I have a question related to the original poster's question:
I very humbly think that the Iraq issue should be the starting point for the green party to finally become a strong, third US party. Your party was clearly opposed to the US/British invasion on Irak, while the Democrats were somehow shy in their criticism before the military actions started, and explicitly supported the so-called war afterwards. Shouldn't you be making it more clear that the two big parties are essentially the same, and that you represent a fundamentally different, actually progressive perspective ? Are you doing it ? What are your thoughts in this regard ?
I gotta try this weed from Anonymous Smoker really;-) (no ofense, just kidding)
1. Easy installation of any Linux software. Don't give me RPM-hell, dependency hell, command-line compiling, proprietary click-n-run depositories, or any other excuses.
Use a front end to RPM or.deb - No dependencies hell, no pain at all. In mandrake, Control Center -> Software Management. I just installed skype. It was even easier. I clicked on the link for the rpm on their webpage, which in turn launched the mandrake software installer. I was asked if I wanted to install the package, I said yes, and that was that.
2. Simple, centralized, user-friendly control panels for *everything*, with smart defaults. Why does Mandrake, arguably the most desktop-ready distro, still have printer settings in PrinterDrake, printer settings in the KDE control center, and another panel full of printer settings in the KDE menu?
Use Yast in SUSE or Control Center in Mandrake. Printer Drake is in the Hardware -> Printers sections. You don't even need to know it exists.
Oh, and Mandrake has a separate package for the duplicate options between Mandrake Control Center and KDE's one. It's called "kdebase-kcontrol-data". It is not installed by default AFAIK.
3. Better support for basic peripherals, like printers and scanners. It's tough shopping for printers at Staples when you know that nothing on the shelf is likely to work.
True, that's why I vote with my wallet. I bought a lexmark printer and got screwed. Wouldn't even print with generic drivers. I went to linuxprinting.org and I found out that Lexmark doesn't cooperate with the linux printing gurus. I returned the printer and got a HP PSC 1210 multifunction (because the same webpage advises that HP support is excellent in general). I pluged in, powered on, launched the Mandrake Control Center, it autodetected it, it installed all the needed software, and that was that.
IMHO, one reason why there are so many duplicate stories in Slashdot is that the search engine is pretty bad. Or is is just me ? I could hardly ever find what I was looking for with the/. search form. I am not bitching (no pun intended), I just wonder if this could be improved...
To me, one of the most exciting developments in KDE 3.3 is the addition of both PyKDE and PyQt to CVS. Hopefully, this will boster the use of these bindings. If you haven't, give them a try.
I humbly think that KDE + KDevelop (or Qt + Designer) give a beautifull Rapid Development tool. Python fits very well with the Object Oriented KDE API. And most of the heavy work is done by Qt anyways, so I would expect that many. many usefull aplications could be written with PyKDE and PyQT, now that they are officially part of the family;-)
While we're all looking at MS, Apple, IBM, Adobe etc and going "tsk! omg!" as they acquire yet another silly patent, they're not necessarily the ones who're going to be a pain in the butt about it, it's the smaller rogues like Forgent, or Acacia etc.
Interesting point. But think of it the other way around: huge companies have about unlimited funds to hold a trial forever if they want to. And the smaller ones have to give up at some point. Big companies can put the small ones belly up any time in the current US (and many other countries) system.
To me the real point here, one that is really hurting the US society, is the industry of Justice. The industry of, instead of producing something useful, being able to base your business on hiring lawyers and ripping money off someone else. This situation brings all sort of pathological social and economical behaviors. It is insane and in the long term self-destructive.
Could anyone on the know tell the rest of us why is it that digital cameras choose JPG as the compressed format instead of PNG ? Looks to me like PNG would typically provide better image resolution for a similar filesize. Maybe I am wrong. And also, is GIF an option now that it is free again ?
Earlier cluster supercomputers at the NCSA used multiple images of the Linux operating system -- one for each node -- along with dedicated memory allocations for each CPU. What makes this system more powerful for researchers is that all of the memory will be available for the applications and calculations, helping to speed and refine the work being done, Pennington said.
"The users get one memory image they have to deal with," he said. "This makes programming much easier, and we expect it to give better performance as well."
So, anyone has any insights as to why/how this matters for the programmers ?
Does this mean that the applications running on the "old" clusters, presumably using some flavor of MPI to communicate between nodes, will have to be ported somehow to become multithreaded applications ? Or maybe they will still run using MPI on the big shared memory pool, and each process will be sent to the appropriate node by the OS on demand ? Thanks !
I meant 1/2 hour the whole installation from the Mandrake 10 CDs. The cable connection was just a matter of starting DHCP on the network card interface:-)
Typical desktop users are not willing and not interested in reading a webpage to get the basics of their OS working. We can't keep telling them "Read the Freaking Manual".
I recently installed Mandrake in a White Box Machine that I got on the mail with no OS. It took less than half an hour to even get my cable connection working. Most everything was auto configured.That I call a desktop distro.
I have a sweet spot for Slack 'cause I started my Linux experience with Slackware almost 10 years ago. And I think there is a certain market niche for Slack, mostly in the server side of things. Some people want a mostly vanilla distro with binary packaging. These people want a stable, standard distro that you can install if you know the business.
That's that. But pointing to a webpage that tells you how to configure things that sould be automatically configured, does not make a distro newbie or desktop friendly. I am not critizing the original webpage, I am critical of the way this news bit has been posted to/.
Then, why do they buy a company that produces
windows only software
? Shouldn't they go for multiplatform / opensource software ? Isn't this Sun's and IBM's and Novell's strategy ?
Besides webmin, the CUPS web interface and some other nice GUIs for generic, cross-distro sysadmin, I would pick the big commercial three,
Fedora,
Suse
and
Mandrake
and showcase their admin tools. They have gotten pretty friendly these days (not so sure about Fedora though), and you can pretty much configure everything from the same "control center" or whatever the name. I think it is important to show people how easy and graphical GNU/Linux has gotten. I still hear people telling me "ah, Linux, yes, it is like DOS, you type everything in a console right ?"
IMHO, Desktop Linux IS already there, with the caveat of doing a bit of research before buying peripherals (most hardware is supported but not all of it is).
From a Business point of view, the Giants (IBM, Sun, Dell) seem to be sharpening their teeth to get ready to provide Linux Desktop solutions for big corporations and Government departments worldwide. IBM and Sun have actually been doing it, and actually deploying solutions, and Novell just jumped in. Intel IS lagging behind IMHO. Not that I care:-)
Thanks for the feedback. Yes, I RTFA, but I probably wasn't clear in my post. What I meant to say is that the two projects mentioned in the FA relate to two different, although related issues. One is about the front end, another for the back end. I meant to clarify this for people who are not into the details of RPM.
And there is the third issue of using URPMI as a front end for intelligent source RPM compiling. Thanks for the feedback on this last point too. It's great to know that there are folks working on that, and I hope the project is successfull:-)
Having URPMI under Fedora will give Fedora users another Front End to RPM. Is this good for them ? Most likely yes.
The other issue, the one about being able to write source RPMS (.src.rpm files) that work in several distros, has to do with the different distros standarizing on the RPM macros, file naming conventions, version schemes and so on. It is all in the Back End. It would be great of course, and it would save a lot of duplicated efforts for the different distributions.
What would be even more interesting as a further step, would be to be able to massively build source RPMS from the RPM front end (be it URPMI or whatever), a la gentoo. This is, to provide a systematic way to get the RPM front end to download and build all the SRC RPMS you need to install the package you need (assuming you can't install these dependendencies from binaries). This would help users compile packages from other distros or from third parties in a painless way.
They seem to be taking the microsoft route. Add features before making it faster and more stable. It's ironic where this is headed.
I have to disagree with almost everything you said:
One of my favorite golden rules for software development is: first make it work, then optimize. NEVER optimize your first shot.
The Microsoft way would be: first make it pretty and somehow useable, release it, then we see about bugs et al.
The KDE way is: alphas are for new features, betas for bug fixing. Released code is stable, not necessarily optimized for speed.
There is a very well known issue with C++ speed in Linux. The overhead for the use of Objects is too high, and there is the object prelink project and other work to find a general solution.
One of the the general features of the upcoming qt vetsion (the library behind KDE) is a general speed up.
One of the main focuses in the latest KDE version has been speed, and it really shows. KDE used to be several times slower to load than GNOME. Now they are comparable. Moreover, if you have lots of memory, after your first login, lots of things will stay on cache, and following logins are much faster.
Sorry for the long post, but there is really no basis for the parent's comparison between MS windows and KDE...
From the Post:
Unfortunately, to install any of these versions without wiping out most Windows installations, you'll need to buy a third-party program to partition your hard drive.
Which is, of course, wrong. Using parted and a graphical frontend like QTparted, you can easily resize partitions. The last SUSE install I helped out with had a partition resizer during the install. Did they take this feature out?
How true. Moreover, go ahead and try to install Windows in a machine with other OS's and tell me how it did. Why is it that it is ok for MS to monopolize your computer (no pun intended), and
it is also ok to bash the Linux distros for presumably not being perfect at trying to interoperate with the OS which is in fact being nasty about interoperability ? Geeess !
A quick an easy option, not to be overlooked, is to put all your home network behind an access point/router. Right now you can get a netgear in Amazon for USD 16 after rebate. This is just an example, the bottomline is: they are affordable, they provide a firewall, and they usually provide a web-interface to configure the firewall and the router in general from any OS (I use Linux exclusively). A quick way to filter content, based on YOUR criteria.
I'd like to think so. However, they chose a BSD license apparently. In my humble opinion, if they want to open Java, they should go GPL. This way, at least they know that nobody (like MS) will use the code they are opening after years of development, turn it into a proprietary, incompatible beast and release it and even perhaps make money out of Sun's effort, without contributing anything back.
In fact, OpenOffice uses a dual licensing scheme that includes the GPL right ?
Bottomline: come on Sun, opensourcing Java will be the final step to make it a Standard (it is almost there, how many IT Jobs requre Java these days ?)
eventually the only PC people will have is a hand-held one
This is about right. I am not sure if people will only use handheld PCs, but there certainly is a golden mountain waiting out there for whoever comes up with something affordable inbetween a PDA and a Sub-Notebook. That means:
TV out for screen presentation
Small and Light (PDA'like)
PDA functionality (PIM software)
Laptop functionality (Office suite, Internet apps)
Wi-Fi
USB host/client
Built-in, perhaps foldable full size keyboard
Decent (800x600) screen resolution
It doesn't need to fit in a pocket, but it should be at least close to that. The Zaurus is the closest thing, and it even runs Linux (which for me is a must). But it lacks a good set of PIM applications AFAIK, and a fullsize keyboard, TV-out and a little more screen resolution perhaps.
It would be great if they would be able to not only strip mozilla down in their embedded version, but also backport changes to the main CVS trunk to help mozilla itself reduce its code-sice and memory/resource foot print.
Think Opera, it is a nice, fast web processor weighting about 5Mb when statically compiled (for Linux). And it also runs embedded. Maybe the folks at Opera managed to capitalize from the parallel development of an embedded and a desktop version of the same browser... of course, they benefit from using Qt/QtEmbedded too I guess!
The primary focus of Minimo to date has been system with ~32-64 MB of RAM, running Linux and using the GTK toolkit. We have been investigating other platforms and toolkits.
In other words, initially it is not intended to be cross-platform, but it might happen in future.
Is it me ? I have the feeling that if Novell keeps opening up (as in GPL) the core of SUSE, then the (perhaps) most polished distro (according to reviews from people I respect) will become extremely popular.
The big picture for commercial distros a couple years back was:
RedHat, open but not very user friendly.
Mandrake, even more open (in its development cycle), more polished and user friendly, not as stable.
SuSE, polished, stable, professional, user friendly, but with closed source bits preventing
widespread use.
With RedHat going Fedora (and resigning its mindshare), it wasn't clear what the new scenario sould be. Mandrake was in my mind the great candidate to be the king of the overall distro (from freelading and home users through enterprise solutions), assuming they released more stable corporate versions.
Mandrake did its homework, and they announced a new development strategy with a community release and a later, more stable official release. They probably should add a slower (once a year) corporate option
But now Novell buys SUSE, opens it up and kicks the hell out of the Chess board. RedHat backs up and announces their (late) return to the desktop. Things are getting hot my friends:-)
Why, oh why, can't there be an official release of the kernel that describes the changes in plain language instead of coder-speak? Look, I know what kind of people it takes to hack the kernel - brilliant people. They think like coders. But when new versions of the kernel are released, why can't there be a summary of what is new/changed? In simple terms.
Not everyone who uses Linux is a kernel hacker, especially nowadays. And yes, there are sites out there that give rundowns of what has changed. But wouldn't it be nice to have an *official* release
statement that outlines the changes made? It seems logical to me that the people managing the changes would be able to articulate this the best. I think it would go a long way in making Linux seem a bit more mature.
The audience of the kernel source, I disagree with you, is hard-core people who compile from source for some reason, and developers who customize it.
End users should follow a Linux distribution, and it should be responsibility of the distribution they choose to give them a list of relevant features of a new kernel when they include it in the distro.
I don't think that the people who can help making Linux more mature really need a simpler changelog. Also, note that the new changelog is automatically generated by the source control software they are using (bitkeeper). Kernel devs can spend more time coding and less time typing boring changelogs... how can this be bad ?
Slashdot Mirror
I very humbly think that the Iraq issue should be the starting point for the green party to finally become a strong, third US party. Your party was clearly opposed to the US/British invasion on Irak, while the Democrats were somehow shy in their criticism before the military actions started, and explicitly supported the so-called war afterwards. Shouldn't you be making it more clear that the two big parties are essentially the same, and that you represent a fundamentally different, actually progressive perspective ? Are you doing it ? What are your thoughts in this regard ?
There is a nice whitepaper on the future of reiserfs and other filesystems for Mac* and Win* OS. The question is: at what level do you need to implement efficient search capabilities ? Filesystem ? Userspace ? Both ?
IMHO, one reason why there are so many duplicate stories in Slashdot is that the search engine is pretty bad. Or is is just me ? I could hardly ever find what I was looking for with the /. search form. I am not bitching (no pun intended), I just wonder if this could be improved ...
I humbly think that KDE + KDevelop (or Qt + Designer) give a beautifull Rapid Development tool. Python fits very well with the Object Oriented KDE API. And most of the heavy work is done by Qt anyways, so I would expect that many. many usefull aplications could be written with PyKDE and PyQT, now that they are officially part of the family ;-)
Kudos and Thank You to everyone involved.
-- Don Inodoro
To me the real point here, one that is really hurting the US society, is the industry of Justice. The industry of, instead of producing something useful, being able to base your business on hiring lawyers and ripping money off someone else. This situation brings all sort of pathological social and economical behaviors. It is insane and in the long term self-destructive.
Could anyone on the know tell the rest of us why is it that digital cameras choose JPG as the compressed format instead of PNG ? Looks to me like PNG would typically provide better image resolution for a similar filesize. Maybe I am wrong. And also, is GIF an option now that it is free again ?
I meant 1/2 hour the whole installation from the Mandrake 10 CDs. The cable connection was just a matter of starting DHCP on the network card interface :-)
I recently installed Mandrake in a White Box Machine that I got on the mail with no OS. It took less than half an hour to even get my cable connection working. Most everything was auto configured.That I call a desktop distro.
I have a sweet spot for Slack 'cause I started my Linux experience with Slackware almost 10 years ago. And I think there is a certain market niche for Slack, mostly in the server side of things. Some people want a mostly vanilla distro with binary packaging. These people want a stable, standard distro that you can install if you know the business.
That's that. But pointing to a webpage that tells you how to configure things that sould be automatically configured, does not make a distro newbie or desktop friendly. I am not critizing the original webpage, I am critical of the way this news bit has been posted to /.
Then, why do they buy a company that produces windows only software ? Shouldn't they go for multiplatform / opensource software ? Isn't this Sun's and IBM's and Novell's strategy ?
Besides webmin, the CUPS web interface and some other nice GUIs for generic, cross-distro sysadmin, I would pick the big commercial three, Fedora, Suse and Mandrake and showcase their admin tools. They have gotten pretty friendly these days (not so sure about Fedora though), and you can pretty much configure everything from the same "control center" or whatever the name. I think it is important to show people how easy and graphical GNU/Linux has gotten. I still hear people telling me "ah, Linux, yes, it is like DOS, you type everything in a console right ?"
From a Business point of view, the Giants (IBM, Sun, Dell) seem to be sharpening their teeth to get ready to provide Linux Desktop solutions for big corporations and Government departments worldwide. IBM and Sun have actually been doing it, and actually deploying solutions, and Novell just jumped in. Intel IS lagging behind IMHO. Not that I care :-)
And there is the third issue of using URPMI as a front end for intelligent source RPM compiling. Thanks for the feedback on this last point too. It's great to know that there are folks working on that, and I hope the project is successfull :-)
Cheers!
The other issue, the one about being able to write source RPMS (.src.rpm files) that work in several distros, has to do with the different distros standarizing on the RPM macros, file naming conventions, version schemes and so on. It is all in the Back End. It would be great of course, and it would save a lot of duplicated efforts for the different distributions.
What would be even more interesting as a further step, would be to be able to massively build source RPMS from the RPM front end (be it URPMI or whatever), a la gentoo. This is, to provide a systematic way to get the RPM front end to download and build all the SRC RPMS you need to install the package you need (assuming you can't install these dependendencies from binaries). This would help users compile packages from other distros or from third parties in a painless way.
I have to disagree with almost everything you said:
-
One of my favorite golden rules for software development is: first make it work, then optimize. NEVER optimize your first shot.
-
The Microsoft way would be: first make it pretty and somehow useable, release it, then we see about bugs et al.
-
The KDE way is: alphas are for new features, betas for bug fixing. Released code is stable, not necessarily optimized for speed.
-
There is a very well known issue with C++ speed in Linux. The overhead for the use of Objects is too high, and there is the object prelink project and other work to find a general solution.
-
One of the the general features of the upcoming qt vetsion (the library behind KDE) is a general speed up.
-
One of the main focuses in the latest KDE version has been speed, and it really shows. KDE used to be several times slower to load than GNOME. Now they are comparable. Moreover, if you have lots of memory, after your first login, lots of things will stay on cache, and following logins are much faster.
Sorry for the long post, but there is really no basis for the parent's comparison between MS windows and KDEA quick an easy option, not to be overlooked, is to put all your home network behind an access point/router. Right now you can get a netgear in Amazon for USD 16 after rebate. This is just an example, the bottomline is: they are affordable, they provide a firewall, and they usually provide a web-interface to configure the firewall and the router in general from any OS (I use Linux exclusively). A quick way to filter content, based on YOUR criteria.
In fact, OpenOffice uses a dual licensing scheme that includes the GPL right ?
Bottomline: come on Sun, opensourcing Java will be the final step to make it a Standard (it is almost there, how many IT Jobs requre Java these days ?)
-
TV out for screen presentation
-
Small and Light (PDA'like)
-
PDA functionality (PIM software)
-
Laptop functionality (Office suite, Internet apps)
-
Wi-Fi
-
USB host/client
-
Built-in, perhaps foldable full size keyboard
-
Decent (800x600) screen resolution
It doesn't need to fit in a pocket, but it should be at least close to that. The Zaurus is the closest thing, and it even runs Linux (which for me is a must). But it lacks a good set of PIM applications AFAIK, and a fullsize keyboard, TV-out and a little more screen resolution perhaps.Now, who is gonna come up with this beauty ?
Think Opera, it is a nice, fast web processor weighting about 5Mb when statically compiled (for Linux). And it also runs embedded. Maybe the folks at Opera managed to capitalize from the parallel development of an embedded and a desktop version of the same browser ... of course, they benefit from using Qt/QtEmbedded too I guess!
The big picture for commercial distros a couple years back was:
With RedHat going Fedora (and resigning its mindshare), it wasn't clear what the new scenario sould be. Mandrake was in my mind the great candidate to be the king of the overall distro (from freelading and home users through enterprise solutions), assuming they released more stable corporate versions.
Mandrake did its homework, and they announced a new development strategy with a community release and a later, more stable official release. They probably should add a slower (once a year) corporate option
But now Novell buys SUSE, opens it up and kicks the hell out of the Chess board. RedHat backs up and announces their (late) return to the desktop. Things are getting hot my friends :-)
The audience of the kernel source, I disagree with you, is hard-core people who compile from source for some reason, and developers who customize it.
End users should follow a Linux distribution, and it should be responsibility of the distribution they choose to give them a list of relevant features of a new kernel when they include it in the distro.
I don't think that the people who can help making Linux more mature really need a simpler changelog. Also, note that the new changelog is automatically generated by the source control software they are using (bitkeeper). Kernel devs can spend more time coding and less time typing boring changelogs ... how can this be bad ?