Yes we. Do. We also remeber lan's reply - while Alan works for Red Hat, I doubt he's one to mince words, as would most people who have had some contact with him would attest.
I know who I would trust with my kernel. Can't remember the link - Google is your friend.
rpm: I think the rpm + up2date combo has all the features you need. If you think there's something we need to add, please let me know..
A centralized repository of packages (unsupported by Red Hat), organized and tested by Open Source maintainers, but providing a wide variety of pakcages.
Users will install unsupported software anyway, but the lack of a single place to find *good* Red Hat packages annoys many Red hat users, myself included (though easily finding the distro better than itys competitors). By using Red Hat's bandwidth and mirroring, it would be *the* first place to look to find packages. With an appropriate warning, up2date could also get packages from these mirrors.
Likewise, I too have been using ReiserFS with 2.2 for around a year. That said, Civilme on Mandrake's Crashtesters list acknowledges ReiserFS under 2.4 )and perhapds 2.2 - I'm not sure) to be unstable. The message is on the crashtester archives from today, but I've deleted it, so no quote...
Red Hat's gcc was severely broken upon release. But so is the current gcc 1.95. The difference is that 2.95 is currently more broken than 2.96 with all the updates (2.95 can't compile glibc 2.2 on non x86).
I don't think Red Hat should complain Mandrake are `stealing' their compiler. That's *very* weak. Mandrake are more acturately using (and legitimizing) Red Hat's compiler.
Unfortunately, a strap-on security model is always very sucky. Look at ACLs on any unix or even NT.
And? What makes them strap on? So Unix wasn't designed to be a secure OS. Neither was it designed to be anything more than a time sharing system.Its evolved beyong those functions and works great. The ACLs on an Unix, NT, 2K, and VMS work very well.
What you want is capabilities.
I want them too, but they don't replace ACLs. Capabilities are already in kernel 2.4.
I have some Word templates on a SMB server. I want:
* The software to be owned by the admin account
* The software group to have read and write permission, and SetGID properties
* The staff group to have
* Other users to have no access.
RWX doesn't allow me to do this in anyway which isn't vastly hackish.
This allows anyone to implement a very simple security model
ACLs are as simple or fine grained as you need them to be, and a single line ACL is quite obviously simpler than an RWX permission.
which is much easier to verify.
Bah. Same thing - again, ACLs can be less compels than RWX if that's all you need.
the dumb ass home grown authentication systems on unix. (ie Each app authenticating people in a funny way)
Damn straight. Things like Squid, and BIND, and...wait, they all use ACLs already.
Capabilities are nice. They don't replace the need for ACLs, and while they work well and should be used, the RWX / Capability combination is more complex than a single tight ACL, combined perhaps with capabilities should you need them.
rwxs wasn't designed for security. The only arguments against ACLs that I've ever encountered are the following:
1) They're too complex.
Er, no. An ACL can be as specific or as simple as you want - rwx permissions are much more complex than a single line ACL. But when you want to be able to set some fine grained permissions, they have the ability to do that. rwxs doesn't.
2) There's no practical use for them.
Oh please. I have some word processor templates on a Linux server for a client. They document templates which contain business sensitive information. This isn't an unusual case:
* The owner should be my own account
* The software group needs read, write, and SetGID permissions
* The staff group needs read permissions
* All other accounts should not have access to these files, as they contain business sensitive information.
I could get around this in a number of ways, but they're all rather hackish.
3) NT uses them and NT sucks.
Grow up. So does Trusted Solaris, TRIX and Trusted BSD. So does Squid and (IIRC) Bind. Regardless of that though, there's no reason to discount a good idea because a company you dislike uses it. In fact, its pretty childish. Its nice to know this usually comes from trolls rather than anyone who works on major Linux based projects I discussed this with at conferences and on mailing lists.
The instruction for acquiring the distribution are unclear. Visiting www.debian.org and trying to obtain ISO images makes me go through a bloody great big questionairre which tells me that cause of my broadband connection I should download it.
There's no `fuck off and let me decide for myself' button. What if I want to install on more than one machine? For a distro that uses `its aimed at technial users' as an excuse for being damned unfriendly, this lack of control is surprising.
As for the installation , I like to think I'm experienced enough with Linux to install just about any distro without RTFM. I know how Linux works. Debian requiores me to read the manual it seems (though I really copuldn't be bothered after my last install). Things like E: for error messages, and `base system' aren't immediately obvious. Prolly my fault for not reading the documentation, but usually, I expoect the documentation to be online. No such luck with the Debian installer.
the choices ambiguous, the order illogical,
or something like that,
In my opinion, its buggy. I tried to do a floppy install on 2 machines. I got driver disk one and loaded it when asked. The installer slowly grabbed the fioles of disk. In the next part, the installer told me there were no modules in its install directory, and to go back to the loading modules bit. Okay, I might have a bad floppy disk. But no error messages. That pisses me off significantly.
There's other issues to. There shoudl be a vertical scroll bar indicating the steps for the menu go off the bottom of the screen. THis is basic GUI fundamentals. There isn't one.
I could see the problem. But, the only "problem" you describe is that it's text based.
Most users have no idea that tab and space can be used to navigate a GUI. they also have no idea what modules are and why they shoul;d be loading them.
Apt-get is great, and installing packages with dependency chains in most distros in bad enough that I would simply call it `broken'. But either using the RPM version of APT or some other new tool with fix that within the next six months. Maybe Red Hat will set up a unsupported mirror with a stack pf packages tested against its distro, and use RHN as a download mechanism (allowing paid subscribers to also get closed source apps and perhaps support for these packages).
Yes, installing packages under most distros sucks. But using Debian isn't the answer for those that want to make this easier on themselves, in my opinion anyway. Give it some time and I bet every distro will do this within a year.
He's barely even started when the first FUD hits "Linux is advertised as free", hinting at vague hidden costs lurking out there to bite the unwary.
Not hidden costs, but certainly a stack of other coists beyond licensing. Eg, quality documentation (sorry, the LDP doesn't fit imnto that category - try getting your parents to use the modem-HOWTO some time costs money. So does support you can have on site at 4:20 AM, who will respond within a half hour. Yes, there's hidden costs to Windows too - its more expensive to own as a server IMO becuase its less reliable, and downtime wates money. But the basic point - that licensing costs mean NOTHING in terms of owning a server is compeltely true.
But he misses that it's Free; that universal access to source code that you are *encouraged* to use elsewhere results in an ever-growing knowledge base - which not only lowers the barriers to (development) entry for those so inclined, but also results in ever-growing numbers of people qualified to provide ad-hoc support - which in turn works against the never-quite-voiced "support costs will kill you" FUDbit.
The quality of free support is limited. It always will be, simply because one cannot wake up a mate on the net at get them on site when the routers down.
He complains about competing desktop standards as being "confusing", but he totally misses that being able to pick, choose, and configure my desktop experience is something I WANT - it makes me more productive.
Yes, but the fact that KDE and GNOME apps aren't compatible (in the sense that drag and drop, menus, panel apps, themes, all common dialoge boxes, and other aspects aren't consistent) means a user must learn the basic of a Linux desktop twice. Competiton doesn't have to mean incompatibility, but riught now, it does.
And he also doesn't seem to grasp that this can be done per-user, so that the complexity of the desktop can be adapted to whoever logs in to the machine without disrupting the others.
You can't eliminate the complexity of File -> Open in Gimp looking and working completely differently from File _> Open on Konqueror.
"Any color you want, as long as it's black" went out with the Model T. Why is MS so intent on reviving it?
Windows has more than one toolkit - MSs MFC, Borlands VCL, and others. They just wrap[ around a common base layer called Win32. This is exactly what Linux should do.
He goes on at length about Linux's lack of a "revenue model", but is completely oblivious that Linux doesn't NEED a revenue model.
But damn stright it *wants* one. A huge reason for the explosion in Linux popularity since 98 was the formation of the Open Source Initiative and its more business friendly message. It doesn't need a softare licensing model beyind the GPL, but it does need a revenue model. X development is being funded by SuSE and VA, GNOME by Eazel and Ximian, KDE by SuSE, Webmin buy Caldera, and more. IF funding for these projects dissappeared the would continue, but at a snails pace. Most larger Open Source projects are commercial in nature.
Linux is NOT about selling software, it's about solving problems.
For Linux companies (and most businesses want commerical support arrangements), it about making money by solving support needs. You're right in that it not about selling software,
"Linux is one of our primary competitors" - no, it's NOT! "Competitor" assumes that both parties are struggling over some territory in a shared space.
The shared space is the server and to a lesser degree the desktop. Just because Linux companies and Microsft have different business models doesn't mean they're not competing.
the success or failure of Linux is not measured on financial scales.
Depends on who you speak to - IMO, its measured by market share.
MS has some very smart people working for it, no question about it. If they were totally incompetant, they wouldn't be where they are today. But the more interviews I see with MS personages, the more I realize that they don't understand the nature of of the beast they're facing. They cannot attack the problem, because they don't comprehend it.
You don't see Windows and Linux competing. IMO, I don't think you comprehend it. IYO, I guess I'm wrong.
Want to know what the dinosaurs said when they saw the first mammal? Ask a MS rep about Linux.
The dinosaurs never asked Mammals to help develop the next versions of dinosaurs (MS tried recruiting via LUGs in Australia recently_. NEither did they hire a team to study mammals in anticipation of the threat.
Choise is great, and the competition between KDE and GNOME enhances the Linux desktop.
However, the complete lack of standardization between the two hurts the Linux desktop much, much more. Sorry, what was that you said, GNOME and KDE apps work in each others desktop? Yes they do. Poorly.
I have
* Completely different file/open dialog boxes in my applications. Users have to navigate both to have any chance to survive.
* Non standardized drag and drop. xdnd is either broken or partially implemented in both toolkits, it seems. Try dragging an FTP file from the latest Konqueror to the latest GNOME desktop, for example.
* Silly companies that think the other desktop is going to go away and conveniently ignore it. Users use apps based on quality. The solution to KDE apps being not quite right in GNOME (and vice versa) is to fix the problem, not ignore it.
* 2 places to configure half my app's pook and feel.
* Different
* 2 sets of menus. Theres no functional difference between the two. Apps add themselves to both, or confuse the user by adding themselves to one. Their icons live in 2 seperate locations.
The near complete lack of standardization between KDE and GNOME hurts the Linux desktop more than competition enhances it. That said, standards can still exist within this competition, and neither desktop will go away
Sorry for the rant. But IMO its important.
Windows also has multiple toolktis, MFC and VCL. Over the years, VCL now looks like MFC. And they both wrap around a common layer called Win32.
Is there a reason why GTK and QT couldn't do the same? They're both GPLed. I want as much as possible coded into a uniform library to be loaded once, and a small GTK and QT library that live on top of it.
Remember how much 56k modems sucked before V90? X2 versus K56Flex. If your ISP used one, you couldn't use the other. Thats KDE and GNOME right now. I install Nautilus on my KDE desktop and it overwrites it with an enhanced GNOME desktop. That sucks.
For the things that need to go one way or the other, find a neutral third party (our ITU equivalent - say, Linux international) and let them decide the specs for a common base, taking the best bits from both. It worked damned well for 56k modems.
Would code it myself, but I don't write applications or code beyond bash and basic Python. By the time I learnt' it will be too late. I do Open Source documentation instead.
When did I say the the content producers weren't rushing out to sell to the new market? We were talking about SAP, if I recall correctly. This is an effort by the content producers to sell to the new market and try and stop people from stealing their works.
Just like the guys who produce thousands of video cassette copies are legislated against, and just like the guys who produce thousands of audio cassette copies are legislated againt, those who thieve music are legislated againt. In no case have all the content producers ever wholeheartedly given up control over their works. Some have make movies for free, some have made cassettes and mp3s for free. But all these new technologies has meant that the people who own the work suddenly stops owning it.
I use Napster just as much as the next guy, and most of my use is theft. Napster (like FTP) isn't illegal, despite what the RIAA say, and the RIAA are generally full of shit. But I, and most other people using Napster to steal music are most definitely committing a crime.
why was I only able to achieve interoperability through samba?
You weren't. You could have chosen you have all your Unix-like NFS servers appear in Network Neighborhood with Microsofts Windows Services for Unix.
The reason you chose Samba is probably because you prefer Open Source software, but also might be because you didn't bother doing much research - PC NFS clients and servers are quite common. So are non-Samba Unix based SMB clients and servers, though I'm not sure they exist on Linux these days.
Please moderate the above post down - I think the answer is obvious and other questions are more pertinent.
For a consumer, SAP means they can actually acquire music from big name label artists and listen to it digitally.
Currently, one can do so, but only if
a) They purchase the entire CD (rather than a few select tracks). This is legal in non DMCA states although the music industry wouldn't have you believe so.
b) You steal the track from someone else over the net. This is illegal in the opinion of lawmakers and most people, and morally wrong in my own opinion.
SAP has the support of the record companies of the artists that have the most people listening to them.
When SAP is implemented, I can get the latest Aerosmith track without having to buy the whole album, and with it being legal.
Given that Microsoft will be developing a product with Transmeta, will we be seeing Microsoft put a kinder face towards Linux?
The first part of the question is a good one. The second part is a load of tripe. Transmeta has nothing to do with Linux. Just because Linux Torvalds works there does not mean MS are about to make more Linux products. Similarly, just because Dave Taylor works there does not mean MS are about to make more first person shooters set in Hell. Or that because Paul Allen works there, Microsoft will make more...Microsoft products.
Yes, Unix permission stop an ordinary user from infecting other users on the system, and destroying the OS and other sopftware on the machine. Destroying the machine is one of the least damagiing things a virus can do.
What's would be worse would be killing all the documents on your home directory, the files which *can't* be replaced off your OS CDs with a simple reinstall. There's absolutely nothing which would stop a virus which says `cool screensaver for Linux (or Unixlike systems)' - download me to your home dir and install me for a single user! going around the net and doing said cool thing for a short amount of time before writing some of/dev/urandom to all the files in your home dir.
And, for that matter, any SetGiD directories you're sharing with other users.
You *can* reinstall postfix if a virus (which needs to run as root to destroy it) wiped it. You CAN'T reinstall your thesis if a virus (which merely needs to run as a USER) wipes it.
And trust me, from the ignorance of the above I've seen in all the posts here, your thesis *will* be wiped.
Yes, Unux permission stop an ordinary user from inmfecting other users on the system, and destroying the OS and other sopftware on the machine.
Destroying the machine is one of the least damagiing things a virus can do. What's would be worse would be killing all the documents on your home directory, the files which *can't* be replaced off your OS CDs with a simple reinstall. There's absolutely nothing which would stop a virus which says `cool screensaver for Linux (or Unixlike systems)' - download me to your home dir and install me for a single user! going around the net and doing said cool thing for a short amount of time before writing some of/dev/urandom to all the files in your home dir.
And, for that matter, any SetGiD directories you're sharing with other users.
You *can* reinstall postfix if a virus (which ran as root) wiped it. You CAN'T reinstall your thesis if a virus which ran as a USER wipes it.
Yeah, it sucks when my dad has to hex edit a binary dll. The DivX player should also not be written in C/C++ either, since my mom doesn't
know how to use a compiler.
This is a troll, but to spell it out: developers aren't end users. The post roystgnr replies to wants his parents to be able to play films, rather than hack the source code.
And apparently sucks for Win32 development for a veriety of reasons according to just about everybody I've met that used it....yes, that one.
Re:This is big news! (For losers, of course.)
on
XFree 4.0.3 Released
·
· Score: 2
Skip Win95, which is two versions behind what MS puts out now, and compare linux v. NT.
I think the reason for the comparison of WIndows 95 with modern Linux is that both are the earliest releases of their OSs with complete working Windows / Icons/Menus / Pointer type GUIs (as opposed to lightweight Window Managers, X desktops are larger and need more RAM). As a Linux user who started using computers in the days of Atari and GEM, and expects a slick GUI to do their work in, I think its a reasonable comparison to make.
full and comprehensive support for *all* Voodoo cards in XFree86 4.0x
Voodoo 3s run at about seventy percent of their Windows performance. Voodoo 4 and 5s (effectively V#s in SLI mode) only support 1 GPU per chip. 3DFX have been promising better drivers for ages now, but neither the company or the Open SOurce community had made them.
In contrast, both ATI and NVIDIA have drivers that peerform between 95 and 105% of Windows performance - check the linuxgames.com benchmarks. ATIs are Open Source, NVIDIAS are closed. However, even with the weak Australian dollar, the Australian distributor of ATI products charges twice as much as their US counterpart so Radeons are basically unaffordable. 3DFXs drivers might be Open Source, but NVIDIAs still allow me to get more fucntionality out of my card. I use Linux because its good, and it happens to be good because its Open Source. I use NVIDIAs drivers because they're good, despite being closed source.
Unless you're a monkey and only use the apps specific to a particular DE
Here here! Its nice to know I'm not the only one who picks their apps based on quality, rather than toolkit religion. Most newcomer end users don't either. I still fail to undertstand (post licensing issues) why someone with more than the tiniest amount of grunt required to load both Qt and GTK will say a `I need a X for GNOME / KDE' when the current effort (based on whichever they've decided to rally against) works great. I use Konquror for the same reason I use Gimp - they'e the best for what they do (in my own opinion). Those who chose based on toolkit religion really fo themselves and the users they `help' a disservice.
Or you could just stop being a cheap bastard and get a real computer
Agreed, but I sort of assumed the biggest application of the technology was for embedded systems.
That said, I'm more interested inseeinga GTK Win32 of MacOSX port. Why? Because QT seems to be the only widget set that runs well on all these platforms, and this means those thinkign about Win32 development now and Linux as a possibility int eh future have a better time of porting when the time comes (so no more dead-slow Corel WPO2Ks to waste your money on thinking the application actually *worked* at a reasonable speed. QT rocks because of this, and GTK should too.
Slashdot Mirror
That's some pretty bad English. But I hope you get the point.
Yes we. Do. We also remeber lan's reply - while Alan works for Red Hat, I doubt he's one to mince words, as would most people who have had some contact with him would attest.
I know who I would trust with my kernel. Can't remember the link - Google is your friend.
rpm: I think the rpm + up2date combo has all the features you need. If you think there's something we need to add, please let me know..
A centralized repository of packages (unsupported by Red Hat), organized and tested by Open Source maintainers, but providing a wide variety of pakcages.
Users will install unsupported software anyway, but the lack of a single place to find *good* Red Hat packages annoys many Red hat users, myself included (though easily finding the distro better than itys competitors). By using Red Hat's bandwidth and mirroring, it would be *the* first place to look to find packages. With an appropriate warning, up2date could also get packages from these mirrors.
Likewise, I too have been using ReiserFS with 2.2 for around a year. That said, Civilme on Mandrake's Crashtesters list acknowledges ReiserFS under 2.4 )and perhapds 2.2 - I'm not sure) to be unstable. The message is on the crashtester archives from today, but I've deleted it, so no quote...
Red Hat's gcc was severely broken upon release. But so is the current gcc 1.95. The difference is that 2.95 is currently more broken than 2.96 with all the updates (2.95 can't compile glibc 2.2 on non x86).
I don't think Red Hat should complain Mandrake are `stealing' their compiler. That's *very* weak. Mandrake are more acturately using (and legitimizing) Red Hat's compiler.
Unfortunately, a strap-on security model is always very sucky. Look at ACLs on any unix or even NT.
...wait, they all use ACLs already.
And? What makes them strap on? So Unix wasn't designed to be a secure OS. Neither was it designed to be anything more than a time sharing system.Its evolved beyong those functions and works great. The ACLs on an Unix, NT, 2K, and VMS work very well.
What you want is capabilities.
I want them too, but they don't replace ACLs. Capabilities are already in kernel 2.4.
I have some Word templates on a SMB server. I want:
* The software to be owned by the admin account
* The software group to have read and write permission, and SetGID properties
* The staff group to have
* Other users to have no access.
RWX doesn't allow me to do this in anyway which isn't vastly hackish.
This allows anyone to implement a very simple security model
ACLs are as simple or fine grained as you need them to be, and a single line ACL is quite obviously simpler than an RWX permission.
which is much easier to verify.
Bah. Same thing - again, ACLs can be less compels than RWX if that's all you need.
the dumb ass home grown authentication systems on unix. (ie Each app authenticating people in a funny way)
Damn straight. Things like Squid, and BIND, and
Capabilities are nice. They don't replace the need for ACLs, and while they work well and should be used, the RWX / Capability combination is more complex than a single tight ACL, combined perhaps with capabilities should you need them.
you cannot possibly inspect the source of every program you install.
However, they overlook one thing: If the source is out, it CAN BE audited
Hrm. Perhaps others capable of source audits can download the soruce while I download the binary? Is it just me or this completely obvious?
That sounds more like an argument for using Open Source software rather than an argument for downloading SRPMs.
rwxs wasn't designed for security. The only arguments against ACLs that I've ever encountered are the following:
1) They're too complex.
Er, no. An ACL can be as specific or as simple as you want - rwx permissions are much more complex than a single line ACL. But when you want to be able to set some fine grained permissions, they have the ability to do that. rwxs doesn't.
2) There's no practical use for them.
Oh please. I have some word processor templates on a Linux server for a client. They document templates which contain business sensitive information. This isn't an unusual case:
* The owner should be my own account
* The software group needs read, write, and SetGID permissions
* The staff group needs read permissions
* All other accounts should not have access to these files, as they contain business sensitive information.
I could get around this in a number of ways, but they're all rather hackish.
3) NT uses them and NT sucks.
Grow up. So does Trusted Solaris, TRIX and Trusted BSD. So does Squid and (IIRC) Bind. Regardless of that though, there's no reason to discount a good idea because a company you dislike uses it. In fact, its pretty childish. Its nice to know this usually comes from trolls rather than anyone who works on major Linux based projects I discussed this with at conferences and on mailing lists.
It has an older version of KDE 2.0.
Last I heard, they were using KDE 1.2 (one and a half years old) in the last beta.
It is really targeted to working with Gnome, not KDE.
That seems rather pathetic of Progeny. So now users are supposed to pick their apps based on toolkit rather than quality?
I though this type of childish bullshit ended a long time ago, once every modern distro decided to let people choose.
Unless Woody and or Progeny are using Glibc 2.96, I understand this is pretty much impossible.
If you find that the instructions are unclear,
The instruction for acquiring the distribution are unclear. Visiting www.debian.org and trying to obtain ISO images makes me go through a bloody great big questionairre which tells me that cause of my broadband connection I should download it.
There's no `fuck off and let me decide for myself' button. What if I want to install on more than one machine? For a distro that uses `its aimed at technial users' as an excuse for being damned unfriendly, this lack of control is surprising.
As for the installation , I like to think I'm experienced enough with Linux to install just about any distro without RTFM. I know how Linux works. Debian requiores me to read the manual it seems (though I really copuldn't be bothered after my last install). Things like E: for error messages, and `base system' aren't immediately obvious. Prolly my fault for not reading the documentation, but usually, I expoect the documentation to be online. No such luck with the Debian installer.
the choices ambiguous, the order illogical,
or something like that,
In my opinion, its buggy. I tried to do a floppy install on 2 machines. I got driver disk one and loaded it when asked. The installer slowly grabbed the fioles of disk. In the next part, the installer told me there were no modules in its install directory, and to go back to the loading modules bit. Okay, I might have a bad floppy disk. But no error messages. That pisses me off significantly.
There's other issues to. There shoudl be a vertical scroll bar indicating the steps for the menu go off the bottom of the screen. THis is basic GUI fundamentals. There isn't one.
I could see the problem. But, the only "problem" you describe is that it's text based.
Most users have no idea that tab and space can be used to navigate a GUI. they also have no idea what modules are and why they shoul;d be loading them.
Apt-get is great, and installing packages with dependency chains in most distros in bad enough that I would simply call it `broken'. But either using the RPM version of APT or some other new tool with fix that within the next six months. Maybe Red Hat will set up a unsupported mirror with a stack pf packages tested against its distro, and use RHN as a download mechanism (allowing paid subscribers to also get closed source apps and perhaps support for these packages).
Yes, installing packages under most distros sucks. But using Debian isn't the answer for those that want to make this easier on themselves, in my opinion anyway. Give it some time and I bet every distro will do this within a year.
He's barely even started when the first FUD hits "Linux is advertised as free", hinting at vague hidden costs lurking out there to bite the unwary.
Not hidden costs, but certainly a stack of other coists beyond licensing. Eg, quality documentation (sorry, the LDP doesn't fit imnto that category - try getting your parents to use the modem-HOWTO some time costs money. So does support you can have on site at 4:20 AM, who will respond within a half hour. Yes, there's hidden costs to Windows too - its more expensive to own as a server IMO becuase its less reliable, and downtime wates money. But the basic point - that licensing costs mean NOTHING in terms of owning a server is compeltely true.
But he misses that it's Free; that universal access to source code that you are *encouraged* to use elsewhere results in an ever-growing knowledge base - which not only lowers the barriers to (development) entry for those so inclined, but also results in ever-growing numbers of people qualified to provide ad-hoc support - which in turn works against the never-quite-voiced "support costs will kill you" FUDbit.
The quality of free support is limited. It always will be, simply because one cannot wake up a mate on the net at get them on site when the routers down.
He complains about competing desktop standards as being "confusing", but he totally misses that being able to pick, choose, and configure my desktop experience is something I WANT - it makes me more productive.
Yes, but the fact that KDE and GNOME apps aren't compatible (in the sense that drag and drop, menus, panel apps, themes, all common dialoge boxes, and other aspects aren't consistent) means a user must learn the basic of a Linux desktop twice. Competiton doesn't have to mean incompatibility, but riught now, it does.
And he also doesn't seem to grasp that this can be done per-user, so that the complexity of the desktop can be adapted to whoever logs in to the machine without disrupting the others.
You can't eliminate the complexity of File -> Open in Gimp looking and working completely differently from File _> Open on Konqueror.
"Any color you want, as long as it's black" went out with the Model T. Why is MS so intent on reviving it?
Windows has more than one toolkit - MSs MFC, Borlands VCL, and others. They just wrap[ around a common base layer called Win32. This is exactly what Linux should do.
He goes on at length about Linux's lack of a "revenue model", but is completely oblivious that Linux doesn't NEED a revenue model.
But damn stright it *wants* one. A huge reason for the explosion in Linux popularity since 98 was the formation of the Open Source Initiative and its more business friendly message. It doesn't need a softare licensing model beyind the GPL, but it does need a revenue model. X development is being funded by SuSE and VA, GNOME by Eazel and Ximian, KDE by SuSE, Webmin buy Caldera, and more. IF funding for these projects dissappeared the would continue, but at a snails pace. Most larger Open Source projects are commercial in nature.
Linux is NOT about selling software, it's about solving problems.
For Linux companies (and most businesses want commerical support arrangements), it about making money by solving support needs. You're right in that it not about selling software,
"Linux is one of our primary competitors" - no, it's NOT! "Competitor" assumes that both parties are struggling over some territory in a shared space.
The shared space is the server and to a lesser degree the desktop. Just because Linux companies and Microsft have different business models doesn't mean they're not competing.
the success or failure of Linux is not measured on financial scales.
Depends on who you speak to - IMO, its measured by market share.
MS has some very smart people working for it, no question about it. If they were totally incompetant, they wouldn't be where they are today. But the more interviews I see with MS personages, the more I realize that they don't understand the nature of of the beast they're facing. They cannot attack the problem, because they don't comprehend it.
You don't see Windows and Linux competing. IMO, I don't think you comprehend it. IYO, I guess I'm wrong.
Want to know what the dinosaurs said when they saw the first mammal? Ask a MS rep about Linux.
The dinosaurs never asked Mammals to help develop the next versions of dinosaurs (MS tried recruiting via LUGs in Australia recently_. NEither did they hire a team to study mammals in anticipation of the threat.
Choise is great, and the competition between KDE and GNOME enhances the Linux desktop.
However, the complete lack of standardization between the two hurts the Linux desktop much, much more. Sorry, what was that you said, GNOME and KDE apps work in each others desktop? Yes they do. Poorly.
I have
* Completely different file/open dialog boxes in my applications. Users have to navigate both to have any chance to survive.
* Non standardized drag and drop. xdnd is either broken or partially implemented in both toolkits, it seems. Try dragging an FTP file from the latest Konqueror to the latest GNOME desktop, for example.
* Silly companies that think the other desktop is going to go away and conveniently ignore it. Users use apps based on quality. The solution to KDE apps being not quite right in GNOME (and vice versa) is to fix the problem, not ignore it.
* 2 places to configure half my app's pook and feel.
* Different
* 2 sets of menus. Theres no functional difference between the two. Apps add themselves to both, or confuse the user by adding themselves to one. Their icons live in 2 seperate locations.
The near complete lack of standardization between KDE and GNOME hurts the Linux desktop more than competition enhances it. That said, standards can still exist within this competition, and neither desktop will go away
Sorry for the rant. But IMO its important.
Windows also has multiple toolktis, MFC and VCL. Over the years, VCL now looks like MFC. And they both wrap around a common layer called Win32.
Is there a reason why GTK and QT couldn't do the same? They're both GPLed. I want as much as possible coded into a uniform library to be loaded once, and a small GTK and QT library that live on top of it.
Remember how much 56k modems sucked before V90? X2 versus K56Flex. If your ISP used one, you couldn't use the other. Thats KDE and GNOME right now. I install Nautilus on my KDE desktop and it overwrites it with an enhanced GNOME desktop. That sucks.
For the things that need to go one way or the other, find a neutral third party (our ITU equivalent - say, Linux international) and let them decide the specs for a common base, taking the best bits from both. It worked damned well for 56k modems.
Would code it myself, but I don't write applications or code beyond bash and basic Python. By the time I learnt' it will be too late. I do Open Source documentation instead.
When did I say the the content producers weren't rushing out to sell to the new market? We were talking about SAP, if I recall correctly. This is an effort by the content producers to sell to the new market and try and stop people from stealing their works.
Just like the guys who produce thousands of video cassette copies are legislated against, and just like the guys who produce thousands of audio cassette copies are legislated againt, those who thieve music are legislated againt. In no case have all the content producers ever wholeheartedly given up control over their works. Some have make movies for free, some have made cassettes and mp3s for free. But all these new technologies has meant that the people who own the work suddenly stops owning it.
I use Napster just as much as the next guy, and most of my use is theft. Napster (like FTP) isn't illegal, despite what the RIAA say, and the RIAA are generally full of shit. But I, and most other people using Napster to steal music are most definitely committing a crime.
why was I only able to achieve interoperability through samba?
You weren't. You could have chosen you have all your Unix-like NFS servers appear in Network Neighborhood with Microsofts Windows Services for Unix.
The reason you chose Samba is probably because you prefer Open Source software, but also might be because you didn't bother doing much research - PC NFS clients and servers are quite common. So are non-Samba Unix based SMB clients and servers, though I'm not sure they exist on Linux these days.
You able to achieve interoperability
Please moderate the above post down - I think the answer is obvious and other questions are more pertinent.
For a consumer, SAP means they can actually acquire music from big name label artists and listen to it digitally.
Currently, one can do so, but only if
a) They purchase the entire CD (rather than a few select tracks). This is legal in non DMCA states although the music industry wouldn't have you believe so.
b) You steal the track from someone else over the net. This is illegal in the opinion of lawmakers and most people, and morally wrong in my own opinion.
SAP has the support of the record companies of the artists that have the most people listening to them.
When SAP is implemented, I can get the latest Aerosmith track without having to buy the whole album, and with it being legal.
Given that Microsoft will be developing a product with Transmeta, will we be seeing Microsoft put a kinder face towards Linux?
The first part of the question is a good one. The second part is a load of tripe. Transmeta has nothing to do with Linux. Just because Linux Torvalds works there does not mean MS are about to make more Linux products. Similarly, just because Dave Taylor works there does not mean MS are about to make more first person shooters set in Hell. Or that because Paul Allen works there, Microsoft will make more ...Microsoft products.
[Bah Submit as HTML button]
/dev/urandom to all the files in your home dir.
Yes, Unix permission stop an ordinary user from infecting other users on the system, and destroying the OS and other sopftware on the machine. Destroying the machine is one of the least damagiing things a virus can do.
What's would be worse would be killing all the documents on your home directory, the files which *can't* be replaced off your OS CDs with a simple reinstall. There's absolutely nothing which would stop a virus which says `cool screensaver for Linux (or Unixlike systems)' - download me to your home dir and install me for a single user! going around the net and doing said cool thing for a short amount of time before writing some of
And, for that matter, any SetGiD directories you're sharing with other users.
You *can* reinstall postfix if a virus (which needs to run as root to destroy it) wiped it. You CAN'T reinstall your thesis if a virus (which merely needs to run as a USER) wipes it.
And trust me, from the ignorance of the above I've seen in all the posts here, your thesis *will* be wiped.
Yes, Unux permission stop an ordinary user from inmfecting other users on the system, and destroying the OS and other sopftware on the machine. Destroying the machine is one of the least damagiing things a virus can do. What's would be worse would be killing all the documents on your home directory, the files which *can't* be replaced off your OS CDs with a simple reinstall. There's absolutely nothing which would stop a virus which says `cool screensaver for Linux (or Unixlike systems)' - download me to your home dir and install me for a single user! going around the net and doing said cool thing for a short amount of time before writing some of /dev/urandom to all the files in your home dir.
And, for that matter, any SetGiD directories you're sharing with other users.
You *can* reinstall postfix if a virus (which ran as root) wiped it. You CAN'T reinstall your thesis if a virus which ran as a USER wipes it.
Yeah, it sucks when my dad has to hex edit a binary dll. The DivX player should also not be written in C/C++ either, since my mom doesn't
know how to use a compiler.
This is a troll, but to spell it out: developers aren't end users. The post roystgnr replies to wants his parents to be able to play films, rather than hack the source code.
And apparently sucks for Win32 development for a veriety of reasons according to just about everybody I've met that used it....yes, that one.
Skip Win95, which is two versions behind what MS puts out now, and compare linux v. NT.
/Menus / Pointer type GUIs (as opposed to lightweight Window Managers, X desktops are larger and need more RAM). As a Linux user who started using computers in the days of Atari and GEM, and expects a slick GUI to do their work in, I think its a reasonable comparison to make.
I think the reason for the comparison of WIndows 95 with modern Linux is that both are the earliest releases of their OSs with complete working Windows / Icons
full and comprehensive support for *all* Voodoo cards in XFree86 4.0x
Voodoo 3s run at about seventy percent of their Windows performance. Voodoo 4 and 5s (effectively V#s in SLI mode) only support 1 GPU per chip. 3DFX have been promising better drivers for ages now, but neither the company or the Open SOurce community had made them.
In contrast, both ATI and NVIDIA have drivers that peerform between 95 and 105% of Windows performance - check the linuxgames.com benchmarks. ATIs are Open Source, NVIDIAS are closed. However, even with the weak Australian dollar, the Australian distributor of ATI products charges twice as much as their US counterpart so Radeons are basically unaffordable. 3DFXs drivers might be Open Source, but NVIDIAs still allow me to get more fucntionality out of my card. I use Linux because its good, and it happens to be good because its Open Source. I use NVIDIAs drivers because they're good, despite being closed source.
Do you enjoy women even though you don't have clue what the operating software looks like?
:)...
Um, actually, I quite enjoy examaining womens internals. But that's probably for to much information than you needed to know
Unless you're a monkey and only use the apps specific to a particular DE
Here here! Its nice to know I'm not the only one who picks their apps based on quality, rather than toolkit religion. Most newcomer end users don't either. I still fail to undertstand (post licensing issues) why someone with more than the tiniest amount of grunt required to load both Qt and GTK will say a `I need a X for GNOME / KDE' when the current effort (based on whichever they've decided to rally against) works great. I use Konquror for the same reason I use Gimp - they'e the best for what they do (in my own opinion). Those who chose based on toolkit religion really fo themselves and the users they `help' a disservice.
Or you could just stop being a cheap bastard and get a real computer
Agreed, but I sort of assumed the biggest application of the technology was for embedded systems.
That said, I'm more interested inseeinga GTK Win32 of MacOSX port. Why? Because QT seems to be the only widget set that runs well on all these platforms, and this means those thinkign about Win32 development now and Linux as a possibility int eh future have a better time of porting when the time comes (so no more dead-slow Corel WPO2Ks to waste your money on thinking the application actually *worked* at a reasonable speed. QT rocks because of this, and GTK should too.