Just to clarify and quantify... what exactly does Red Hat contribute to?... here's their list:
https://community.redhat.com/s...
Not only do they contribute by funding development here and there, they have created many projects themselves and sustained them until they successfully made them into an upstream community. I'm not sure where to find that list but it is a subset of the above link. Any of the upstream project web sites that have a "Powered by Red Hat" type logio in the top right is a good indicator.
Red Hat over the years also purchased / acquired a few handful of companies... and if those companies already had open source products, they remained open source... and if those companies had some proprietary products, in most all (if not all) cases, even though it took quite a bit of work... over time those proprietary products were opened up... and upstream projects were created for them. Red Hat Directory Server used to be Netscape's proprietary product but is now upstreamed as 389. Quranet's SolidIce management system was all Windows server/technology-based and although it took a couple of years to completely re-write it in Java... it became RHEV and later RHV... and upstreamed as oVirt. The list goes on and on.
Then there are the big Enterprise class products Red Hat started from scratch and then over multiple releases adapted to use other's technologies where appropriate (they aren't afraid of not-invented here) like with OpenShift. At first they had their own container like entity that was called a "gear" and mostly namespaces, cgroups and SELinux policies strung together. Then they adopted Docker... and later Kubernetes.
When they found issues with the development model of Docker Inc, they started developing completely open alternatives like podman, buildah, skopeo, etc.
In any event, their story has always been combined with Free and Open Source software and believing in it. They did have one challenge when they made their own Red Hat Network backend based on Oracle DB... but that was fixed over time and eventually completely open sourced for use with other databases... and even forked to be used as a commercial product by SUSE.
And then there was the whole Oracle Enterprise Linux fork of RHEL... and CentOS... (in no specific order)... and their eventual sponsoring of CentOS. And of course, don't forget Fedora. It took them a while to get to Fedora but they did.
BTW, a commenter's mention of systemd... Red Hat did not originally agree with the idea of starting systemd... but Lennart P. continued developing it on his own time... and got it into Fedora and it proved itself... and was eventually adopted by RHEL... but it was never a given. Some people think systemd is a cancer others really appreciate the wide range of features it enabled (welcome modern world!)... and still others seek to emulate it on other OSes. One thing is for sure, systemd makes sysadmin easier to teach, learn and do across all of the distros that have adopted it.
I'm sure there are many developers who have gotten paid by various companies to write both proprietary as well as open source applications on top of RHEL... and to sysadmin RHEL... and to devops it... and to support it... as an alternative to Red Hat's paid support. It has created all kinds of opportunities outside of RHEL and their various other products. As previously mentioned a whole lot of that work has made its way into the greater Linux and FOSS ecosystems and is enjoyed by pretty the vast majority of Linux users everywhere.
I test out quite a few Linux distros and even though I'm a systemd fan, I thought I'd give Devuan a try as a KVM virtual machine... something I would expect a lot of Linux users to do considering that KVM has been the native Linux kernel hypervisor since 2007. Anyhoo... no matter that I did, the partitioner would allow me to partition/dev/vda but when making the mount points it couldn't see anything because it was expecting/dev/sda. Wha, wha, what?
Doesn't seem ready for prime time.
I'm a long time OpenVZ user. I don't know if your "VPS" is a container or a full-blown VM running under KVM or Xen... but if you want graphical access to a desktop environment I'd recommend using xrdp. xrdp is an RDP service for Linux. xrdp uses VNC underneath but it is transparent to the user who uses any RDP client they like. Many distros package xrdp.
I've used xrdp on physical systems, KVM virtual machines, and inside of OpenVZ containers and it seems to work equally well.
If you have a very limited set of packages installed, you'll probably have to install a desktop environment first. Get to know your package manager from the command line... rpm/dpkg or yum/apt-get.
The command line is still the preferred way to do things in some environments. Period. This is especially true in servers. In fact the next release of Windows Server supposedly has a non-GUI aka command line oriented interface by default. Linux is becoming more like Windows and vice versa.
As someone else said, some command line tools offer features that just aren't available from a GUI... and again... vice versa. I say use the best tool for the job... even if it is a command line tool.
I still prefer to run package managers from the command line even when there are reasonable GUI front-ends for them. The command line isn't always a failure, it's a feature.
I don't think Red Hat puts logos or branding in the kernel.
This was talked about earlier in the week on LWN.net and the patches ARE still available but for RHN (Red Hat Network) subscribers... and they modified the RHN service agreement to include a clause that makes it a violation to share the individual patches. While that might seem like a violation of the GPL, it isn't because they do make the entire thing publicly available (no RHN subscription required) in the new (no separate patches) kernel package release.
Packt Publishing Website has eBook version on sale
on
Drupal 7
·
· Score: 1
I really wanted this book since I'm a Druapl user and want to find out about the upcoming changes / features in Drupal 7. I found out that the Packt Publishing website has a special going... and I was able to pick up the electronic release for $9.99. I bought myself a copy.
They let you download a PDF that has your name and address embedded in the footer of each page. I like to have printed copies so I fired it off to the laser printer. Great value. Of course you can order the print version, or the print / ebook combo from them too if desired.
I'm a long time Prince fanatic. I *LOVE* Prince's music but the stuff he says when not performing music, not so much. I'll admit that in many areas, Prince is a nut. Aren't most of us?
Anyway... I think his point about the Internet is that, generally speaking, there aren't a whole lot of artists making money off of their works by distributing them over the Internet. There are a few monopoly sites / services who are getting the Lion's share of the money. Yes, there are a number of free music systems, or donation music sites where you can donate or pay what you want, but not really what Prince is looking for.
As many of you already know, Prince was an early Internet pioneer... creating several music clubs online to distribute his music. I think his main reason for stopping those is that people have very little control / discipline when it comes to pirating his music. With a later iteration of his online music club DRM was added but what a major pain that was for the end user (speaking for myself). Music without DRM gets pirated too much, and music with it is painful.
While I believe media piracy acts as a superior form of marketing, I'm guessing it would be hard to convince Prince of that.
So, from one aspect... online media content distribution has become just as corrupted as the pre-online... with only a handful of players making most of the money and to a certain extent, dictating the terms... and I think that is what Prince is talking about... and I agree with him... although I am encouraged by many of the free/donation type sites. I think Prince should be one of the first major artists to distribute content via some of our preferred services that are primarily independent music... to see how it goes before he calls it quits. I mean, giving away your CD in a UK newspaper is going to help your sales (compared to non-DRM Internet-based, easily pirated delivery) how?
I think if Prince gave his fans a chance, they'd be happy to donate... and just maybe he'd do as well or better as those efforts from Radio Head and Nine Inch Nails.
Here is what I use for the classes I have run (two so far): A server running an OpenVZ kernel with a public IP address that can be accessed from on or off campus. Then I recommend making a course container with a public IP address and give each student a user account on that container.
Then create a container for each student with a private IP address. A simple iptables line will allow you to NAT the containers with a private IP address through the host node so each container can reach the outside world.
To access their containers students would just ssh into the course server and then ssh from there to their containers. Containers don't take up a whole lot of resources nor disk space compared to a full virtual machine.
That will provide you with a virtualized environment for each student where they have full root access, can install software, and if they screw it up somehow, you can easily repair or replace it. You aren't limited to what Linux distribution you can run in a container so if you want to give them access to multiple distros, you can.
Of course that only gives you a command line only environment which is suitable for a sysadmin type class (what I teach). If you absolutely need a GUI environment, you can easily install one or more desktop environments in your containers which would be accessible over the LAN with VNC. Routing private IPs over a LAN isn't too difficult, you'd just need a static route to the host node.
Of course you could do the same thing with KVM on a server but it would require a whole lot more resources.
If that still will not meet your requirements because they turned down a server-based solution, I'd recommend external USB hard drives. You can get a fairly large one for about $80 and you'll have way less problems than with LiveUSB media that tend to mess up easily. Then you can either run a regular OS from it OR uses it to store disk images for VirtualBox or whatever virt solution the students would have to use on their personal computer.
Yep, I definitely meant olpc... darn fingers.
You don't think they are coming anytime soon? I don't have any inside information but I'm guessing within a year or less. Why? Because as Mary Lou points out, unlike many new technologies, her design doesn't use any new materials nor require any new manufacturing process... as they use existing materials and can be made in existing plants. They are going to be inexpensive too... and available on netbooks initially I would guess.
What more could you want?
Have you seen the videos on oplc.tv of the new screen technology coming?
Much cheaper and better... no special materials or new manufacturing facilities needed.
Let me clarify before people jump down my throat... OpenVZ (www.openvz.org) is OS Virtualization (aka containers) and NOT machine / hardware virtualization... so it can only run Linux on Linux... but it has been able to do live migrations from one processor family to another since they initially added checkpointing.
OpenVZ is fairly CPU agnostic and it has been ported to a number of CPU families. In fact the project leader recently ported it to ARM (Gumstix Overo). See: http://community.livejournal.com/openvz/24651.html
Once Firefox 3.1 comes out and includes support for playback of Theora videos and Ogg audio... I hope there will be in influx of new content published (using the more simple tags) using Theora and Ogg. Hopefully that will cause some momentum and give Flash some competition.
I realize that Flash is used for a lot more than just video and audio but it is the dominant thing Linux users care to use Flash for.
Of course that isn't going to cause YouTube to switch everything over to Theora / Ogg but you have to start somewhere.
Yeah, so you have a fantastic sysadmin who compiles everything and tweeks it just so... and then (s)he goes away... and you are left with an unmaintainable mess that has to be completely figured out by the next person. Sounds like something I want... NOT.
Your suggestion to change the subject of the post to remove "Closed-Source" is unfounded. There *IS* actually an open-sourced driver for nVidia and the problem is only with the closed (accellerated) driver.
I have to admit that I didn't read the article (yet)... but this begs the question... what about "global dimming". Haven't heard of global dimming? most people haven't. For a good overview, visit:
Ok, now we know what makes a distro suck. As long as it supports reiserfs and includes xcdroast, it must be great.
The differences between WS, ES and AS are annoying at first until you realize that originally they just had AS, making everyone pay for support for stuff they didn't need. I'm glad they broke it into three. Of course it would great if the add additional options with even lower cost of entry.
I think Red Hat will reduce prices over time and perhaps they picked their current pricing stucture to give them more flexibility in future Microsoft pricing wars.
Umm, what are you talking about specifically? Heard of kickstart? Red Hat still has a graphical and a text based install. The graphical can even be done over VNC. Remote kickstarts are also possible by passing some install parameters (kickstart filename, network location, etc) via DHCP.
> Never use the command line to change settings again, use X!
Again what specifically are you talking about? Most tools still have text-gui interfaces, in addition to the X-gui ones. Editing config files with one's preferred text editor is also an option... unless you are claiming that Red Hat has somehow altered network server apps to use binary config files. Ok, you could be referring to Gnome and GConf... and their registry like system... but I'm not familiar with Gnome enough to comment... but again, Gnome isn't specific to Red Hat.
> Don't pick specific software you want installed, pick broad categories [etc]
Ok, that is somewhat of a valid complaint. I think they removed the individual package selection from the install for reasons other than an attempt to limit users. Of course you can always install or remove any software after the initial install... so the validity of the point is weakened greatly.
> up2date -u
Yeah, Red Hat does sell a service. up2date does still use rpm and all of the source for both apps has been released under the GPL. Of course the source for their RHN has not be released... but alternative systems are available... yum and apt-get.
Even in RHEL, source code to all distro packages is freely available, but not the binary packages... which still follows the rules of the GPL to the best of my knowledge. Red Hat has not taken any negative action against those taking the RHEL source packages and building a binary install path for them.
Red Hat also employs many top level kernel, gcc, glibc, and filesystem developers and are directly responsible for much of the aggressive development we see in all distros... because everything is given back to the community. To have a sustainable business model, Red Hat had to find some way to "take" and I think they have a good blance... especially since Fedora is still a viable option... rather than being the myth some folks make it out to be... as a substandard beta-test for RHEL.
My attitude has been that Linux is about preferences and choice and I don't fault anyone for using other distros... so why fault me for using Red Hat? I'm not trying to spread FUD here, but at least with Red Hat you have an idea of where they are going in the future. With SuSE, who knows what Novell will eventually do to grab onto a sustainable business plan? Where will Gentoo go? Mandrake? It is hard to say... but at least they are all working together to make Linux better. I see other commercial distros following in Red Hat's footsteps and ain't it grand we have literally hundreds of distros to pick from?!
I'm not sure why everyone seems fit to pick on Fedora Core 2 alone... when other distros share many of the same problems. I'm not a Mandrake user but I know a few people who are... and everyone I've talked to who tried Mandrake 10 switched back to the 9 series because they had noticable differences in hardware compatibility. I'm not trying to pick on Mandrake here but it seems to me that the 2.6.x kernel simply isn't finished. I'm not trying to bad mouth the kernel developers... but it is a fact that Linus has not started a 2.7.x devel tree... and that even now the kernel developers are making major changes to important subsystems... in what is supposedly a production kernel.
I'm confident that in a couple of months, once the 2.6.x kernel has been weened from the developers... and all of the issues get worked out at the distribution level... it'll be a clear winner.
On the flipside of the coin, I've installed FC2 on about a dozen machines and have actually found that some hardware that didn't work in any previous distribution release, now works great in FC2. For me, FC2 works quite well on a variety of hardware and I am confident that as some of the minor issues are resolved, it will just get better and better.
I don't know if this is just a mis-perception, but I feel that the Fedora Core team is taking even bolder steps to mainstream Linux than Red Hat was... and Red Hat has always been aggressive in promoting new software technologies. I see this as a good thing. Without that pushing, Linux would not continue to improve and mature at the impressive rate it has enjoyed thus far.
Using Fedora Core 1 and now FC2, I can actually start to see the not too distant future where Linux has a good fighting chance at, dare I say it serioulsy?... the DESKTOP MARKET!
While it is true that FC2 isn't perfect, no new major release (new kernel, new releases of the desktop environs, etc) is born perfect... and it is unrealistic to think any will or even should. Regardless of the amount of people submitting bugs during the test-releases, in the real world an initial production release is just the next step in shaking out the bugs. It is that way even with Microsoft and Apple... even if they don't want to admit it. The difference is that our community is more open about the bugs and as a result, most of them get fixed and fixed faster.
In summary, quit picking on the fruits of Red Hat simply because you have some resentment about their change in marketing with Red Hat Enterprise Linux and their success in the marketplace. If you want to debate those, do so directly. I don't expect everyone to love them, but give them the fair shake you give everyone else... and have realistic expectations. Long live Linux.
Red Hat has had information on their web site regarding Red Hat Academy for some time... but they are just getting it rolling... and haven't really announced it in a big way.
I passed the RHCE back in April of 2002 and I know that the RHCE program is a high quality program... especially the test.
I happen to work for a college in Montana and I was interested in learning what it would take to get Red Hat Academy / RHCT training and testing in our area... hoping to eventually turn it into an RHCE training program (not currently in Red Hat's cards). After talking to Joel Jackson (I think that was his name) in the Sales department... he emailed me a PDF with the details.
First year cost:
$15,000 for training - 2 RHCE Instructors and 1 RHCE Administrator $22,000 for a year of Red Hat Academy curriculum, support, manuals, distribution, etc.
So... the first year investment is $37,000
Additional years are $22,000 (unless you lost your trained people and need to train more).
I wonder if that $22,000 figure is negotiable? That is good for up to 5,000 students per year.
They only allow non-profit educational institutions (high schools, colleges, universities, etc) in the Red Hat Academy program... and while they say you can charge the student a fee (call it whatever you want... book fee, lab fee, etc) they really don't give you any guidelines as to what you can charge... except that Red Hat wants $150 for each test administered. How much do you charge for tests? Again, that's up to your institution.
Being from a small college in Montana, $37,000 is a sizable first year investment. It wouldn't be any big deal if we could get 500 students interested in the program... but that would be a real challenge. Considering the fact that Red Hat is "planting seeds" with the Academy program... you'd think the seeds would be cheaper... or that they'd have a program priced close to cost. I have no idea how much it will take Red Hat to administer. I'm sure it is a quality program, judging by the RHCE.
I bought my own copy of Free as in Freedom but I couldn't help but be offended by the very last page. You know the one... after all of the Appendixes... after the Index. The one that says (please excuse any typos):
The cover of this book was designed and produced in Adobe Photoshop 5.5 and QuarkXPress 4.1 with Interstate and Sabon fonts. The cover photograph of Richard Stallman was taken by Sam Ogden / Photo Researchers, Inc.
The interior of the book is set in Adobe's Sabon font and was produced with FrameMaker 5.5.6. Sabon was designed by Jan Tschichold in 1964. The roman design is based on Garamond; the italic is based on typefaces created by Robert Gronjon, one of Garmond's contemporaries. Sabon is a registered trademark of Linotype-Hell AG and / or its subsidiaries.
Many people contributed to this project, including Tim O'Reilly, Laurie Petrycki, Jeffrey Holcomb, Edie Freedman, Hanna Dyer, Emma Colby, Melanie Wang, David Futato, Sheryl Avruch, Claire Cloutier, Joe Wizda, Rachel Wheeler, and Leanne Soylemez.
Who remembers the live interview with RMS where it was being broadcast using Real Audio Server (or whatever the proper name of it is)? Remember how Richard told them to stop the broadcast because he refused to be used (through the broadcast of his interview) to encourage people to use proprietary software (RealPlayer)? Well, this isn't exactly the same because the book isn't directly making people use proprietary programs to read the text... but since it was made with proprietary software... I'd think he would discourage it. I guess TeX, LaTeX, etc weren't good enough?!?...not that I know how to use any of those.:)
No, the command line isn't dead. People who touch type and like to read... actually prefer the command line... not for everything... but for many, many things.
Apple is very much like Atari. Before Atari became a logo to stick on Hasbro remakes of aging Atari games it was often said that, "Atari couldn't market immortality." My point here is that while Steve Jobs might have helped save Apple from an earlier grave, he certainly is no magician.
It is estimated (and I'm not going to quote the sources here because they are well known) that Linux holds a 4% share of the desktop market... barely trailing the Macintosh's 5%. Please note that the tiny 5% marketshare that the Mac has is ONLY AFTER a period of record sales figures. How the Macintosh can be seen as a potential force in the desktop world with 5% of the market, and Linux is seen by many (currently) as a desktop failure with only 1% difference is beyond me.
Ok, ok... Apple has decided that they can't bandaid the MacOS like Microsoft has done with Windows 95/98... and they FINALLY decided to start with a new, Unix based, foundation. I'll give them credit for that BUT... Apple seems to take longer to make things mature... and until Apple makes some serious (as in measurable) inroads into the server market, I don't see it being a major contender in the desktop market... because people often times want to bring their work home with them and make their home systems mini versions of their business systems.
Slashdot Mirror
Just to clarify and quantify... what exactly does Red Hat contribute to?... here's their list: https://community.redhat.com/s... Not only do they contribute by funding development here and there, they have created many projects themselves and sustained them until they successfully made them into an upstream community. I'm not sure where to find that list but it is a subset of the above link. Any of the upstream project web sites that have a "Powered by Red Hat" type logio in the top right is a good indicator. Red Hat over the years also purchased / acquired a few handful of companies... and if those companies already had open source products, they remained open source... and if those companies had some proprietary products, in most all (if not all) cases, even though it took quite a bit of work... over time those proprietary products were opened up... and upstream projects were created for them. Red Hat Directory Server used to be Netscape's proprietary product but is now upstreamed as 389. Quranet's SolidIce management system was all Windows server/technology-based and although it took a couple of years to completely re-write it in Java... it became RHEV and later RHV... and upstreamed as oVirt. The list goes on and on. Then there are the big Enterprise class products Red Hat started from scratch and then over multiple releases adapted to use other's technologies where appropriate (they aren't afraid of not-invented here) like with OpenShift. At first they had their own container like entity that was called a "gear" and mostly namespaces, cgroups and SELinux policies strung together. Then they adopted Docker... and later Kubernetes. When they found issues with the development model of Docker Inc, they started developing completely open alternatives like podman, buildah, skopeo, etc. In any event, their story has always been combined with Free and Open Source software and believing in it. They did have one challenge when they made their own Red Hat Network backend based on Oracle DB... but that was fixed over time and eventually completely open sourced for use with other databases... and even forked to be used as a commercial product by SUSE. And then there was the whole Oracle Enterprise Linux fork of RHEL... and CentOS... (in no specific order)... and their eventual sponsoring of CentOS. And of course, don't forget Fedora. It took them a while to get to Fedora but they did. BTW, a commenter's mention of systemd... Red Hat did not originally agree with the idea of starting systemd... but Lennart P. continued developing it on his own time... and got it into Fedora and it proved itself... and was eventually adopted by RHEL... but it was never a given. Some people think systemd is a cancer others really appreciate the wide range of features it enabled (welcome modern world!)... and still others seek to emulate it on other OSes. One thing is for sure, systemd makes sysadmin easier to teach, learn and do across all of the distros that have adopted it. I'm sure there are many developers who have gotten paid by various companies to write both proprietary as well as open source applications on top of RHEL... and to sysadmin RHEL... and to devops it... and to support it... as an alternative to Red Hat's paid support. It has created all kinds of opportunities outside of RHEL and their various other products. As previously mentioned a whole lot of that work has made its way into the greater Linux and FOSS ecosystems and is enjoyed by pretty the vast majority of Linux users everywhere.
I've been using Linux as my primary OS since 1995 and do installs a few times a week... so no... your "probably" didn't pan out this time.
I test out quite a few Linux distros and even though I'm a systemd fan, I thought I'd give Devuan a try as a KVM virtual machine... something I would expect a lot of Linux users to do considering that KVM has been the native Linux kernel hypervisor since 2007. Anyhoo... no matter that I did, the partitioner would allow me to partition /dev/vda but when making the mount points it couldn't see anything because it was expecting /dev/sda. Wha, wha, what?
Doesn't seem ready for prime time.
I'm a long time OpenVZ user. I don't know if your "VPS" is a container or a full-blown VM running under KVM or Xen... but if you want graphical access to a desktop environment I'd recommend using xrdp. xrdp is an RDP service for Linux. xrdp uses VNC underneath but it is transparent to the user who uses any RDP client they like. Many distros package xrdp.
I've used xrdp on physical systems, KVM virtual machines, and inside of OpenVZ containers and it seems to work equally well.
If you have a very limited set of packages installed, you'll probably have to install a desktop environment first. Get to know your package manager from the command line... rpm/dpkg or yum/apt-get.
The command line is still the preferred way to do things in some environments. Period. This is especially true in servers. In fact the next release of Windows Server supposedly has a non-GUI aka command line oriented interface by default. Linux is becoming more like Windows and vice versa. As someone else said, some command line tools offer features that just aren't available from a GUI... and again... vice versa. I say use the best tool for the job... even if it is a command line tool. I still prefer to run package managers from the command line even when there are reasonable GUI front-ends for them. The command line isn't always a failure, it's a feature.
I don't think Red Hat puts logos or branding in the kernel. This was talked about earlier in the week on LWN.net and the patches ARE still available but for RHN (Red Hat Network) subscribers... and they modified the RHN service agreement to include a clause that makes it a violation to share the individual patches. While that might seem like a violation of the GPL, it isn't because they do make the entire thing publicly available (no RHN subscription required) in the new (no separate patches) kernel package release.
Jim said the same stuff during his keynote at the recent Red Hat Summit 2010. If you want to see it, watch the video here: http://www.redhat.com/videos/summit2010/Jim_Whitehurst.html
I really wanted this book since I'm a Druapl user and want to find out about the upcoming changes / features in Drupal 7. I found out that the Packt Publishing website has a special going... and I was able to pick up the electronic release for $9.99. I bought myself a copy. They let you download a PDF that has your name and address embedded in the footer of each page. I like to have printed copies so I fired it off to the laser printer. Great value. Of course you can order the print version, or the print / ebook combo from them too if desired.
I'm a long time Prince fanatic. I *LOVE* Prince's music but the stuff he says when not performing music, not so much. I'll admit that in many areas, Prince is a nut. Aren't most of us?
Anyway... I think his point about the Internet is that, generally speaking, there aren't a whole lot of artists making money off of their works by distributing them over the Internet. There are a few monopoly sites / services who are getting the Lion's share of the money. Yes, there are a number of free music systems, or donation music sites where you can donate or pay what you want, but not really what Prince is looking for.
As many of you already know, Prince was an early Internet pioneer... creating several music clubs online to distribute his music. I think his main reason for stopping those is that people have very little control / discipline when it comes to pirating his music. With a later iteration of his online music club DRM was added but what a major pain that was for the end user (speaking for myself). Music without DRM gets pirated too much, and music with it is painful.
While I believe media piracy acts as a superior form of marketing, I'm guessing it would be hard to convince Prince of that.
So, from one aspect... online media content distribution has become just as corrupted as the pre-online... with only a handful of players making most of the money and to a certain extent, dictating the terms... and I think that is what Prince is talking about... and I agree with him... although I am encouraged by many of the free/donation type sites. I think Prince should be one of the first major artists to distribute content via some of our preferred services that are primarily independent music... to see how it goes before he calls it quits. I mean, giving away your CD in a UK newspaper is going to help your sales (compared to non-DRM Internet-based, easily pirated delivery) how?
I think if Prince gave his fans a chance, they'd be happy to donate... and just maybe he'd do as well or better as those efforts from Radio Head and Nine Inch Nails.
Here is what I use for the classes I have run (two so far): A server running an OpenVZ kernel with a public IP address that can be accessed from on or off campus. Then I recommend making a course container with a public IP address and give each student a user account on that container.
Then create a container for each student with a private IP address. A simple iptables line will allow you to NAT the containers with a private IP address through the host node so each container can reach the outside world.
To access their containers students would just ssh into the course server and then ssh from there to their containers. Containers don't take up a whole lot of resources nor disk space compared to a full virtual machine.
That will provide you with a virtualized environment for each student where they have full root access, can install software, and if they screw it up somehow, you can easily repair or replace it. You aren't limited to what Linux distribution you can run in a container so if you want to give them access to multiple distros, you can.
Of course that only gives you a command line only environment which is suitable for a sysadmin type class (what I teach). If you absolutely need a GUI environment, you can easily install one or more desktop environments in your containers which would be accessible over the LAN with VNC. Routing private IPs over a LAN isn't too difficult, you'd just need a static route to the host node.
Of course you could do the same thing with KVM on a server but it would require a whole lot more resources.
If that still will not meet your requirements because they turned down a server-based solution, I'd recommend external USB hard drives. You can get a fairly large one for about $80 and you'll have way less problems than with LiveUSB media that tend to mess up easily. Then you can either run a regular OS from it OR uses it to store disk images for VirtualBox or whatever virt solution the students would have to use on their personal computer.
Yep, I definitely meant olpc... darn fingers. You don't think they are coming anytime soon? I don't have any inside information but I'm guessing within a year or less. Why? Because as Mary Lou points out, unlike many new technologies, her design doesn't use any new materials nor require any new manufacturing process... as they use existing materials and can be made in existing plants. They are going to be inexpensive too... and available on netbooks initially I would guess. What more could you want?
Have you seen the videos on oplc.tv of the new screen technology coming? Much cheaper and better... no special materials or new manufacturing facilities needed.
Let me clarify before people jump down my throat... OpenVZ (www.openvz.org) is OS Virtualization (aka containers) and NOT machine / hardware virtualization... so it can only run Linux on Linux... but it has been able to do live migrations from one processor family to another since they initially added checkpointing. OpenVZ is fairly CPU agnostic and it has been ported to a number of CPU families. In fact the project leader recently ported it to ARM (Gumstix Overo). See: http://community.livejournal.com/openvz/24651.html
Once Firefox 3.1 comes out and includes support for playback of Theora videos and Ogg audio... I hope there will be in influx of new content published (using the more simple tags) using Theora and Ogg. Hopefully that will cause some momentum and give Flash some competition. I realize that Flash is used for a lot more than just video and audio but it is the dominant thing Linux users care to use Flash for. Of course that isn't going to cause YouTube to switch everything over to Theora / Ogg but you have to start somewhere.
Yeah, so you have a fantastic sysadmin who compiles everything and tweeks it just so... and then (s)he goes away... and you are left with an unmaintainable mess that has to be completely figured out by the next person. Sounds like something I want... NOT.
Your suggestion to change the subject of the post to remove "Closed-Source" is unfounded. There *IS* actually an open-sourced driver for nVidia and the problem is only with the closed (accellerated) driver.
I have to admit that I didn't read the article (yet)... but this begs the question... what about "global dimming". Haven't heard of global dimming? most people haven't. For a good overview, visit:
http://en.wikipedia.org/wiki/Global_dimming
Reducing the amount of sunlight that hits the earth is already happening with some negative side effects.
Ok, now we know what makes a distro suck. As long as it supports reiserfs and includes xcdroast, it must be great. The differences between WS, ES and AS are annoying at first until you realize that originally they just had AS, making everyone pay for support for stuff they didn't need. I'm glad they broke it into three. Of course it would great if the add additional options with even lower cost of entry. I think Red Hat will reduce prices over time and perhaps they picked their current pricing stucture to give them more flexibility in future Microsoft pricing wars.
> Never do another headless install
Umm, what are you talking about specifically? Heard of kickstart? Red Hat still has a graphical and a text based install. The graphical can even be done over VNC. Remote kickstarts are also possible by passing some install parameters (kickstart filename, network location, etc) via DHCP.
> Never use the command line to change settings again, use X!
Again what specifically are you talking about? Most tools still have text-gui interfaces, in addition to the X-gui ones. Editing config files with one's preferred text editor is also an option... unless you are claiming that Red Hat has somehow altered network server apps to use binary config files. Ok, you could be referring to Gnome and GConf... and their registry like system... but I'm not familiar with Gnome enough to comment... but again, Gnome isn't specific to Red Hat.
> Don't pick specific software you want installed, pick broad categories [etc]
Ok, that is somewhat of a valid complaint. I think they removed the individual package selection from the install for reasons other than an attempt to limit users. Of course you can always install or remove any software after the initial install... so the validity of the point is weakened greatly.
> up2date -u
Yeah, Red Hat does sell a service. up2date does still use rpm and all of the source for both apps has been released under the GPL. Of course the source for their RHN has not be released... but alternative systems are available... yum and apt-get.
Even in RHEL, source code to all distro packages is freely available, but not the binary packages... which still follows the rules of the GPL to the best of my knowledge. Red Hat has not taken any negative action against those taking the RHEL source packages and building a binary install path for them.
Red Hat also employs many top level kernel, gcc, glibc, and filesystem developers and are directly responsible for much of the aggressive development we see in all distros... because everything is given back to the community. To have a sustainable business model, Red Hat had to find some way to "take" and I think they have a good blance... especially since Fedora is still a viable option... rather than being the myth some folks make it out to be... as a substandard beta-test for RHEL.
My attitude has been that Linux is about preferences and choice and I don't fault anyone for using other distros... so why fault me for using Red Hat? I'm not trying to spread FUD here, but at least with Red Hat you have an idea of where they are going in the future. With SuSE, who knows what Novell will eventually do to grab onto a sustainable business plan? Where will Gentoo go? Mandrake? It is hard to say... but at least they are all working together to make Linux better. I see other commercial distros following in Red Hat's footsteps and ain't it grand we have literally hundreds of distros to pick from?!
I'm not sure why everyone seems fit to pick on Fedora Core 2 alone... when other distros share many of the same problems. I'm not a Mandrake user but I know a few people who are... and everyone I've talked to who tried Mandrake 10 switched back to the 9 series because they had noticable differences in hardware compatibility. I'm not trying to pick on Mandrake here but it seems to me that the 2.6.x kernel simply isn't finished. I'm not trying to bad mouth the kernel developers... but it is a fact that Linus has not started a 2.7.x devel tree... and that even now the kernel developers are making major changes to important subsystems... in what is supposedly a production kernel.
I'm confident that in a couple of months, once the 2.6.x kernel has been weened from the developers... and all of the issues get worked out at the distribution level... it'll be a clear winner.
On the flipside of the coin, I've installed FC2 on about a dozen machines and have actually found that some hardware that didn't work in any previous distribution release, now works great in FC2. For me, FC2 works quite well on a variety of hardware and I am confident that as some of the minor issues are resolved, it will just get better and better.
I don't know if this is just a mis-perception, but I feel that the Fedora Core team is taking even bolder steps to mainstream Linux than Red Hat was... and Red Hat has always been aggressive in promoting new software technologies. I see this as a good thing. Without that pushing, Linux would not continue to improve and mature at the impressive rate it has enjoyed thus far.
Using Fedora Core 1 and now FC2, I can actually start to see the not too distant future where Linux has a good fighting chance at, dare I say it serioulsy?... the DESKTOP MARKET!
While it is true that FC2 isn't perfect, no new major release (new kernel, new releases of the desktop environs, etc) is born perfect... and it is unrealistic to think any will or even should. Regardless of the amount of people submitting bugs during the test-releases, in the real world an initial production release is just the next step in shaking out the bugs. It is that way even with Microsoft and Apple... even if they don't want to admit it. The difference is that our community is more open about the bugs and as a result, most of them get fixed and fixed faster.
In summary, quit picking on the fruits of Red Hat simply because you have some resentment about their change in marketing with Red Hat Enterprise Linux and their success in the marketplace. If you want to debate those, do so directly. I don't expect everyone to love them, but give them the fair shake you give everyone else... and have realistic expectations. Long live Linux.
Red Hat has had information on their web site regarding Red Hat Academy for some time... but they are just getting it rolling... and haven't really announced it in a big way.
I passed the RHCE back in April of 2002 and I know that the RHCE program is a high quality program... especially the test.
I happen to work for a college in Montana and I was interested in learning what it would take to get Red Hat Academy / RHCT training and testing in our area... hoping to eventually turn it into an RHCE training program (not currently in Red Hat's cards). After talking to Joel Jackson (I think that was his name) in the Sales department... he emailed me a PDF with the details.
First year cost:
$15,000 for training - 2 RHCE Instructors and 1 RHCE Administrator
$22,000 for a year of Red Hat Academy curriculum, support, manuals, distribution, etc.
So... the first year investment is $37,000
Additional years are $22,000 (unless you lost your trained people and need to train more).
I wonder if that $22,000 figure is negotiable? That is good for up to 5,000 students per year.
They only allow non-profit educational institutions (high schools, colleges, universities, etc) in the Red Hat Academy program... and while they say you can charge the student a fee (call it whatever you want... book fee, lab fee, etc) they really don't give you any guidelines as to what you can charge... except that Red Hat wants $150 for each test administered. How much do you charge for tests? Again, that's up to your institution.
Being from a small college in Montana, $37,000 is a sizable first year investment. It wouldn't be any big deal if we could get 500 students interested in the program... but that would be a real challenge. Considering the fact that Red Hat is "planting seeds" with the Academy program... you'd think the seeds would be cheaper... or that they'd have a program priced close to cost. I have no idea how much it will take Red Hat to administer. I'm sure it is a quality program, judging by the RHCE.
Who remembers the live interview with RMS where it was being broadcast using Real Audio Server (or whatever the proper name of it is)? Remember how Richard told them to stop the broadcast because he refused to be used (through the broadcast of his interview) to encourage people to use proprietary software (RealPlayer)? Well, this isn't exactly the same because the book isn't directly making people use proprietary programs to read the text... but since it was made with proprietary software... I'd think he would discourage it. I guess TeX, LaTeX, etc weren't good enough?!? ...not that I know how to use any of those. :)
Just food for thought.
Ummm, no.
Current estimates are that Linux commands 25% of the Server market, which is 2nd. NT is 1st with 38%... and Novell has moved down to 3rd with 19%.
Numbers can lie but in this case...
--
Scott Dowdle
No, the command line isn't dead. People who touch type and like to read... actually prefer the command line... not for everything... but for many, many things.
Apple is very much like Atari. Before Atari became a logo to stick on Hasbro remakes of aging Atari games it was often said that, "Atari couldn't market immortality." My point here is that while Steve Jobs might have helped save Apple from an earlier grave, he certainly is no magician.
It is estimated (and I'm not going to quote the sources here because they are well known) that Linux holds a 4% share of the desktop market... barely trailing the Macintosh's 5%. Please note that the tiny 5% marketshare that the Mac has is ONLY AFTER a period of record sales figures. How the Macintosh can be seen as a potential force in the desktop world with 5% of the market, and Linux is seen by many (currently) as a desktop failure with only 1% difference is beyond me.
Ok, ok... Apple has decided that they can't bandaid the MacOS like Microsoft has done with Windows 95/98... and they FINALLY decided to start with a new, Unix based, foundation. I'll give them credit for that BUT... Apple seems to take longer to make things mature... and until Apple makes some serious (as in measurable) inroads into the server market, I don't see it being a major contender in the desktop market... because people often times want to bring their work home with them and make their home systems mini versions of their business systems.
--
Scott Dowdle
Another Linux Advocate - http://linuxadvocate.b-squared.net
--
Scott Dowdle