RedHat 7.2 Beta: Roswell
LinuxNews.pl writes: "Few days ago RedHat uploaded new Beta release of their distro -
Roswell. If you want to find out more about it just go to the LinuxNews.pl"
And I won't even make a snide comment about how I haven't run Red Hat in 2 years!
I have done a few sudden power-off tests with XFS-SGI/RedHat and Reiser/Mandrake. I had disk corruption with Reiser once, but once is all it takes to ruin your weekend. XFS has been rock solid. My employer has been using XFS on SGI servers for years without problems. The XFS for Linux site is here: http://oss.sgi.com/projects/xfs/ I have only done one power-off test with ext3 and it performed as it should. I have not done any speed benchmarking between the different types yet.
If you think Taco's an asshole, try being on the receiving end of the amount of crap email he has to reply to every day.
"Like a madman shooting firebrands or deadly arrows is a man who deceives his neighbor and says, 'I was only joking!'" -- Proverbs 26:18-19
Sometimes the Bible is surprisingly relevant.
(BTW, this comment is not intended as a troll; it's just that few people know the Bible has anything to say about the whole "just kidding" thing. Though I fear I may spark a long offtopic thread anyway. Apologies in advance if such a thread ensues.)
Graham "Teach" Mitchell, computer science teacher, Leander HS
I installed it on a spare box a few days ago. The default filesystem is ext3. It also has migrations tools to convert existing ext2 partitions.
Here is a list of gcc-3.0 improvements over gcc-2.95.3. As for myself, I really appreciate the libstdc++-3.0 support.
I am less than enthusiatic about the
New warnings for C code that may have undefined semantics because of violations of sequence point rules in the C standard (such as a = a++;, a[n] = b[n++]; and a[i++] = i;), included in -Wall.
OMS/LiVid code, in particular, trips this up.
He's a decent guy in person. I spent a couple of days around him, Chris Dibona, Krow, Pudge, and Patrick G. as well. They're all decent guys, and at least half of them are brilliant.
Maybe it's different 'cuz of my business dealings, but Malda sure wasn't a whiny prima-donna. He's got his opinions and all, but doesn't take himself too seriously (really!). I now have the impression half the stupid article comments are meant to poke fun at himself.
Besides that, he bought a round of drinks one night and supper another. (Of course, he probably did it to pay back Jim Gettys for the years of work on X11.)
I still wouldn't hire him to write software that anyone else would ever maintain, though.
how to invest, a novice's guide
;-)
I love the style of the writing of the press release:
Etc.Too bad we can't say the same about XP
"It is a greater offense to steal men's labor, than their clothes"
7.2 includes KDE 2.2 and GNOME 1.4 - both of them make life easier for new users.
/bin/sh".
;)
7.2 includes more packages than any of the previous releases, so chances you'll actually have to compile something on your own are lower.
File permissions etc. are still there (and will stay) - but they're not really complicated once you've understood them. Basically, a file can be read, written and executed (think of the third as renaming a file from test to test.exe on that other operating system - it's done differently, but the effect is pretty much the same) - each of the operations can be allowed or denied. The file managers in KDE and GNOME give you a GUI frontend to changing permissions, so you don't need to remember commands like "chmod o+rw test" or "chmod 4777
Try it, and let me know if you're seeing any usability problems - I'd like to fix them, but noticing them after you've used Linux for 10 years is quite tricky.
This message is provided under the terms outlined at http://www.bero.org/terms.html
They didn't have an option on this. Red Hat have always maintained binary compatibility throughout major version numbers, so shipping with anything other than 2.96-RH as the default compiler just wasn't an option. Naturally, they've supplied 3.0 for those that want to use it. I can't really fault RH about this -- they've done exactly as they said they would. Can you imagine the uproar if they'd broken their promises on binary compatibility?
"The invisible and the non-existent look very much alike." -- Delos B. McKown
I have to disagree with redhat...
/usr/local, and I use it with great frequency. I have had no problems with its binaries. I dont pass binaries from system to system so unless you make RPMS i dont think anyone really cares, particularly when it comes to workstation machines, we all hack our own the way we like, and no one hacks up stuff around 2.96 - at least i wouldnt.
:0)
:0)
I link to kgcc on all redhats I use. Yes, egcs/gcc 2.9.1/112. or whatever kgcc is.
The first reason is the linux kernel is recommended to be compiled with that release and the second is that 2.96 was an experimental 3.0alpha fork. It is broken, deprecated and gcc and stallman were so pissed redhat had used it in a wide release product.
I have redhat 7.1 with gcc 3.0 in
I like Slackware 8.0 for workstation use, at work we stuck with redhat 7.0 (with all the updates of course). I recommend that for server use (7.1 did some weird stuff around java). I also use kgcc -> gcc/cc sym links as the binaries produced by that compiler are "real," that 2.96 is a horror show.
The 3.0 gcc is also better at ansi/posix/whatever else, i currently have 2.4.7 running perfectly compiled from that, as well as xfree 4.1.0. I think KDE is a nice UI, but it has "dirty closed roots" and do not consider it a valid project (YET
CHeers to redhat for trying, but i vote NO one the lame 2.96 fork
-Z
Legalize the constitution. Think for yourself question authority.