Window order is not something that the gimp can do. That job belongs to the (duh) window manager, which may or may not follow the hints that gimp provides.
The old Nedap voting machines use obsolete hardware, and those machines are often not stored in a secure way (so they could be tampered with).
The new machines run Windows and a wireless modem. That doesn't sound like a safe combination to me.
As far as I'm concerned, a voting machine should at least make an immediate print-out of each vote (a good old-fashioned line printer would do), so that a recount can be done to check the machine's results.
Their plan (as admitted in interviews) was to withhold "evidence" to the last moment to prevent IBM from preparing a good defence. This is unfair and not allowed of course, which is part of the reason some of their "evidence" was thrown out.
They presented these items as evidence, but when the items were rightly challenged (basically, SCOG didn't supply file, version and line information that the judge ordered of code/methods they accused IBM of misusing), they changed their story (as they are accustomed to) and told the judge the items were claims.
If you take a good look at SCOG's filings, you can see that they cherry-pick those parts of the evidence and declarations that seem to support their claim, often quoting incomplete paragraphs of contracts and correspondence, and using parts of declarations out of context.
It's just plain lies, dressed up to enable SCOG to present them without sanction.
Having used AutoCAD, Unigraphics and Pro/Engineer, I must say that what I use varies.
For a quick but accurate sketch, I tend to use AutoCAD Light. Basically an upgraded version of my old drawing board, and with a better eraser.:-)
For 3D stuff, I prefer Pro/E over Unigraphics, especially the wildfire versions. The wildfire interface takes much less time to get used to than the old Pro/E interface.
But... 3D feature based modelling isn't without it's own warts. You have to think carefulley about your model beforehand. You want to have the features that you are likely to change as far as possible in the "branches" or even the "leaves" of the "tree". It you put them too near the root, and link other feature to them, you run a great risk of a model regeneration failure if you change a dimension of that feature. Fixing that can be a major PITA.
Mind you, even Pro/E has troubles with _really_ large models (like a complete X-ray scanner, for instance). You hav e to hide a lot of parts and sub-assemblies to be able to move the model around on screen in real time. And you can easily go get a cup of coffee waiting for it to load.
Which is no surprise of course. No CAD supplier wants to make it easier for you to switch to another package.
Having said that, packages that use the same modelling kernel (e.g. unigraphics and solid edge both use the parasolid kernel) can usually exchange data very well.
Mind you, IGES and STEP is fine for sending stuff to suppliers; you don't want them mucking around with your models anyway!
AutoCAD is by far the industry standard CAD tool for engineering drawings.
Not in the mechanical engineering world. 3D packages like CATIA, Pro/Engineer and Unigraphics have long eclipsed it in the high-tech industry.
It's still usefull for making quick sketches etc.
For exporting drawings, dxf and especially pdf are much more important than the dwg format. For 3D data, IGES and STEP are most often used, especially because they're open standards.
The author points to MS's secret codebase. This has nothing to do with patents.
Besides, if MS tries to sue or extort money from someone for use of it's patents, they'd have to specifiy the patents in question, and be sure that these patents would survive being challenged.
It's also pretty useless in this case. Since the developers are probably a tiny subset of the people using the driver, his data does not enable them to know "who and where" there customers are.
If they treally wanted to know that, they should register their chip sales, _and_ force their OEM customers to supply them with the details of everybody who buys a product with said chip in it.
No system that someone else has physical access to is secure.
If the laptop is stolen for its data most of the protection effort is in vain. The thief can rip out the harddrive and read it's contents on another machine. Unless the harddrive or the files are well encrypted. And even in that case, the laptop user could yield to rubber hose cryptanalysis.
If the "trusted" computing chip uses keys to sign/encrypt things, those keys will get leaked/hacked eventually.
The premise of the article seems to be that a secure system has no need for backups. That is pretty silly. Your data can get lost in more ways than being wiped by a virus, e.g.;
accidental deletion by the user
hardware failure
A car-related analogy would be to say that if you have good tires, to don't have to carry a spare. It sounds pretty good, but it's beside the point. The point is not that the tire will fail in normal use. It will fail from abnormal circumstances like shards of glass or a nail on the road.
The second point is that viruses on Windows rely on some shortcomings of Outlook and IE that allow execution of programs downloaded from the internet without user intervention. UNIX programs tend not to do that.
The best way to run a program on UNIX without a user starting it is to start it via cron, the startup scripts or by having it replace an often used system program. But since write access to all of these is restricted to root, so a normal user can't touch them or infect them with a virus.
So a UNIX virus pretty much needs to be explicitly started by a user. Which is a case of "pilot error", and not a shortcoming of the system.
There are good reasons for not giving it wide-spread attention.
First is that the judge sealed the hearing transcript. Probably because one of the SCO lawyers tried to read sealed material into the record. Sealed material is not for the public eye. Helping to spread it could even be an offence, but IANAL.
Second, links to stories that are debunked on Groklaw have a tendency to disappear, so the links could very well be dead soon.
Groklaw is dedicated to accurate reporting. Providing links to bogus articles could give the articles more credit than they deserve.
Before the Trinity test during the Manhattan project, a test
was done with en explosion of 108 tons of Composition B (RDX/TNT). Among the explosives were tubes containing fission products.
Looks like the biggest "dirty" bomb ever made.
Of course the fireball would not be as hot as a nuclear one, but the pictures look very impressive nontheless.
I wouldn't touch anything that MS has a patent or copyright stranglehold over with a 10 ft pole!
If you do use it, you are inviting MS to screw you over whenever it suits them. Whatever their intent might be at this moment doesn't matter, it's the capability that counts. IHMO that's not a position a Free Software developer wants to be in.
As to anybody who wants to do kernel development in C#, go talk to Linus for a good LARTing.:-)
Slashdot Mirror
Window order is not something that the gimp can do. That job belongs to the (duh) window manager, which may or may not follow the hints that gimp provides.
Maybe you should try another window manager?
Kensington Expert mouse.
Why?
- usable with both hands
- four buttons
- scroll ring
It's not just the radiation.
The old Nedap voting machines use obsolete hardware, and those machines are often not stored in a secure way (so they could be tampered with).
The new machines run Windows and a wireless modem. That doesn't sound like a safe combination to me.
As far as I'm concerned, a voting machine should at least make an immediate print-out of each vote (a good old-fashioned line printer would do), so that a recount can be done to check the machine's results.
Paraphrased: "The next two versions will get all the deleted features we promised for vista, honest!"
So essentially, vista won't be launched until 2008?
Their plan (as admitted in interviews) was to withhold "evidence" to the last moment to prevent IBM from preparing a good defence. This is unfair and not allowed of course, which is part of the reason some of their "evidence" was thrown out.
They presented these items as evidence, but when the items were rightly challenged (basically, SCOG didn't supply file, version and line information that the judge ordered of code/methods they accused IBM of misusing), they changed their story (as they are accustomed to) and told the judge the items were claims.
If you take a good look at SCOG's filings, you can see that they cherry-pick those parts of the evidence and declarations that seem to support their claim, often quoting incomplete paragraphs of contracts and correspondence, and using parts of declarations out of context.
It's just plain lies, dressed up to enable SCOG to present them without sanction.
Having used AutoCAD, Unigraphics and Pro/Engineer, I must say that what I use varies.
:-)
For a quick but accurate sketch, I tend to use AutoCAD Light. Basically an upgraded version of my old drawing board, and with a better eraser.
For 3D stuff, I prefer Pro/E over Unigraphics, especially the wildfire versions. The wildfire interface takes much less time to get used to than the old Pro/E interface.
But... 3D feature based modelling isn't without it's own warts. You have to think carefulley about your model beforehand. You want to have the features that you are likely to change as far as possible in the "branches" or even the "leaves" of the "tree". It you put them too near the root, and link other feature to them, you run a great risk of a model regeneration failure if you change a dimension of that feature. Fixing that can be a major PITA.
Mind you, even Pro/E has troubles with _really_ large models (like a complete X-ray scanner, for instance). You hav e to hide a lot of parts and sub-assemblies to be able to move the model around on screen in real time. And you can easily go get a cup of coffee waiting for it to load.
Which is no surprise of course. No CAD supplier wants to make it easier for you to switch to another package.
Having said that, packages that use the same modelling kernel (e.g. unigraphics and solid edge both use the parasolid kernel) can usually exchange data very well.
Mind you, IGES and STEP is fine for sending stuff to suppliers; you don't want them mucking around with your models anyway!
Not in the mechanical engineering world. 3D packages like CATIA, Pro/Engineer and Unigraphics have long eclipsed it in the high-tech industry.
It's still usefull for making quick sketches etc.
For exporting drawings, dxf and especially pdf are much more important than the dwg format. For 3D data, IGES and STEP are most often used, especially because they're open standards.
The author points to MS's secret codebase. This has nothing to do with patents.
Besides, if MS tries to sue or extort money from someone for use of it's patents, they'd have to specifiy the patents in question, and be sure that these patents would survive being challenged.
I'd say it is cheaper to FUD than to sue.
It's also pretty useless in this case. Since the developers are probably a tiny subset of the people using the driver, his data does not enable them to know "who and where" there customers are.
If they treally wanted to know that, they should register their chip sales, _and_ force their OEM customers to supply them with the details of everybody who buys a product with said chip in it.
On my FreeBSD I compile almost everything from ports. That way you have maximum control about the options and CFLAGS with wich your app is compiled.
It's as easy as 'cd /usr/ports/category/portdir; make install clean'.
If you have a number of identical machines, you could compile and build packages on one machine, and then install them on all the others.
On Linux I think that Gentoo's portage comes closest.
No system that someone else has physical access to is secure.
If the laptop is stolen for its data most of the protection effort is in vain. The thief can rip out the harddrive and read it's contents on another machine. Unless the harddrive or the files are well encrypted. And even in that case, the laptop user could yield to rubber hose cryptanalysis.
If the "trusted" computing chip uses keys to sign/encrypt things, those keys will get leaked/hacked eventually.
Sensitive data does not belong on a laptop.
The premise of the article seems to be that a secure system has no need for backups. That is pretty silly. Your data can get lost in more ways than being wiped by a virus, e.g.;
A car-related analogy would be to say that if you have good tires, to don't have to carry a spare. It sounds pretty good, but it's beside the point. The point is not that the tire will fail in normal use. It will fail from abnormal circumstances like shards of glass or a nail on the road.
The second point is that viruses on Windows rely on some shortcomings of Outlook and IE that allow execution of programs downloaded from the internet without user intervention. UNIX programs tend not to do that.
The best way to run a program on UNIX without a user starting it is to start it via cron, the startup scripts or by having it replace an often used system program. But since write access to all of these is restricted to root, so a normal user can't touch them or infect them with a virus.
So a UNIX virus pretty much needs to be explicitly started by a user. Which is a case of "pilot error", and not a shortcoming of the system.
Yeah, and 'Emacs Makes A Computer Slow' and 'Eventually Munches All Computer Storage' aren't what they used to be either.
But see http://www.ahajokes.com/com027.html for more emacs jokes.
After cutting my Linux teeth on Slackware I tried different versions of Red Hat, Debian and SUSE. In every case I came back screaming to Slackware.
Whenever I need Linux, I get Slack.
There are good reasons for not giving it wide-spread attention.
First is that the judge sealed the hearing transcript. Probably because one of the SCO lawyers tried to read sealed material into the record. Sealed material is not for the public eye. Helping to spread it could even be an offence, but IANAL.
Second, links to stories that are debunked on Groklaw have a tendency to disappear, so the links could very well be dead soon.
Groklaw is dedicated to accurate reporting. Providing links to bogus articles could give the articles more credit than they deserve.
Before the Trinity test during the Manhattan project, a test was done with en explosion of 108 tons of Composition B (RDX/TNT). Among the explosives were tubes containing fission products. Looks like the biggest "dirty" bomb ever made.
Of course the fireball would not be as hot as a nuclear one, but the pictures look very impressive nontheless.
Three to go. :-)
They slip into their asbestos underpants :-)
I wonder how long it will take before someone develops a program that kills the ads or hides their windows?
Those who live by the sword die by the arrow.
I wouldn't touch anything that MS has a patent or copyright stranglehold over with a 10 ft pole!
:-)
If you do use it, you are inviting MS to screw you over whenever it suits them. Whatever their intent might be at this moment doesn't matter, it's the capability that counts. IHMO that's not a position a Free Software developer wants to be in.
As to anybody who wants to do kernel development in C#, go talk to Linus for a good LARTing.
Not to mention an inherent instability from the not insignificant weight of the pilot riding atop of this thing.
Unless the engine is heavy enough to make sure that the plane is inherently stable, I'd say the full-scale version would be a flying deathtrap.
Accurate mechanical clocks were veritable engineering milestones.
:-)
They made it possible to determine longitude.
The available replicas (see e.g. http://www.clockmakers.com/john_harrison_sea.php) should make any engineer drool.