Don't you LIKE having support for MD5 passwords? That's one of my biggest beefs with pre-5.9. It just bugs the hell out of me when you can still crack passwords with 10 year old tools.
with a Celeron 700 in it. The system boots into networked linux in seconds from a stock BIOS and the system consumes less than 45W idle with a normal, 7200RPM hard disk spun up. Pretty good, huh? It's essentially my firewall/router/WAP/webserver.
And you could easily do a lot better in the power department by using flash storage and lower the clock speed and chip voltage, or by going to a newer CPU (coppermine).
I'm, I get your point, but it's a bad example.
What needs to be emphasized is the difference between a 12W draw in well engineered case with x86, and the 2W draw from a typical ARM based WAP. It won't make or break a sale, but it's a consideration, especially since higher power handling necessitates bulkier hardware construction and maybe passive heatsinks (heavy, thus expensive in production)
It'd still be bayesian, except that word frequencies and graph connectivity of sender would _both_ be considered for additional spam probability. I don't have a filter to check, but don't most Bayesian classifiers also include other metrics besides top 20 word frequency, like length or presence of attachments, etc.?
can post in under 2 seconds if not in hardware-change-check mode.
Besides, if you were making an appliance, I'm sure you could write your own BIOS (take LinuxBIOS, for example). That'd make it boot instantly into any size kernel image you care.
They would be expending far less effort than you would believe. The PPro and netburst architectures are pretty flexible. You update the chip to for larger width of registers in the register file, change the width of some parts of address decoding stages, add more address pins, and update the Microcode to understand the new mode/instructions.
I'm sure centrino (Pentium 4-M) was a more expensive endeavor then this Plan B.
even SLES 8.x and RHES 3.0 are not rock solid on AMD64 servers. Sure, single CPU, fine. But you try to use 4-way on the AMD8XXX and you will have an interesting time.
I think Microsoft was having similar trouble trying to adapt Windows to run properly and without issue on this brand new hardware. Maybe AMD was dragging their feet addressing errata exposed by the effort?
But even then I feel that's a bit of a stretch. It's been in beta state for a long time now. At least they're offering security updates for it... that's a sign of commitment.
AMD has done the heavy lifting designed an IA at the front end, and put the arch in the marketplace for people to start targeting code at it.
A year from now, people will be buying 64-bit Xeons running at 4GHz which will fly on the x86_64 optimized code... they were probably buying AMD before because Intel didn't have a similar offering, but now Intel can get their Gold-plated name back into the newly created market.
...of bytes so that it will overwrite the stack frame as needed, but not cause an access violation. Furthermore, a read past the end of a file does not cause the Read to terminate, but to read until EOF and then return the number of bytes it didn't get... (usually you store that result, and then do a select/poll loop to look for more data written to the end later on, until you get 0 back, which means you're done).
So yeah, it actually does exactly what you want. A Read with an unchecked buffer and a very large size parameter is a h4xx0r's wet dream.
And when it gets hacked, it's usually because someone is using an extension that tries to do a lot, or isn't as popular as the default ones, or is non-standard. Go figure.
that's fucking awesome.
on
Hack Your Car
·
· Score: 1
What did I say wrong? That I said something that if read quickly, without full comprehension, might be construed as criticism of the Samba project? Way to jump the gun, Moderator! Did you read any of my followup posts before getting all hot under the collar?
And -1 overrated, so no one can M2. I can't believe this childishness.
But it doesn't matter what you or I know or think.
It matters what other people think, lawyers, judges, IT workers who have not PERSONALLY been involved with or followed Samba's development.
Microsoft can claim that aliens stole their RPCSS stuff and beamed it into Andrew Tridgell's head. But as soon as they leak that to the press, suddenly Samba's in a world of shit, whether or not it's true. That's why it's called FUD. If you don't know any better, then you'll likely bet on the side with more lawyers.
All Microsoft has to do is hint that Samba had the leaked stuff all along, by claiming they wouldn't have gotten as far as they have gotten without it. Then if they can convince a judge in Australia that they plan on filing for infringement, the Samba project will have to go on the defensive. They'll have to (trivially) prove they didn't have the code, but by then the accusations will be all over the press.
I mean, Samba specifically deviates from what published specs on CIFS are to be bug compatible. If I were a lawyer pursuing this angle, I'd latch onto that as evidence that there was something fishy going on. (Evidence of source code paraphrasing, which is "easy", as opposed to the result of hundreds of man hours of testing and debugging, which is "hard")
It doesn't have to be true, just plausible. We've done the research. Lawyers and trial judges in Australia have no such intimate knowledge of the subject matter, and thus would probably agree to at least hear evidence, and grant Microsoft many motions. I mean, look at SCO! They don't have to show any copying, they just have to show evidence that they think it happened, and then go into fact finding mode. Boom! Instant clusterfuck, for a good while. And it scares off people at the margins.
And this "plausible" copyright and/or trade secret violation scenario is at least somewhat lucid (compared to SCO)... at least up to the point that you do a code comparison and see that this is patently false. Again, they know they'd lose in court. It's just the act of calling it into question (which you absolutely could do) that is the problem.
Slashdot Mirror
Don't you LIKE having support for MD5 passwords? That's one of my biggest beefs with pre-5.9. It just bugs the hell out of me when you can still crack passwords with 10 year old tools.
I don't see the problem.
And I can't live without ^W/^/ and indent mode.
It's just uncivilized without them!
It's Bill... the frustrated mainframe guy!!!
j/k
Love, Rei
with a Celeron 700 in it.
The system boots into networked linux in seconds from a stock BIOS and the system consumes less than 45W idle with a normal, 7200RPM hard disk spun up.
Pretty good, huh?
It's essentially my firewall/router/WAP/webserver.
And you could easily do a lot better in the power department by using flash storage and lower the clock speed and chip voltage, or by going to a newer CPU (coppermine).
I'm, I get your point, but it's a bad example.
What needs to be emphasized is the difference between a 12W draw in well engineered case with x86, and the 2W draw from a typical ARM based WAP. It won't make or break a sale, but it's a consideration, especially since higher power handling necessitates bulkier hardware construction and maybe passive heatsinks (heavy, thus expensive in production)
A lot better. I use it almost all the time.
When pico gets mouse support, highlighting, bracket match, regex search/replace, and multiple buffers, let me know.
We call it "eating your own dog food".
It'd still be bayesian, except that word frequencies and graph connectivity of sender would _both_ be considered for additional spam probability. I don't have a filter to check, but don't most Bayesian classifiers also include other metrics besides top 20 word frequency, like length or presence of attachments, etc.?
can post in under 2 seconds if not in hardware-change-check mode.
Besides, if you were making an appliance, I'm sure you could write your own BIOS (take LinuxBIOS, for example). That'd make it boot instantly into any size kernel image you care.
It's called SSE.
Yeah, integrating a dual channel memory controller was probably a good idea.
Instant resume, configuration stored in the app itself, etc. etc.
They would be expending far less effort than you would believe. The PPro and netburst architectures are pretty flexible. You update the chip to for larger width of registers in the register file, change the width of some parts of address decoding stages, add more address pins, and update the Microcode to understand the new mode/instructions.
I'm sure centrino (Pentium 4-M) was a more expensive endeavor then this Plan B.
even SLES 8.x and RHES 3.0 are not rock solid on AMD64 servers. Sure, single CPU, fine. But you try to use 4-way on the AMD8XXX and you will have an interesting time.
I think Microsoft was having similar trouble trying to adapt Windows to run properly and without issue on this brand new hardware. Maybe AMD was dragging their feet addressing errata exposed by the effort?
But even then I feel that's a bit of a stretch. It's been in beta state for a long time now. At least they're offering security updates for it... that's a sign of commitment.
AMD has done the heavy lifting designed an IA at the front end, and put the arch in the marketplace for people to start targeting code at it.
A year from now, people will be buying 64-bit Xeons running at 4GHz which will fly on the x86_64 optimized code... they were probably buying AMD before because Intel didn't have a similar offering, but now Intel can get their Gold-plated name back into the newly created market.
Just change it in /etc/passwd and /etc/shadow /etc/security and /etc/pam.d
Also, look for instances of it in
(some scripts/PAM config stuff in those directories like to refer to "root" the username for stuff)
...of bytes so that it will overwrite the stack frame as needed, but not cause an access violation. Furthermore, a read past the end of a file does not cause the Read to terminate, but to read until EOF and then return the number of bytes it didn't get... (usually you store that result, and then do a select/poll loop to look for more data written to the end later on, until you get 0 back, which means you're done).
So yeah, it actually does exactly what you want. A Read with an unchecked buffer and a very large size parameter is a h4xx0r's wet dream.
And when it gets hacked, it's usually because someone is using an extension that tries to do a lot, or isn't as popular as the default ones, or is non-standard.
Go figure.
I want that Jetta now. Screw the Subaru.
Do they even bother numbering the license plates?
^_^
We made five of them.
What did I say wrong? That I said something that if read quickly, without full comprehension, might be construed as criticism of the Samba project?
Way to jump the gun, Moderator! Did you read any of my followup posts before getting all hot under the collar?
And -1 overrated, so no one can M2. I can't believe this childishness.
Jesus you people!
:P
It's like every five minutes...
I know. I KNOW. I KNOW!
But it doesn't matter what you or I know or think.
It matters what other people think, lawyers, judges, IT workers who have not PERSONALLY been involved with or followed Samba's development.
Microsoft can claim that aliens stole their RPCSS stuff and beamed it into Andrew Tridgell's head. But as soon as they leak that to the press, suddenly Samba's in a world of shit, whether or not it's true.
That's why it's called FUD. If you don't know any better, then you'll likely bet on the side with more lawyers.
All Microsoft has to do is hint that Samba had the leaked stuff all along, by claiming they wouldn't have gotten as far as they have gotten without it. Then if they can convince a judge in Australia that they plan on filing for infringement, the Samba project will have to go on the defensive. They'll have to (trivially) prove they didn't have the code, but by then the accusations will be all over the press.
I mean, Samba specifically deviates from what published specs on CIFS are to be bug compatible. If I were a lawyer pursuing this angle, I'd latch onto that as evidence that there was something fishy going on. (Evidence of source code paraphrasing, which is "easy", as opposed to the result of hundreds of man hours of testing and debugging, which is "hard")
It doesn't have to be true, just plausible. We've done the research. Lawyers and trial judges in Australia have no such intimate knowledge of the subject matter, and thus would probably agree to at least hear evidence, and grant Microsoft many motions. I mean, look at SCO! They don't have to show any copying, they just have to show evidence that they think it happened, and then go into fact finding mode. Boom! Instant clusterfuck, for a good while. And it scares off people at the margins.
And this "plausible" copyright and/or trade secret violation scenario is at least somewhat lucid (compared to SCO)... at least up to the point that you do a code comparison and see that this is patently false. Again, they know they'd lose in court. It's just the act of calling it into question (which you absolutely could do) that is the problem.