At this point we really are not much better off than with distro-specific standards, especially as there are very few newbie distros (Linspire's the only one I can think of off the top of my head).
There is also Xandros, Ubuntu, Mandriva. But if you can't think of more than one, then I doubt you're familiar with the pain that 'newbies' (or oldhands who don't want to become CLI-adled ciphers) go through. Frankly I don't know why you responded at all. Maybe you, an adept Linux user, will feel shame that your 8yr old machine isn't complaint, but I and most everyone else couldn't care less. No one is revoking your right to use a custom configuration.
The CAT5 spec has constraints, but that does not prevent me from taking apart an ethernet cable to create a custom pair of headphones. Not am I prevented from starting with some of the same raw materials that go into a CAT5 in the first place. I lose only the ability to claim the headphones are a CAT5 device, which I never had to begin with.
Once you have internet access available, it seems to make much more sense to do program installation through the distribution-specific package repositories, such as gentoo's emerge and debian's apt.
Good, now you convince creative types who code games, video editors, CAD, or the next killer-app that learning to package for all those distros is a good use of their energies. Mac OS X is much newer than SuSE or RedHat, yet it has far more of these end-user apps. That is because it represents a stable platform that engenders confidence that work put into a product won't be broken or marginalized by an inconsistent environment.
That way the people who know the distribution best can make the best install, and dependencies can best be resolved.
Even on the 'newbie' distros, frequently does the installation of a GUI app result in no appropriate icon being placed in the main menu or on the desktop. There are also a relatively high number of untested details that remain broken in applications as a result of this process.
Pardon me, but I didn't say vendors should write drivers. I said they should open up their hardware and give developers a chance to write drivers.
OK... Why should ATI do anything for us? They already support one "two-percent" system, Mac OS X. And thankfully for Mac users, Apple has staff who will discriminate in their purchases based on the quality of driver implementation.
Who shows us Linux users where the best hardware and driver implementation is? RedHat? No. SuSE? No. Anyone?
No.
The Linux community is not serious about this, therefore neither is ATI.
A substantial part of the usability and power of systems like Linux comes from the fact that it can be installed in the exact way that is needed for a specific purpose.
I think you mean "control". Linux has no credibility where the usability if its desktops are concerned, and we frankly cannot even give them away (according to marketshare statistics). As for power, it can be comparable to Solaris and Winodws in their respective market segments-- *IF* you're an IT pro and know intimately the myriad component-level standards that you require; everyone else is left out on the cold, must hire YOU, or must cling to a mere brand.
Rather, we need a desktop specific distribution that addresses those needs. Linux itself won't. It serves thousands of other needs that conflict with this idea.
No one ever suggested that the Linux kernel itself must embody LSB functionality, so I think you are debating an imaginary issue. As for addressing needs, see above (don't think that users will distro-shop based on hundreds of piecemeal specs that you and I eat-and-breathe every day).
Might be nice, but also will stop development right in its tracks.
It hasn't stopped other OSes. Nor did the MultiMedia PC spec published by MS in the early 90s stop the evolution of PCs; On the contrary it helped adoption to surge ahead and ensured that audiovisual apps like Netscape and Real would just work. Standards haven't stifled the web, although lack of them did: Do you also think the acid2 test, HTML validators and the W3C are halting browser innovation?
Standardization on the desktop essentially amounts to whittling a repository down to the bare essentials for LSB compliance. In this respect, the DCCA is on the right track. How that prevents prefessionals from installing non-LSB Linuxes for very specific purposes, you haven't explained, and I'm likely to think this is merely a phantom imagined by someone who is comfortable with an elite status-quo.
If Linux/BSD wants more market share, more hardware needs to be supported out of the box. Now, that being said, I will fault chipset makers for not helping out the FOSS community by opening up their specs. They need to get their shit in gear as well.
Pardon me, but the above is just whining. Why should HW vendors write drivers when there's almost no pressure to do so?
If the community were serious about desktop HW compatability, then we would have an easy-to-use Hardware Compatability List/Database that we could visit before making purchasing decisions. The best one I've seen it as linuxquestions.org, and to put it nicely the models listed are mostly out-of-date and the GUI makes my teeth itch!
It appears the general consensus is that linux needs to be more like MacOS X. But everyone's been trying to make Linux more like Windows. No wonder linux is a flop as a desktop OS.
I have to disagree.
Refusing to define which libraries and functionality are a standard part of the OS, and then amassing huge interlinked OS/Middleware/Applications repositories to compensate for that deficiency seems a uniquely Linux phenomenon to me.
IMO when an installer looks at "uname -a" it should be able to parse "LSB 3.0" from that string if the distro is really LSB 3.0 compliant...similar to the way websites tell if they're dealing with "Mozilla 5.0". But right now they get back "GNU/Linux" and a kernel version, which mean almost nothing to someone packaging their software for distribution from their website or via CD-ROM.
This is the way Xandros' Control Center modules treat manual changes to xorg.conf and other files: A warning box pops and and you have to supply the root password in order to change those settings.
A more intelligent dynamic linking system is good.
I would settle for being able to tell which libraries are optional and which are part of the OS proper.
On a GNU-Linux system, checking the OS version gets you a kernel version number plus some distro-specific gobbledygook-- almost meaningless. It doesn't even tell you if a GUI is present never mind what 3D capabilities exist. And Linux distros don't even identify to programs whether they are LSB compliant!
The lack of committment to a standard for desktop functionality manifests as strangeness and problems to coders and end-users in many ways, both large and subtle; They get scared away from a landscape that looks like rapidly-shifting sand. As much as I like APT and Debian 'universe' as a power user, it is not an appropriate software management model for a personal computer. RPM-Yast-Urpmi-whatever doubly so.
Just like websites check for "Mozilla 5.0" compliance, basic Linux GUI/game programs should be able to check ONE dependency like "LSB 3.0" and have done with it.
Why must my package manager check to see if it has to update/install some of the most basic components on my system everytime I install a chat program or a calendar??? That should not be happening! The install process for any user application should keep its nose the HELL out of my system folders. I don't care if its APT doing the legwork. Except for system updates intended as such, leave my system libraries alone dammit.
Debian is more of a meta-distribution: A Linux that distributors can mould into something made for their end-users. By itself, Debian is pretty raw.
I recommend Xandros Linux if you're used to Windows. It consistently gets the highest marks from PC Mag, PC World, Cnet and others. And its based on Debian. I've used it for years and it has possibly the best hardware detection available, along with simple and complete integration with Windows Domains, VPN and home folder encryption setup, and a Control Center that actually makes sense (to me).
You can also buy Xandros with Crossover Office pre-installed. I haven't needed to run Windows programs for some time, but DVD Shrink runs well so I use that as a more convenient alternative to DVD::Rip.
SuSE (with KDE) and Linspire are two others that you may feel comfortable with.
Finally, do remember this: Once you leave Windows, no longer can you take hardware compatability for granted. Even if you get a bit lucky (as happened with me) and your chosen Linux distro works with 100% of the equipment you already have, careless purchasing could easily land you with a device that is driverless or a configuration nightmare. Check on the net (particularly with an HCL) before you buy.
I would agree. Which is why the static page dumps I have gotten from the download site will not be going to the printer. Instead they are being processed for eventual use on a PDA or notebook (sans discussions, and graphics for now). I am not sure yet, but will probably choose either wik2dict (creates a compressed dict database) or Plucker to generate the portable Wikipedia.
A version is available for download in the proprietary TomeRaider format, which can be read on Windows and Palm.
I don't like seeing these summaries and being left to think that my OS X and Linux systems could be compromised, then having to scour the linked article just to be sure.
This is becoming a common occurance on Slashdot: Articles about viruses and other Windows exploits are posted with no hint as to their platform-specific nature. "Systems" are attacked. Is it so difficult to write "Windows systems"?? And then of course, when vulnerabilities of non-MS stuff like Linux are reported, the platform in question is Big News. So on top of vagueness WRT Windows, I get bias. Its like reading the front page of ZDnet.
Please just mention the friggin platform, thank you.
Seems like part of the author's message is pleading for a return to consumers who cannot easily defend themselves against against corporate swindliers and incompetence... and right-wingers. Are we surprised?
Yes, paid shills are an odious problem. But why not simply call them that? Could the author himself be a shill? One has to wonder.
No target is too mighty, or too obscure,
Its great being both mighty and obscure, isn't it? Rich crooks are under attack by concerned citizens and consumers; Now that small-fry crooks have joined the fray, the blogosphere must be litigated into oblivion??
He's apparantly realized that his only hope of keeping his job indefinitely is to convince his bosses that having one's arguments meticulously dissected by flaw-finding weblogs is a meaningless annoyance...
Or it's a way to drum-up pagehits from the blogosphere.
I'm sure the sensationalism helps Forbes' ad revenue, in the short-term at least.
I think this suggestion has been made before, and it may a good idea. However the initial response mentioned that the code wasn't owned by the project or the university, but is on-loan from the UK's Met Office. The CPDN-distributed models are even named "Hadley", the Met's climate research center that has developed these models over the course of decades.
You seem prepared to make a well-reasoned argument for open sourcing the models, so you may want to start at the discussion board. Perhaps project members can offer insight on dealing with the Met Office.
I think it is becoming apparent that Excel uses a file-loading scheme that does not scan the whole file, nor load many portions of the file into memory until they are used.
With Calc and Gnumeric, there seems to be an assumption that the whole spreadsheet needs to be loaded into RAM before using it.
OTOH, I wonder how Corel Quattro would perform. It's another spreadsheet with a pedigree that goes back to the memory-constrained era, and competed directly with Excel on the basis of performance. I have a Linux version around here somewhere...
Another comparison: AbiWord vs Word vs Writer 2.0r
on
OpenOffice Bloated?
·
· Score: 1
Someone should mod the parent way up.
See http://www.geocities.com/typopl/bug5291.html#2005O ct3 for a speed and memory comparison. OOWriter is almost as fast as MS Word. But both AbiWord and OOWriter are getting slower with each new version.
That's what I've noticed as well. KDE has been improving its performance.
However OOo doesn't really use KDE, or QT or GTK. It has its own widget set and abstraction layer, which means its an OS within an office suite. Oddly enough, it still cannot be easily ported to a platform like Mac OS X without first having to install X11.
Do not forget there was the Atari 5200 Supersystem (and the 7800 later on). The 5200 was basically an Atari 800 system, with some decent graphics ability, inside a big slab of shiny black acrylic!
Slashdot Mirror
First, you'd be interested to see a response I gave to another poster here:9 42831
http://slashdot.org/comments.pl?sid=167131&cid=13
At this point we really are not much better off than with distro-specific standards, especially as there are very few newbie distros (Linspire's the only one I can think of off the top of my head).
There is also Xandros, Ubuntu, Mandriva. But if you can't think of more than one, then I doubt you're familiar with the pain that 'newbies' (or oldhands who don't want to become CLI-adled ciphers) go through. Frankly I don't know why you responded at all. Maybe you, an adept Linux user, will feel shame that your 8yr old machine isn't complaint, but I and most everyone else couldn't care less. No one is revoking your right to use a custom configuration.
The CAT5 spec has constraints, but that does not prevent me from taking apart an ethernet cable to create a custom pair of headphones. Not am I prevented from starting with some of the same raw materials that go into a CAT5 in the first place. I lose only the ability to claim the headphones are a CAT5 device, which I never had to begin with.
Once you have internet access available, it seems to make much more sense to do program installation through the distribution-specific package repositories, such as gentoo's emerge and debian's apt.
Good, now you convince creative types who code games, video editors, CAD, or the next killer-app that learning to package for all those distros is a good use of their energies. Mac OS X is much newer than SuSE or RedHat, yet it has far more of these end-user apps. That is because it represents a stable platform that engenders confidence that work put into a product won't be broken or marginalized by an inconsistent environment.
That way the people who know the distribution best can make the best install, and dependencies can best be resolved.
Even on the 'newbie' distros, frequently does the installation of a GUI app result in no appropriate icon being placed in the main menu or on the desktop. There are also a relatively high number of untested details that remain broken in applications as a result of this process.
Pardon me, but I didn't say vendors should write drivers. I said they should open up their hardware and give developers a chance to write drivers.
OK... Why should ATI do anything for us? They already support one "two-percent" system, Mac OS X. And thankfully for Mac users, Apple has staff who will discriminate in their purchases based on the quality of driver implementation.
Who shows us Linux users where the best hardware and driver implementation is? RedHat? No. SuSE? No. Anyone?
No.
The Linux community is not serious about this, therefore neither is ATI.
A substantial part of the usability and power of systems like Linux comes from the fact that it can be installed in the exact way that is needed for a specific purpose.
I think you mean "control". Linux has no credibility where the usability if its desktops are concerned, and we frankly cannot even give them away (according to marketshare statistics). As for power, it can be comparable to Solaris and Winodws in their respective market segments-- *IF* you're an IT pro and know intimately the myriad component-level standards that you require; everyone else is left out on the cold, must hire YOU, or must cling to a mere brand.
Rather, we need a desktop specific distribution that addresses those needs. Linux itself won't. It serves thousands of other needs that conflict with this idea.
No one ever suggested that the Linux kernel itself must embody LSB functionality, so I think you are debating an imaginary issue. As for addressing needs, see above (don't think that users will distro-shop based on hundreds of piecemeal specs that you and I eat-and-breathe every day).
Might be nice, but also will stop development right in its tracks.
It hasn't stopped other OSes. Nor did the MultiMedia PC spec published by MS in the early 90s stop the evolution of PCs; On the contrary it helped adoption to surge ahead and ensured that audiovisual apps like Netscape and Real would just work. Standards haven't stifled the web, although lack of them did: Do you also think the acid2 test, HTML validators and the W3C are halting browser innovation?
Standardization on the desktop essentially amounts to whittling a repository down to the bare essentials for LSB compliance. In this respect, the DCCA is on the right track. How that prevents prefessionals from installing non-LSB Linuxes for very specific purposes, you haven't explained, and I'm likely to think this is merely a phantom imagined by someone who is comfortable with an elite status-quo.
If Linux/BSD wants more market share, more hardware needs to be supported out of the box. Now, that being said, I will fault chipset makers for not helping out the FOSS community by opening up their specs. They need to get their shit in gear as well.
Pardon me, but the above is just whining. Why should HW vendors write drivers when there's almost no pressure to do so?
If the community were serious about desktop HW compatability, then we would have an easy-to-use Hardware Compatability List/Database that we could visit before making purchasing decisions. The best one I've seen it as linuxquestions.org, and to put it nicely the models listed are mostly out-of-date and the GUI makes my teeth itch!
It appears the general consensus is that linux needs to be more like MacOS X. But everyone's been trying to make Linux more like Windows. No wonder linux is a flop as a desktop OS.
...similar to the way websites tell if they're dealing with "Mozilla 5.0". But right now they get back "GNU/Linux" and a kernel version, which mean almost nothing to someone packaging their software for distribution from their website or via CD-ROM.
I have to disagree.
Refusing to define which libraries and functionality are a standard part of the OS, and then amassing huge interlinked OS/Middleware/Applications repositories to compensate for that deficiency seems a uniquely Linux phenomenon to me.
IMO when an installer looks at "uname -a" it should be able to parse "LSB 3.0" from that string if the distro is really LSB 3.0 compliant
This is the way Xandros' Control Center modules treat manual changes to xorg.conf and other files: A warning box pops and and you have to supply the root password in order to change those settings.
A more intelligent dynamic linking system is good.
I would settle for being able to tell which libraries are optional and which are part of the OS proper.
On a GNU-Linux system, checking the OS version gets you a kernel version number plus some distro-specific gobbledygook-- almost meaningless. It doesn't even tell you if a GUI is present never mind what 3D capabilities exist. And Linux distros don't even identify to programs whether they are LSB compliant!
The lack of committment to a standard for desktop functionality manifests as strangeness and problems to coders and end-users in many ways, both large and subtle; They get scared away from a landscape that looks like rapidly-shifting sand. As much as I like APT and Debian 'universe' as a power user, it is not an appropriate software management model for a personal computer. RPM-Yast-Urpmi-whatever doubly so.
Just like websites check for "Mozilla 5.0" compliance, basic Linux GUI/game programs should be able to check ONE dependency like "LSB 3.0" and have done with it.
Why must my package manager check to see if it has to update/install some of the most basic components on my system everytime I install a chat program or a calendar??? That should not be happening! The install process for any user application should keep its nose the HELL out of my system folders. I don't care if its APT doing the legwork. Except for system updates intended as such, leave my system libraries alone dammit.
Look at it? No... I think I'd go blind if I did. :-)
...like extra chrome, a built-in DVD player, mobile office, and crummy MPG.
:-D
Sorry, did I say "creampuff"? I meant "American".
Try the free version of Kivio.
Most people who have tried both prefer it much to Dia.
OTOH, if this came down as a Microsoft service pack or an OS upgrade, then the rootkit would no longer be considered a rootkit.
Debian is more of a meta-distribution: A Linux that distributors can mould into something made for their end-users. By itself, Debian is pretty raw.
I recommend Xandros Linux if you're used to Windows. It consistently gets the highest marks from PC Mag, PC World, Cnet and others. And its based on Debian. I've used it for years and it has possibly the best hardware detection available, along with simple and complete integration with Windows Domains, VPN and home folder encryption setup, and a Control Center that actually makes sense (to me).
You can also buy Xandros with Crossover Office pre-installed. I haven't needed to run Windows programs for some time, but DVD Shrink runs well so I use that as a more convenient alternative to DVD::Rip.
SuSE (with KDE) and Linspire are two others that you may feel comfortable with.
Finally, do remember this: Once you leave Windows, no longer can you take hardware compatability for granted. Even if you get a bit lucky (as happened with me) and your chosen Linux distro works with 100% of the equipment you already have, careless purchasing could easily land you with a device that is driverless or a configuration nightmare. Check on the net (particularly with an HCL) before you buy.
I would agree. Which is why the static page dumps I have gotten from the download site will not be going to the printer. Instead they are being processed for eventual use on a PDA or notebook (sans discussions, and graphics for now). I am not sure yet, but will probably choose either wik2dict (creates a compressed dict database) or Plucker to generate the portable Wikipedia.
A version is available for download in the proprietary TomeRaider format, which can be read on Windows and Palm.
I don't like seeing these summaries and being left to think that my OS X and Linux systems could be compromised, then having to scour the linked article just to be sure.
This is becoming a common occurance on Slashdot: Articles about viruses and other Windows exploits are posted with no hint as to their platform-specific nature. "Systems" are attacked. Is it so difficult to write "Windows systems"?? And then of course, when vulnerabilities of non-MS stuff like Linux are reported, the platform in question is Big News. So on top of vagueness WRT Windows, I get bias. Its like reading the front page of ZDnet.
Please just mention the friggin platform, thank you.
FFS, *please* stop saying that awful marketing speak buzzword BLOGOSPHERE. Gah.
;-)
OK, if you stop typing it in CAPS.
Yes, paid shills are an odious problem. But why not simply call them that? Could the author himself be a shill? One has to wonder.
Its great being both mighty and obscure, isn't it? Rich crooks are under attack by concerned citizens and consumers; Now that small-fry crooks have joined the fray, the blogosphere must be litigated into oblivion??
He's apparantly realized that his only hope of keeping his job indefinitely is to convince his bosses that having one's arguments meticulously dissected by flaw-finding weblogs is a meaningless annoyance...
Or it's a way to drum-up pagehits from the blogosphere.
I'm sure the sensationalism helps Forbes' ad revenue, in the short-term at least.
But do not inquire as to why. Because I said so.
Now run along and play with your HummerDinger.
I think this suggestion has been made before, and it may a good idea. However the initial response mentioned that the code wasn't owned by the project or the university, but is on-loan from the UK's Met Office. The CPDN-distributed models are even named "Hadley", the Met's climate research center that has developed these models over the course of decades.
You seem prepared to make a well-reasoned argument for open sourcing the models, so you may want to start at the discussion board. Perhaps project members can offer insight on dealing with the Met Office.
Let's see, the entire multi-year experiment at ClimatePrediction.net could be completed in about... oh 13 days. :^)
;) ).
(Not really; I made that up. But if you're curious about how much crunching power we have on tap, visit the project website
I think it is becoming apparent that Excel uses a file-loading scheme that does not scan the whole file, nor load many portions of the file into memory until they are used.
With Calc and Gnumeric, there seems to be an assumption that the whole spreadsheet needs to be loaded into RAM before using it.
OTOH, I wonder how Corel Quattro would perform. It's another spreadsheet with a pedigree that goes back to the memory-constrained era, and competed directly with Excel on the basis of performance. I have a Linux version around here somewhere...
That's what I've noticed as well. KDE has been improving its performance.
However OOo doesn't really use KDE, or QT or GTK. It has its own widget set and abstraction layer, which means its an OS within an office suite. Oddly enough, it still cannot be easily ported to a platform like Mac OS X without first having to install X11.
Do not forget there was the Atari 5200 Supersystem (and the 7800 later on). The 5200 was basically an Atari 800 system, with some decent graphics ability, inside a big slab of shiny black acrylic!
(Awful game controllers though...)
Interesting comparison of Colecovision and Atari 5200.