My next video card purchase will be from a company that makes GPL-compatible drivers. Suggestions anyone?
We're basically screwed. Nvidia's success with binary-only drivers has made ATI do the same thing. Unless the Weather Channel decides to fund development of drivers for the newer cards, Radeon 8500 is the end of the open-source line. And no one else really exists in the high-performance 3D graphics card market.
I hope Red Hat uses some of their new half a billion to do something about this, but they've probably got other goals to hit first.
Okay, fair enough -- FUD might have been a bit strong. Never attribute to malice and all that.
Anyway: Red Hat / Fedora doesn't have the nifty "upgrade between releases while the system is running" thing you get with Debian. (Although you could try it with apt-rpm or yum, and probably get decent results, it just won't be as clean.) But the installer is able to detect previous installations and relatively smoothly update them.
Since you pretty much have to reboot to make a completely-upgraded distro really take effect (new kernel, new libc, new everything), it's not really a big deal.
Red Hat Linux has been able to do this since forever, and it's the primary reason I switched from Slackware back in the day. Well, that and SysV init scripts.:)
Fedora Core says 6 months of support. [...] Fedora Core's policy is explicity to upgrade to the latest packages.
Take a peek at Fedora Legacy. This addresses your first concern directly. And, although I haven't heard anything of it, if it turns out that Fedora Core's updates policy is too disruptive, I wouldn't be surprised if Fedora Legacy picks up the slack there. (In the meantime, there isn't really any indication that the updates policy will be as disruptive as threatened. Time will tell.)
Actually, Red Hat folks release Fedora updates for current releases. Here's the one from Monday for slocate, for example. Notice the redhat.com return address, and if you look at the package, you'll see it was built by a Red Hat engineer on a Red Hat system.
Older releases will be handled by the Fedora Legacy project, and while it'll take a little bit for that to get settled in, I'm highly confidant that it'll be a success. Again, see Debian -- "hobbyists" can do a good job of keeping security updates current.
[...] or b) move to this untested hobby distro (fedora) that requires a complete re-install anyway, people start looking at other distros.
I'm going to have to call FUD on this. Why would installing Fedora Core require a complete re-install? Doing an upgrade from Red Hat Linux 9 works fine.
For that matter, what's untested about it? Red Hat has to take some of the blame for this confusion, but in actuality, Fedora Core has gotten just as much pre-relese testing as previous consumer-level Red Hat distributions -- probably more, with the more-open development model.
It's also not *really* a hobby distro, any more than Debian is.
so, yeah, i'll be migrating our twelve servers from red hat to suse sometime in the next month or so.
Now *that* will take a complete reinstall. SuSE is a great distro so there's nothing wrong with that, but I suggest you take a second look at Fedora first.
That's right -- the last time around, these guys were attacking anti-spam projects -- basically members of our community. I'm willing to believe Bruce Perens' hypothesis that this is just more of the same, except indirectly.
You really find it easier to believe that this is a plot to bring down linux than that some high school kid who doesn't like sco did something stupid, as high school kids tend to do?
Maybe. But consider that the people who brought us earlier flavors of this worm aren't the old-fashioned "make a big fuss and break stuff aren't I l33t" variety -- they're in it for the money.
The people who are behind this know what they're doing. So, high school kid or not, I'm willing to think for a second about possible motives beyond the first glance.
Polite letters to CNN are probably in order. The line you've quoted is hardly the worst in that article -- the bit about "takes the Linux Wars to a new intensity" and so on is shameful.
Trust me, I've tried both. I got so fed up with the messages from Klez that I went on a little crusade. More people doing it can't hurt, but it's basically a tilting-with-windmills sort of exercise.
I wish some sysadmins would get a clue and realize that with viruses spoofing the From: address, there is no fscking point in sending the "you sent me a virus" panic mail.
I've been trying to complain to admins about this ever since Klez. You wouldn't believe the abuse I've gotten back -- and I've been very polite and nice. Generally, sites feel that it's adequate to add the newly found spoofing viruses to a don't-mail-notices blacklist after it's "realized" that yet another one can't be trusted. GET A CLUE, people -- you can't trust *viruses* at all.
The *real* problem is the antivirus software -- notices should only be sent for "known honest" viruses -- if at all. There should be *no* option to send these notices by default. But the antivirus companies *love* this -- they get to send out *millions* of advertisements for the effectiveness of their product, and no one is allowed to call it spam -- even though it *is*.
For some reason, people tend to read the message that recent nslookup spits out and glaze over once it gets to suggesting dig. Dig is really only useful for diagnostics -- for day to day use, the second replacement program nslookup suggests, host, is much prettier and simpler.
Using dig spits out about a page of info for aol; `host -t txt aol.com` just gives back a nice one-line response containing only the information asked for.
Yeah, if AOL is respecting SPF, and someone forges your domain name as part of the return address for spam destined for aol.com, they can know to drop it without bouncing. So it'd help. The spammer's mail server doesn't -- can't -- do anything about it. That's the whole point.
Well, I chose it because I like that band, and yeah, it makes a good example of what's broken. But remember, it's *or*, so it's not like it comes up with anything particularly bad -- just too many results, all ridiculous. The problem gets even worse with longer phrases -- good thing I'm not into emo bands....
(In addition to not being as slow as dirt, I mean.)
One of the obvious and natural things for a site like this is to try and link people together via shared interests, quirks, ideals, memberships, and so on.
Friendster lets you list these things, but has a terrible search interface -- if you like the band Poster Children, you can say so, but then trying to find other people with the same interet will reveal everyone who mentions either "poster" or "children". Basically, broken.
People try to work around this by creating fake "people" for abstract ideas -- there's a whole article about it at Salon. But the Friendster site people, instead of capitalizing on this, decided t hat this "subversion" is a plague trying to destroy their system. And, this work-around does make the social network part less useful -- having no way of distinguishing links between real people from those via Mickey Mouse is a problem.
Instead of trying to kick every abstract concept, cartoon character, university mascot, and geek web site logo off of the network, these things should be *encouaged*, but defined as separate from real people. It's both more fun *and* more useful. And it's exactly what Orkut does -- in addition to your own entry, anyone can create a "Community", and join as many such communities as they like.
So, Orkut is cooler because they have this feature -- but even more because they understand *why* to have it.
Are you aware of the history of this? There actually *was* a whole huge "intellectual property" squabble. Calculus was actually invented by *Leibniz* mostly-independently at about the same time or slightly earlier. Newton neglected to publish his original work for twenty years -- apparently for fear of criticism. Leibniz also didn't publish right away, but when he did, a huge PR fight ensued -- which Newton basically won, given that most people think of Newton as the inventor these days -- even though the modern version owes much to Leibniz.
(A quick google search will turn up more; too lazy to link myself.)
Slashdot Mirror
Right, unless it's being used descriptively to *actually refer directly to your product*. Which it clearly would be, in this case.
My next video card purchase will be from a company that makes GPL-compatible drivers. Suggestions anyone?
We're basically screwed. Nvidia's success with binary-only drivers has made ATI do the same thing. Unless the Weather Channel decides to fund development of drivers for the newer cards, Radeon 8500 is the end of the open-source line. And no one else really exists in the high-performance 3D graphics card market.
I hope Red Hat uses some of their new half a billion to do something about this, but they've probably got other goals to hit first.
Okay, fair enough -- FUD might have been a bit strong. Never attribute to malice and all that.
:)
Anyway: Red Hat / Fedora doesn't have the nifty "upgrade between releases while the system is running" thing you get with Debian. (Although you could try it with apt-rpm or yum, and probably get decent results, it just won't be as clean.) But the installer is able to detect previous installations and relatively smoothly update them.
Since you pretty much have to reboot to make a completely-upgraded distro really take effect (new kernel, new libc, new everything), it's not really a big deal.
Red Hat Linux has been able to do this since forever, and it's the primary reason I switched from Slackware back in the day. Well, that and SysV init scripts.
Fedora Core says 6 months of support.
[...]
Fedora Core's policy is explicity to upgrade to the latest packages.
Take a peek at Fedora Legacy. This addresses your first concern directly. And, although I haven't heard anything of it, if it turns out that Fedora Core's updates policy is too disruptive, I wouldn't be surprised if Fedora Legacy picks up the slack there. (In the meantime, there isn't really any indication that the updates policy will be as disruptive as threatened. Time will tell.)
Actually, Red Hat folks release Fedora updates for current releases. Here's the one from Monday for slocate, for example. Notice the redhat.com return address, and if you look at the package, you'll see it was built by a Red Hat engineer on a Red Hat system.
Older releases will be handled by the Fedora Legacy project, and while it'll take a little bit for that to get settled in, I'm highly confidant that it'll be a success. Again, see Debian -- "hobbyists" can do a good job of keeping security updates current.
Fedora is the renamed Raw Hide.
The development tree of Fedora is the new Raw Hide. The *releases* are the new Red Hat Linux. The process doesn't seem to have changed that much.
does it run on dell computers or HPs?
MS Internet Explorer? Well, no, not very well.
[...] or b) move to this untested hobby distro (fedora) that requires a complete re-install anyway, people start looking at other distros.
I'm going to have to call FUD on this. Why would installing Fedora Core require a complete re-install? Doing an upgrade from Red Hat Linux 9 works fine.
For that matter, what's untested about it? Red Hat has to take some of the blame for this confusion, but in actuality, Fedora Core has gotten just as much pre-relese testing as previous consumer-level Red Hat distributions -- probably more, with the more-open development model.
It's also not *really* a hobby distro, any more than Debian is.
so, yeah, i'll be migrating our twelve servers from red hat to suse sometime in the next month or so.
Now *that* will take a complete reinstall. SuSE is a great distro so there's nothing wrong with that, but I suggest you take a second look at Fedora first.
This may come as a shock, but MS IE isn't the only browser available to run on MS operating systems -- or better for your Mom, perhaps, Mac OS X.
First of all, it should read, some moronic Linux fans are cheering on MyDoom. Take an English class, buddy.
:)
If it's a halfway-decent English class, you'll learn that the first form is fine, and your suggested fix is wrong. (Look up: ambiguous phrasal verb.)
You Microsoft people. Sheesh.
Oh, wait: I meant to get this link in there, but hit submit too soon:
MiMail Virus Attacking Anti-Spam Groups
That's right -- the last time around, these guys were attacking anti-spam projects -- basically members of our community. I'm willing to believe Bruce Perens' hypothesis that this is just more of the same, except indirectly.
You really find it easier to believe that this is a plot to bring down linux than that some high school kid who doesn't like sco did something stupid, as high school kids tend to do?
Maybe. But consider that the people who brought us earlier flavors of this worm aren't the old-fashioned "make a big fuss and break stuff aren't I l33t" variety -- they're in it for the money.
The people who are behind this know what they're doing. So, high school kid or not, I'm willing to think for a second about possible motives beyond the first glance.
Polite letters to CNN are probably in order. The line you've quoted is hardly the worst in that article -- the bit about "takes the Linux Wars to a new intensity" and so on is shameful.
Trust me, I've tried both. I got so fed up with the messages from Klez that I went on a little crusade. More people doing it can't hurt, but it's basically a tilting-with-windmills sort of exercise.
Well, it's a start. C'mon, spread the word!
I wish some sysadmins would get a clue and realize that with viruses spoofing the From: address, there is no fscking point in sending the "you sent me a virus" panic mail.
I've been trying to complain to admins about this ever since Klez. You wouldn't believe the abuse I've gotten back -- and I've been very polite and nice. Generally, sites feel that it's adequate to add the newly found spoofing viruses to a don't-mail-notices blacklist after it's "realized" that yet another one can't be trusted. GET A CLUE, people -- you can't trust *viruses* at all.
The *real* problem is the antivirus software -- notices should only be sent for "known honest" viruses -- if at all. There should be *no* option to send these notices by default. But the antivirus companies *love* this -- they get to send out *millions* of advertisements for the effectiveness of their product, and no one is allowed to call it spam -- even though it *is*.
Wow. I never thought about that. Good thing I read Slashdot, so I could discover this valuable lesson about life.
Oh wait.
Ok, we've all been screwed over by the first two movies. They sucked. And there's nothing we can do about it: we have to go see the third one too.
But here's my feeble little geek protest: don't go opening night. Go the second night. Wouldn't it be cool if the first day, *no one* showed up?
For some reason, people tend to read the message that recent nslookup spits out and glaze over once it gets to suggesting dig. Dig is really only useful for diagnostics -- for day to day use, the second replacement program nslookup suggests, host, is much prettier and simpler.
Using dig spits out about a page of info for aol; `host -t txt aol.com` just gives back a nice one-line response containing only the information asked for.
Yeah, if AOL is respecting SPF, and someone forges your domain name as part of the return address for spam destined for aol.com, they can know to drop it without bouncing. So it'd help. The spammer's mail server doesn't -- can't -- do anything about it. That's the whole point.
Well, I chose it because I like that band, and yeah, it makes a good example of what's broken. But remember, it's *or*, so it's not like it comes up with anything particularly bad -- just too many results, all ridiculous. The problem gets even worse with longer phrases -- good thing I'm not into emo bands....
The notice is there after you log in.
(In addition to not being as slow as dirt, I mean.)
One of the obvious and natural things for a site like this is to try and link people together via shared interests, quirks, ideals, memberships, and so on.
Friendster lets you list these things, but has a terrible search interface -- if you like the band Poster Children, you can say so, but then trying to find other people with the same interet will reveal everyone who mentions either "poster" or "children". Basically, broken.
People try to work around this by creating fake "people" for abstract ideas -- there's a whole article about it at Salon. But the Friendster site people, instead of capitalizing on this, decided t hat this "subversion" is a plague trying to destroy their system. And, this work-around does make the social network part less useful -- having no way of distinguishing links between real people from those via Mickey Mouse is a problem.
Instead of trying to kick every abstract concept, cartoon character, university mascot, and geek web site logo off of the network, these things should be *encouaged*, but defined as separate from real people. It's both more fun *and* more useful. And it's exactly what Orkut does -- in addition to your own entry, anyone can create a "Community", and join as many such communities as they like.
So, Orkut is cooler because they have this feature -- but even more because they understand *why* to have it.
Are you aware of the history of this? There actually *was* a whole huge "intellectual property" squabble. Calculus was actually invented by *Leibniz* mostly-independently at about the same time or slightly earlier. Newton neglected to publish his original work for twenty years -- apparently for fear of criticism. Leibniz also didn't publish right away, but when he did, a huge PR fight ensued -- which Newton basically won, given that most people think of Newton as the inventor these days -- even though the modern version owes much to Leibniz.
(A quick google search will turn up more; too lazy to link myself.)
You can't acronymize acronyms...
, and PINELMINELMINELMINELMINELMINELMINELM....?
You can't?
So... GNUNUNUNUNUNUNUNUNUNUNUNUNUNUNUNUNNUNUNU.../Linux