One of the founders of the open-source software initiative, Bruce Perens, is planning to release a new version of Linux to challenge Red Hat’s enterprise version of the operating system, and fill the hole left in the consumer market since Red Hat announced it would no longer sell a consumer version in retail stores.
User Linux: Yet Another Rendition of Linux
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Replacing RedHat Linux aye? that could be hard.. you’d need to add lotsa chunky GUI’s which I dislike about RedHat and why I prefer Debian > RedHat.
This is just a modified debian w/ commercial support aimed at the corporate market?
“Replacing RedHat Linux aye? that could be hard.. you’d need to add lotsa chunky GUI’s which I dislike about RedHat”
You mean those which you don’t have to install in RedHat ?
It sounds interesting however I believe that by not allowing Linux to be commercialized than freedom is being restricted. Is the GPL against freedom? I suspect that there is a solution for both businesses who want to use Linux as a commercial product as well as the open source community who want Linux to be an open platform (and in some ways accessible). The solution is a software layer that can be commercialized as a product line. This simply satisfies all parties and it enables the GPL to promote freedom rather than to restrict the freedom of proprietary interest in Linux.
What would become of Linux if some of the 8,000 Linux distros were to go away and the authors began to work in other areas.
If anything, this should be wonderful for Debian. Debian is a huge project, and has a very high-quality back-end, but the goals of the project are conservative and sometimes not in line with those of the mainstream. If UserLinux makes Debian more accessible, and more usable for businesses, it’d be wonderful. I, for one, would love to recommend a Debian-based distro over RedHat, if only for the enormous central APT repository, and the number of high-quality third-party repositories.
Many, multi-million dollar donations. Bruce Perens must be one hell of a speaker. I really admire him. He’s got good connections with business, and is level-headed and practical, but he still belives strongly in the ideals of open source.
Nothing. These authors work on their distros because they want to. Its highly unlikely that they’d just go work on something else if you didn’t allow them to maintain their own distro.
What is the benefit of using this operating system? It keeps spliting up and becoming more fragmented.
It sounds interesting however I believe that by not allowing Linux to be commercialized than freedom is being restricted. Is the GPL against freedom?
No. The GPL ensures that the product will remain free, even if it is commercialized. Commercial organizations will be forced to play fair, if they want to play at all.
restrict the freedom of proprietary interest in Linux.
Proprietary interests restrict the freedom of their end users to modify, copy and redistribute the product. That is why the GPL exists. Read it sometime.
We’re playing a new game with intellectual property. The new game is a fair game. Everyone gets access to the technology. The leaders will be the people who put the most effort into making the technology great. The people who innovate. Unfortunately commercial capitalist American has a problem with this game. Tough titty!
Sorry, but there is absolutely NO way he will be able to produce a distro that will be able to compete with RedHat. Out of all of the other distros, Suse is the only one that is even in the same ballpark. Why do all these people waste their time reinventing the wheel, when they can be more productive in other areas? If he was doing something new and totally different, I could understand, but this is just the same crap as about 100 other distros.
-G
I wasn’t aware that being fragmented was a weakness. As long as interoperability is there (not there just yet, but its making progress) fragmentation is a strenght. Here in the USA, we call it the free market. I know that after being under the oppressive shadow of a monopoly for so long, its hard to adjust to freedom, but you’ll get used to it, just as the Russians did when the Soviet Union fell…
I haven’t tried Debian yet, might look into Knoppix. I have tried Redhat 8, Suse 8.2 and Mandrake 9.1, personally I didn’t like them. I reverted back to Slackware 9. I like having lots of distro’s to choose from. It gives me options and choices.
yeah fragmentation is a strength. its what gets you half a dozen directory layout structures each not compatible with the other. 8000 people selling the same car but with different colours, nobody inovating, each building the wheel the other already built.
Actually, the ideal of capitalism *is* everyone selling the exact same thing. Quixotically, that leads to the fastest innovation, because its basically innovate or die.
If UserLinux makes Debian more accessible…
Aren’t Libranet, Knoppix, and some other already doing that? To me (but that’s just my opinion), it’s just one more beast in the jungle.
Did he miss the Fedora Announcement? I think instead of re-inventing the wheel his and a lot of others energy could be best spent contributing to the 1000 other projects we got running around, rather than starting A N O T H E R distro. I dont care what they work on even if it is Fedora or not. please.
Hmmm…and here I thought this rather nice piece of work called Fedora Core that I’ve been using is the follow-on to the RedHat desktop shrinkwrapped products. Free, too. Beats the pants, and more, off a Debian Woody install.
UserLinux smells a lot like the current crop of Debian-based distributions that somehow manage to survive the inclusion all that “risky” code Debian says is unstable.
Perens, among others, seems to base his sales pitch on the ideology of open source. That’s all well and good, but that isn’t the way to put Linux on either corporate or home desktops. (However, as Debian shows, that is the way to put behind-the-curve tools on the desktops of ideologues.)
For desktop users — especially in the home market — Linux needs to offer them something that they can’t get from the Windows machines they already own. Free as in beer won’t cut it, because they are not in the market for a new OS at any price. Free as in RMS won’t cut it, because they aren’t developers and have never heard of RMS. OpenOffice won’t do it, because they already have the real Office. (It’s a bit like saying to someone who’s just bought a Big Mac: Here, throw away that Big Mac you’ve already purchased and eat the free and almost-as-good Mirror Image Mac cooked in our open source kitchen. Why would anyone do that?)
The home desktop market is not the corporate desktop market. Linux won’t conquer the suburbs simply by being cheap, free, and just as good as Windows. It needs to deliver something people want and can’t get from Windows. Apart from ideology and geek appeal, it doesn’t, yet.
I totally agree with all the commentators against yet another pointless, fragmented Linux distro when there’s trillions out there already suitable for every purpose imaginable.
And why stop there? Why have more than one car manufacturer? Why have more than one bottled cola? Why have more than one type of clothing to wear? Why not have ONE manufacturer of everything, and leave it at that!!! That’s capitalism, isn’t it? Why reinvent the wheel, eh?
And then the state could take over the company that makes these goods for us (because it would be too important to be left to a single organisation). Brilliant!
And then remove democracy, because then we only need one party to vote for! Yay!
Why do they reinvent the wheel? Why???
It’s almost like going to a news forum and repeating “stop reinventing the wheel” ad nauseu, just like everyone else – duplicating effort. Oh…
</sarcasm> If we didn’t “reinvent the wheel”, we’d still have stone wheels that weighed several tons each. If developers want to spend their own personal free time doing stuff that someone has already done, then it is entirely up to them, not to anyone else.
If you don’t like the idea of another distro, then don’t. But please don’t insist that everyone think the same as you.
The GPL is indeed against freedom in some ways that make it unrealistic. Many people choose to lose their freedom by using a commercial product, they do not care that they are not in control of the factors of production. I think that the product world and the open source world can both move forward but in order to do that they have to find the way. The bases of this effort by Bruce Perens is illogical at this stage, because it does not fit into the big picture.
I agree with you but I believe that too many choices is like not enough…
bruce perens knows about fedora, and he’s no dummy either.
if this isn’t just preliminary “hot air”, and if he’s really able to pull that off, i’m curious how this project will relate to the commercial distros as well as fedora and debian and the smaller ones.
Synergy, unless someone writes some unique and compelling code for UserLinux, it can’t escape being just like every other distribution: 95 percent identical with the other 5 percent devoted to installation, update, and administration.
And, unless that code isn’t GPL’d, everyone in the world can use it whether or not they run UserLinux.
Given the nature of open source, it’s unlikely that we’ll see any compelling applications developed by any single distribution, because that shiny new GPL’d app would give them a market edge for about 5 minutes.
“UserLinux smells a lot like the current crop of Debian-based distributions that somehow manage to survive the inclusion all that “risky” code Debian says is unstable.”
Migration of software into Testing from Unstable is bug-driven. Bugs of a release-critical severity (http://bugs.debian.org/release-critical) will stop packages from migrating until the issue is resolved. Affected software is not “said to be unstable” – it’s *proven to be broken*. This proof is always publicly available in the BTS.
Sure, Debian has high standards. But if something “managing to survive” is your sole criterion for judging quality, it really sounds like pretty much anything would be good enough for you.
That’s one of the points of the GPL. Innovation gets moved out so quickly, that you have to innovate to survive. If you don’t, then you die. Evolutionary development at its finest.
As for UserLinux, it has a few of very important things that none of the other Debian-based distros do:
1) A very high-profile leader.
2) Extensive corporate backing.
3) A lot of development cash.
These will probably be enough to get UserLinux in the top ranks, alongside Fedora, RHEL, and SuSE.
Perhaps we’re evolving to a point where there will be no more distros, they’ll all be like Linux From Scratch where you simply mix and match however you want it and simple enough where a newbie could do it…
Wasn’t Progeny Debian meant to be something like this?
Syntaxis:
What I’ve never understood about Debian is why code that the rest of the Linux world uses without apparent problems is labelled as “proven to be broken”. If that’s the actual situation, why aren’t non-Debian Linux users screaming about their broken distributions?
Rayineer:
I’m not so sure about the connection between the GPL and innovation. Commercial distributions have no financial incentive to develop unique high-quality applications under the GPL because they will never be the sole source of those applications.
Non-commercial developers will likely develop whatever happens to interest them. Since the interests of software developers do not necessarily parallel the interests and needs of home desktop users, most desktop apps in the Linux arena are very closely modeled on successful commercial software. Meanwhile, famous open source software like emacs, apache, perl, etc., have no relevance in the home desktop space.
To me, that’s an issue: There’s no incentive for the commercial side of Linux to develop new kinds of applications for desktop users, and the non-commercial side of Linux is populated with developers who don’t want what desktop users want. If the commercial coders aren’t going to write those applcations, and if the non-commercial coders aren’t going to write those applcations, who will?
I’ll grant you that Perens has some noteriety within the Linux community (but none outside it), and that he can probably line up cash and corporate backing. But none of that, I think, means that Perens’ distribution will offer any applications that aren’t available elsewhere. It can’t so long as it’s GPL. Absent those apps, what’s going to distinguish UserLinux from any other distribution?
Each new distro takes (on the order of) the same amount of work, but adds less and less to the community on the whole.
Of course, not everything is about “the community”. There is much to be said about “scratching the itch” – but unfortunately, many of these itches are born out of arrogance. There’s a lot of trying to one-up other distros, and we get a bunch of half-assed distros.
It’s incorrect to think that Linux development is all about the developers. If the users disappeared, so would the development. It is about serving the end user (OF WHICH the developers are a part of, but do not supercede).
I’ve used many distros – from Slackware to Red Hat to SuSE to Mandrake to Gentoo to Debian to Libranet, and a couple others. I currently use Gentoo, but in the long run, I wish that a lot of the work among various distros would merge and create maybe a trio of very sound, less flawed distros. We have problems like Mandrake releasing a point release full of bugs (NOT counting the LG drives flaw – that’s the hardware’s fault). Many distros have “gotchas” at some point that shouldn’t happen.
I like being able to check out different Linux distros, but I would like even more to have fewer distros that work better. SuSE detects my printer networking great, but can’t handle my humble wheel mouse. Others handle video, mouse, networking, printer, etc with varying levels of success. Anything can be configured manually, but one of the main differences between distros is the install routine, of which hardware detection is a big part.
Fewer, better distros would only be a big positive for every area of Linux use, from desktop to enterprise. The only people that lose out are the arrogants that want THEIR baby to lord over everyone else’s baby.
i’m not convinced. you have opinion and nothing backing it up.
-an NT/2k/Redhat admin
what difference does it make?
the question i ask is: who here is tired of working for large companies, where management has it’s head up it’s rear end, and you are surrounded by people who play politics, don’t care or are just collecting a paycheck?
i was. so i started my own business.
part of that business is supporting MS environments…and part of my job is supporting Unix/Unix-like environments.
guess which one pays more?
guess which area is impossible to swing a dead cat without hitting an mcse, or windows experienced techie?
it’s clear to me folks. the writing is on the wall. the dot-bomb kaboom was all i needed to see the freaken masses of unemployed M$ programmer and support types.
here in san antonio, you could be in a pool of 300 applicants for a microsoft domain admin, beat ALL 299 other applicants, and your 7 years of admin experience, your NT4.0 mcse, and 2k mcse will net you $15/hr.
i’ve seen it happen.
want to talk capitalism?
i’m living it.
From the previous post;
‘here in san antonio, you could be in a pool of 300 applicants for a microsoft domain admin, beat ALL 299 other applicants, and your 7 years of admin experience, your NT4.0 mcse, and 2k mcse will net you $15/hr’.
I do not agree with any of this, if you worked and PAID for your own college education like I did and have a degree and experience then this does NOT apply. It all depends on the person, work ethic, character ect. MOST professions will pay for the certifications, because they want to maximize your potential. Where I work, they pay for your certifications, such as Oracle, MSCE, RHCE, ect….. And a degree is the field you work in is required.
Does the prefect desktop GNU/Linux exist?
No.
So where is the harm in another group of people having a go? This could be a very, very good product. I suspect Bruce Perens knows what he is doing here.
‘Sorry, but there is absolutely NO way he will be able to produce a distro that will be able to compete with RedHat’
Compete how? uhhm packages (11,000+), architectures (x86/68k/spark/mips(el)/alpha/pa-risc/arm/powerpc/ia-64/amd64, kernels (free/hurd/linux/net). I think debian wins in all of those categories.
Oh you mean mindshare ala windows don’t you?
pfft.
“What I’ve never understood about Debian is why code that the rest of the Linux world uses without apparent problems is labelled as “proven to be broken”. If that’s the actual situation, why aren’t non-Debian Linux users screaming about their broken distributions?”
Because the part of the world that truly cares is already using Debian? Or the non-Debian users just don’t know what they’re missing? 😉 Lol! But, more seriously:
I think we’re talking at cross purposes. My bad, I didn’t make it clear. A release-critical bug does not necessarily render the package unusable; it merely indicates that it is not of a suitably high standard for inclusion in a Stable release. This is an important distinction to bear in mind. See http://people.debian.org/~ajt/sarge_rc_policy.txt for the specific criteria.
You again make it sound as though a release-critical bug filed against a package is a form of slander, an unfounded accusation. To the contrary: the proof is always there for all to see. Check the BTS (http://bugs.debian.org/release-critical) for specifics, as I said.
Syntaxis, I’m not trying to slander Debian. I’m just curious why Debian seems to feel the need for higher standards that the rest of the Linux market and community appears to think aren’t necessary. Why are other distributions including code that Debian says have release-critical bugs, without apparent negative impact?
I’m not criticizing Debian’s standards; that’s their business, not mine. But it does seem at odds with Perens’ intention to base a desktop-oriented distribution on Debian, since much of the software he presumably needs to include is not in Debian’s stable release.
“I’m not trying to slander Debian.”
Ok. 🙂 I didn’t say you were.
“I’m just curious why Debian seems to feel the need for higher standards that the rest of the Linux market and community appears to think aren’t necessary.”
I’m not really sure what you’re asking. The level of integration and quality control is IMO one of Debian’s best features, and the vast number of people who use the distribution (whether stock Debian, or one of the many distributions that are Debian-based) reap the benefits of it.
“Why are other distributions including code that Debian says have release-critical bugs, without apparent negative impact?”
Debian includes said code, too, in the Unstable/Sid distribution (which is, despite the name, really quite usable). It’s just prevented from migrating into Testing whilst rc bugs remain open against the package(s).
But, nonetheless, possible answers to your question (why other distros may disagree with Debian over whether a given piece of software is suitable for a *release*) include:
a) the bugs don’t affect their particular set-up, or else they’ve judged that the number of users who *will* be affected is acceptably small
b) the bugs are wrt breaking binary compatibility with earlier iterations, or other things which the other distributions may not care about
c) the release-critical bug is against the *package* and not the code inside, e.g. serious violations of packaging policy – other distros have no equivalent of the Policy Manual (http://www.debian.org/doc/debian-policy/)
d) they feel that the bug is offset by a new feature or bugfix
e) they include the software when it first comes out, so they can’t even *know* what bugs there are in it
f) they don’t even bother packaging/including/supporting it, and just let third parties (e.g. FreshRPMs) do all the work
g) they want people to have to buy the Enterprise version of their product to enjoy real stability
h) they assume that if the upstream developers deem it suitable for release, it must be
I’m sure there are plenty of other possible explanations, too. To be honest, though, the question’s really just too broad in its scope. One’d have to check out each piece of software on a case-by-case basis, one rc bug at a time.
“But it does seem at odds with Perens’ intention to base a desktop-oriented distribution on Debian, since much of the software he presumably needs to include is not in Debian’s stable release.”
Why is this at odds? Such a distribution would be free to pick and choose packages from the Stable, Testing, Unstable and Experimental trees as the developers see fit, just as the existing Debian-based distributions already do.
Debian says “testing” is “not completely tested and has no official support from Debian security team” and “unstable” is under “active devlopment”. That would lead me to believe that Debian is telling us that neither testing or unstable is approrpriate for reliable server or desktop use. In other words, that using stable involves forgoing current software in return for acceptable reliablity. Why do they take that stand, when clearly much software that Debian tags for testing or unstable is used, quite reliably, by many, many Linux users. If I understand you correctly, this is partly attributable to delays internal to Debian.
If Debian means for us to understand that they stand behind the “testing” release as reliable and appropriate for daily use, they should say so. If they believe that the inclusion of many contemporary and up-to-date programs in their “stabe” release is delayed because these programs are insufficiently reliable, they shoudl say so. If those programs are delayed because Debian’s wheels grind slowly or because of something related to Debian’s packaging scheme, they should say so, too.
Re: Perens’ release — If he takes software from outside the stable release, it will, by Debian’s definition, not be ready for reliable use. Unless he, or someone else, writes some new and interesting software for the release, it can’t be anything more than “just another distribution” with yet one more kind of installation routine.
you don’t agree. that’s fair.
but you mention you paid for your own college…so that’s different.
i’ll agree that my description is not the majority of cases, but 9 months ago, it was looking bleak for a lot of my friends, and people I knew.
Currently, I see no end to the flood of M$ technical people into a market that is no where near the size pre-dot-com bust.
there will always be consulting for dynamic ppl, but even that piece of the pie isn’t that big anymore.
that’s they way i’m calling it. you can disagree…but i’m moving my chips (in advance) to a different number.
not to mention, i enjoy it more.
Sorry, anyway crap no wonder linux can’t compete with Windows. All the linux guys yell that their distro is the best. How about getting some standerds?
“If they believe that the inclusion of many contemporary and up-to-date programs in their “stabe” release is delayed because these programs are insufficiently reliable, they shoudl say so.”
They do say so. The entire triple-tiered Unstable–>Testing–>Stable migration system says so. In a nutshell:
Stable has a feature freeze policy. With few exceptions, the only things to make it into Stable between releases are backported security updates and critical bugfixes – no new functionality is permitted. It is stable as in “unchanging” and “constant”, as well as in “rock solid”. This is part of what makes Debian such a breeze to maintain and administrate.
The migration of packages from Unstable –> Testing, which will become the next Stable release, is bug-driven. Whether the problem lies within the upstream code itself or is related to the packaging is immaterial; either way, the package is of insufficient quality for inclusion in the next Stable release.
“Why do they take that stand, when clearly much software that Debian tags for testing or unstable is used, quite reliably, by many, many Linux users.”
You’ve asked this exact same question already, and I’ve already answered. If my response did not satisfy you, feel free to ask a *different*, follow-on question to elicit clarification on my part. Don’t just ask the same question again.
“If he takes software from outside the stable release, it will, by Debian’s definition, not be ready for reliable use.”
As I have already stated, Debian-based distributions are free to incorporate packages from whichever tree they wish. Who’s stopping them? Knoppix, for instance, makes liberal use of packages from Unstable.
Did I miss something? If there are essentially four styles of distros out there (RPM, Apt, Ports/Portage and source-based), and if UserLinux is to be based on Debian (Apt), then how is it that this proposed distro can be considered a competitor to RedHat or Fedora? Seems like they’ll be competing against Debian, but maybe that’s just too obvious.
Can anyone clarify that for me?
They’ll be building on top of stock Debian, not seeking to replace it. The idea is that the Project takes care of all the packaging work already, leaving the derivative distributions able to focus their energies on differentiating themselves and adding value.
It makes sense for the developers of derivative distros to actively partake in Debian development, since the more they can collaborate with the Debian package maintainers, the less work they’ll need to shoulder themselves.
>> ” Knoppix, for instance, makes liberal use of packages from Unstable.” — says Syntaxis.
Yes, of course, but that’s the same software that Debian says isn’t good enough for them to release. Why will it be be good enough for Perens’ customers, but not good enough for Debian customers? Does Perens care less about shipping reliable software?
I still don’t get it. If the only Linux software that is reliable enough to use is the software that ships with Debian Stable (as implied by Debian), why aren’t all the other distributions crashing and burning? If they aren’t, why is Debian Stable behind the curve? Are Debian’s standards over-zealous?
“Yes, of course, but that’s the same software that Debian says isn’t good enough for them to release.”
They judge that it isn’t good enough *for them* to release, and the rationale (what constitutes an rc bug) can be found in one of my earlier posts. Others are free to make their own decisions.
“Does Perens care less about shipping reliable software?”
Please see my earlier post listing a number of reasons why other distributions may consider software suitable for use in a release whilst Debian proper does not.
“If the only Linux software that is reliable enough to use is the software that ships with Debian Stable (as implied by Debian)”
This has got to be a troll. In no way does Debian imply that all Linux software other than that included in the Stable tree is unusable.
“why is Debian Stable behind the curve?”
You cannot have a feature freeze without the software that is being frozen becoming out of date as a consequence.
“Are Debian’s standards over-zealous?”
I think the end product of the release cycle (the Stable releases) speaks for itself, and additionally I find Unstable to be a great desktop distribution. Debian’s dedication to quality is what makes it my distribution of choice.
I think we’re going to have to agree to disagree on this one. We’re both now making the same points over and over, with neither convincing the other.
All theory, philosophy and trolling aside, I gave Redhat a year, Mandrake a year, and briefly fooled with Suse.
Debian Stable is, hands down, the most bulletproof distro I have run on my servers. The quality control is easily identifiable. NOTHING I have wanted to do with stable was undo-able, and I have had none of the problems related to badly-packaged, badly-controlled software on other distros.
Not to mention that RPM is, on the whole, fairly frustrating to use.
I wanted a distro that did Samba+LDAP+Postfix+Courier-Imap+Amavis out of the box, was easily upgradable, would not ‘expire’ within a short period (a la Mandrake or Rehdat, if you don’t pony up for support), and would install and perform well on enterprise level hardware.
Debian stable did that for me, Redhat and Mandrake did not, would not, and now with Fedora, definitely will not.
That’s proof enough for me that the Policy nazis are doing their job, and that it has value. Everything else is allegorical.
Syntaxis: No problem. I recognize the reliability of Debian stable. Within a desktop a desktop context, however, I think that sticking with stable is too high a price to pay.
Starbane: Desktop users won’t be running Samba+LDAP+Postfix+Courier-Imap+Amavis. Desktop functionality, useability, visual design, and professional aesthetics count for much more. People who expect their computer to behave like an appliance (most people) and aren’t at all interested in what makes it work will form their first, and probably lasting, impression based on what they see on the screen. It’s rather like the way a lot of people buy cars.