Some of the delays are probably caused by the sale of StrongARM to Intel from Digital. But Intel has been promising the SA-2 core for a long time. Originally the promise was 600 MHz at around 500mW in 0.18 micron, but a recent EE Times story said now they're shooting for 1 GHz at the same power consumption.
That idea sounds like what Jordan Pollack was talking about during his recent Slashdot interview... But I'm not sure I like the idea of computer games that are covered by SEC regulations.
IPSec can tell you what machine you're talking to, but not which user AFAIK. So it isn't as powerful as ssh's RSA authentication or SSL client certificates.
For commercial vendors (ie E-commerce sites), they plan on charging a negotiable amount per "model-view"
I wonder if I'm the only person who gets nostalgic for the old days when you could buy (actually license) some software and use it as much as you want.
OK, I see that you're splitting hairs here. Yes, technically when I buy a copy of Red Hat, I'm only paying them for the cost of distribution. But I'm paying them, and that's the common-sense notion of selling.
So my real point was that software is not Open Source unless I can put it on a CD and sell the CD. I don't have anything against the GPL, since it allows that.
AFAIK, there are three kinds of intellectual property: copyrights, patents, and trademarks.
Is the OpenStep API copyrighted? From what I've read, an API (i.e. a set of function signatures) is purely functional and thus cannot be copyrighted. (The documentation for the API is copyrighted, so just write your own.)
Is the OpenStep API patented? APIs are not patentable, but maybe Apple has some patents on algorithms that are required to implement OpenStep.
Is the OpenStep API trademarked? I assume the name "OpenStep" is a trademark of Apple, so don't call your implementation OpenStep. Call it GNUStep. cf. the OpenGL/Mesa issue.
If all of these issues are satisfied, then you don't need any kind of license to implement an API.
QuickTime 4 still supports the ability to create and play back normal movies. If certain Web sites are only offering streaming then your gripe is with those sites, not with QuickTime.
Also, QuickTime uses the IETF standard RTSP and RTP protocols for streaming, so there's nothing stopping someone from writing something like Streambox VCR for QuickTime.
Sure, reiserfs, ext3, XFS, and JFS all have journaling, but what does that have to do with attributes, queries, and node monitoring? (which, as far as I can tell, is what Brian was talking about)
Henry Massalin tried this about a decade ago in his Synthesis kernel for 68k; it was much faster than SunOS and Mach on the same hardware. (But then Linux is probably faster also, so it's hard to get a good comparison.)
I've been thinking about this, too. It seems like there are two ways to go:
Smart: Put as much of the logic and work as possible in the server. NeWS was an example of this; Berlin is possibly an extreme example.
Stupid: Put as little in the server as possible (events, region management) and let all the apps draw directly to the framebuffer. Y seemed to be taking this approach, and I hear MacOS X is also.
(Note that I don't mean "stupid" in a bad way. The Internet is a really stupid network, and that's why it's the best one around.)
Slashdot Mirror
Some of the delays are probably caused by the sale of StrongARM to Intel from Digital. But Intel has been promising the SA-2 core for a long time. Originally the promise was 600 MHz at around 500mW in 0.18 micron, but a recent EE Times story said now they're shooting for 1 GHz at the same power consumption.
You mean like one of these? It's a little underpowered, but they claim 15 hours of battery life.
That idea sounds like what Jordan Pollack was talking about during his recent Slashdot interview... But I'm not sure I like the idea of computer games that are covered by SEC regulations.
IPSec can tell you what machine you're talking to, but not which user AFAIK. So it isn't as powerful as ssh's RSA authentication or SSL client certificates.
I guess what I really meant was: If I can do Twofish at >100Mbps in software, I see a lot less need for hardware implementations.
Will these crypto chips still be a good idea when AES comes out? It's going to be much faster than 3DES.
QuickTime Streaming is based on the IETF standard RTSP and RTP protocols. Go here and here to download the specifications yourself.
AFAIK, you're right about RealSystem and Windows Media though.
For commercial vendors (ie E-commerce sites), they plan on charging a negotiable amount per "model-view"
I wonder if I'm the only person who gets nostalgic for the old days when you could buy (actually license) some software and use it as much as you want.
OK, I see that you're splitting hairs here. Yes, technically when I buy a copy of Red Hat, I'm only paying them for the cost of distribution. But I'm paying them, and that's the common-sense notion of selling.
So my real point was that software is not Open Source unless I can put it on a CD and sell the CD. I don't have anything against the GPL, since it allows that.
If the code is really Open Source, then anyone can sell it. If only Asynchrony has the rights to sell the code, then it just ain't Open Source.
No, that wasn't the assumption in my post.
Is there anyone who prefers Darwin to Linux or BSD? Why? It's not a rhetorical question; I'm honestly curious.
AFAIK, there are three kinds of intellectual property: copyrights, patents, and trademarks.
Is the OpenStep API copyrighted? From what I've read, an API (i.e. a set of function signatures) is purely functional and thus cannot be copyrighted. (The documentation for the API is copyrighted, so just write your own.)
Is the OpenStep API patented? APIs are not patentable, but maybe Apple has some patents on algorithms that are required to implement OpenStep.
Is the OpenStep API trademarked? I assume the name "OpenStep" is a trademark of Apple, so don't call your implementation OpenStep. Call it GNUStep. cf. the OpenGL/Mesa issue.
If all of these issues are satisfied, then you don't need any kind of license to implement an API.
Of course, porting Carbon would be non-trivial.
Why? It's just a shared library written in C AFAIK. It shouldn't be any trickier than, say, porting Gtk+ from Linux to BSD.
Darwin doesn't have any GUI and it doesn't support Classic, Carbon, or Cocoa Mac apps.
So what is its benefit over Linux or BSD? I'm still looking...
Darwin does not have a GUI at all. It's important not to confuse Mac OS X (which uses Quartz) with Darwin.
How does IPv6 stop DDoS attacks?
QuickTime 4 still supports the ability to create and play back normal movies. If certain Web sites are only offering streaming then your gripe is with those sites, not with QuickTime.
Also, QuickTime uses the IETF standard RTSP and RTP protocols for streaming, so there's nothing stopping someone from writing something like Streambox VCR for QuickTime.
MPEG-4 uses the QuickTime file format (or maybe a modified version of it), but the audio and video coding layers are new.
Sure, reiserfs, ext3, XFS, and JFS all have journaling, but what does that have to do with attributes, queries, and node monitoring? (which, as far as I can tell, is what Brian was talking about)
This story is about Linux, remember? If the kernel starts using AltiVec for memcpy, TCP checksumming, etc. all apps will benefit.
Likewise, if some of the crucial libraries like libart and libjpeg get AltiVectorized then many apps will get faster with no changes.
Here you go
Of course, since Moto is not in the desktop market, these are embedded mobos...
Henry Massalin tried this about a decade ago in his Synthesis kernel for 68k; it was much faster than SunOS and Mach on the same hardware. (But then Linux is probably faster also, so it's hard to get a good comparison.)
http://citeseer.nj.nec.com/mass alin92synthesis.html
If anyone's curious, more info on confinement and covert channels is here:
at erights.org
The computer security fact forum
There was a paper (at SOSP, I think) about SLIM, which is the protocol that the Sun Ray uses. In their benchmarks, SLIM beats VNC by a wide margin.
I've been thinking about this, too. It seems like there are two ways to go:
Smart: Put as much of the logic and work as possible in the server. NeWS was an example of this; Berlin is possibly an extreme example.
Stupid: Put as little in the server as possible (events, region management) and let all the apps draw directly to the framebuffer. Y seemed to be taking this approach, and I hear MacOS X is also.
(Note that I don't mean "stupid" in a bad way. The Internet is a really stupid network, and that's why it's the best one around.)
Which way should we go?