said was technically true. Just like Bill Clinton didn't get busted for his lie in court, and just like a lot of advertisers don't even get sued for their lies.
But they're not going to quit lying unless they lose, which is unlikely. It adds more than 7% to their size figures, and they don't want to give that up.
Unit Prefix Abbreviation 2^10 kibi Ki 2^20 mebi Mi 2^30 gibi Gi 2^40 tebi Ti 2^50 pebi Pi 2^60 exbi Ei
Examples and comparisons with SI prefixes
1 Kibit = 2^10 bit = 1024 bit 1 kbit = 10^3 bit = 1000 bit 1 MiB = 2^20 B = 1 048 576 B 1 MB = 10^6 B = 1 000 000 B 1 GiB = 2^30 B = 1 073 741 824 B 1 GB = 10^9 B = 1 000 000 000 B
In particular, 20 GB = 18.6 GiB. So, they're telling the truth, albeit in a not-so-honest way; it's really the disk info page that's lying.
It's also worth noting that EXT2 and some other UNIX-based filesystems reserve a certain percent of the space; this makes their available capacity smaller for non-root users.
... because it's a heap buffer. Furthermore, it's not a simple buffer overrun, but an error in reallocation. As far as I've seen, there are no known exploits of it either. If there are, please link.
Slashdot Mirror
My problem is that Apple broke Bluetooth in a MAJOR way with 10.2.8, and with Panther right around the corner, it looks like it'll never get fixed. That's practically illegal- "we broke it, so just buy the update." Um, no- and as a result, I think I'll be downloading Panther, not buying it.
Apple broke a lot of things in a MAJOR way in 10.2.8, and they've fixed bost of them in 10.2.8.1. I bet they fix bluetooth if it's broken so bad.
And if not, you still aren't forced to upgrade. You can reinstall from 10.2.6, which is sort of a bitch to do, but it's still a way out. You can also compile your own SSH fixes (i do stuff like that sometimes).
Have you actually tried out your own sig?
/bin/rm in the history.
.sig, it's time to change it anyway.
It worked (and did the obvious thing) on whatever shell I was using when I made that sig, because I pasted that from the shell. However, you're right, it doesn't seem to work now. Maybe it was a really old shell (from my operating systems project) or there was an instance of
Given the number of "That doesn't work if . is at the end, dumbass!" flames I get with this
If you got a crappy experience out of the film because it was intentionally ruined, you have a right to a refund.
IIRC, which I may not, you could set up handlers in HyperCard which were preserved across stacks. Which is a problem if you open a trusted stack, and a malevolent handler is still lying around from an untrusted one.
And yes, in this case you've sandboxed it wrong. But sandboxing for scripted apps like that can be much harder than it sounds, especially if they have complex features. That was just an example of how easily you could screw it up and allow CSS.
I would call it USB 2.0 compliant. But I would not talk about how USB 2.0 is capable of 480Mbps (in huge red letters) on my package, when my device does not go anywhere near that fast. Such branding creates in the minds of consumers the following equation:
MyProduct = USB 2.0 = 480Mbps
Which is false.
If it complies with the USB2.0 protocols, but isn't any faster than 12Mbps ("full-speed"), it's still USB 2.0 and can be branded as such. And therein lies the problem, because companies claim that their slow drives, cameras or whatever are "full-speed" USB 2.0, and extol the virtues and great speed of USB 2.0. They probably can't get sued for this bullshit though, because they didn't technically lie.
It is not technically false, but it is intentionally misleading, which is pretty much the same thing everywhere outside of the courts.
I've seen products marked "Full-Speed USB 2.0! USB 2.0 can transfer data at up to 480Mbps..." etc, etc. And of course if you look at the fine print in the specs on the other side, it says their max data transfer rate is 12Mbps. I've bought a few of said products anyway, because they were cheap and shoddy and I was looking for something cheap, but always with a little guilt for contributing to such a bullshit company.
On a side note, does anyone know if this drive actually supports hi-speed USB (ie, anything more than 12Mbps data transfer)? It uses evasive language and descriptions, has a restocking fee for returns, and no tech specs available...
Imagine if you could easily pull up cards from stacks on other computers across an AppleTalk network -- it would have very much resembled an early version of HTML -- only more powerful.
Yeah, but remember that just like Word, hypercard has them crazy macroviruses... it would be a bitch to get a trust-model worked out that would protect against macrovirus and cross-site-scripting vulnerabilities. Even under OSX, where you could chroot / su it into a very small sandbox, you have to worry about CSS: if it could redefine a procedure in memory, used by more trusted stacks, you could end up screwed anyway.
This feature shouldn't really need hardware support. It should be able to make a memory image in a file on the hard disk, and when it reboots, if the image exists, to reload it.
Really? I thought that went with cheap.
But it's easy to do that if you don't mind your app running slow. Just write it in Java, or some other language with no buffer overflows; that's n fewer things to check.
Here lies the problem:
1) Cheap
2) Fast
3) Secure
Pick 2
This is simply not true. If you add in some other commonly requested attributes, like "full of overly complex GUI iCandy," "every feature under the sun," or the like, then you might have to decide.
The key is simplicity. A simple, well-designed, and carefully coded solution can be cheap, fast, and secure; the simplicity of the design reinforces all three of these.
QMail, for instance, is free, fast, and secure.
I'll take "Documents You'll Never See Again", for $500, Alex
Good thing the constitution forbids double jeopardy.
So? The site is slashdotted! We're getting the same 1.5k/s as you are...
Diablo II LoD has never worked right for me in full-screen mode. It runs fine at 800x600 in a window, but lags horribly at 640x480 fullscreen.
... didn't do good enough job first time.
What?! What exactly wasn't true about what was said?
These guys are all famous security researchers, and what they say in this report is mostly if not entirely true, but how they say it is intentionally inflammatory. They bash Microsoft left and right, and regard them throughout the paper as an evil empire. While Microsoft is in many ways an evil empire, the analysts sacrificed objectivity and even-handedness for additional force to drive their point. Very rarely did they point out mitigating circumstances. Furthermore, they did not mention the dangers of heterogeneity in computer systems, such as:
More training required to use the software.
More training required to securely administer it, meaning more incompetent admins.
More ports of software, which may be rushed or otherwise buggy. (This is especially problematic with patches).
More difficulty in creating and applying patches.
More difficulty in researching bugs, as ports may be substantially different.
More divided development efforts, with increasingly divergent platforms.
Fewer security appications per platform, for above reasons.
Furthermore, heterogeneity does not retard the spread of a Flash worm, only a worm of Warhol speed or below (although it does limit the damage caused). Granted, we haven't seen any worms even of the Warhol type, but the paper does warn of future attacks...
... and if anyone post one of those no-carrier jokes, I'll... umm... let's see, I can't mod them down...
Not meaningless, but certainly unscientific.
Disclaimer: I am Christian, but I believe that the creation account in the bible is metaphorical.
This orchestration of life is almost certainly bullshit. Even if a life-form could evolve from his bubbles, it would not share many of the features of life on earth. These things are pretty much miniature ball lightning.
However, many of the experiments into the origin of life are quite reasonable. Scientists have a pretty good idea of the environment about the time that life arose (at least, the time it arose if you trust fossil evidence). So they try to simulate things like lightning strikes or tidal pools in a similar environment, and they find that it creates many of the prerequisites for life "as we know it," including amino acids, nucleic acids, and microscopic spheres bounded by structures siimilar to prokaryotic membranes (no, I'm not talking about the plasma experiments).
Such experiments do not have humans "designing" life, but merely trying to recreate conditions which could have started it.
In any case, development of life this way can still be consistent with a God that created the universe, and possibly guided the development of life.
It's not just EXT2, though, unless you're speaking of some particular use for that reservation (the journalling perhaps?).
Ext2 doesn't have journalling. That's Ext3. But no, I'm not talking about fragmentation (internal in the case of Ext2), or data structures, or journalling. However, Ext*, FFS, UFS, etc all reserve 5% of the disk space to sit free, to make it easier to find free contiguous blocks when they allocate things; this is why defragging rarely helps a Unix partition. Root can use the extra 5%, so as to make disks report that they are over 100% full. I think that this is a silly choice for personal machines; the system should tell the user that it'll be slow, but still let them use it.
I wish I could find an FS which could flexibly handle differing file sizes and so the File Size vs. Space Used were always within a small amount, able to handle 10-byte files and 100+MB files without leaving lots of unused HD space. That would make the FS work harder though, right?, plus make it more difficult if you ever tried defragging.
http://www.namesys.com/v4/v4.htmlReiser4 is supposed to be released this summer. ReiserFS (at the same site) accomplishes this pretty well. They are log-structured and therefore have one heck of a time defragging, but they have greatly increased performance between defrags. So if your load is bursty (say, during the day), then these might be a good choice.
You have been warned: Reiser4 is beta! Don't put a beta filesystem on a production machine! Although it is designed with crash-protection in mind (easier in an LFS), it might corrupt everything!
What I want to know is, is there a filesystem designed for USB Flash drives? They have very different characteristics from spinning media: large blocks must be erased at the same time, but nonlocal accesses are perfectly OK.
No, they are lying.
Of course they're lying! Advertisers always lie. But the thing is, they won't get busted in court, because what they said was technically true. Just like Bill Clinton didn't get busted for his lie in court, and just like a lot of advertisers don't even get sued for their lies.
But they're not going to quit lying unless they lose, which is unlikely. It adds more than 7% to their size figures, and they don't want to give that up.
The NIST suggestion is just confusing for too many users. How many times do you actually hear someone say "mebibyte?"
I know that. They're being dishonest. It's called advertising. But they have a legal defense, which is that what they said is technically correct.
It's also worth noting that EXT2 and some other UNIX-based filesystems reserve a certain percent of the space; this makes their available capacity smaller for non-root users.
... because it's a heap buffer. Furthermore, it's not a simple buffer overrun, but an error in reallocation. As far as I've seen, there are no known exploits of it either. If there are, please link.
neighboring