Not being able to unmount a removable storage device (CD, my digital camera, whatever) because some process had the bright idea of keeping an open file on it, or hanging around with it as its cwd. Nautilus used to be especially bad in this regard.
It's a 386 compatible CPU featuring pipelined execution and an onboard FPU. Pretty neat, huh? Those puppies can be pushed to speeds in excess of 100MHz.
This is probably wrong. In most cases when I came across stub functions that the application was trying to use, the application just failed to work properly in one way or another.
The core API functionality is there, and working. You can't reasonably claim that there's a performance benefit to be gained from stubbing out functionality.
Can we build it? Yes we can!
on
Replacing SMTP?
·
· Score: 1
With all the loopholes in the current SMTP specification, is it possible for the Slashdot collective to come up with another one?
Yes, I'm sure that Slashdot is up to the task of coming up with another loophole.
Good luck. From dealing with Comcast I know that they're slow to adopt new technology, and even if they buy from you, it doesn't mean they'll continute to buy. Especially if they call what they're doing "a trial".
Of course, having to roll out HDTV might force them to act quickly, but who knows?
Microsoft, like all other technology corporations, has its own bulky patent portfolio --- which is useless defense against a company that makes no use of its own patents, much less anyone else's.
It's hard to use your own patents against an IP litigation company, but you can cause them some pain - Microsoft, I'm sure, has its own DRM patents that it can enforce. Microsoft could declare publicly that, for the industry's sake, it refuses to license its own patents to anyone who licenses InterTrust's patents, and that it will seek to enforce its own patent rights.
This should throw a nice fat wrench in the works. If nothing else, it would make licensing InterTrust's patents pointless for the purpose of implementing a DRM product. if all goes well. It would also throw DRM down the toilet, giving everybody involved a compelling reason to play nice.
I could live with a 27MHz CPU. It's fast enough, really. The problem with the DCT2000 is its graphics bus, which is so slow that you can hear the pixels strolling across it to video memory. Or something like that.
Now, the Scientific Atlanta boxes - those are really nice, even at 50MHz.
As for where you work, I haven't the slightest clue. I'm not familiar enough with the European cable/settop industry to make an educated guess, but never mind - let's all stay very mysterious...
Hah! I wrote code for these boxes. They're based on 68K microcontrollers - dig around the comments here for a link to a page with (very little) hardware and hacking information on them.
The DCT5000s are a completely different design, but most of the Motorola boxes in the field are various DCT2000 flavors.
Could the MS IPG SW be a starting point for hacking these cheap embedded video processors to some open source OS, like Linux, QNX or *BSD?
Not without a significant investment in headend hardware and a lot of reverse engineering (unless you're already an authorized developer for the platform, in which case you're probably sick of it by now). Those things are as closed as they get - the MPEG hardware won't do anything but decode MPEG coming from the RF connection, the firmware is prorietary, the hardware specs are ridiculous. Getting code to even run on the box is an unbelievably painful chore.
And, assuming you manage to hack a box and get it to run your code - it's going to be pretty much useless without a server at the headend sending it data.
These things are useless. If you want to work on a box with similar hardware specs, get a 68k based Palm device, and have fun. At least you'll have a decent amount of memory to play with.
Open sourcing OS code has proven to be a good way to keep ailing systems relevant in the current marketplace. It kept BeOS and VMS from dying in obscurity
What are you talking about? BeOS and VMS were never open sourced.
You're getting out of your car with your briefcase and your bag of groceries and you have this eerie feeling you're forgetting something important. You stand up and reach for the door and give it a shove. As the door careens toward closure, that little switch in your brain flips and that little voice starts screaming "Take it back! Take it back! Your keys are in there!!!" In an instant, your body wretches trying to catch the car door closing but to no avail. That little voice in your head then says "Awww shit, you really fucked up now, and you're beyond the point of no return."
And the time all this takes is called an Ohnosecond, boys and girls.
Maybe because it will make the cost of maintaining a server impossible to project, and accountants don't like that. A contract that promises to maintain n servers for x dollars for a year can be more attractive to a client than a contract that promises to fix m problems for x dollars, even if the average number of problems experienced by n servers over a year is m.
Your recollection is impressive. The legality of Linux DVD players hadn't been determined yet. I'm not convinced the courts will rule that Xine+libdvdcss is illegal because of the DMCA.
Mplayer is a different story - mencoder might be seen as a tool designed primarily for piracy.
You are misguided. nvidia's drivers are proprietary, and they won't release hardware information that would allow anyone to write open source drivers.
ATI, besides supporting Linux with their proprietary drivers, also releases hardware documentation, and there are open source drivers for their cards as a result.
Now go take that nvidia card back to the store, and let your wallet talk to the right company...
Still, I don't remember the GNOME developers jumping up and down with joy over Bluecurve. Sure, they weren't as vocal as (some of) the KDE developers, but I don't remember a warm endoresement.
But then, I don't remember what I had for breakfast.
Slashdot Mirror
Not being able to unmount a removable storage device (CD, my digital camera, whatever) because some process had the bright idea of keeping an open file on it, or hanging around with it as its cwd. Nautilus used to be especially bad in this regard.
Damn! My secret identity was revealed!
I'll go sulk now. And then sculpt, or read Proust, or watch American Idol...
I'm really glad I don't have a TV.
Haven't you been reading the spam lately? Mortgage interest rates are at an all time low! Refinance now!
Yep. But the question was:
:)"
"What the hell is a 486?
I think we're both taking this way too seriously.
It's a 386 compatible CPU featuring pipelined execution and an onboard FPU. Pretty neat, huh? Those puppies can be pushed to speeds in excess of 100MHz.
This is probably wrong. In most cases when I came across stub functions that the application was trying to use, the application just failed to work properly in one way or another.
The core API functionality is there, and working. You can't reasonably claim that there's a performance benefit to be gained from stubbing out functionality.
With all the loopholes in the current SMTP specification, is it possible for the Slashdot collective to come up with another one?
Yes, I'm sure that Slashdot is up to the task of coming up with another loophole.
Good luck. From dealing with Comcast I know that they're slow to adopt new technology, and even if they buy from you, it doesn't mean they'll continute to buy. Especially if they call what they're doing "a trial".
Of course, having to roll out HDTV might force them to act quickly, but who knows?
Microsoft, like all other technology corporations, has its own bulky patent portfolio --- which is useless defense against a company that makes no use of its own patents, much less anyone else's.
It's hard to use your own patents against an IP litigation company, but you can cause them some pain - Microsoft, I'm sure, has its own DRM patents that it can enforce. Microsoft could declare publicly that, for the industry's sake, it refuses to license its own patents to anyone who licenses InterTrust's patents, and that it will seek to enforce its own patent rights.
This should throw a nice fat wrench in the works. If nothing else, it would make licensing InterTrust's patents pointless for the purpose of implementing a DRM product. if all goes well. It would also throw DRM down the toilet, giving everybody involved a compelling reason to play nice.
Oh, but I liked it so much better when we were mysterious...
So, are these boxes any good?
I could live with a 27MHz CPU. It's fast enough, really. The problem with the DCT2000 is its graphics bus, which is so slow that you can hear the pixels strolling across it to video memory. Or something like that.
Now, the Scientific Atlanta boxes - those are really nice, even at 50MHz.
As for where you work, I haven't the slightest clue. I'm not familiar enough with the European cable/settop industry to make an educated guess, but never mind - let's all stay very mysterious...
Hah! I wrote code for these boxes. They're based on 68K microcontrollers - dig around the comments here for a link to a page with (very little) hardware and hacking information on them.
The DCT5000s are a completely different design, but most of the Motorola boxes in the field are various DCT2000 flavors.
Which opposition would that be?
probably a RISC
Definitely not a RISC.
RAM
Could have fooled me...
Could the MS IPG SW be a starting point for hacking these cheap embedded video processors to some open source OS, like Linux, QNX or *BSD?
Not without a significant investment in headend hardware and a lot of reverse engineering (unless you're already an authorized developer for the platform, in which case you're probably sick of it by now). Those things are as closed as they get - the MPEG hardware won't do anything but decode MPEG coming from the RF connection, the firmware is prorietary, the hardware specs are ridiculous. Getting code to even run on the box is an unbelievably painful chore.
And, assuming you manage to hack a box and get it to run your code - it's going to be pretty much useless without a server at the headend sending it data.
These things are useless. If you want to work on a box with similar hardware specs, get a 68k based Palm device, and have fun. At least you'll have a decent amount of memory to play with.
And QNX isn't open source. So there.
Just what I thought. There's also an effort to write an open source VMS, but that doesn't count, either.
Open sourcing OS code has proven to be a good way to keep ailing systems relevant in the current marketplace. It kept BeOS and VMS from dying in obscurity
What are you talking about? BeOS and VMS were never open sourced.
You're getting out of your car with your briefcase and your bag of groceries and you have this eerie feeling you're forgetting something important. You stand up and reach for the door and give it a shove. As the door careens toward closure, that little switch in your brain flips and that little voice starts screaming "Take it back! Take it back! Your keys are in there!!!" In an instant, your body wretches trying to catch the car door closing but to no avail. That little voice in your head then says "Awww shit, you really fucked up now, and you're beyond the point of no return."
And the time all this takes is called an Ohnosecond, boys and girls.
Maybe because it will make the cost of maintaining a server impossible to project, and accountants don't like that. A contract that promises to maintain n servers for x dollars for a year can be more attractive to a client than a contract that promises to fix m problems for x dollars, even if the average number of problems experienced by n servers over a year is m.
That's fine with me. Thanks for clarifying things.
Plus, he smells kinda funny.
Not that I doubt you, but is this bit of information based on personal observation?
Your recollection is impressive. The legality of Linux DVD players hadn't been determined yet. I'm not convinced the courts will rule that Xine+libdvdcss is illegal because of the DMCA.
Mplayer is a different story - mencoder might be seen as a tool designed primarily for piracy.
No they won't. Crossover Office doesn't require any Windows code.
You are misguided. nvidia's drivers are proprietary, and they won't release hardware information that would allow anyone to write open source drivers.
ATI, besides supporting Linux with their proprietary drivers, also releases hardware documentation, and there are open source drivers for their cards as a result.
Now go take that nvidia card back to the store, and let your wallet talk to the right company...
Still, I don't remember the GNOME developers jumping up and down with joy over Bluecurve. Sure, they weren't as vocal as (some of) the KDE developers, but I don't remember a warm endoresement.
But then, I don't remember what I had for breakfast.
Twelve, according to the article. You must have missed that when you read it.
They're on LTSP, and loving it.