If you read the "findings of fact" in the DOJ/Microsoft case, you'll see that one of Microsoft's key concerns is to prevent cross-platform APIs that would allow other OSes to be used on an equal footing with Windows, and allow easy migration. While Office isn't an API as such, it would certainly allow some projects to be done entirely on Linux that currently are deemed to require Windows.
And of course it would add to the credibility of Linux amongst those who currently take the view that "no-one got fired for buying Microsoft".
Several people have commented on the possibility of writing native Crusoe programs, and how this is just what Transmeta don't want - because they want to be able to change the hardware incompatibly. But unlike almost any other company now, they can do something else: design a new instruction set.
This would be an instruction set chosen to be particularly efficiently translatable into the Crusoe native instructions, but would remain constant as the processor changed.
Why can they do this when others can't? Because they can have x86 compatibility at the same time, and cheaply. The system could simultaneously run both formats, just by having multiple translators.
And they could do other things, in particular they could join with AMD and implement its as yet unrevealed 64-bit instruction set.
The argument that open source is more secure is based on fact that many people will see problems by looking at the source, and will fix bugs that they encounter.
In the case of the SETI@Home project, the security is needed by the project, who don't get to see the modified versions. So it doesn't apply.
Of course, if people distribute their modified versions, other people will find the bugs. But how will the already-submitted results be retracted? And what about modified versions that people keep to themselves?
I suspect that future projects of this kind are going to have to include more dummy data to verify that incorrect algorithms are not being used.
As has been pointed out elsewhere (comp.arch I think), the patents that are available on the net reflect what Transmeta was doing a couple of years ago, when the patents were filed. So don't be sure that they are central to their current plans.
Every international telephone call, fax, e-mail, or radio transmission can be listened to
but it also suggests that the system works by monitoring satellites. So what about international phone calls that don't go via satellite?
Certainly phone calls within Europe don't go by satellite, and not all transatlantic ones do. There's even a prefix you can dial to ensure you get a cable connection - and you can tell it works because it doesn't have the characteristic satellite delay (you can work out the height required for geostationary satellites and calculate the speed-of-light delay - it's about a quarter of a second if I remember correctly).
So, do they have separate monitoring for the cables?
The widely-used Yamaha YMF24-based cards are still not supported, because Yamaha won't release the programming information. Perhaps they might be more amenable to persuasion now that more of their competitors are opening up?
Netscape already allows you to provide a javascript function to determine the proxy for a given URL. I use this to filter out ads by having a dummy proxy for known ad URLs.
What it should also do is let you provide a function to determine whether a cookie should be accepted, so we can reject cookies based on the URL (and mime type - a much more general solution than saying "html only").
This is exactly the sort of thing I want an open source browser for.
Quite apart from the problem of finding a backup device with enough capacity, it's going to take longer and longer to just copy all the data off these ever bigger devices.
The capacity goes up because the recording density increases. This increases the transfer rate too, but only by the square root of the capacity increase (because reading speed depends on the linear, rather than area, density).
So unless spin speeds increase further (which is a problem because of heat), a disk of double the capacity takes 41% longer to copy.
Given the amount of FP processing power and memory on recent graphics cards, they could probably run a variety of non-graphics tasks faster than the host computer.
For example, you could run seti@home on your graphics card, instead of just using it to display the results.
How about it NVIDIA? You could leap to the top of the seti@home CPU statistics!
For those who don't know, Jeremy Paxman is the interviewer who recently asked Henry Kissinger if he felt like a fraud for accepting a Nobel peace price considering his support for Pinochet, the Chinese government and the bombing of Cambodia.
If I understand this correctly, the originator can only "proprietarize" derived work if the deriver agrees to it by choosing the TGPL instead of the GPL.
But a deriver of a GPL'd work can already agree to allow the originator to use his contribution in a proprietary version.
I suppose this makes it a bit more straightforward, though.
Another advantage of using XML + (DTDs or schemas) is that XML editors will provide structured editing "for free". How many times have you run vipw only to get a complaint about a syntax error at line 137? An editor (or an emacs mode) that only allows you to enter valid XML will catch many of these errors as you type them. Or to look at it another way, you can automatically derive a form-style interface from a DTD or schema.
So rather than it being "another thing to learn" it can provide a simpler user interface.
I think that XML provides an obvious first-choice syntax for anyone defining new files structures, and unix (not just Linux) configuration files are a good example. Here are a few thoughts I've had on the subject:
It must be simple for applications to read the data into directly useful structures - when you read the password file you want to get a pointer to a struct passwd rather than an element whose children and attributes you have to examine. XML Schemas should help with this (and it will be schemas rather than DTDs).
There's a small annoyance concerning control characters. Despite XML having a syntax for numeric character references, these can only refer to legal XML characters, which means not control characters (so you won't be able to use ! for escape in termcap.xml). The library will need to provide a mechanism on top of plain XML parsing to handle this.
It should be possible to make the change invisibly to many programs (getpwent will just read the XML syntax into the same old structures) but it will still be necessary to provide conversion tools.
Validation (probably schema-validation) will provide a handy way to (at least partially) check files before installing them.
Slashdot Mirror
So you need a new motherboard.
And of course it would add to the credibility of Linux amongst those who currently take the view that "no-one got fired for buying Microsoft".
This would be an instruction set chosen to be particularly efficiently translatable into the Crusoe native instructions, but would remain constant as the processor changed.
Why can they do this when others can't? Because they can have x86 compatibility at the same time, and cheaply. The system could simultaneously run both formats, just by having multiple translators.
And they could do other things, in particular they could join with AMD and implement its as yet unrevealed 64-bit instruction set.
Slashdot moderating is done by the readers, not the site owners. The moderation just conveys other readers views.
NFS works by having a process accepting commands through a socket, so you can do this (rather inelegantly) on most operating systems.
In the case of the SETI@Home project, the security is needed by the project, who don't get to see the modified versions. So it doesn't apply.
Of course, if people distribute their modified versions, other people will find the bugs. But how will the already-submitted results be retracted? And what about modified versions that people keep to themselves?
I suspect that future projects of this kind are going to have to include more dummy data to verify that incorrect algorithms are not being used.
And don't trust the predicted peak time too much. Last year it was about 18 hours earlier than predicted. We had an excellent display in Scotland.
As has been pointed out elsewhere (comp.arch I think), the patents that are available on the net reflect what Transmeta was doing a couple of years ago, when the patents were filed. So don't be sure that they are central to their current plans.
Certainly phone calls within Europe don't go by satellite, and not all transatlantic ones do. There's even a prefix you can dial to ensure you get a cable connection - and you can tell it works because it doesn't have the characteristic satellite delay (you can work out the height required for geostationary satellites and calculate the speed-of-light delay - it's about a quarter of a second if I remember correctly).
So, do they have separate monitoring for the cables?
The widely-used Yamaha YMF24-based cards are still not supported, because Yamaha won't release the programming information. Perhaps they might be more amenable to persuasion now that more of their competitors are opening up?
What it should also do is let you provide a function to determine whether a cookie should be accepted, so we can reject cookies based on the URL (and mime type - a much more general solution than saying "html only").
This is exactly the sort of thing I want an open source browser for.
The capacity goes up because the recording density increases. This increases the transfer rate too, but only by the square root of the capacity increase (because reading speed depends on the linear, rather than area, density).
So unless spin speeds increase further (which is a problem because of heat), a disk of double the capacity takes 41% longer to copy.
The data sheet for this is now available at IBM's web site http://www.storage.ibm.com/hardsoft/diskdrdl/ultra /72zxdata.htm .
For example, you could run seti@home on your graphics card, instead of just using it to display the results.
How about it NVIDIA? You could leap to the top of the seti@home CPU statistics!
For those who don't know, Jeremy Paxman is the interviewer who recently asked Henry Kissinger if he felt like a fraud for accepting a Nobel peace price considering his support for Pinochet, the Chinese government and the bombing of Cambodia.
Typical - on the one hand we have people saying that the GPL is communism, and on the other people blaming communism for violation of the GPL.
Answers: "16" and "Z8000".
(From the "AmZ8000 Microprocessor Family", 1978)
But a deriver of a GPL'd work can already agree to allow the originator to use his contribution in a proprietary version.
I suppose this makes it a bit more straightforward, though.
Maybe AMD plan to license Transmeta's technology to emulate IA-64.
As for the size, that will depend on what's available when it arrives.
Yes, it's expat (already mentioned in other comments).
So rather than it being "another thing to learn" it can provide a simpler user interface.
You do know that song is still in copyright, don't you?