Uh. The SCSL is Yet Another software license which has come to the fore with Sun "releasing" source. There are plenty of us for whom the majority of other source-available software comes out under the GPL and other Open-Source arrangements. And if you were half-way awake during the 'sun releases solaris source' debate on here, you'd be well aware that it caused some speculative confuddlement. I say, good on Linus for setting the record straight. S'there:)
I'd agree lots: just because something is American, it doesn't necessarily deserve a.com address. We have.us for American stuff, and it's *high* time that got enforced. That'd clean-up.com no end. Second, people putting their names as.com domains seem to be selling their soul a bit - I am not a commercial entity, and hope never to be "for sale". (In much the same way as, say, microsoft.com is a company, selling "stuff", and theoretically open for potential sale as an entity - I'll never want PigleT.com on the same grounds!).
Roll on IPv6 with a *new* domain name scheme, I say.
I'd agree that either the US' justice system is screwy, and with you point about corporeality though. If it doesn't *exist*, who cares how much it's worth?
a) Linux is a kernel - but we've been down THAT line before. b) Linux the OS is as much ideology ("free software", "free beer", "free speech", Open-Source, GNU) as it is software, it seems. Or at least the ideology is a big thing behind it all. c) Freeware gets around but has no clout with PHBs who don't understand "No Money", so someone starts selling CDs of it. d) Somehow, RedHat make enough money from the combined box-shifting and code-writing and support and distro-maintenance things, that there's a "linux market"? Oh pants.
What is the precise relationship between Mandrake and RedHat linux distributions these days?
Perhaps Mr Leibovitch ought to wake up to the idea that it was us "geeks" who brought him linux in the first place, and if we don't credit a Macmillan distribution (I'd never even *heard* of it until today!) then that's the way it goes, whether it sells boxes or not.
It's called doing the Right Thing, and I suspect merely cloning distributions doesn't deserve as much "respect" as writing your *own*, surprisingly enough...
As for we shun it "because it represents what [we] don't understand: marketing and sales", bollocks. I understand "Marketing" and "sales" as far as they're useful - they get credit for supplying my salary, and that's about it. For selling their souls to push something and having no brains (as per Dilbert stereotype), no credit. The phrase is "box-shifter", and we've never yet had any respect for them, if they lack any developers - brains! - of their own...
Yes, but how much stuff at university is geared towards you getting a job in an industry at the end of it? I know I was a tad horrified when a fairly large corporation opened up shop outside my old Uni in the past couple of years, and they started tailoring the MSc in Electrical Engineering towards folks wanting to work in the suburbs thereafter...
It begs the question, what's research worth of its own, if it doesn't lead to an industry?
I thought MIT was a Good Place(TM)? As in, PGP... GNU?! Hang around: % nslookup Name: prep.ai.mit.edu Address: 209.81.8.252 Name: ftp.gnu.org Address: 209.81.8.252
Uh-oh bigtime. Paws off our FTP server, you evil M$loth scum you!
> Now with linux, the newbie has to learn all this from manuals written for the wrong culture. I know...it's maddening.
Don't say "wrong", if you want to live;)
There are still those of us who read and understand manpages, surprisingly enough - and it's not "with linux", but "with the linux 'revolution'" that I think you meant. (Not just as pedantic "it's only a kernel" point, but that even as an OS, why should it be subject to a rave review process and lots of hype [albeit well-founded]?)
Newbies start with newbie documentation. Get some fundamental *understandings* (how to use your favoured text editor, what a panel, launcher and applet are in gnome, how to use your chosen mailer, the list is endless) and then *move on*. Don't fall into the trap of blaming linux for things which aren't its own fault (I'm glad to see you've managed to avoid this one).
POP3 to download stuff, SMTP to send it out. You knew that:) Format of ~/.fetchmailrc is pretty damned simple - stick all the keywords together on a line per POP3 host, and you're away. Oh, and use something like fetchmail -k -M '/usr/bin/procmail %T' to do the actual delivery. If you don't know what the -M thing does, definitely read the manpage! ;)
Well said! Security, like firewalls, is not something that comes in a big box to be applied en-lump. (Firewalls: which do you want, an ipchains that lets 110/tcp through as filtered regardless of setting (beats me!) or which advertises what sort of firewall it is when you telnet into it??)
Anyone here been to sunsolve recently? I mean, I diff'd someone else's solaris 2.6 box against one of ours, and found a meagre 98 patches different... it's not as though "Real OSs" don't have individual patches to apply...
OTOH: games are very male-orientated, especially the shoot-em-ups, some strategy, football... etc. How many programs are designed to make anyone have 'softer' feelings like wanting to cry? (Wuss alert!:)
There's got to be a compromise though, somewhere - I think if women want to be as clued-up as guys in the computing field, they should clue-up, not blame the IT world though - as long as the IT scene exists independent of any sex bias (which it should).
In that case, it wants integrating with the regular rpm command as shipped with RedHat, unless I'm much mistaken. (Doesn't *have* to be the same executable, but it should be linked with it, so that folks end up thinking in terms of rpmfind to get their stuff, rather than the ol' rpm command itself).
Been there, tried that. rpm -Uhv is my standard way of installing them, and I do do them in bulk. But not off a remote site - that's a pain in the butt to organize on the commandline, even with glob-style filename generation (e.g. http://site.net/path/{a,b,c}.rpm , that kind of thing). The question is, how do I know what RPMs to get other than by trawling through rpmfind.net all the time?
And until that's fixed, RPM will still be second fiddle to deb/apt...:)
No. Rpmfind.net is not a central repository, it's a quick search tool for the RPM you wanted. You have to resolve all the dependencies *yourself*, which is evil. I don't want to have to navigate a tree of these dependencies and then install them in the right order all by hand, when I can do a quick 'apt-get install whatever' against any Debian mirror and it works. In fact, with apt you can have several repositories listed, like the KDE site, Gnome site, unstable and stable debian mirrors, and then in two commands, get the whole lot of updates at once. With RPM, fine so you can use ftp or http if you know the URL but as for automating the whole thing... just forget it.
One word: glibc2. I can upgrade libc6 on the fly here, with no problems at all. I've never yet seen that happen on a non-debian (rpm-based) box (redhat OR suse - both seem to require a reboot or two and it might not even work..)
I guess RedHat had better shape-up their package *management* (as distinct from install/update/freshen) tool... then I'll consider RPM remotely valid.
I don't know when 2.2.13 is expected, and it can't be that long off, but there's nothing wrong with wanting to get a new release out now, and if they have to *test* the stuff first, waiting on a new kernel and then testing/everything/ to make sure it still works is likely to take absolutely ages.
What worries me is the same response to the microsoft 'crack this box please' challenge: *why* should "our lot" of open-source developers work on something for Sun, for free? We don't get anything out of it, really - we have our own OSs, we don't need to fix Sun's bugs for them!
Given this, it's a nonsense to hope "we'll trust solaris because we can see the source". We can just see the bugs in the thing some of us used to trust anyway...
Besides which, the restrictions on commercial use probably violate one or other open-source license...
As a general discrimination "thing", *why* must there be 50% male/female ratios in linux, *why* must there be equal proportions of coloured / white / ethnic (whatever) in business, *why* do people seem to think 1/(total_number_of_options) in general, as though everyone were the same??
If females don't like linux, they don't have to use it. If a female wants to get into linux, it's here & waiting. We don't have to have a 'gurlLinux' (whatever the heck that is) for the purpose, do we?
How's about instead of having to live in an "equal-opportunity" world we had "equal-ineptitude"? That strikes me as more fair...
>What do people think would be hard to test about putting a satellite in orbit around another planet?
Your testing model only extends as far as a tree structure, where Component A relies on A.1, A.2, say, so you're testing only that reliance works. In practice, the project components are far more likely to be a graph, ie with C.3 relying on A.2 and so on. ("Relies on" could be "uses", or "conflicts with"; it matters not, the point is there's a relationship between diverse bits of soft- or hardware.)
Hence you don't have a simple test process where you can say "A.1 works and A.2 works therefore A as a whole works", because for every other thing out there with any chance of relating to other components, you've got to test it against them all.
In practice, how many people would do n**2 tests when they think they can get away with log2(n)?
It might be a design feature that they wanted some results to come out in olde English units, on the grounds that "miles" are a much better-understood unit.
Honestly... just because you can't understand pounds, shillings & pence doesn't mean you have to inflict a rebellious "metric" system on the rest of the world...
b) vmware is not proven for testing purposes; it could *fix* bugs in real installations, not necessarily just introducing more of the old 'windoze is unstable' crap.
c) if it's windoze, it's more virus-prone than linux, whether you run the sucker in vmware or not.
I think I agree entirely. If folks think that Christianity is in some way related to having t-shirts that are more provocative than Christian in nature (whatever makes a Christian t-shirt I have no idea), they're seriously wrong. And the general effect can be to cheapen the real thing.
It's those who can't distinguish the attempt at a joke and think *everything* has to be serious that get to me...
I dunno... I don't think it's a "trend", as the excessively-"spiritual" folks lacking in humour have been with us for years; chances are they'll never go away either.
The way to deal with it all, I think, is to see the funny side of it if you can, or ignore it if you can't.
Slashdot Mirror
Uh. :)
The SCSL is Yet Another software license which has come to the fore with Sun "releasing" source. There are plenty of us for whom the majority of other source-available software comes out under the GPL and other Open-Source arrangements.
And if you were half-way awake during the 'sun releases solaris source' debate on here, you'd be well aware that it caused some speculative confuddlement.
I say, good on Linus for setting the record straight.
S'there
I'd agree lots: just because something is American, it doesn't necessarily deserve a .com address. We have .us for American stuff, and it's *high* time that got enforced. That'd clean-up .com no end. .com domains seem to be selling their soul a bit - I am not a commercial entity, and hope never to be "for sale". (In much the same way as, say, microsoft.com is a company, selling "stuff", and theoretically open for potential sale as an entity - I'll never want PigleT.com on the same grounds!).
Second, people putting their names as
Roll on IPv6 with a *new* domain name scheme, I say.
I'd agree that either the US' justice system is screwy, and with you point about corporeality though. If it doesn't *exist*, who cares how much it's worth?
What is a linux "market", anyway?
a) Linux is a kernel - but we've been down THAT line before.
b) Linux the OS is as much ideology ("free software", "free beer", "free speech", Open-Source, GNU) as it is software, it seems. Or at least the ideology is a big thing behind it all.
c) Freeware gets around but has no clout with PHBs who don't understand "No Money", so someone starts selling CDs of it.
d) Somehow, RedHat make enough money from the combined box-shifting and code-writing and support and distro-maintenance things, that there's a "linux market"? Oh pants.
But read "the scary bit" by Awel further down. How does MIT benefit at all from it? It appears to me that M$loth win, or win. D'oh... :(
What is the precise relationship between Mandrake and RedHat linux distributions these days?
Perhaps Mr Leibovitch ought to wake up to the idea that it was us "geeks" who brought him linux in the first place, and if we don't credit a Macmillan distribution (I'd never even *heard* of it until today!) then that's the way it goes, whether it sells boxes or not.
It's called doing the Right Thing, and I suspect merely cloning distributions doesn't deserve as much "respect" as writing your *own*, surprisingly enough...
As for we shun it "because it represents what [we] don't understand: marketing and sales", bollocks. I understand "Marketing" and "sales" as far as they're useful - they get credit for supplying my salary, and that's about it. For selling their souls to push something and having no brains (as per Dilbert stereotype), no credit.
The phrase is "box-shifter", and we've never yet had any respect for them, if they lack any developers - brains! - of their own...
Yes, but how much stuff at university is geared towards you getting a job in an industry at the end of it?
I know I was a tad horrified when a fairly large corporation opened up shop outside my old Uni in the past couple of years, and they started tailoring the MSc in Electrical Engineering towards folks wanting to work in the suburbs thereafter...
It begs the question, what's research worth of its own, if it doesn't lead to an industry?
I thought MIT was a Good Place(TM)?
As in, PGP... GNU?! Hang around:
% nslookup
Name: prep.ai.mit.edu
Address: 209.81.8.252
Name: ftp.gnu.org
Address: 209.81.8.252
Uh-oh bigtime. Paws off our FTP server, you evil M$loth scum you!
>"If you can't explain what it does in one sentence, it will fail."
Erm, yeah. Sounds like one of those Dilbert PHB sayings, on a bad day...
They did say what it does in one or two sentences - as an animated GIF thing in the middle. And then it disappeared and was replaced by yet more crap.
Mind you, it might be interesting to see what we Brits have to put up with...
> Now with linux, the newbie has to learn all this from manuals written for the wrong culture. I
;)
:)
know...it's maddening.
Don't say "wrong", if you want to live
There are still those of us who read and understand manpages, surprisingly enough - and it's not "with linux", but "with the linux 'revolution'" that I think you meant.
(Not just as pedantic "it's only a kernel" point, but that even as an OS, why should it be subject to a rave review process and lots of hype [albeit well-founded]?)
Newbies start with newbie documentation. Get some fundamental *understandings* (how to use your favoured text editor, what a panel, launcher and applet are in gnome, how to use your chosen mailer, the list is endless) and then *move on*. Don't fall into the trap of blaming linux for things which aren't its own fault (I'm glad to see you've managed to avoid this one).
POP3 to download stuff, SMTP to send it out. You knew that
Format of ~/.fetchmailrc is pretty damned simple - stick all the keywords together on a line per POP3 host, and you're away. Oh, and use something like fetchmail -k -M '/usr/bin/procmail %T' to do the actual delivery. If you don't know what the -M thing does, definitely read the manpage!
;)
Well said!
Security, like firewalls, is not something that comes in a big box to be applied en-lump.
(Firewalls: which do you want, an ipchains that lets 110/tcp through as filtered regardless of setting (beats me!) or which advertises what sort of firewall it is when you telnet into it??)
Anyone here been to sunsolve recently? I mean, I diff'd someone else's solaris 2.6 box against one of ours, and found a meagre 98 patches different... it's not as though "Real OSs" don't have individual patches to apply...
Quite so!
:)
OTOH: games are very male-orientated, especially the shoot-em-ups, some strategy, football... etc.
How many programs are designed to make anyone have 'softer' feelings like wanting to cry?
(Wuss alert!
There's got to be a compromise though, somewhere - I think if women want to be as clued-up as guys in the computing field, they should clue-up, not blame the IT world though - as long as the IT scene exists independent of any sex bias (which it should).
In that case, it wants integrating with the regular rpm command as shipped with RedHat, unless I'm much mistaken.
(Doesn't *have* to be the same executable, but it should be linked with it, so that folks end up thinking in terms of rpmfind to get their stuff, rather than the ol' rpm command itself).
Been there, tried that. rpm -Uhv is my standard way of installing them, and I do do them in bulk.
... :)
But not off a remote site - that's a pain in the butt to organize on the commandline, even with glob-style filename generation (e.g. http://site.net/path/{a,b,c}.rpm , that kind of thing).
The question is, how do I know what RPMs to get other than by trawling through rpmfind.net all the time?
And until that's fixed, RPM will still be second fiddle to deb/apt
No. Rpmfind.net is not a central repository, it's a quick search tool for the RPM you wanted.
You have to resolve all the dependencies *yourself*, which is evil. I don't want to have to navigate a tree of these dependencies and then install them in the right order all by hand, when I can do a quick 'apt-get install whatever' against any Debian mirror and it works.
In fact, with apt you can have several repositories listed, like the KDE site, Gnome site, unstable and stable debian mirrors, and then in two commands, get the whole lot of updates at once.
With RPM, fine so you can use ftp or http if you know the URL but as for automating the whole thing... just forget it.
One word: glibc2. I can upgrade libc6 on the fly here, with no problems at all. I've never yet seen that happen on a non-debian (rpm-based) box (redhat OR suse - both seem to require a reboot or two and it might not even work..)
I guess RedHat had better shape-up their package *management* (as distinct from install/update/freshen) tool... then I'll consider RPM remotely valid.
Just my two pfennigs' worth...
Why jump to this conclusion?
/everything/ to make sure it still works is likely to take absolutely ages.
I don't know when 2.2.13 is expected, and it can't be that long off, but there's nothing wrong with wanting to get a new release out now, and if they have to *test* the stuff first, waiting on a new kernel and then testing
And the coke will be flat... but at least it'll have green cheese! ;)
Erm, me, I pronounce it something like '(d)'vor-zhak' with a relatively quiet 'd'...
:)
YMMV though
One of these days, I'll actually get to play with a dvorak keyboard then..
I guess you didn't read the article...
What worries me is the same response to the microsoft 'crack this box please' challenge: *why* should "our lot" of open-source developers work on something for Sun, for free?
We don't get anything out of it, really - we have our own OSs, we don't need to fix Sun's bugs for them!
Given this, it's a nonsense to hope "we'll trust solaris because we can see the source". We can just see the bugs in the thing some of us used to trust anyway...
Besides which, the restrictions on commercial use probably violate one or other open-source license...
A slightly idealist comment...
As a general discrimination "thing", *why* must there be 50% male/female ratios in linux, *why* must there be equal proportions of coloured / white / ethnic (whatever) in business, *why* do people seem to think 1/(total_number_of_options) in general, as though everyone were the same??
If females don't like linux, they don't have to use it. If a female wants to get into linux, it's here & waiting. We don't have to have a 'gurlLinux' (whatever the heck that is) for the purpose, do we?
How's about instead of having to live in an "equal-opportunity" world we had "equal-ineptitude"? That strikes me as more fair...
>What do people think would be hard to test about putting a satellite in orbit around another planet?
Your testing model only extends as far as a tree structure, where Component A relies on A.1, A.2, say, so you're testing only that reliance works. In practice, the project components are far more likely to be a graph, ie with C.3 relying on A.2 and so on.
("Relies on" could be "uses", or "conflicts with"; it matters not, the point is there's a relationship between diverse bits of soft- or hardware.)
Hence you don't have a simple test process where you can say "A.1 works and A.2 works therefore A as a whole works", because for every other thing out there with any chance of relating to other components, you've got to test it against them all.
In practice, how many people would do n**2 tests when they think they can get away with log2(n)?
Crap.
It might be a design feature that they wanted some results to come out in olde English units, on the grounds that "miles" are a much better-understood unit.
Honestly... just because you can't understand pounds, shillings & pence doesn't mean you have to inflict a rebellious "metric" system on the rest of the world...
a) It's not emulation, for goodness' sake
b) vmware is not proven for testing purposes; it could *fix* bugs in real installations, not necessarily just introducing more of the old 'windoze is unstable' crap.
c) if it's windoze, it's more virus-prone than linux, whether you run the sucker in vmware or not.
Is it really "linux as a windows app"??
:)
I thought it was a linux-on-umsdos thing, with a windoze installer & starter. Big difference.
We abandoned umsdos installations (with slackware) some years ago, let's keep it that way, OK?
I think I agree entirely.
If folks think that Christianity is in some way related to having t-shirts that are more provocative than Christian in nature (whatever makes a Christian t-shirt I have no idea), they're seriously wrong. And the general effect can be to cheapen the real thing.
It's those who can't distinguish the attempt at a joke and think *everything* has to be serious that get to me...
I dunno...
I don't think it's a "trend", as the excessively-"spiritual" folks lacking in humour have been with us for years; chances are they'll never go away either.
The way to deal with it all, I think, is to see the funny side of it if you can, or ignore it if you can't.