And are you seriously claiming that this feature in Tomboy is bigger than Expose would be?
It is for me, but I'm not claiming my opinion to be some god-given truth either... I don't care for Expose at all. Gimp has image windows thumbnails as icons -- distinctive enough. I don't have many instances of Inkscape or Blender open at once. And don't get me into a rant about thumbnailing windows full of text...
Dell diagnostics are pretty good at detecting hardware problems - at least to the point that Dell is willing to ship a replacement part. If the video card's busted, it should be possible to tell if it's hardware. If it isn't hardware, then the paid support for the distro comes into play.
And if you say they can't detect that with the random distro that is now on the computer then imagine a seperate diagnostics boot partition or diagnostics live cd. Linux is flexible that way.
1. Not all software exists in repositories - and even if it does, finding out that a certain program exists (by taking some task or function as a starting point) would be equally, if not more difficult in Synaptic for a grandmother.
This isn't a problem that needs a new package managing system to solve. Distributions could create a system that allows application websites to provide one-click install initiation from the repositories.
2. A user should not know about package manager. The task a user faces in the real world is to "install this software", not to "manage packages". Therefore installation should not involve more than a click on the downloaded software packages (maybe with license acceptance and some initial setup). Moreover, if we talk about an apt-like system, the user shouldn't have to explicitly add sources - information about a third party repository should be included in the package itself and added automatically.
And remove it. And update it. The second is solved with existing architecture very well already. And I'm confident that the first task can be as well, Ubuntu is moving in that direction with the "Add/Remove..." menu item. Downloadable packages that provide repository information should be doable as well, with some sort of jailling to minimize the possibility of non-repository packages to tamper with the main system. This should be possible with current packaging systems.
3. There are various incompatibilities between distros where packaging is concerned.
That's because of the incorrect assumption that there is such a thing as as "Linux OS". Just like there are different packages for Windows (or Win9x, 2000/XP, XP x64...), Mac OS, OS X the software vendor is the one who has to provide specific (or not so specific) packages for the distribution(s) they choose to support... Or let the maintainers handle it if it fits, as it does with most FLOSS projects. Clever packaging should make it possible to handle a large variety of disrtibutions with few packages.
If he had taken the month to make Git before Bitkeeper it would have shown pragmative foresight, reactive pragmatism isn't a reason to trust somones far-reaching design-decisions though, and that is what Linus has mostly shown outside of kernel development.
Is spot on, there would be no enterprise level kernel adoption without RMS, it was originaly licensed under a non-comercial license and IIRC changed because Linus was thankful for GCC.
I'd say we'll be able to evaluate it in 5-10 years, no telling yet how sticking to GPL2 will play out.
Having IBM, the open source and the free software communities taking care of the issue for him is more lack then foresight.
Needing software with everchanging license agreements for kernel developers and later kernel development in VCS limbo aren't things you want for your enterprise level kernel.
IMHO he was better at it before he developed into the strong leader he is today.
Beeing able to patch UI on the software level is altogether different then making a good interface, the GNOME developers have mostly learned as it goes from the first steps in improved usability in 2.0, having used all versions since then I'd say they are getting good at it and continue moving in the right direction, unless Linus has a natural talent for UI design he has a few years of intensive learning to do to catch up.
Different permissions on music bought from the same store are going to be confusing for both consumers and music labels.
You certainly don't want to confuse customers with all that freedom, why, they might start asking why they can't do all the nice things with all tracks they buy. And we wouldn't want them to learn about the evils of DRM, no sir, they should just think thats the way the world now works.
The author simply uses "digital content protection" where we would use DRM, and DRM as an all encompasing term for encryption, digital signatures, computer readable copyright licenses and probably a bunch of other things he didn't directly meantion. From this he concludes that DRM is both widely used and accepted. Now whether he is trying to convince poeple that the fight is allready lost and we should work on interoperable lockdowns or is just confused himself I don't know.
For as long as I can remember I've had a subtle effect on machines. I've heard similar things described here many times, in many discussions. When friends and relatives ask me to fix something, and I come over to help them out, the thing just starts working. Mostly it's with computers.
[..] I attach a note to the device saying "this is a sound monitoring device for project XXX. If you have any questions, call John Smith at (617) 555-8944."
Sweet cell phone activated bomb trap you have there!
Slashdot Mirror
Besides it's not giving to the poor, it's taking as much as you can take from everyone.
No matter what format you choose create error correction data.
Some are helpfull, some just point at something missed and often more isn't required.
Keyword: medium.
You can't do that with GPL either.
The latest info would be "The state of GEGL".
Not by default no, but take a look at File/Preferences/Window Management/Window Manager Hints.
I gather that GP was pointing out that the wok only saves $70 of that, but is for some reason still made the focus point of the article.
If he had taken the month to make Git before Bitkeeper it would have shown pragmative foresight, reactive pragmatism isn't a reason to trust somones far-reaching design-decisions though, and that is what Linus has mostly shown outside of kernel development.
I must not be real. The other ~30-40% of Linux desktop users mustn't be real as well or utter idiots as you like to put it.
The author simply uses "digital content protection" where we would use DRM, and DRM as an all encompasing term for encryption, digital signatures, computer readable copyright licenses and probably a bunch of other things he didn't directly meantion. From this he concludes that DRM is both widely used and accepted. Now whether he is trying to convince poeple that the fight is allready lost and we should work on interoperable lockdowns or is just confused himself I don't know.
That explains why they apply DRM even to music that is sold in DRM-less versions elsewhere...