The constant concern, of course, is that the fragmentation of the linux market into so many distributions makes the product difficult to market to the end user, especially as there is relatively little usability and interface consistency when you jump between distros. Tools, package managers, configuration wizards, installers, desktop managers as a whole... all vary from distro to distro. I'm not saying it's the obstacle preventing Linux from being a strong competitor to Windows, but I don't think it helps any.
One thing you can be sure about is that all 7 windows flavors will have the same UI, installer, and bugs.
It's not a problem of the machine miscounting. Part of the concept involved here is to allow us to be sure the machine is not intentionally lying about the results.
I think the boxes have expiration dates printed on them, but as annoying as HP's ink level checks have ever been (notifying me months before the ink is depleted), I've never run into a situation where it wouldn't let me run the ink all the way to the actual empty point.
The current major problem is that a clean installation of Windows XP can actually become seriously infected before the user has a chance to install certain major security fixes. This is very troubling for a lot of people.... The (annoying) workaround is to get the security fix onto the box before connecting it to the Internet, but this requires having the fix ready to go on a CD or other media.
Certainly not to say that this couldn't happen with a clean Linux installation, but I don't know of any problems quite like the one that affects Windows.
The problem, of course, is that forkers often fork as an angry reaction to a rejection from the developer team without realizing what kind of commitment it is running a project. It gets worse if you try to parallel an existing project.
In order for a fork of Firefox to be successful, you'd have to gather a team of developers, and actually find your own means to decide who gets access and who doesn't, but you'd also have to merge all the changes going into Firefox's own tree at the same time as you accept lots of contributions from your fork community (assuming you get enough press to be receiving any).
I think a further complication is that sometimes with these forks, the mindset of being more open lets contributors get patches through with less quality control, leading to a product which fails to offer the same degree of stability and code quality as the original project.
Xorg seems to be a decent example of a fork team that got these problems reasonably ironed out. Perhaps they're a good place to turn for advise for people serious about forking a major project.
I suspect a lot of the negative comments in this thread are from viewers that haven't watched any episodes this season. I am enjoying a lot of the shows that have taken the opportunity to finally tie in to Trek history and future.
Previously a lot of the storylines deviated from anything that's been recorded in the Trek documented history. But now, they've changed direction a bit. I'd really hate to lose this series at this point, because it feels to me like Trek's one and only shot to fill in the pre-Original Series blanks in the timeline and strongly develop an understanding of how the Federation and Starfleet came to be.
Going back any further would set the show at about right now, which doesn't work, and going much ahead puts them back in Kirk's days, which has been done, and probably can't be again without systematically rewriting everything we know about that time period.
You've more or less got it now. Features and plugins are not frozen, exactly, but now you'll be able to tell when you should need to upgrade your plugins to new versions, because Gaim's version number contains that information.
Gaim is preparing version 2.0 which will include a rewrite of the API involving presence states (away, idle, etc). This doesn't indicate that 2.0 will be a total rewrite, but anything that uses the old away API will get broken, and 2.0 will indicate that a signification API change has been made.
This isn't a PR stunt. We've been dreading this posting on Slashdot, because we knew it would mean our IRC channel would become overrun with people thinking 1.0.0 was significant, and that this thread would run away in all the wrong directions.
People have (and we knew this way ahead of the actual release) this horrible misconception that 1.0.0 is Gaim's first major stable release, but for those too lazy to read the discussion on the mailing list, Gaim's versioning is very simple, and 1.0.0 is completely insignificant. We could have started at 12.8.17, but 1.0.0 is a nice place to start numbering from and not break package managers version comparison logic or seeming too completely random....
Example versioning:
1.0.0: First release with this numbering system 1.0.1: No changes to Gaim's API; New features, bugfixes 1.1.0: New Gaim API added; none removed or changed 1.2.0: Same as above 2.0.0: Gaim API changed or removed breaking plugins or anyone using libgaim. 2.0.1: No changes to API; New features & bugfixes 2.0.2: et cetera
As I placed in the topic of #Gaim:
1.0.0! Not special in any way, other than that you should be using it.
From what I understand, you're right. The commercials will feature people who are lower-paid staffers. My question is: Why can't the actors spare a half million dollars in a film? It could pay 10 members of the crew 50,000 each for the film easily, and I'm sure there are lots of crew that don't make even that much.
If the MPAA wasted less money on this nonsense, they could probably save enough money to give out raises to the production crew of most movies... or they could afford to stop showing me damn commercials before movies in the theaters.
At $8.50 for a 2 hour movie, I think I deserve not to have to watch commercials.
I used their tuned kernel, but there's something wrong with their NTFS driver. I get applications freezing when they try to access the partitions (intermittently). These apps won't even quit on a kill -9!
Consequently, I switched to vanilla sources, which don't have such a problem.
I don't see why the Kernel needs to be marketed in the first place. Distos include the kernel, prebuilt, but no consumers ever bother to find, download, and compile kernel source, unless they're advanced enough to care about (or perhaps even understand) the kernel improvements.
Anyone that's daring enough to compile the kernel himself, should know what's new, and not just compile for the sake of a rounded version number.
I'm inclined to agree with the sentiment of Linus. No 3.0 until the binary compatibility is broken. This is how they did it with Gnome 2, and they really broken binary compatibility there...
dolby2, for Linux ME, that would be the kernel released in 998 years, correct? Are you planning to be around to see it?;-)
Slashdot Mirror
The constant concern, of course, is that the fragmentation of the linux market into so many distributions makes the product difficult to market to the end user, especially as there is relatively little usability and interface consistency when you jump between distros. Tools, package managers, configuration wizards, installers, desktop managers as a whole... all vary from distro to distro. I'm not saying it's the obstacle preventing Linux from being a strong competitor to Windows, but I don't think it helps any.
One thing you can be sure about is that all 7 windows flavors will have the same UI, installer, and bugs.
It's not a problem of the machine miscounting. Part of the concept involved here is to allow us to be sure the machine is not intentionally lying about the results.
But the inkjet heads are in the cartridges.... If the dried ink destroys the head, you have to replace the cartridge anyway.
I think the boxes have expiration dates printed on them, but as annoying as HP's ink level checks have ever been (notifying me months before the ink is depleted), I've never run into a situation where it wouldn't let me run the ink all the way to the actual empty point.
The current major problem is that a clean installation of Windows XP can actually become seriously infected before the user has a chance to install certain major security fixes. This is very troubling for a lot of people.... The (annoying) workaround is to get the security fix onto the box before connecting it to the Internet, but this requires having the fix ready to go on a CD or other media.
Certainly not to say that this couldn't happen with a clean Linux installation, but I don't know of any problems quite like the one that affects Windows.
The problem, of course, is that forkers often fork as an angry reaction to a rejection from the developer team without realizing what kind of commitment it is running a project. It gets worse if you try to parallel an existing project.
In order for a fork of Firefox to be successful, you'd have to gather a team of developers, and actually find your own means to decide who gets access and who doesn't, but you'd also have to merge all the changes going into Firefox's own tree at the same time as you accept lots of contributions from your fork community (assuming you get enough press to be receiving any).
I think a further complication is that sometimes with these forks, the mindset of being more open lets contributors get patches through with less quality control, leading to a product which fails to offer the same degree of stability and code quality as the original project.
Xorg seems to be a decent example of a fork team that got these problems reasonably ironed out. Perhaps they're a good place to turn for advise for people serious about forking a major project.
I suspect a lot of the negative comments in this thread are from viewers that haven't watched any episodes this season. I am enjoying a lot of the shows that have taken the opportunity to finally tie in to Trek history and future.
Previously a lot of the storylines deviated from anything that's been recorded in the Trek documented history. But now, they've changed direction a bit. I'd really hate to lose this series at this point, because it feels to me like Trek's one and only shot to fill in the pre-Original Series blanks in the timeline and strongly develop an understanding of how the Federation and Starfleet came to be.
Going back any further would set the show at about right now, which doesn't work, and going much ahead puts them back in Kirk's days, which has been done, and probably can't be again without systematically rewriting everything we know about that time period.
You've more or less got it now. Features and plugins are not frozen, exactly, but now you'll be able to tell when you should need to upgrade your plugins to new versions, because Gaim's version number contains that information.
Gaim is preparing version 2.0 which will include a rewrite of the API involving presence states (away, idle, etc). This doesn't indicate that 2.0 will be a total rewrite, but anything that uses the old away API will get broken, and 2.0 will indicate that a signification API change has been made.
This isn't a PR stunt. We've been dreading this posting on Slashdot, because we knew it would mean our IRC channel would become overrun with people thinking 1.0.0 was significant, and that this thread would run away in all the wrong directions.
Just to be clear, I am not a Gaim developer. I am a patch contributor and plugin writer for Gaim that happens to be very familiar with the project.
People have (and we knew this way ahead of the actual release) this horrible misconception that 1.0.0 is Gaim's first major stable release, but for those too lazy to read the discussion on the mailing list, Gaim's versioning is very simple, and 1.0.0 is completely insignificant. We could have started at 12.8.17, but 1.0.0 is a nice place to start numbering from and not break package managers version comparison logic or seeming too completely random....
Example versioning:
1.0.0: First release with this numbering system
1.0.1: No changes to Gaim's API; New features, bugfixes
1.1.0: New Gaim API added; none removed or changed
1.2.0: Same as above
2.0.0: Gaim API changed or removed breaking plugins or anyone using libgaim.
2.0.1: No changes to API; New features & bugfixes
2.0.2: et cetera
As I placed in the topic of #Gaim:
1.0.0! Not special in any way, other than that you should be using it.
Fresh installs of Windows XP contain a compressed dllcache folder by default if I'm not mistaken.
For reference, port 23 is telnet. SMTP is port 25.
There are advantages to emerging the openoffice-bin package. :)
From what I understand, you're right. The commercials will feature people who are lower-paid staffers. My question is: Why can't the actors spare a half million dollars in a film? It could pay 10 members of the crew 50,000 each for the film easily, and I'm sure there are lots of crew that don't make even that much.
If the MPAA wasted less money on this nonsense, they could probably save enough money to give out raises to the production crew of most movies... or they could afford to stop showing me damn commercials before movies in the theaters.
At $8.50 for a 2 hour movie, I think I deserve not to have to watch commercials.
They'll demand something that just works.
Then Windows definitely isn't for them either...
Perhaps they should try a Mac? Apple's been working that "It just works" angle for a while now....
I used their tuned kernel, but there's something wrong with their NTFS driver. I get applications freezing when they try to access the partitions (intermittently). These apps won't even quit on a kill -9!
Consequently, I switched to vanilla sources, which don't have such a problem.
I don't see why the Kernel needs to be marketed in the first place. Distos include the kernel, prebuilt, but no consumers ever bother to find, download, and compile kernel source, unless they're advanced enough to care about (or perhaps even understand) the kernel improvements. Anyone that's daring enough to compile the kernel himself, should know what's new, and not just compile for the sake of a rounded version number. I'm inclined to agree with the sentiment of Linus. No 3.0 until the binary compatibility is broken. This is how they did it with Gnome 2, and they really broken binary compatibility there... dolby2, for Linux ME, that would be the kernel released in 998 years, correct? Are you planning to be around to see it? ;-)