Maybe it's harsh to say this, but anyone buying something from a site with a name like INTERNETSHOP.COM deserves what they get. It's even worse than dealing with a company in the.info or.biz domains!;)
Rip off Apple by using their code in your project without permission -> get sued by the copyright holder (Apple) Rip off Microsoft by using their code in your project without permission -> get sued by the copyright holder (Microsoft) Rip off a free software developer by using their code in your project without permission -> get used by the copyright holder (the developer)
Which part of this simple process don't you understand?
Bullshit. The reason you can't statically link Glibc because the developers don't think it's worth their time to solve the NSS dynamic module version skew problem (read the Glibc FAQ for details).
But why does it matter? Libc6's ABI/API has been stable for ages. Link against something old like 2.1, and your binary will work almost everywhere. You say you need to link against libc5; I don't know about that, but AFAIK, every distribution moved to Glibc years ago.
A new binary every month? Please. Here's a hint--you can pick between static/dynamic linking for each library you link to! If you want to use experimental/unstable library libvolitile, you can link that in statically, while still dynamically linking to Glibc and other stable libraries.
FYI, quietpc.com have an excellent selection of hardware, including PSUs. They trade all over the world, too!
IME the loudest things in a computer are the tiny graphics card, and motherboard chipset fans. They have to spin faster to shift the same amount of air; I ripped mine off my graphics card and replaced it with an 80mm case fan, and the computer is a lot quieter. I'm replacing the whole thing with a GV-N68128DH (passively cooled GeForce 6800!) soon though, for even more golden silence.
Of course, at the end of the day, noise is caused by motion. If I could afford it, I'd load up with a fanless PSU and solid state hard drives.:)
Thanks for the replies... I guess it's true what they say, about how to get an answer on the Internet: post some incorrect information and someone will correct you.:)
In theory it's piss all. In practice, developers are lazy. Readers of planet.gnome.org will have noticed the recent drive to reduce Gnome's memory footprint. For example, until a couple of days ago, every copy of nautilus maintained several copies of the desktop background in memory, so there's an instant 10-15 MB penalty per session. Pango and freetype were likewise egregious offenders.
I just came up with this rather nasty shell script to find out how much memory on my machine is being used for non-code segments in my processes:
sum=$(for pid in $(ps aux | awk '$1 == "sam" {print $2}'); do cat/proc/$$/maps | while read addr prot junk; do test $prot = 'r-xp' && continue; start=$(echo $addr | sed 's/-.*//'); end=$(echo $addr | sed 's/.*-//'); echo $(( 0x$end - 0x$start )); done; done | while read x; do echo + $x; done) echo $(( ($sum) / 1024 / 1024 )) 155
So my session is currently using up 155 MB of non-sharable memory. Actually, this seems rather low, given that I have firefox, thunderbird and azureus all chugging away. Maybe/proc/$pid/maps doesn't show the mmap'd sections that libc creates for large mallocs or something...
Can code segments even be shared between processes on i386? I seem to remember reading somewhere that they can't (or, in Linux, aren't)... and looking in/proc/something/maps I can't see *any* segments marked as sharable. They're all r-xp (readable, executable, private; code segments) or rw-p (readable, writable, private; data segments).
I reckon the people who still like (and use) CDE do so because they are happy with how it is at the moment. Those who want something else have already moved away to XFCE, Gnome, KDE, etc. So since the only users of CDE are happy with it... why change what works?:)
You appear not to have noticed... it might seem strange, surprising, maybe even a little confusing to you, but Slashdot has more than one reader.
Some of the members of this plurality of readers even have differing opinions. That's why (in theory) the discussions that acompany each story are interesting.
Of course in practice, story discussions consist entirely of whiners, trolls and clueless fucktards who don't even read the fucking article.
Fair enough--I apologise for the tone of my reply.
I wouldn't worry too much about unscrupulous companies making a habit of ripping off code from GPL projects--since each project has a different copyright holder, it's hard to say in advance who is likely to retaliate with a lawsuit and who is not.
Slashdot Mirror
Maybe it's harsh to say this, but anyone buying something from a site with a name like INTERNETSHOP.COM deserves what they get. It's even worse than dealing with a company in the .info or .biz domains! ;)
Maybe you should read the license that the software is distributed under, instead of trolling Slashdot.
Rip off Apple by using their code in your project without permission -> get sued by the copyright holder (Apple)
Rip off Microsoft by using their code in your project without permission -> get sued by the copyright holder (Microsoft)
Rip off a free software developer by using their code in your project without permission -> get used by the copyright holder (the developer)
Which part of this simple process don't you understand?
And rightly so. If Microsoft is so against software patents, why are they lobbying for them in the EU?
Just because they haven't gone on the offensive yet doesn't mean they won't in the future. Why give them the chance?
Bullshit. The reason you can't statically link Glibc because the developers don't think it's worth their time to solve the NSS dynamic module version skew problem (read the Glibc FAQ for details).
0 347.html for more.
But why does it matter? Libc6's ABI/API has been stable for ages. Link against something old like 2.1, and your binary will work almost everywhere. You say you need to link against libc5; I don't know about that, but AFAIK, every distribution moved to Glibc years ago.
A new binary every month? Please. Here's a hint--you can pick between static/dynamic linking for each library you link to! If you want to use experimental/unstable library libvolitile, you can link that in statically, while still dynamically linking to Glibc and other stable libraries.
This came up recently on debian-devel; read http://lists.debian.org/debian-devel/2005/03/msg0
FYI, quietpc.com have an excellent selection of hardware, including PSUs. They trade all over the world, too!
:)
IME the loudest things in a computer are the tiny graphics card, and motherboard chipset fans. They have to spin faster to shift the same amount of air; I ripped mine off my graphics card and replaced it with an 80mm case fan, and the computer is a lot quieter. I'm replacing the whole thing with a GV-N68128DH (passively cooled GeForce 6800!) soon though, for even more golden silence.
Of course, at the end of the day, noise is caused by motion. If I could afford it, I'd load up with a fanless PSU and solid state hard drives.
Erm, whoops. I replied to the wrong commont or something. Which I've now closed the tab for. Argh. Much gnashing of teeth.
Care to recommend something else? There are about a hojillion books of this nature, and I am afflicted with the subtle tyranny of choice. :)
That's a problem with Glibc, not GCC. If you chose to go against the developers and statically link to Glibc, be prepared to handle the consequences!
Thanks for the replies... I guess it's true what they say, about how to get an answer on the Internet: post some incorrect information and someone will correct you. :)
In theory it's piss all. In practice, developers are lazy. Readers of planet.gnome.org will have noticed the recent drive to reduce Gnome's memory footprint. For example, until a couple of days ago, every copy of nautilus maintained several copies of the desktop background in memory, so there's an instant 10-15 MB penalty per session. Pango and freetype were likewise egregious offenders.
/proc/$$/maps | while read addr prot junk;
/proc/$pid/maps doesn't show the mmap'd sections that libc creates for large mallocs or something...
/proc/something/maps I can't see *any* segments marked as sharable. They're all r-xp (readable, executable, private; code segments) or rw-p (readable, writable, private; data segments).
I just came up with this rather nasty shell script to find out how much memory on my machine is being used for non-code segments in my processes:
sum=$(for pid in $(ps aux | awk '$1 == "sam" {print $2}');
do cat
do test $prot = 'r-xp' && continue;
start=$(echo $addr | sed 's/-.*//');
end=$(echo $addr | sed 's/.*-//');
echo $(( 0x$end - 0x$start ));
done;
done | while read x;
do echo + $x;
done)
echo $(( ($sum) / 1024 / 1024 ))
155
So my session is currently using up 155 MB of non-sharable memory. Actually, this seems rather low, given that I have firefox, thunderbird and azureus all chugging away. Maybe
Can code segments even be shared between processes on i386? I seem to remember reading somewhere that they can't (or, in Linux, aren't)... and looking in
I reckon the people who still like (and use) CDE do so because they are happy with how it is at the moment. Those who want something else have already moved away to XFCE, Gnome, KDE, etc. So since the only users of CDE are happy with it... why change what works? :)
It's not all text. Each library will have a per-process segment for stack and heap, which therefore can not be shared between users.
everything2.com?node=stupid+unix+tricks
Some of the members of this plurality of readers even have differing opinions. That's why (in theory) the discussions that acompany each story are interesting.
Of course in practice, story discussions consist entirely of whiners, trolls and clueless fucktards who don't even read the fucking article.
They don't need to give themselves a right to do that at all.
Have you agreed to the terms and conditions of every mail server your mail is relayed to?
I assumed that was just audiophiles getting ripped off.
"yeah i use these special rare oxygen impregnated CDs. my recordings sound so much more natural"
That's talking about the 1.4 release of Mozilla. The page you reference is about 35 years out of date.
Ben, thanks for taking the time to look at all the spyware and EULA bullshit that these programs hit us with.
I think a good way to improve your page would be to state, for each tested program:
* Download URL
* Version downloaded
* MD5sum of the downloaded file
Yes, you can. Next time, try using a program before you bash it.
Didn't it occur to you that the lameness filter was doing its job?
Christ, what is it with you people? Did you all manage to remove libgnomevfs by accident or something?
I just ran "gnome-gv http://www.marcusevans.com.au/pdf/413.pdf" in Gnome 2.8 and it worked fine, just as it's done for ages.
"Slashdot" is not a single, monolithic being, you pillock.
Try http://writetothem.com/.
Fair enough--I apologise for the tone of my reply.
I wouldn't worry too much about unscrupulous companies making a habit of ripping off code from GPL projects--since each project has a different copyright holder, it's hard to say in advance who is likely to retaliate with a lawsuit and who is not.