Ransom Love's Answers About UnitedLinux

We posted your questions for Ransom Love, Caldera's CEO, on June 10. Here are his answers, which ought to clear up a few things about what UnitedLinux is doing, along with some discussion of Linux advances and how to (hopefully) make money selling Linux services and support. This is one of the best "CEO interviews" we've ever posted, with lots of straight answers and hardly any buzzwords or marketspeak. Lots of good insight about the "business of Linux" here.

1) LSB
by Anonymous Coward

From the description of UnitedLinux it seems to me that it is simply a commercialized LSB. How is what you are offering different from the LSB project?

Ransom:

LSB is a specification and does not describe a complete distribution. UnitedLinux is combining the efforts of four major Linux providers around the world to create a common, best of breed server implementation (full Linux Server distribution). It will be a full implementation of the LSB standard, but it will go beyond and include components not currently defined in LSB.

LSB does not attempt to solve the business model around Linux. UnitedLinux is not only providing packaged bits and bytes of a distribution, but also giving developers a global infrastructure of support in local languages and channels to sell their products on a global basis. UnitedLinux solves nearly all of the impediments facing the commercialization of Linux, such as market fragmentation, the availability of applications (by simplifying certification of hardware and software solutions), and a valid business model for Linux (Supportable business quality product through limited binary distribution and 12 month maintenance agreement with every license sold).

2) What will you give back to the community?
by dbarclay10

In a completely selfish vein, what will you give back to the community? Caldera doesn't have the greatest track record (I can think of a few specific cases but I'll omit them here for brevity) for providing some return to those people who have coded the _VAST_ majority of Linux, GNU, and everything else. Aside from, of course, providing jobs for developers.

Ransom:

I am a little frustrated by this question as it implies that somehow Caldera or any other Linux company is making millions of dollars off of Linux. Every Linux provider has spent far more on promoting Linux than they have ever received. Not even Red Hat is profitable and a lot of their revenues are generated from non-Linux technologies. Millions of dollars have been spent in recruiting applications, advertising, and tradeshows to promote Linux, not to mention the millions spent in employing engineers as well as the innovations that have been given back to the community. The actual development cost of producing a product is only about 20-30%; marketing, sales and support constitute the majority.

I am also troubled by your impression that Caldera's contributions to the community have been scarce. Caldera architected and paid Red Hat to enhance the original RPM. We were the first to address a graphical desktop; the first to provide a graphical installer; the first to provide a management system. I could go on and on. All of these innovations were given back to the community. You can see a more detailed list at www.caldera.com/developers/community/contrib/.

Please don't get me wrong - I firmly believe that what is produced by the community is vitally important. But it is not what creates a product or a solution that businesses have to deploy. Caldera has been a company for profit. Profit enables continued investment in Linux. UnitedLinux is for profit so that there is money that can continue to be invested in Linux itself. I do not believe in a Linux model that requires ongoing charity to survive.

That said, Caldera and all the UnitedLinux partners will continue to provide all of the changes and enhancements that are made back to the community. I believe that all of the participants have a very good track record of so doing. In addition to providing the source back, UnitedLinux will offer development programs that will provide continued access and updates for the serious developer.

3) Source and binary distributions
by RGRistroph

There has been some confusion on your statement in the UL teleconference to the effect that while source code would be available to meet the requirements of the GPL, "binaries would not be freely available." Could you clarify what that means? Is it possible that UL will distribute only source, or only distribute source and binaries to it's member companies? (Who will then be responsible for making sure they meet the license requirements on software which is in their distributions?) Surely UL or it's members don't intend to distribute binaries compiled from GPL code and assert the recipient can't re-distribute them?

Ransom:

The binaries that are certified by the major ISVs and OEMs will not be made freely available for distribution by anyone. This is to limit the support liability for these companies and to ensure a high quality, consistent product around the world for support purposes. The UnitedLinux product produced is not just a binary, but 12-months of maintenance. That maintenance is for a single system and therefore has limited distribution. The source code for the server will be made freely available for all in compliance with all of the Open Source licenses.

There will be programs for developers who need access to the binaries and they will include options for ongoing updates and patches to ensure continued certification compliance. Our desire is to make UnitedLinux easily available for serious developers, and give them means to make the development process easier.

4) Commercial Development
by Marx_Mrvelous

It seems to me that a group like UnitedLinux could bring a lot of commercial development to the Linux platform. Are there any efforts to bring companies who have so far neglected developing for Linux due to support costs, like most hardware venders, into UnitedLinux?

Ransom:

Certainly. By enabling one certification for hardware and software and then facilitating a global distribution of the solution through established global channels with support in local languages around the world, UnitedLinux should increase the number of hardware and software participants supporting Linux.

5) Future of Linux
by micro-colonel

Where do you see the true future of Linux being? Will it remain mostly in the enterprise and web server market, or do you think that it will also make large gains in the desktop market? Also, to what end does the goals of UnitedLinux fit into your predictions for the future of Linux?

Ransom:

Linux has great potential in moving beyond the web server market into the mainstream of the application server market. There is a lot of work that still needs to be done, however, to allow Linux to be a dominant application server platform. The objectives of UnitedLinux are to take that first step: enable Linux to be used by mainstream businesses. Accordingly, the initial effort is focused at the server.

I also believe that web services will become the dominant method for outsourcing IT. As the Internet becomes the primary business platform and Internet client interfaces become dominant, Linux will continue to make inroads into the client. (Take, for example, the fact that we now spend more time in email than in an Office suite.)

Finally, one of the keys to desktop penetration of Linux is in ease of management and provisioning. Making Linux easy to configure, deploy, manage and interoperate with Microsoft alternatives will greatly enhance Linux's acceptance at the desktop. Clearly, several of the UnitedLinux companies are addressing these issues individually. UnitedLinux may address this after enabling Linux to be used as a mainstream application platform for business. Another key, of course, would be the number of applications with which end-users are familiar. With the current balance weighted towards Microsoft, the need is for new software that makes end users more comfortable with and consequently more accepting of Linux on the desktop.

6) Business Model...
by powerlinekid

Mr. Love, I'm curious as to how you'll make money from this? By not giving away binaries it seems as if your group is trying to sell Linux, and probably service and support with it. Now you appear to be in competition with Red Hat (on server) and Mandrake (on desktop) who both give their software away. Red Hat makes it's money from service contracts and Mandrake from special software for paying customers. I guess my question is how can you compete against them, when they are just as good and give it away for free or cheaper? What is the incentive you will give consumers to actually purchase your software as opposed to downloading isos from other companies?

Ransom:

It should be noted, first off, that Red Hat has moved to a model on advanced server where they are not giving away the binaries and they are charging around $800+ for their advanced server product.

Going forward, there will only be two platforms certified by the major hardware and software vendors, Red Hat and UnitedLinux. For Linux to move from the peripheral of the business network into mainstream application server market, businesses must be assured that their platform is certified and will work with other applications and hardware solution in their environment. What the UnitedLinux customer is paying for is 1) the assurance that his applications will work together, and 2) the ongoing maintenance and support of that certified platform. The restriction on binaries is to ensure product quality and consistency of the brand for hardware and software vendors and for the quality of support within the business organization. I believe that Red Hat is moving to a similar model with business customers. The majority of the value will be in product assurance and maintenance. Both of these are of tremendous value to the business customer.

As mentioned earlier, binaries will also be made available to developers, but they will be through programs that can keep them updated and in sync with all changes. More information on this will be forth coming.

Desktop derivatives will be made available by each of the respective Linux companies. Since they will not carry the UnitedLinux brand and do not need application certification, their binaries may be available based on the individual company's policies.

7) Documentation
by forgoil

Will there be some form of initiative to work together on online documentation for both end users and developers? For instance making sure that there is up to date information on all applications and APIs in a common format (for example XML, that can be used to generate info, man pages, html, etc)? I personally don't think the distributions as a whole are well documented enough, and I think it would be one area where everybody would gain from co-operating.

Ransom:

We will be offering developer programs to address many of these needs and your suggestion here is a very good one. Clearly not having to duplicate efforts will allow us to create a much higher quality combined product and this is an important area to improve.

8) Who certifies compliance?
by Rogerborg

Who will certify compliance for each vendor provided distro, and who will pick up the pieces when (not if) an application appears that borks on one or more of the distros? If it's UnitedLinux, is each vendor prepared to pay to fix snafus committed by the others? If it's the individual vendors, what happens when one of them screws it up and wrecks confidence in UnitedLinux?

Ransom:

Every company will be shipping a common CD that will include a complete Linux distribution including installer and desktop. This is the UnitedLinux aspect of the distribution. All the additional value-add will be on separate CDs. Consequently, there will be a common quality check on the base components. The testing of the value-added components will be the responsibility of the individual companies.

9) Patents
by Rogerborg

Given the ongoing uncertainty over whether Red Hat's actions regarding patents will actually match its rhetoric, what is UnitedLinux's position on patents? Specifically which of the following will you do?

• Eschew patents altogether.
• Obtain your own patents.
• License, trade or buy outright patents from other companies.
• Oblige your members to hand over or license patents to UnitedLinux or to all other members.
• Match Red Hat's current stated intent and express a non-binding intention to stay enforcement for a given type of open source development as long as it is convenient for you to do so.
• Agree to explicitly license your patents at no cost, for a limited time or in perpetuity, to a given type of development (as sharply distinct from merely staying enforcement and leaving a Sword of Damocles dangling over developers' heads).
• Obtain and reserve the right to use patents freely against any target, as any other commercial software companies (e.g. Sun, Microsoft) would do.

Ransom:

The four companies have not discussed their position on patents. The official statement will have to be forth coming.

We live in a day when patents have not become a tool to protect, but a weapon to wield. Since the patent office lacks the technical expertise to discern between what is valid and what is clearly an attempt to blackmail, and because innovation continues at a rapid pace, this is a serious problem for our industry and one that will need to be addressed.

10) On the Relationship between Companies
by the-banker

How is the membership into the United Linux group going to work, and how much flexibility will there be? Can any distribution join? Are there significant costs to becoming a member? Can members set their own policies with regard to per-seat-licensing? In sum, how much freedom do the member companies have in how they market, contribute and license United Linux?

Ransom:

The membership will be open. Any Linux company will be able to join, but they will have to pay the fee to become a joint owner of the UnitedLinux LLC. That fee is to offset the cost of development and marketing the UnitedLinux product and brand and to have the upside potential of profits.

There is no per seat license for UnitedLinux. The restriction is per server and it is the 12-month maintenance fee for that system. Other companies can set their own pricing, but they are under obligation to deliver the same product and maintenance deliverable per system and will be responsible for the fees back to UnitedLinux. There may be several ways of participating with UnitedLinux. The details on membership and different options will be forth coming.

Comments

And I wanted to know about his name...

[Anonymous Coward]

Guess that question wasn't ranked high enough. :(

Re:And I wanted to know about his name...

[YOND R BOY]

I think its pretty obvious... he was named by a kidnapper and a hippy

Business-speak

[reverius]

Mr. Love's responses are so riddled with corporate jargon I can barely understand them... can someone translate please? :)

From what I can tell, he's so intent on selling a "secure, reliable, and hippie-free" product to other corporate-minded folks, he's fallen completely out of touch with the entire linux community.

This is the equivalent of Apple making computers without sound cards or high-end video, and Jobs saying that they think it's more "professional" that way (but ignoring the majority of Mac users who do audio/video work!)

Re:Business-speak

[Sheetrock]

Here's a tip: whenever a business is trying to tell you they're doing something for your protection or to continue the high level of service you've come to expect from them, watch your ass. As an example, the following response to "Why will source be offered for free but not binaries?"

The binaries that are certified by the major ISVs and OEMs will not be made freely available for distribution by anyone. This is to limit the support liability for these companies and to ensure a high quality, consistent product around the world for support purposes.

The question itself is typical from the community that has grown to embrace Open Source, but the answer is pure PR fluff. Keep an eye out for this technique; you'll see it in play whenever cable rates go up or a business is about to make it harder for you to cash a check.

Re:Business-speak

[bilbobuggins]

he's so intent on selling a "secure, reliable, and hippie-free" product to other corporate-minded folks, he's fallen completely out of touch with the entire linux community.

i suppose that would make the Linux community a bunch of insecure unreliable hippies then?;)

Re:Business-speak

[Sloppy]

This is the equivalent of Apple making computers without sound cards or high-end video, and Jobs saying that they think it's more "professional" that way (but ignoring the majority of Mac users who do audio/video work!)

Provided that more "traditional" variants on the Mac were still available (just as other Linux distros will remain available), I don't see what would be wrong with a move like that. The whole idea would be to get the Mac into a new market, not to sell that model to the people who already use Macs to do audio/video work. If Jobs were to try to sell a supercheap audio-less "BusinessMac" to people who run spreadsheets all day, I would applaud.

I think Love is up to the same thing. He's not interested in you "Linux community" people. You'll go on running Slackware or Debian or whatever. Corporate jargon, on the other hand, is the language that the people running NT know, and those are his prospective customers.

Re:Business-speak

[Thing 1]

OT: your sig:

My heart is human, my blood is boiling, my brain IBM...

When I first heard Mr. Roboto, I interpreted the last part as "my brain I B.M." (where "B.M." is short for bowel movement, i.e. "I shit out my brain").

Surprisingly, it seems that Love has done the same in order to rid himself of hippies. ;-)

Re:Business-speak

[HiThere]

In the early days of the Mac, Apple did do something rather like that. They wanted to differentiate the Mac as a business machine, as opposed to the Apple II/III, so they did various things that discouraged the porting of game. It didn't stop them, but it did slow them down.

This may be one of the factors in the dominance of the IBM PC. It sure isn't the only one, of course. Color screens were also an important factor. Here I think that the "anti-game" argument was used as a justifier when the real reason was the cost, and that Apple couldn't agree on how to integrate color into the Mac system. HyperCard never was properly converted.

But at first the Mac was intentionally designed to not be a gamers machine, so that businesses wouldn't think of it as a "toy computer".

Re:Business-speak

[HiThere]

The real reason is that under the GPL you are allowed to sell the binaries, but after you have done so, you are compelled to distribute the source (to those customers) for free.

It is interesting that his explanation is basically unintelligible, or possibly just silly, when there really is a sensible reason.

A subsidiary reason is that for many of their potential customers the thought of doing their own compiling is ... uncomfortable. So by selling the binaries they are catering to that market.

Another person has brought up the possibility that the binaries may also have the United Linux trademark embedded in them, which would separately restrict the right of any third party to redistribute them. A further barrier to competition.

I think that there are many obvious reasons for their choices, and his obfuscation is merely an indication of how trustworthy Mr. Love is.

Re:Business-speak

[Theodrake]

You people just don't get it. A company has to be profitable to stay in business. They are trying to create a model that makes money and meets the requirements of free software. They offer the software for free, but if you want certified support, they provide a binary. If you don't want support go get the source and build it yourself or find a binary someone else built.

And the idea that this restriction on the binary is a barrier to competition is just plain stupid. You have the source, build your own version and then sell support. How is this a barrier to competition?

Re:Business-speak

[warriorpoet]

Maybe I'm missing something here, but the answers above don't seem to bear any resemblance to obfuscation at all. The case he represents makes perfectly valid business sense - they're not distributing normal free binaries because they're basing a large part of their business on selling people discs with binaries on it that are guaranteed to work. What they're selling people is that guarantee - and the work that's gone into the verification and compatability testing/development.

I don't see where you get that there's anything wrong with this... if they were selling to companies the binaries that you and I normally download, then prohibiting re-distribution, then we would have a legitimate beef. And the entire argument about "anti-hippy" and not respecting the community, is in my opinion - no offence - bollocks.

The issue is that big companies buying servers and server OS' have to be able to say to the decision makers that a) we're buying something that is certified to run with the other stuff we have and b) come with full support and maintenance package. It's another way of representing a linux distribution company in a more corporate packaged way - which is what appeals to corporates.

I love the linux and opensource communities, and I prefer their ideals to any other in the industry. But I don't care what anyone says - the vast majority of major corporates who are going to buy server OS' will balk at buying anything that isn't fully guaranteed and warranteed and professionally supported. Nobody at the top of those chains gives a toss about the community benefit or anti-microsoft ideals or even security in some cases.

From the limited amount available to read about it, and the above answers, it at least seems as though UL are heading in the right direction for business. At the end of the day, they've guaranteed that any innovations or enhancements will be passed on to the community for free, and they're not going to force you to buy their distro, so what does it matter. Calling the guy a liar or untrustworthy just because he's finding a way to build a good business from linux distributions is insane - it's exactly this kind of thing we need if we're ever to get Linux over the acceptance hurdles it currently faces...

Just my .02

Re:Business-speak

[einhverfr]

The real reason is that under the GPL you are allowed to sell the binaries, but after you have done so, you are compelled to distribute the source (to those customers) for free.

And those customers can then redistribute those binaries for free. The only way they can get around this is by copyrighting the ISO images (but do these constitute a derivative work?) or by adding proprietart software (OK, so I strip out the proprietart components, remaster the CD-ROM and redistribute that?).

THis will either flop or the binaries will be redistributed shortly.

Re:Business-speak

[Sheetrock]

I wasn't thinking that big. I'm saying that the answer doesn't line up properly with the question.

Think about it: what kind of 'support liability' would these certifying OEMs be in for if people are pulling the binaries for free? Hell, I can drop a few bills on commercial software and the company I'm buying the software from feels no obligation to provide support; not without charging per call, at least. I'd be more comfortable if he said "We're not providing binaries because we expect to try to make money from this thing." because that's the simplest logical rationale (not necessarily correct, of course, but less likely to trip up my mind while reading it).

I'm not in any way trying to assert that he is being less than truthful, because I know very little about their business strategy or arrangements between the companies involved. I just wanted to point out that this answer didn't seem logical based on what I've seen in the industry in general and that for me this is usually a red flag to look at things more closely to ensure I understand everything that is going on.

Answer Two Inaccuracy

[SN74S181]

Mr. Love claims that Caldera was the first to provide a 'graphical' install.

That's certainly incorrect. Yggdrasil had a graphical install in the Fall of 1993.

I wonder how much of the other stuff he mentions in his answers is incorrect?

Re:Answer Two Inaccuracy

[einer]

How is this essentially correct? He's wrong! Un-Right! Incorrect! You're basically saying that "well, it wasn't first, but it was the first one 'of consequence' and the 'best one' (even though it wasn't first)." Well hell then! "Microsoft had the first gui!" (of consequence, of that quality, etc..etc..)

You're not even splitting hairs here!

12 month maintenance fee

[Senjiro]

What is that supposed to mean? I have to pay for the United Linux product to get upgrades (free with any other linux distro) and tech support (mailing lists, faqs, deja, etc)? I must assume from this model that they are really only targeting the numb masses of people who don't like the nuts and bolts of linux. Seems clear that UL will become not a distro for linux users, but a crossover distro for people who don't like to think.

Re:12 month maintenance fee

[Jeppe Salvesen]

If the quality is higher than the free alternatives, there will be people willing to buy for that extra reliability.

Re:12 month maintenance fee

[GrenDel Fuego]

Not exactly.

They're targeting corporations. Companies that can afford to pay large amounts of money to have someone else worry about the operating system for them.

Much like Redhat's enterprise edition.

Re:12 month maintenance fee

[thesolo]

I have to pay for the United Linux product to get upgrades (free with any other linux distro) and tech support (mailing lists, faqs, deja, etc)? I must assume from this model that they are really only targeting the numb masses of people who don't like the nuts and bolts of linux. Seems clear that UL will become not a distro for linux users, but a crossover distro for people who don't like to think.

You're missing the point entirely. UnitedLinux is meant for businesses. (Whether or not you equate businesses with "people who don't like to think" is another topic altogether!)

A large corporation is not going to want to be told to check mailing lists and newgroups. A large corporation is going to want 24/7 phone support, and guaranteed system integrity and maintenance. This is a good thing! After all, it's what Microsoft tries to sell to their business customers. (Whether or not they deliver...again, another topic) Linux needs something like that to ever make it to some large corporations.

Re:12 month maintenance fee

[spudnic]

Apparently the corporation you admin for isn't very big. In almost any organization that depends on it's computer resources, a fee per server wouldn't even raise a question. Would I pay for UL when I could get Red Hat for free? Only if there was some credible reason to do so. But I assure you, If UL was more stable, or robust, or whatever, I would be the first in line to pay for. Most professional sysadmins would.

When I worked for small businesses, I ran Linux on their servers to save money. Now that I'm no longer in that situation and have a little money to spread around, I still choose to run Linux because it's the best tool for the job.

Re:12 month maintenance fee

[fiftyfly]

Not exactly. They're targeting corporations. Companies that can afford to pay large amounts of money to have someone else worry about the operating system for them.

Not only that but companies _want_ to pay someone else to do that kind of maintenance. Why? because a fee is a _known_quantity_ - you can budget around it. The last thing the comptroller wants to hear is "Oh, I don't know how much the TCO will be, who cares it was free! Now pay me some value, for an unkown number of overtime hours, to keep it running".

Re:12 month maintenance fee

[Tony-A]

Why would I as a corporate admin, ever choose to pay for what I can get free?
The blunt answer is so that it keeps working next year and the year after and so on. So that if/when you do have a problem you get an answer better than "Fix it yourself" or "Read The Fine Manual". Basically, corporate wants to pay so that if/when they have a problem someone is interested, and not just because it's an interesting problem. The free support may even be better than paid support, but if nobody is interested, you're SOL.

Re:12 month maintenance fee

[lysurgon]

I must assume from this model that they are really only targeting the numb masses of people who don't like the nuts and bolts of linux.

Bingo! Give that man a ham. Obviously this is what they are doing. They're trying to take Linux and sell it in to big business server markets. You don't get united linux for your hack box at home, you get it when you need to run a robust server and you don't have your own crack team of propellorheads to build it up for you. They're attacking microsoft and sun's server territory, and unfortunately that battle is fought on M$/Sun's terms: biz-speak, FUD and memebers only fee-based support.

Until we "community types" (damn hippies) get our collective act together and break down the massively inhumanizing and undemocratic loci of power known as corporations, this is what it will take to get said institutions to use GNU/Linux. In the best of all possible worlds it will work like a seed from within, but that's probably a little too optomistic, since as you point out the whole point of UL is to releive the client of the need or inclination to look under the hood.

Me? I'm a borne hot-rodder. If I make it big I'm buying an old Mustang I can repair myself. But the prevailing coporate culture favors the Lexus. Ever seen a lexus engine? It's a typical feat of blackbox engineering. No way you could fix it on your own, and why would you want to. So sad that so many people never get to feel that power of creating something, fixing something, hacking something with your own ingenuity.

Re:12 month maintenance fee

[gol64738]

I have Win2k boxes under my control with 400 days of uptime, but average about 95% availability (bandwidth, software defects from 3rd parties, etc).

huh? how do you apply all of your security patches on win2k without rebooting??
i have a mandrake linux machine with almost 500 days uptime, and the only thing it needs is a kernel upgrade. oh, and this machine runs NIS/NFS, samba, apache, sendmail and cvs for a 20+ man company... so this machine isn't exactly idle.

Anyone who claims to offer five 9's availability for an applicatio without taking alot of your money first is probably smoking something - it requires redudant power, physical locations, bandwidth, cooling/heating, best of breed software and administrators, and a healthy dash of luck.

well, i agree that hardware plays a large part in total availbility time, but so does the software (OS). we have recently installed vmware's GSX server on an IBM netfinity server. we have 3 virtual win2k servers running under linux. the other day, one of the win2k machines locked up after 3 weeks of uptime, so the virtual machine had to be restarted. when our IT department heard about this, most of the windows admins were like, 'cool, 3 weeks without a reboot.. we rock!'
in other words, the windows admins were not surprised that win2k only lasted 3 weeks before a dump.

on a side note, vmware's GSX server completely kicks ass, and i highly recommend it to all linux admins that have whiny windows IT depts breathing down their necks.

After showing them usuage charts, they realized. Our customers all use the system during normal business hours, all in EST time zone, all very predictably. After a few late stragglers NO ONE had ever logged in between the hours of 1:30 am and 4:30 am. Ever. It had never happenned. So that became maintenance time, everday. Instead of rushing off closely timed scripts to make maintenenace work, I spread it out, did a bit more maintenance than needs to be done daily.

jesus christ, daily maintenence? you really are a windows admin, aren't you? in linux, it's sometimes hard to get a service configured just right (sendmail anyone?).. but once you do, you never have to mess with it again.
using windows, it's easy to configure anything, but then you find yourself reconfiguring over and over and over..how disgusting..

The bottom line is that very few places need five nines. Very, very, very few. And you're not going to get it for free (or for $800 from redhat).

no, you won't get it for free, but i guarentee you can get it cheaper using linux than with windows. that really is the bottom line.

Re:12 month maintenance fee

[Tony-A]

for people who don't like to think.

More accurately, for people who have better things to think about.

Re:12 month maintenance fee

[RatFink100]

You're right - my bad - pesky percentage sign confused me!

9 hours (8.76) is only 3 9's

ClueStick this is MrLove, Mr.Love meet ClueStick

[SubtleNuance]

But it is not what creates a product or a solution that businesses have to deploy.

I hope the dark irony of that statement doesnt escape this audience. The fact is GNU/Linux was created exactly BECAUSE of this concept. Sad, Ransom Doesnt Get It.... or he thinks we are all stupid.

What a wanker

[photon317]

He dodged most of the important questions. What little of his views came to light showed that he is just trying to reap (or is it rape) profits off the backs of the open source community. Profit motives have a place in the Linux community, but this isn't the way to do it.

UnitedLinux and Sun are both going the wrong way here. Heed the advice that has been around for years now - Give Away the Software (that means code and binaries, in a usable form), and Sell Hardware, Support, and Maintenance.

Re:What a wanker

[photon317]

Or... perhaps it's not about being established, but just about doing things right. Perhaps Redhat (especially having no hardware) and LNUX just weren't doing it well. IBM is doing a good job so far, they're poised quite well. Sun is more than capable of doing it better than IBM, but they refuse to.

The past and current failures of a handful of companies don't invalidate the idea that software should be commoditized, and that computer companies should make money from hardware/support/maintenance. They may not make the stellar killing that Microsoft made, but then again they probably won't end up being Convicted Criminals either.

Re:What a wanker

[Z4rd0Z]

What is up with you people? I honestly cannot see the problem with what this guy is doing. The GPL does not require people to give away binaries of their products. The source still is available. Use a different license if you don't like it. Use a different distro if you don't like it. But, for Christ's sake, stop acting like paranoid schizophrenics. This guy is not out to get you!

Re:What a wanker

[photon317]

He's not out to get anyone, and I didn't imply it. He is out to exploit the fruits of others' labor. He is a hollow shell marketing droid, another symptom of the empty american economy, which creates fake wealth out of thin air. I'm putting him down because he and his type are lame and statistically doomed to failure in the long run, not because he poses any direct threat to me.

Re:What a wanker

[Sloppy]

He's not out to get anyone, and I didn't imply it.

Then quit using words like "rape."

Re:What a wanker

[photon317]

Take your own advice. I can't believe someone with a sig that claims that can't code and have only used linux since '99 is giving me advice. You'll understand some day I hope.

Re:What a wanker

[photon317]

I've been an avid hacker and coder on linux for about 9 years now, I think that fairly well qualifies me for having used linux and followed through.

Buisness model....

[$carab]

First off, theres a formatting error in Question 4...needs to be bold and italicized, etc.

Pardon me if im wrong, but if that interview was viewsed as very "straight up", then /. must have been interviewing advertising pamphlets in the past.

Dont see what I mean? Take this phrase from the answer to question 4- "By enabling one certification for hardware and software and then facilitating a global distribution of the solution through established global channels with support in local languages around the world,". "enabled" "facilitating" "global".....at least he didnt say "paradigm".

Desktop derivatives will be made available by each of the respective Linux companies. Since they will not carry the UnitedLinux brand and do not need application certification, their binaries may be available based on the individual company's policies.

I think that Mr Love's defence of his Buisness Model is very weak. I appreciate the Red Hat/Mandrake/Gentoo. etc. ability to download an ISO, and then burn it onto a CD. Not FTP'ing the files, not being given a limited time trial....And I thought that it was interesting that United Linux, essentially has forsaken the Desktop. It looks to me like United is just the last ditch alliance to stop Red Hat from complete buisness market domination. The companies making up United Linux dont really have the best Desktop/ISO policies, and as far as I could see there was no asssurance at all that the Desktop Distros would be LSB compliant. I came away from reading this very underwhelmed....

Just my .02$

Re:Buisness model....

[dublin]

It's really pretty hilarious to hear the /. crowd talking badly about people talking in buzzwords! Have a look around - most of the people here seem incapable of having a conversation with ordinary people. (Most of who are more intelligent, although less technically inclined that the self-righteous crowd that hangs out here.)

Believe it or not, there *is* a real world out there, and people live and work in it. It's that world that produces the surplus that gets wasted on salaries for ungrateful Stallman slaves.

It's crap like this that really makes me wonder if supporting Linux is a good idea. (after many years of supporting Linux, I am seriously pondering whether it is wise to continue to do so.)

If Ransome Love had any sense, he'd recognize this crowd of backstabbers for what it is and go pursure his plan based on BSD, where at least he has a chance at being successful. Linux is doomed to play along the edges of the market so long as the "community" insists on bludgeoning to death anyone that tries to bring it sucessfully into the real world.

Re:Buisness model....

[GigsVT]

We don't use buzzwords to hide our intent like corporate types do.

Some of us use buzzwords to save time, or to sound smart, but never, ever, to hide our intent.

Insightfulization of the message content

[PD]

These are actual quotes from the article. People who talk like this can only relate to two things: other people who talk like this, and their huge egos. Clearly, Ransom Love has fought no battles, written no code, has not earned any glory for his name. His heart is not truly Klingon.

create a common, best of breed
giving developers a global infrastructure
local languages and channels
sell their products on a global basis
Supportable business quality product
address a graphical desktop
innovations
By enabling one certification
facilitating a global distribution
solution through established global channels
hardware and software participants
dominant application server platform
dominant method for outsourcing IT
Internet client interfaces become dominant
continue to make inroads
ease of management and provisioning
peripheral of the business network
quality and consistency of the brand
quality of support within the business organization
a much higher quality combined product
to have the upside potential of profits.
maintenance deliverable

Re:Insightfulization of the message content

[Jerf]

I'm gonna have to disagree.

If you want to swim in the financial waters, you need to speak the financial language.

Being able to sling around financial buzzwords no more implies technical incompetence then being able to sling around Linux buzzwords implies programming competence. (Not to mention that it is not a given that technical competence is necessary to run a technical business; it's just so few CEOs, or indeed people, can stand to trust the tech people enough to make good decisions just based on their word.)

While it can annoy people like us to slog through a language with a lower informational density (caused by the higher density of words meant to signal things to the listener's subconcious like 'conformity' and 'if I sound like this, I must know what I'm doing', annoying to us but CEOs who don't use this language aren't CEOs for long), the true test of the value remains the content of the speech.

On that measure, this CEO did quite well, and your harshness is incredibly unfair, and your moderation probably undeserved. They equally justifiably think the same thing about people who are only concerned about tech, and never the business. ("What actually puts food on the table? Do the techies think they can eat their server?")

Look past the surface.

Re:Insightfulization of the message content

[kitzilla]

I've translated a few of these jargonites:

*create a common, best of breed* It'll have a cool box.

*giving developers a global infrastructure* We'll have a cool website.

*Supportable business quality product* It'll have a cool box.

*innovations* It'll have a cool box.

*facilitating a global distribution* We'll have a cool website.

*quality and consistency of the brand* It'll have a cool box.

Re:Insightfulization of the message content

[Zathrus]

Yes, the parent was modded as funny (rightfully), but...

People who talk like this can only relate to two things: other people who talk like this

Welcome to the real world. The real world does not consist of coders who ignore everyone else in their insular little universes and think that everyone should be as smart as they are. There's a reason that there's more MBA's than engineering degrees out there -- more people think in a management/marketing perspective than an engineering one.

Is Mr. Love spewing a lot of horseshit? Sure. But it's horseshit that other horses will sniff, recognize and find comfort in rather than the bullshit that a lot of the Linux community spouts. Bullshit is just confusing to horses - it smells kinda similar, but the nuances are missing and it makes the horses nervous.

Insert sheep or lemmings for horse if that makes you feel better.

It still doesn't change the reality that Linux is still far, far away from serious corporate acceptance. IBM, Redhat, and other companies are making headway here (hooray), but it's largely through the same kind of crap that Mr. Love is talking. It's not based off some ethical superiority, and it sure as hell isn't because businesses think OSS is a good model for their core logic.

We'd love to move our systems over to Linux here... but upper management starts looking nervous every time it's mentioned. So instead we're on AIX (gack) and horribly overpriced PPC boxes. Maybe once Linux gets some more positive press from companies like Redhat, IBM, Caldera, etc. we'll finally be able to move to something better than AIX.

Re:Insightfulization of the message content

[dvdeug]

If you want to swim in the financial waters, you need to speak the financial language.

Slashdot isn't financial waters. Any good speaker knows that you pick your language to match what the audience will respond to. If you can't speak techie, then you probably can't think techie. If can't figure out when it's inappropiate to speak financial, you probably can't figure out when it's inappropiate to think financial.

Re:Insightfulization of the message content

[warkda rrior]

Ransom Love had his anwers posted into a widely-watched public forum (i.e. Slashdot), so he had to be careful what and how he says what he says. Anything he mentions publicly can affect UnitedLinux and/or Caldera in the eyes of the business world. Adapting to the audience is not really possible, at least not to a large degree.

Re:Insightfulization of the message content

[Snootch]

In fact, as far as Tech's who will recommend purchasing of UL, I think those are pretty few and far between as well.

I'm still not sure you get it - this is not aimed at us! By the time you've got your average Slashdotter on your staff (and certainly by the time you've got someone sure on their feet enough to raise a ruckus at something like this) with enough influence over the purchase, you've got someone who won't mind maintaining a Debian/Slak/Pink Tie/whatever install for you, and is trusted enough to do so. This is aimed at corporations where managers are making the purchasing decisions.

Re:Insightfulization of the message content

[Zathrus]

I'd counter by saying that Mr. Love wasn't an appropriate selection for a Slashdot interview... he's a CEO and was chosen (by Caldera) to speak CEO-ese. Is the talk inappropriate for techies? Sure. But that's not his job. We're not the ones he needs to convince that Linux is good.

BTW, I'm a coder. I have absolutely zero aspirations to be a manager. I'd royally suck at being a manager. But I've had good managers (and CIO's) in the past and know their value. I've also worked in Dilbert-esque environments and know that not all managers are worth the air they breathe.

Re:Insightfulization of the message content

[extrasolar]

"There's a reason that there's more MBA's than engineering degrees out there -- more people think in a management/marketing perspective than an engineering one."

I honestly thought that reason is because any business degree is a couple order of magnitudes easier to get than an engineering degree.

"It still doesn't change the reality that Linux is still far, far away from serious corporate acceptance."

Caldera may be begging for corporate dollars but I don't want anything. Corporations are strange things that I don't think geeks understand. Not because they are efficient beasts but because when they fall, they tend to land in the strangest positions.

Re:Insightfulization of the message content

[Jerf]

Can you speak both financial and tech speak? I can't.

Insisting on only listening to people who speak your exact sub-dialect is a great way to fatally narrow your worldview.

Re:Insightfulization of the message content

[dvdeug]

Can you speak both financial and tech speak? I can't.

No. But I'm not in charge of a Linux corporation - Ransom Love is. It's not an important skill for us, but it is for him.

Re:Wheres Question 5 and 8?

[bje2]

you didn't read very closely...it's actually 4) and 8) that *appear* to be missing, but if you look more closely, you'll see that the HTML is just screwed up, so the headings aren't bolded, and they are missing a line break right before them....just look right at the end of questions 3 & 7...

Business-speak to Programmer-speak translation

[Mr. Sketch]

Allow me to translate a summary of all his answers into simple programmer speak:

bandwagon[Linux].push_back(RansomLove);

For anyone

[llamalicious]

questioning fees or the idea of paid binaries:
If you don't work in a large corporate environment with some heavy iron or reliability/QoS contracts with customers, shut your yap now and read on.

Allow me to explain.

Current primary linux distros (RH, Mandrake et. al.) provide a set of binaries and source for multiple target platforms.
Each of these "should" work, or will "probably" work on a given platform, and the free updates, or access to source from which you can upgrade your system will also probably work; none of these distros go through any heavy-duty regression testing or certifications for uptime guarantees or reliability.

What Mr. Love is getting at, and the idea behind UL, is to provide a serious platform for linux in the server market.

By maintaining strict control of the binaries (both Linux base and applications) UL can fully test and certify that "out-of-the-box" the software and server combos are completely compatible. No guesswork. No hoping that source compiles and runs ok the first time on platform x,y or z.
They can also begin to provide guarantees on the availabilty of critical applications.

That's what business customers are going to pay for. They'll demand nothing less. To get something for free in a quality-controlled environment is useless if you aren't 100% certain that it will all function as-promised.

Re:For anyone

[stipe42]

Working at a small company with only a couple servers and over a hundred clients, I can testify to how much money we would be willing to pay for binaries guaranteed to work because they have been compiled for our exact setup without having to actually compile them on our setup.

stipe42

Re:For anyone

[wfrp01]

By maintaining strict control of the binaries ... UL can fully test and certify that "out-of-the-box" the software and server combos are completely compatible.

There is certainly a value-add here, and I don't question the validity of this business model. What I'd like to know, though, is whether these "guarantees" and "certifications" imply that UL will assume liability for product performance? If not, I don't see how this "certification" is anything other than yet another piece of paper with colorful scrollwork you can put in your office. Is this for real, or (Roblimo's introduction to the contrary) is this just more marketspeak gobbledygook?

Re:For anyone

[Ami Ganguli]

Current primary linux distros (RH, Mandrake et. al.) provide a set of binaries and source for multiple target platforms. Each of these "should" work, or will "probably" work on a given platform, and the free updates, or access to source from which you can upgrade your system will also probably work; none of these distros go through any heavy-duty regression testing or certifications for uptime guarantees or reliability.

I think this is a misconception. RedHat, and I think most major distributions, do extensive hardware testing. Whenever you hear about IBM/HP/etc. providing Linux on their servers, you know that they've made a deal to certify their hardware with the given distribution. So sure, if you buy a random server and load RedHat it may not work (although I've never had a problem on normal server hardware), but if you ask RedHat or your hardware vendor what's guaranteed to work, they'll have no problems telling you. In fact it's probably in the marketing material already. A few seconds of clicking on IBM's web site revealed this handy little chart.

The real story behind United Linux is the fact that the companies involved want to save development $$ and they had to come up with some sort of license that all the parties could agree to. More power to them - I hope they succeed. But they're not providing anything fundamentally different than what's already on the market.

Re:For anyone

[HiThere]

I'm sure that there is a market for software that is guaranteed to work. Provided that the guarantee is worth anything.

Unfortunately, most guarantees in the software world are basically guarantees that you received some bits on some medium. They make no guarantee as to what those bits are, or what they will do. Just that they exist, and will be replace (without penalty) if they don't. (Even this guarantee isn't always kept. I've gotten some blank CDs, with no replacement offer.)

So what is he proposing to back his guarantee with? And what is he going to guarantee? This is the kicker. And if he provided any answers, I didn't notice them.

Re:For anyone

[Nailer]

To get something for free in a quality-controlled environment is useless if you aren't 100% certain that it will all function as-promised.

MD5, or Red Hat's `check' option in the installer, works for me.

Here's where I stopped reading:

[ivan256]

I am a little frustrated by this question as it implies that somehow Caldera or any other Linux company is making millions of dollars off of Linux. Every Linux provider has spent far more on promoting Linux than they have ever received. Not even Red Hat is profitable and a lot of their revenues are generated from non-Linux technologies. Millions of dollars have been spent in recruiting applications, advertising, and tradeshows to promote Linux, not to mention the millions spent in employing engineers as well as the innovations that have been given back to the community. The actual development cost of producing a product is only about 20-30%; marketing, sales and support constitute the majority.

If marketing and sales are that high a percentage of your costs, you're doomed. What he means is that millions of dollars have been wasted promoting linux to the wrong people. As much as I like free beer and booth babes, you don't need to spend millions of dollars selling a free product to a bunch of geeks who already have it, and don't intend to buy support. Furthermore, they should take a lesson from microsoft, and have virtually zero support costs. All support should be billed by the hour. Somebody needs to stop giving this guy money.

The worst of it though is that he feels that we, the developers out here writing the actual code, owe him something because of all this "promotion" that Caldera has done. It's like the guys who used to wash your windshield at the toll booths and then ask for a tip. Did we ask for this "promotion"? No, so go away. At least now we know not to hold our breath when looking for community support from UL.

Linux will be sold the same way all other platforms are sold: by the applications. When there are applications you need that run on linux, you get linux. I see it every day as the purchase orders come in. No distribution provider out there looking to make a buck understands yet.

Re:Here's where I stopped reading:

[sigwinch]

If marketing and sales are that high a percentage of your costs, you're doomed.

To make money on infrastructure software, you have to convice hundreds of thousands of individual businessmen to buy a copy or two each. *They* control the purse strings, and if you don't convince them you die.

...you don't need to spend millions of dollars selling a free product to a bunch of geeks who already have it, and don't intend to buy support.

No shit. That's why they're going after businessmen who have money to burn and problems to solve.

Furthermore, they should take a lesson from microsoft, and have virtually zero support costs.

1. Microsoft spends tens of millions on Windows Update, support websites and documentation, and quality assurance to ward off the need for support.

2. Supported retail Win2k costs $200. Unsupported OEM Win2k costs $85. You do the math.

The worst of it though is that he feels that we, the developers out here writing the actual code, owe him something because of all this "promotion" that Caldera has done.

The single largest category of software is custom/customized business-automation apps. Getting Linux into businesses, and making them believe it is a valid enterprise platform, creates jobs for Linux developers.

Linux will be sold the same way all other platforms are sold: by the applications. When there are applications you need that run on linux, you get linux.

Yeah, applications like turnkey mail servers, heavy-duty DBMS clusters, network monitoring appliances, SMB/CIFS-compatible terabyte fileservers. Those are the sort of thing businesses are willing to burn money on, to the tune of billions of dollars. And the great thing is that they hate replacing a working system, so they'd rather come back year after year to keep that box running.

Re:Here's where I stopped reading:

[Hard_Code]

"As much as I like free beer and booth babes, you don't need to spend millions of dollars selling a free product to a bunch of geeks who already have it, and don't intend to buy support."

No. You spend millions of dallors selling it to corporate clients. That means your boss. Or your boss's boss. etc. Obviously they are not selling it to us who will just download an ISO and slap it on some box under the desk (or whatever we can get our hands on).

Re:Here's where I stopped reading:

[ivan256]

You've discovered the unspoken meaning of that sentence!

Corporate managers and decision makers don't go to LinuxWorld Expo or any such thing. Usually software speciffic conventions and trade shows spring up after software gains a market. At trade shows dedicated to a single software product, people who make 3rd party add-ons and accessories sell their stuff. When you want to sell linux to corporate clients that have never heard of it, you don't spen $100k on a party at LWE, you go to general computing shows, hardware platform shows, you do direct mailing campaigns, you hire a sales team with established contacts. Getting the geeks drunk gets you nowhere. Why am I saying this? Because these linux only events are where these companies are spending their money.

Screw this UnitedLinux crap, write some applications! There's no money to be made selling distributions, and there's no need to sink money into developing one when you can just use Debian. These "big" linux companies won't make any money until they realize that the OS itself won't earn them anything. They certainly won't get something for nothing, which is what it sounds like this Love guy is looking for.

Re:Here's where I stopped reading:

[ivan256]

2. Supported retail Win2k costs $200. Unsupported OEM Win2k costs $85. You do the math.

The bill still says that you have to pay by the hour when you call them up. The extra $115 reflects the lack of a 50% bulk discount (practically industry standard), and printed documentation.

1. Microsoft spends tens of millions on Windows Update, support websites and documentation,

Really? Show me where Microsoft spends 10s of millions on Windows Update. They're smarter then that. Besides, I hardly consider security updates "support", and I hardly consider $10s of millions a significant percentage of their overhead. It's less then their monthly engineering payroll.

The single largest category of software is custom/customized business-automation apps. Getting Linux into businesses, and making them believe it is a valid enterprise platform, creates jobs for Linux developers.

I've interviewed for, bid on, consulted on, or written code for almost 100 seperate custom software projects using linux in the last three years. Guess which percentage of them cared which distribution we were supporting. Only about 10% (None requested Caldera). Guess who the biggest promoters of custom linux development jobs are. Microsoft and Wind River. Nobody wants to pay their licensing fees anymore. They also don't want to rely on another company for support. Caldera hasen't, and UL will not promote many engineering positions for linux developers.

Yeah, applications like turnkey mail servers, heavy-duty DBMS clusters, network monitoring appliances, SMB/CIFS-compatible terabyte fileservers. Those are the sort of thing businesses are willing to burn money on, to the tune of billions of dollars. And the great thing is that they hate replacing a working system, so they'd rather come back year after year to keep that box running.

This is exactly why UL is barking up the wrong tree. Write some applications, and forget about your never was distribution. If they're not careful, IBM will be making all the money...

Re:Here's where I stopped reading:

[ivan256]

Where does it say that UnitedLinux is going to be sold to the "bunch of geeks" out there? UnitedLinux is not targeted at geeks! Isn't this a glaringly obvious point yet?

This will be a recurring theme in the rest of this post: It's not what he says, it's what actually happens that counts.

I understand what their target market is, and if they truly believe that, then they shouldn't spend a dime at LinuxWorld or the likes. That's not where the uninitiated corporate types are. They are wasting their money there.

Where does it say that developers owe UnitedLinux anything?

He doesn't say it, he implys it. He talks like we should be grateful or something. The only thing Love's corporation has done for me (I'm a for hire linux developer) is give me headaches. I've got one right now.

If anything, UnitedLinux is looking to entice those who haven't looked at Linux in the past. That in and of itself would make me happier as a developer!

Bullshit. They're only looking to line their pockets and salvage something from the wreckage. They're using this "united" crap to wrangle as much control of the platform away from the developers as possible so that they can have a stable business model.

And UnitedLinux looks to bring the corporate application developers on board who wouldn't even consider Linux before. Sounds like you're actually in agreement on this one!

I've talked to Love on the phone when I worked for Lotus Development. (Why they wouldn't let me talk to an engineer that could have then known about and fixed the problem, and instead put me on the phone with the CEO is beyond me.) He's great at making promises to bring initial support on board, but he doesn't deliver. He almost single-handedly turned IBM off to a Domino port to linux, even though all the code was already written. (He's also the one that talked them into it in the first place if I recall.) If I had to choose someone to attract corporate developers to Linux, it certainly would not be him. I wouldn't get your rally flag out.

Re:Here's where I stopped reading:

[deblau]

[Love]
The actual development cost of producing a product is only about 20-30%; marketing, sales and support constitute the majority.
[/Love]

[ivan256]
If marketing and sales are that high a percentage of your costs, you're doomed.
[/ivan256]

Sir, have you ever run a corporation? I have, and I support Mr Love's numbers. Furthermore, so you don't think I'm blowing smoke, I will give you an example.

Example: Coca-Cola (KO), FY 2001
URL: (EDGAR report 10-K, annual summary)

From the EDGAR report under Consolidated Statements of Income (p 57):

Gross receipts: $ 20,092 M (1)
Cost of goods: 6,044 M (2)
Gross profits: 14,048 M (3) [(1)-(2)]
Operating expenses: 8,696 M (4)
Operating income: 5,352 M (5) [(3)-(4)]

Line (4) can be further broken down into Selling Expenses and Administrative/General Expenses, which is done in the text of the Management's Discussion and Analysis of Operations (p 48):

Operating expenses: $ 8,696 M (4)
Selling expenses: 6,930 M
Admin/gen expenses: 1,766 M

For Coke in year ended 12/31/01, they spent almost $7 billion on selling their product. That works out to 80% of their total operating expenditure for the year. Since Coke ended up with income of more than $5.3B last year, I'd hardly say they're "doomed".

Re:Here's where I stopped reading:

[deblau]

Good point, for their primary product, you are correct. For the most part, their costs have been due to marketing their primary product. However, Coke still does R&D (their recent entry into the "flavored water" and juice markets, for example).

Nevertheless, I fail to see how the process of software development is any different, in broad business strokes, from developing any other product, save that the cost of goods in software production is much, much lower than manufacture (as a relative percentage of overall outlay). The reason being that there are no raw material purchases. Coke spent over $6 billion on raw materials, plant upkeep, actually making their product. That's a full third of their gross receipts. It costs the UL companies maybe some electricity for code compiles and a few other, incidental charges. Consequently, the relative percentage of marketing expense for UL should be a higher percentage of the overall expense.

I agree with a statement I assume you were making but didn't state: that a higher percentage of money than 20% should be spent on development in a startup. Nevertheless, it can't be too much higher than that, or UL will fall into the classic trap of great product that no one knows about. As I'm sure you know, this is a large contributing factor in the failure of a majority of companies run by techies.

As far as Coke's games in the market, I agree that they are in trouble, which is why they're branching out into the aforementioned flavored water and juice markets.

Re:Here's where I stopped reading:

[SectoidRandom]

Really? Show me where Microsoft spends 10s of millions on Windows Update. They're smarter then that. Besides, I hardly consider security updates "support", and I hardly consider $10s of millions a significant percentage of their overhead. It's less then their monthly engineering payroll.

Have you seen Technet? Sure the CD in a folder version costs you $1000/year, but all of that information is free on the web; technical notes, service packs, problem fixes, installation guides, deployment guides, migaration tools / guides, etc, etc. Which all cost's money.

If you havn't used technet, then frankly you have not used the single best support resource available for free in the world. (But that's not the point here) I would think MS's tech-support costs are also increased up by phone support, as contarary to common knowledge when an MS bug causes your server to crash, those support calls are free. Also when you call for a fix that is not freely available, those calls are free. (nb, i said server there for a reason, I don't mean your home copy of Win98! But it would include your work pc.) That all add's up to ten's of millions of dollars...

Re:Here's where I stopped reading:

[fferreres]

Holy god, this does then mean that we pay them a lot more so that they can can actually make that sale? It's something like:

company: The product costs $10. Ordering?

client: For that cheapo non-brand crap? NO WAY

company: Ok, REMEBER, YOU BROGHT THIS ON YOURSELF!! [Marketing expense increase. Price goes to $15]

company: So? You buying?

client: nope.

company: Whaaaa?! You don't want to pay the now $15 bucks? You ASKED FOR IT ASSH^H^H^H^HDAMNIT!! [HUGE Marketing expense increase, price goes to $70]

client: please charge mi credit card!! (man, it's only $70...! I saw it at $80 at some other shop!)

...

Negra Modelo Operating System

[rice_burners_suck]

I have to give Ransom credit for his good answers. Especially in the area of whether Caldera will make billions off Linux without giving back to the community. Ransom makes it pretty clear that Caldera's development of Linux, even though done for profit, is in itself something big that's being given to the community--it's the effect that takes place when a product has a lot of support from reputable companies more than which lines of code Caldera writes and puts wherever. Even in Caldera didn't write a single apostrophe of code but only packaged other people's hard and free work and sold it for profit, they'd still be opening a market window for the Linux community that wouldn't have existed otherwise.

I believe that for Linux to really be successful, both in terms of code quality and widespread acceptance, a team of companies needs to get together with a specific purpose to fulfill: To develop a desktop version of Linux similar in functionality to Windows XP or Mac OS X. In other words, other than a few options here and there, it wouldn't really have the possibility of configuring every detail. The bare-bones system, which basically means a desktop with a few icons on it for basic functionality (and no applications) would basically be integrated into the kernel. When the computer boots up, it boots directly into a nice graphical desktop, kind of like the way BeOS did, except Linux will actually stick around. Most of the configuration would be built in, heavily reducing the number of "system" files. No desktop system needs to drown in its own configuration. I would say that a really good bare-bones desktop Linux OS shouldn't take more than 15 to 20 megs of space, and that includes all the beautiful graphics it would have. To make this system useful, a number of applications made specifically for this version of Linux would be included, increasing the size to about 60 or 80 megs. This might include the standard "fun" stuff that consumers look for, like video editing, audio editing, image editing, and the more mundane crap, like word processing and whatnot. Obviously, major formats would need to be recognized. Only when something that's actually easy to use becomes available, Linux will skyrocket in popularity and Windows will go into the dumpster. Yeah, most of the code for this exists. Someone just needs to recognize why Linux tends to be so damn complicated, and rearrange these pieces into something that's really well designed and coherent.

As a disclaimer, just so someone doesn't say, "You stupid fsck! If you think Linux is too hard, go back to Windows and enjoy crashing all the time. Otherwise, learn how to use the damn thing!" I've been using Linux for years now, since an old Yggdrasil distro I picked up at some computer trade show, and a 8 CD set of crap they downloaded from a bunch of mirrors and stuck on some CDs. (Back in those days, those CDs were really handy, especially considering that most folks had a 14.4 connection. And that was considered fast! Even 56k was far off! So that was a loooooooooooooooooooooooong time ago.) Anyway, I've since realized that Linux is too fragmented. There's way too much duplication of effort. The Linux file system "standard" is totally wrong, IMO, because everything's in the wrong place. (/home should be under /usr, for example.) And the configuration is a nightmare! There are literally hundreds of configuration files strewn all over the system! So I since moved on to FreeBSD. All ya'll Linux folks out there can learn a shitload of great stuff from the BSD folks. Just look at how much of the system configuration for a FreeBSD system is in a single file (rc.conf). In Linux, this would be scattered across 10,000 different files in a monster directory structure under /etc/rc or something like that, as far as I remember. As you can easily tell, I haven't touched a Linux system in ages, and there's a reason. That's why I say this desktop thing needs to be made. Linux can kick Windows' ass in server stuff. Now it needs to start kicking Windows' ass in the desktop, because only when Microsoft loses that battle, they'll no longer be able to embrace and extend, a practice that fscks up all our code because we have to be compatible with their bugs, flawed design decisions, and programming features probably designed by lawyers and marketeers rather than programmers. Ooooooooooh well. Oh yeah, I had all kinds of Negra Modelo last night. It was fun. (If Cerveceria Modelo made software, their operating system would kick everyone's ass!)

Re:Negra Modelo Operating System

[rice_burners_suck]

Dude, you just said it yourself: " In bsd you have rc.conf yes, but that is just equivalent to the init scripts in linux"... May I point out that "init scripts" is plural?!?!?! That means that FreeBSD elegantly deals with all this crap in one file, when Linux has a bunch of init scripts! And that's just one of the areas that configuration in Linux takes place in a million billion places when it's more organized in FreeBSD. Sure, FreeBSD ain't perfect, but it works damn good, if I may say so myself.

Oh yeah... and you mentioned that Linux gives you flexibility through its Sys V style runlevels and whatnot. But honestly, who the hell uses runlevels? When I used Linux, my "server" boxes were always in runlevel 3 (IIRC, multiuser with network services) and my "desktop" boxes were always at runlevel 5 (everything in 3 plus X). The runlevels just made all the configuration into a big mess. Why do you need on the order of 100 start/stop scripts in rc when about 5 centralized scripts will do the work? When do you use runlevels anyway? Once in a year when you want to check something out on your server? Just use kill -9 and do it yourself! And suppose you do actually use runlevels more than once in a lifetime... I assure you that a few well-designed scripts will get the job done without the mess that everyone deems necessary in a Linux distro. (And who said Linux can't use FreeBSD-style init scripts?)

Re:Negra Modelo Operating System

[rice_burners_suck]

Hey, that's cool. I'd just like to point out that I never asked anybody to agree with my viewpoint. I only described my reasons for switching from Linux to FreeBSD. That's all.

Re:Negra Modelo Operating System

[rice_burners_suck]

Nice analysis, man, nice! I would like to point out a few things:

Windows has tons of configuration options, with most of those being hidden it clsid's and other crap. It's either configuration TEXT files or a kludge of a registery.

Uh, that's because Microsoft is STUPID. If all that crap is hidden in all sorts of places, then the damn values should just be hard-coded (in a global header or something, so it can be changed in one place if the need arises) because nobody can change them anyway! There's no need for all that crap they have in there. It's part of the reason Windows is such a piece of crap. There's no need for it to take up a flippin' gigabyte. Microsoft is just really wasteful, mostly because they have to HURRY HURRY HURRY to finish implementing the next version and get it out the door. As if anybody was bitching that Windows 95 was getting boring because it wasn't new anymore, so they had to hurry and make Windows 98. I personally wouldn't have minded one bit if they'd have taken a few more years putting together Windows 98, but released something that's rock solid. Of course, this is Microsoft we're talking about.

Funny thing is Linux is sort of the P2P os. All the parts are everywhere, there are multiple 'fixes' to the same problem, and developers are around the world. Asking some standards body to make a "UnifiedLinux" is asking for trouble.

And who said anybody would have to merge UnitedWhatever's changes back into the main sources? Who said they can't take a shitload of GPL'd stuff, analyse it, figure out where the duplication is, rip it apart, put it back together, and sell it as another product? Yeah, the dreaded code fork. But hey, this is GPL'd software, and any shmoe out there can do a code fork. If it's successful, and that company sells it as "Linux" (because it contains a ton of "Linux" code), a lot more people will get interested. Already, 99% of users hate Windows. I see it everywhere Windows is installed. People hate it because it's clunky, no matter how good Microsoft makes the user interface. It's crash prone. People don't trust their computers. Business owners pay a ton of money for licenses and upgrades, and Windows creates all sorts of headaches for them. Literally everybody I've spoken to about the subject (mostly non-programmer types, too) say, in one form or another, that Windows (or more generally, that their computer) SUCKS! And it's because of the unreliability that Microsoft trained everyone to accept as a fact of life. I disagree with this outlook. I believe software should be damn reliable, because it's used in so many ways and whatnot. It's an engineering job that's no less serious than one of building bridges or buildings or whatnot, and it's getting increasingly more serious with each passing day. The time for "who cares about the bugs, they'll just reboot or something" is long over. Computing is a serious matter, and it must be treated as such.

I dont agree with that... You live in a home, so /home it is. If you have problems with .conf's everywhere (like I did), just cp the files to /etc and sym-link to the original location. Now ALL your configs are in 1 place. Now you can back up your /etc dir (and sym links across your whole FS).

First of all, I said home goes under user, so it'd be like /usr/home/rice_burners_suck or whatever. Not /usr/rice_burners_suck. That would be stupid. UNIX lets you mount crap wherever you want, so why not mount a home partition under /usr?!?!?!??! And the other comment about the symlinks... That's a good one! I'll have to try it someday.

Well, it's 2:00, and I actually have to get some crap done today.

Re:Negra Modelo Operating System

[josh crawley]

---"Snip off of good stuff ;_)

... I believe software should be damn reliable, because it's used in so many ways and whatnot. It's an engineering job that's no less serious than one of building bridges or buildings or whatnot, and it's getting increasingly more serious with each passing day. The time for "who cares about the bugs, they'll just reboot or something" is long over. Computing is a serious matter, and it must be treated as such."

That's the thing. Architechural engineers don't just come up with new schmeatics for every project they build. They have basic layouts of every major work. When a client comes to them, they just go to an archive of "SOLID" projects and modify to custom specs.

In the software world, there's good ways to do things and bad ways to do things. Most intelligent ccomputer programmers agree that POSIX is the best standard to do computing right. In this, Unix, Linux, Xenix, Lunix, **ix came from it. The Unix way seems to be the stablest, secure way of doing things. Everything's locked down with user permissions and such. Well, until Linux came around, there wasn't much work done with *Nix as most required beefy systems and/or mucho $'s. Well, Linus came around with his pet project hacking at a minix kernel. His little pet project will probably be the "computer revolution".

Well, this is where Dos/Windowss came in. You had a split of philosophies. Power or KISS. Windows was KISS. There was no/little security. In the Win9x line, you could read/write directly to videoRAM, something you could NEVER do in a REAL os. Well, where this has lead us is to a Windows platform with Games Galore (with good plability) versus Linux where games lag like hell (3d games). First, its a software issue. Linux 3d renderer is too young and the security takes a bit of overhead (with secure DRI).

Sadfully, MS will probably be gone in 20 years or so. They can either adapt or die. They also have too much pride to touch Linux, so they'll die. Apple was smart in that they're using the tried and true OS'es.

As a last note. I like those FreeBSD trolls. Don't they realise that BSD is now the MOST used *Nix out there? Apple would have used Linux, but the GPL was too strong. Oh well. I guess BSD is dying (g).

If you want socialism, go to China

[Kenneth Stephen]

The GPL doesnt ask for any of which you are demanding from businesses. Indeed, some of the other Open source licenses (notably, the BSD license) are even less demanding than the GPL. Noone forced the authors of open-source software to pick a specific license - they did so of their own volition. One of the consequences of that choice is that other companies can make money off of that work. How is this immoral?

Redhat (and some other companies) have done more than what is required off them by the license. I wont go into their motivation - which is irrelevant to the point I am trying to make - that is : just because some companies may not go the extra mile, doesnt make them immoral. If you cant stand the consequences of the open-source licenses, dont pick them to license your code in.

Re:If you want socialism, go to China

[GigsVT]

Then lets come up with a UL-Free GPL.

It will basically say "Anyone is free to use this in a way consistant with the GPL and copyright law, except United Linux member companies, they may not use it period."

Then lets see how far they get with their bullshit business-speak.

Fragmentation is horrible

[Sheetrock]

Too many times, I have had minor filesystem incompatibilities shifting even from one Linux kernel version to the next. But even with a perfect kernel, everything above it is constantly shifting and hopelessly complex.

For example, I am now forced to interpret MiB (Men in Black) units when I check my network interface. Did I miss a meeting? Trying to get fonts working -- perhaps one of the simplest operations in Windows next to Solitare operation, BTW -- is a two-day adventure, and don't even think about trying to operate a printer. I use Debian, and while testing Gentoo next to it I find that I am capable of burning a CD-R by using 'cdrecord dev=ATAPI:0,0,0 speed=16 -v' (have to figure out how to enable BurnFREE someday)... under Debian, I still can't get it to recognize a standard ATAPI burner!

These are the things I hope UnitedLinux can address. Give us one of everything to make support problems easier to diagnose -- one window manager (KDE), one MP3 player (XMMS), one video app (MPlayer), etc. Test everything thoroughly to guard against trojanned Open Source applications and only release upgrades a CD at a time to reduce dependency breakage. Then we'll have something to compete with Microsoft.

Re:Fragmentation is horrible

[Rebel Patriot]

Give us one of everything to make support problems easier to diagnose -- one window manager (KDE), one MP3 player (XMMS), one video app (MPlayer), etc

Sounds an awfull lot like "One Microsoft Way".

One window manager, one mdia player, one web browser, one video app....

Re:Fragmentation is horrible

[BreakWindows]

Then we'll have something to compete with Microsoft.

I agree, but it's not a good thing.

1) Why compete with Microsoft, when you should be aiming to blow them out of the water (aiming, not necessarily doing).
2)The Microsoft way is some corporation deciding what is best for you. No, it is some corporation deciding what would cater to the lowest rung of the userbase. The GNU/Linux way is giving everyone their choice, which they can base on their skill level or personal ideas of what is "good".

I can sum it up easily: "one MP3 player (XMMS)". What? XMMS is garbage! Try Noatun, since you already have KDE...you'll never look back. Or maybe you will, but I don't. Unless United Linux comes to my place and asks what apps I want in a distribution, I don't want them only bundling one of each type of app.

If they're defaulting to KDE or Gnome, it isn't difficult to find a way to do what you want. In Windows, it is Programs=>Huge list of all your programs , which can be confusing (or annoying). In KDE or Gnome, it is Multimedia=>Multimedia apps, Office=>Office Apps, Internet=>Internet Apps, etc. Just keeping things like this is a far better approach than just leaving out what I might think is really good.

Not making millions?

[Telastyn]

It's not as though Caldera is making millions off of Linux?

This indicates Mr Love is making $277k/yr off of Linux, so I'm pretty certain they aren't strugling...

Re:Not making millions?

[vinnythenose]

Perhaps you didn't see the net loss of $17.6 million for the company.

Whether a company is making profits or losing money, the employees still get paid (be them coders or CEOs).

Caldera is not making money. The employees are, but then again, would you work for a company if you were losing money working there?

United Linux trying to reduce choice?

[FreeUser]

I found this more than a little disturbing:

Going forward, there will only be two platforms certified by the major hardware and software vendors, Red Hat and UnitedLinux. For Linux to move from the peripheral of the business network into mainstream application server market, businesses must be assured that their platform is certified and will work with other applications and hardware solution in their environment. What the UnitedLinux customer is paying for is 1) the assurance that his applications will work together, and 2) the ongoing maintenance and support of that certified platform.

One of the most frustrating things about using some commercial products under GNU/Linux has been the Red Hat centric methods of distribution: binary RPMS designed for Red Hat that may or may not work with other RPM based distros, and will require quite a bit of hoop-hopping on non-RPM distros such as Slackware, Debian, and all of the excellent Source Based Distros.

Not all products do this ... some of the more intelligent software manufacturers release their products as tarballs, and are thus easilly installed onto any distro whatsoever. That, coupled with a short list of required dynamic libraries and versions, is enough to get the product working with any distro out there (even if you have to retroactively install some older libraries yourself, by hand).

Now it looks like United Linux is trying to reduce the choice of anyone wanting to run a commercial product on their system to two choices: Red Hat or United Linux. How likely is a product released as a cross-distro compatable tarball going to be to get "certification" vs. a Red Hat/United Linux RPM. It looks like the behometh forming here has every intention of dictating standards and shoving de-facto norms down the community's throat, rather than taking the consensus-based approach we have used up until now. If this impression is correct, this is anything but a positive development for GNU/Linux or free software and will likely be quite detrimental to the communities which surround and support them.

Having tried Red Hat, Suse, Mandrake, Debian, Source Mage, Gentoo, and others, I can unequivocably say that, in my place of work as well as in my home, the kind of prior restraint on my choice they seem to be aiming for (with their 'certification' requirements) is such that it will eliminate commercial products we would have considered purchasing from contention altogether (We are a Debian shop, currently transitioning to a source based distro for performance/reliability reasons. Red Hat and United Linux are not, and never will be, under consideration for deployment here).

It would be unfortunate if this were to become the norm ... we actually do use commercial products under GNU/Linux where I work, but we have never, and will never, allow a commercial product to dictate the distribution we deploy. To do so would be absurd ... and it sounds very much like this 'certification' process they're talking about is designed to coerce any and all commercial GNU/Linux vendors to target Red Hat and United Linux, to the detriment of other, likely better, distros out there.

This reduces choice, and is a bad thing for the Linux community IMHO. The result is more likely to be less willingness to deploy commerical products if a marginally usable free alternative exists that happens to be usable on the platform/distro of choice (whatever that may be). Contrast this to current conditions, where many vendors are sufficiently neutral that one can deploy their product anywhere, and the free community drives most standards, not Red Hat and Caldera.

I started reading this interview with a very positive "I'm glad to hear the negative rumors are likely wrong" and by the end of the interview have reluctantly concluded that this isn't going to be a positive development for GNU/Linux at all, and many of the worst fears expressed by others earlier are quite probably entirely well founded.

Worse still, the answer given to the patent question was downright chilling...thank [deity] it only affects the United States, and not (yet) the entire world. This is one way they could very effectively steal our work and cut off our access to the products of our own labor, perhaps even in spite of the GPL.

Re:United Linux trying to reduce choice?

[elflord]

You're contradicting yourself left right and center. First you come up with this ...

One of the most frustrating things about using some commercial products under GNU/Linux has been the Red Hat centric methods of distribution: binary RPMS designed for Red Hat that may or may not work with other RPM based distros,

The reason why this is the way it is has to do with the fact that different distributions have subtle differences at the core. They have slightly different naming conventions, and put files in slightly different places.

With that in mind, let's scroll down ...

Now it looks like United Linux is trying to reduce the choice of anyone wanting to run a commercial product on their system to two choices: Red Hat or United Linux.

This is simply false. United Linux is not a single distribution, it's a common foundation. You can't have it both ways -- you either have a common foundation for different distributions, or you have incompatibilities. What this actually does is give you more choice, because the United Linux based distributions should be binary compatible.

It looks like the behometh forming here has every intention of dictating standards and shoving de-facto norms down the community's throat, rather than taking the consensus-based approach we have used up until now.

Not true. UnitedLinux is based on LSB, which was indeed developed by the community.

and it sounds very much like this 'certification' process they're talking about is designed to coerce any and all commercial GNU/Linux vendors to target Red Hat and United Linux, to the detriment of other, likely better, distros out there.

A package that targets UnitedLinux is LSB compatible. Here's the bottom line-- you can make your distribution binary-compatible with market leaders, by making it Redhat or LSB compatible, otherwise you're kind of stuck. I don't see how this constitutes a change in the state of affairs, besides the fact that the commercial Linux distribution market is getting its act together (at last)

Re:United Linux trying to reduce choice?

[sulli]

Going forward, there will only be two platforms certified by the major hardware and software vendors, Red Hat and UnitedLinux

Sounds like wishful thinking to me. "Okay, everyone certifies Red Hat, but they'll also certify us! Really they will!" We'll see.

Re:United Linux trying to reduce choice?

[FreeUser]

You're contradicting yourself left right and center.

No, actually I'm not contradicting myself at all. You, however, are doing quite a bit of creative editing on my comments, construing one meaning where there was another, modifying the implications thereby and attributing them back to me.

I won't bother to go through this point by point (the original post is online and available for anyone to read who wishes to do so, unabridged), but will cite one example of your editorial ... ah ... creativity:

One of the most frustrating things about using some commercial products under GNU/Linux has been the Red Hat centric methods of distribution: binary RPMS designed for Red Hat that may or may not work with other RPM based distros,
(what you snipped out and ignored was)
[...] Not all products do this ... some of the more intelligent software manufacturers release their products as tarballs, and are thus easilly installed onto any distro whatsoever. That, coupled with a short list of required dynamic libraries and versions, is enough to get the product working with any distro out there.

The reason why this is the way it is has to do with the fact that different distributions have subtle differences at the core. They have slightly different naming conventions, and put files in slightly different places.

As I pointed two sentences later (which you chose to completely remove, since it completely rebuts the point you made based upon my edited comments), binaries can in general be made to work with any distribution, regardless of packaging system or file organization, if the binary is provided in a tarball, along with a list of required dynamic libraries and their versions.

It looks like the behometh forming here has every intention of dictating standards and shoving de-facto norms down the community's throat, rather than taking the consensus-based approach we have used up until now.

Not true. UnitedLinux is based on LSB, which was indeed developed by the community.

Perhaps the suspicion I expressed isn't true, but it is certainly well founded, and may in fact be true. If United Linux is based upon the LSB, then why not certify things as "LSB" compliant and allow them to work with any distro? The answer is rather obvious, if unpleasant: because UL intends to have additional, distro-specific standards to which software vendors will be required to adhere for certification, which may well mean (and, perhaps, even be designed such) that such software will be less compatable with other distributions, whether or not they adhere to the LSB. I hope this suspicion does in fact turn out to be wrong, but your comments certainly do nothing indicate that is so (beyond simple denial, which frankly isn't enough).

This is certainly not an unreasonable suspicion to have in light of the statement I originally quoted, to wit: "Going forward, there will only be two platforms certified by the major hardware and software vendors, Red Hat and UnitedLinux."

You even go so far as to confirm this, when you say: "Here's the bottom line-- you can make your distribution binary-compatible with market leaders, by making it Redhat or LSB compatible, otherwise you're kind of stuck."

You go on to say that you do not see how this constitutes a change in the state of affairs.[1] Perhaps for Caldera software it doesn't, but for a wide range of commercial products (VMWare, numerous commercial games, several office products we use, etc.) it is a change of affairs.

When releasing binary packages, one can package them in a distriubution-neutral format (tar.gz for example, NOT rpm or deb), and provide in the README an exhaustive list of the required libraries and versions said package needs to run, with or without hints for installation on various distros (the latter are nice, but not required for functionality). This is the CORRECT way of doing things, and doesn't coerce one's customer into using a particular distribution in order to use the software. VMWare is good about this, so are several other software vendors.

OR, one can package their software such that it is designed ONLY to work with [Red Hat | UL], taking choice away from their customer (and in our case, and in probably many other cases, forcing those who have chosen to use other distributions to jump through numerous hoops and quite probably losing a customer in the process. Most of the software we write is in-house, and we choose the distribution we like based on technical merit, performance, and stability. If a third party commercial product we'd like to use requires a conflicting distro, most commonly Red Hat, then we use something else. We do not let a vendor's product dictate our infrastructure ... too much of our important work would be affected adversely if we did. We are not alone in this, and if commercial products take on the tack your comments suggest, these commercial products will be rejected in favor of less capable, but more compatable, products instead, be they freeware or more intelligently packaged proprietary products).

UL seems designed to incorrectly indicate to commercial software vendors that, rather than packaging their product up in a generic manner usable on any distro (provided the end-user insurs all necessary libraries are in place, as documented in the software release), they should simply target one or two preferred distros and leave everyone else in the cold. Worse, the effort seems designed to incorrectly tell commercial vendors that packing their wares for UL and/or Red Hat, rather than in a more generic and widely compatable way, is the only correct way to do so.

While this may be a win for UL and Red Hat, it is very bad for the commercial vendor who is suddenly out a large chunk of the GNU/Linux marketplace, and a big loss for the GNU/Linux community which faces a behometh who is intent on taking away their freedom of choice through market coercion.

This is the last thing the GNU/Linux community needs, and, frankly, as one who works at a company that has deployed GNU/Linux widely, and uses third party, commercial products on various GNU/Linux distributions, the last thing we the customer want.

[1]The precise quote, so they reader may judge if my characterizatino is fair or not, was: "I don't see how this constitutes a change in the state of affairs, besides the fact that the commercial Linux distribution market is getting its act together (at last)"

Re:United Linux trying to reduce choice?

[elflord]

As I pointed two sentences later (which you chose to completely remove, since it completely rebuts the point you made based upon my edited comments), binaries can in general be made to work with any distribution, regardless of packaging system or file organization, if the binary is provided in a tarball, along with a list of required dynamic libraries and their versions.

This does not advance your argument, because it is simply incorrect. I was do you a favor by snipping this (-; Of course, the binaries themselves may work, but then again, may not-- for example, what if libstdc++ was compiled with different C++ compilers ? Or if a library does something has an ABI change without a change of soname ? What if different vendors choose different sonames for the same library ? (eg I recently used libqt-gcc-3.1.so.3 for a gcc 3.1 based Qt. Someone else might use a different name for a gcc 3.1 based qt. ) Assuming these issues don't prove to be a problem, and you do actually have binary compatible packages, you're still some way from having a package that the average user will find useful. For example, what if the package needs to install init scripts, or configuration files, or files for KDE/GNOME, or run some post-install scripts to work properly, etc ? Having the actual binaries ius only a small part of the picture. If you'd done any packaging, you'd know this.

So no, I'm not snipping context to make you look bad, I'm snipping it to address the main line of argument.

If United Linux is based upon the LSB, then why not certify things as "LSB compliant and allow them to work with any distro ?

If you read the article, you'd be able to answer this correctly, instead of spouting speculative conspiracy theories. The answer is that LSB does not address some issues like internationalisation, for example. So as far as your definition of "work" is concerned (meaning binaries link properly), they would "work with any distro", but for the intents and purposes of United Linux's more ambitious goals (internationalisation, for example), they would not.

The answer is rather obvious, if unpleasant: because UL intends to have additional, distro-specific standards to which software vendors will be required to adhere for certification, which may well mean (and, perhaps, even be designed such) that such software will be less compatable with other distributions

I don't see how it could be less compatible. At worst, it could handle things like internationalisation, installing configuration/desktop files, etc a little differently. But you've stated that these are things that you don't consider to be very important anyway (and things that don't work correctly today, for that matter), so I don't see why you're making such a fuss about it.

When releasing binary packages, one can package them in a distriubution-neutral format (tar.gz for example, NOT rpm or deb), and provide in the README an exhaustive list of the required libraries and versions said package needs to run, with or without hints for installation on various distros (the latter are nice, but not required for functionality). This is the CORRECT

Correct to you maybe. All this amounts to is that you have binary packages which are equally broken on all distributions. The point is that there's no such thing as "distribution neutral". An RPM package can easily be converted to a binary tarball (slackware ships a utility for this), but without more integration, you're still stuck with a package where the install requires a fair degree of manual intervention. BTW, without some degree of standardisation, it also means commerical software is stuck with libc-only, static binaries, (basically this is the way it's been) or incompatibility (because no-one will have all the right libraries)

OR, one can package their software such that it is designed ONLY to work with [Red Hat | UL], taking choice away from their customer

If Redhat were binary compatible with everyone else , I don't see how you could credibly claim that the packages "only worked with Redhat", unless you expect some distribution integration, which you don't get today. And no, it's not a problem if the packages are RPM only, because it is trivial to convert an RPM to a binary tarball. What is highly nontrivial is actually doing something useful with that binary tarball.

Re:United Linux trying to reduce choice?

[FreeUser]

As I pointed two sentences later (which you chose to completely remove, since it completely rebuts the point you made based upon my edited comments), binaries can in general be made to work with any distribution, regardless of packaging system or file organization, if the binary is provided in a tarball, along with a list of required dynamic libraries and their versions.

This does not advance your argument, because it is simply incorrect. I was do you a favor by snipping this

Based on my experience (I've been using GNU/Linux since 1993) it is correct. Even ill-behaved binaries packaged specifically for some ancient version of Red Hat, like the Sybase 11.9.2 binaries, can be made to work with, for example, the current release of Gentoo in spite of their deliberate 'Red Hat Only' packaging and configuration assumptions. I could list a dozen other examples, not to mention a plethora of much better behaved and more intelligently packaged proprietary binaries, such as VMWare and various commercial games. To get any of these working all you need to do is insure the proper libraries have been installed in the proper location.

The only exception that comes to mind was Word Perfect, which was so poorly packaged that it required their own distriubution, a failing that reflected their incompetence, or the poor porting job that was done, far more than it did the interoperability of diverse distributions even then.

As an aside I should mention that we are moving away from Sybase to Postgres because of the hoops their idiotic RPM-only packaging scheme, coupled with a Red Hat centric installer that tries to be intelligent (but isn't), force us to jump through. We can get it working, but it is more trouble than it is worth because of the way Sybase has chosen to package it, such that Sybase is now in the process of losing a customer.

If you'd done any packaging, you'd know this.

I have done packaging, and I do know this. I know this to be absolutely incorrect. At worst, your installation script needs to ask a human being (the person installing the software) where the other software it requires is located (be it KDE, Gnome, or what have you). It should be doing this regardless, and getting permission before it does any mucking about with the installed (and presumably functioning) system, and it most certainly should not be making assumptions about any system it is installing upon. For all you know the system has been customized anyway, for another compelling reason, and the fact that it is a Red Hat or UL system may mean absolutely nothing WRT the validity of a particular assumption about where things are located, or what versions are installed.

Furthermore, if your packaging, or your binary, requires modification to existing init scripts, then it is broken. Period. That is the job of the system administrator, and NOTHING should ever be rewriting or modifying existing init scripts. Placing a script in /etc/init.d (or wherever), and suggesting that it should be added to your default runlevel, is as far as any installation should ever go. And if a distribution is really so out of sync with the LSB and LFSS that the installation script can't make an intelligent guess as to where to put the script, then it should put it into a directory under opt and inform the installer that THEY need to add that script to their default init environemt, something that is there responsiblity regardless.

Software installation scripts that go about modifying an installed UNIX or GNU/Linux system like you imply are a recipe for Microsoft-Style disaster with or without a One True Distro, and frankly a profoundly bad idea.

libstdc++ was compiled with different C++ compilers ? Or if a library does something has an ABI change without a change of soname ? What if different vendors choose different sonames for the same library ?

libstdc++ is the only real problem, and even it is easilly resolvable using either chrooted environments, or environments with a modified ld.so.conf configuration in userspace.

An even easier solution is to distribute a statically linked binary in addition to the dynamically linked version. This is how Blender very neatly avoided any problems of this kind., and is a wise thing to do regardless.

And that is the most extreme example. The other examples you cite can be addressed with symbolic links (ideally in a subdirectory added to the app's LD_LIBRARY_PATH specific to that app, so that the overall system isn't cluttered).

An easy solution is to bundle commercial apps into /opt (which is what the LFS standard requires anyway) along with the libraries they need and a startup script that sets the appropriate library environement up for them regardless of distribution. I've done this myself more than once, and it isn't terribly difficult.

The answer isn't to create a One True Distribution(tm) and replace consumer choice, competition, and the free market that GNU/Linux has created with the Linux equivelent of a Microsoft Monopoly, it is to get vendors to intelligently package their products so that they can run with most if not all of the various distros out there.

Correct to you maybe. All this amounts to is that you have binary packages which are equally broken on all distributions. The point is that there's no such thing as "distribution neutral".

This simply isn't true. RPM is most emphatically NOT distribution neutral, even if volunteer developers have spent many man-hours writing packages like "alien" in order to make RPMs accessible to the majority of distros out there which do not use RPM. tar.gz IS distribution neutral. A well packaged binary, with all necessary libraries installed in its own /opt subdirectory and a run script that sets the library resolution environment appropriately goes a very long way toward making a binary distribution neutral. Offering a statically linked binary in parallel to a dynamically linked binary, for those few distros that may be so out of sync with the state of affairs when the binary was created would finish the job.

And that goes for future proofing the binary as well ... what if you need to run that binary 5 years from now to access some data, and no distribution on the planet (including Red Hat 75.5 and UL 10.2) has the archaic libraries it was linked against anymore? Well, a statically linked binary will run anyway ... no binary-only, closed source commercial product should ever ship without a statically linked version of their binaries, regardless.

If Redhat were binary compatible with everyone else , I don't see how you could credibly claim that the packages "only worked with Redhat"

This is a null argument. It is akin to [insert favorite religion here] saying that if everyone believed as they do the world would have peace.

First, Red Hat (and UL) change, and they change fairly quickly. It is unreasonable, and inappropriate, to demand the rest of the GNU/Linux community, to whome they owe the very existence of the software they are selling, to subordinate themselves in such a way and follow suit every six months.

Second, there are other approaches which, in some cases (like the source based distros such as Source Mage and Gentoo) work better than anything Red Hat or UL is going to be able to come up with because their very paradigm is superior. Coercing them to adhere to Red Hat's notion of how things should be means giving up much of the technnical merit that makes Source Mage and Gentoo signficiantly more stable and better performing than, for example, Red Hat (or any other binary-based distro, for that matter).

We have some degree of standardization already (LSB compliance, LFS, etc.). We do not need proprietary standards in addition to that, and we do not need to trade one set of masters (Microsoft, Sun, or Apple, depending on which camp you come from) for another (Red Hat or UL).

Re:United Linux trying to reduce choice?

[elflord]

Based on my experience (I've been using GNU/Linux since 1993) it is correct. Even ill-behaved binaries packaged specifically for some ancient version of Red Hat, like the Sybase 11.9.2 binaries, can be made to work with, for example, the current release of Gentoo

I suppose can be made to work is the key point that I missed the first time. This is true, provided that the binaries are really compatible. The reason that binaries are usually compatible is that vendors either don't use any libraries besides libc and raw Xlib, or they statically link. Neither solution is terribly satisfactory.

We can get it working, but it is more trouble than it is worth because of the way Sybase has chosen to package it

The same argument applies with binaries packaged in this "distribution neutral" manner. You can make it work, but it would be nicer if it just worked without manual intervention.

t worst, your installation script needs to ask a human being (the person installing the software) where the other software it requires is located (be it KDE, Gnome, or what have you). It should be doing this regardless, and getting permission before it does any mucking about with the installed (and presumably functioning) system, and it most certainly should not be making assumptions about any system it is installing upon.

You're entitled to an opinion on the degree to which installation should be automated, but it's worth noting that your opinion does not appear to reflect either the opinion of the market which purchases distributions that make such assumptions in post-install scripts, or the users who drive that market.

Furthermore, if your packaging, or your binary, requires modification to existing init scripts, then it is broken.

I didn't say it should. However, it could certainly modify other files (eg: /etc/ld.so.conf, ) As for installing init scripts, at present different distributions put them in different places. There are similar issues with other things.

Another problem with installing binary tarballs is that once you do that, you're outside the package management system. There are obvious problems with this.

ibstdc++ is the only real problem, and even it is easilly resolvable using either chrooted environments, or environments with a modified ld.so.conf configuration in userspace.

libstdc++ dependencies cascade to all C++ programs installed on your system, and all libraries that are C++ based. And it's not the only problem. Any library potentially has the same issues. As for "easily", I don't think a separate chrooted environment per application is an "easy" solution. Modifying the configuration of ld.so.conf doesn't help a whole lot, because it's not going to automatically create a parallel subsystem of libraries based on the other libstdc++ version, and it's going to make the other libstdc++ version unavailable.

nd that is the most extreme example. The other examples you cite can be addressed with symbolic links (ideally in a subdirectory added to the app's LD_LIBRARY_PATH specific to that app, so that the overall system isn't cluttered).

You can't address different soname conventions with symbolic links. If an application wants a library called libqt-gcc3.1.so.3, you're not going to get away with relabelling some other libqt. The soname is built into the library. You could have a special LD_LIBRARY_PATH workaround for cases where you merely have two incompatible libraries with the same soname (eg compiled against different g++ versions), but that means you're going to have to install all the required libraries for that application in that directory -- so you're essentially back to the statically linked solution.

Even this doesn't always work. For example, KDE creates a .qtrc that tells qt to load certain plugins. So if you're running a gcc 3.1 Qt and a gcc 2.9x qt, they're going to clash because the gcc 3.1 qt will try to load plugins built for the gcc 2,9x Qt and/or vice versa.

This simply isn't true. RPM is most emphatically NOT distribution neutral, even if volunteer developers have spent many man-hours writing packages like "alien" in order to make RPMs accessible to the majority of distros out there which do not use RPM.

I didn't say that it was "distribution neutral".

tar.gz IS distribution neutral. A well packaged binary, with all necessary libraries installed in its own /opt subdirectory and a run script that sets the library resolution environment appropriately goes a very long way toward making a binary distribution neutral.

If it doesn't live under the distributions packaging system, it's not distribution neutral. How do I know that uninstalling or upgrading library (X) won't break my "distribution neutral" tarball ? I don't, unless the tarball was statically linked, or ships its own libraries, or ships with such a spartan set of dependencies that I can remember them.

This is a null argument. It is akin to [insert favorite religion here] saying that if everyone believed as they do the world would have peace.

No it isn't. it makes an important point-- that binary compatibility addresses the issues that you keep whining about.

First, Red Hat (and UL) change, and they change fairly quickly.

Bogus assumption. If UL is based on LSB and the other distributions are based on LSB, it's reasonable to expect that the LSB based distributions have a substantial overlap in compatibility with united. On the other hand, if distributions want to follow Redhat, they are, as you say, on a treadmill.

Second, there are other approaches which, in some cases (like the source based distros such as Source Mage and Gentoo) work better than anything Red Hat or UL is going to be able to come up with because their very paradigm is superior. Coercing them to adhere to Red Hat's notion of how things should be means giving up much of the technnical merit that makes Source Mage and Gentoo signficiantly more stable and better performing than, for example, Red Hat (or any other binary-based distro, for that matter).

They're within their rights to do things differently, but that's a tradeoff you make. You're either compatible with the market leaders, or you're technically superior (whatever that means) and pay the price for being incompatible.

We have some degree of standardization already (LSB compliance, LFS, etc.). We do not need proprietary standards in addition to that,

You're "heap closing". You have not demonstrated that United doesn't add any value, or that LSB and LFS address the problems that United are addressing. You haven't shown that United is going to add incompatibilies, while I have pointed out that United supplements, and not replaces the standards you mention. All you've done is offered dark mumblings, unsupported speculation, and conspiracy theories.

Re:United Linux trying to reduce choice?

[FreeUser]

All you've done is offered dark mumblings, unsupported speculation, and conspiracy theories.

Actually, no. All I've done is point out the publicly stated intentions of UL, then expounded on why they are not good for the community. Your responses have largely been "you're wrong!" and when pressed by additional detail, revised to say "ah, I overlooked what you're saying there." All of this of course interlaced with dismissive insults and innuendos, and allegations of contradictions within my arguments, all of which I have thoroughly rebutted.

This final dismissal, labelling the very valid concerns I raise as "dark mumblings, unsuported speculation, and conspiracy theories" I will now rebut by presenting quotes from yourself and the UL article which show that they are in fact not only supported, but require no theory as such, and indeed speak for themselves. Although presented without context, the meaning is IMHO preserved, and the reader can follow the footnoted links to view the context if they are interested.

Going forward, there will only be two platforms certified by the major hardware and software vendors, Red Hat and UnitedLinux. [Reference]

Here's the bottom line-- you can make your distribution binary-compatible with market leaders, by making it Redhat or LSB compatible, otherwise you're kind of stuck. [Reference]

The point is that there's no such thing as "distribution neutral". [Reference]

If Redhat were binary compatible with everyone else , I don't see how you could credibly claim that the packages "only worked with Redhat" [Reference]

If it doesn't live under the distributions packaging system, it's not distribution neutral. [Reference] (Redefining distribution neutral as integrated into all distributions packaging systems, something I'd already demonstrated in two posts was unnecessary and, indeed undesireable.)

They're within their rights to do things differently, but that's a tradeoff you make. You're either compatible with the market leaders, or you're technically superior (whatever that means) and pay the price for being incompatible. [Reference

It is clear that the intent is to replace and supplant community standards based on consensus with standards driven by whatever UL (perhaps with Red Hat) decides. This intent has been reiterated in virtually every response you've made in this discussion. It is irrelevant that this extended standard is based upon open standards if it is being modified and imposed back upon the community in a non-open fashion as has been described here. If this were Microsoft everyone would be shouting "embrace and extend," for indeed that is exactly what is proposed here (though I must point out: sans the "destroy" clause).

It is my opinion that any non-open approach to defining standards is a very negative thing for the community. The process is as important as the result, and the open, collaborative process here is being abandoned and supplanted by a coercive, closed process that is designed to benefit one or two entities, under the guise of serving the community. This is a concern, and indeed if the standards process for the community is so supplanted, one of the most compelling reasons for switching to GNU/Linux (to get out from under one's vendors' thumbs, ie. freedom) goes away almost completely.

As for accusations of "conspiracy theory" you asked me for examples of how such a closed process could be detrimental to the community and I provided several scenerios which have occurred repeatedly, both in the Wintel environment and the (prior to GNU/Linux) fragmented UNIX world. Perhaps you, as an advocate of United Linux, can dismiss these concerns, but as one who makes a living using and deploying Free Software, and as one whose business has benefited greatly from the Freedom provided by Free Software, I cannot dismiss these concerns so easilly, particularly given the fact that every statement made to rebut those concerns has done nothing but confirm them.

It is a collassal mistake in judgement for the likes of Caldera (even coupled with HP and others) to approach the GNU/Linux community, to whome they owe the very product they are leveraging, and arrogantly assert something to the effect of:

As your industry leader we are now getting your act together and defining your standards for you. We will base our work upon the consensus you've formed thus far, but now you will follow our lead or you will be stuck.

I think you are going to find the community singuarly unresponsive and unfriendly to this sort of thing, and rightly so. It is my fervent hope that you do not lead too many software vendors down the garden path in the process ... it would be a shame to see the waters poisoned for commercial software vendors because they mistakenly followed the advice of a corporate entity to package their wares for only one or two distros (their own, as it happens) and a non-consensual, imposed standard, rather than packaging them in a form usable by everyone, agnostically, such as VMWare, Blender, and numerous other products have, and then mistakenly reading the community's rejection of the UL imposted "standards" as a rejection of their products.

Re:United Linux trying to reduce choice?

[elflord]

Actually, no. All I've done is point out the publicly stated intentions of UL, then expounded on why they are not good for the community.

The only concrete issue you've pointed to that has some legitimacy is their choice of RPM format. My counter arguments are:

• Binary compatibility issues, relating to choice of library version, choice of C++ compiler, , sonames (or rather, so-renaming), and package location pose a greater problem than RPM format. You did present a counter argument of sorts to this (though I don't buy it)
• Going outside the package management system means you're not "distribution neutral", because you're not taking advantage of the distributions dependency tracking capabilities. (I understand that you disagree with this, but that is my argument)

It is irrelevant that this extended standard is based upon open standards if it is being modified and imposed back upon the community in a non-open fashion as has been described here. If this were Microsoft everyone would be shouting "embrace and extend," for indeed that is exactly what is proposed here

I agree that this is an instance of "embrace and extent" (sans destroy). I think there is an important distinction between this and simply attempting to subvert the standard by using a competing one -- to me, it's not irrelevant at all, because LSB compliant distributions will be binary compatible with United. The difference is, I don't really have a problem with that. This is the way standards evolve. Design by a committee doesn't work very well, and you need someone to blaze a trail and experimenting before committing something to a standard. For example, gcc has "embraced and extended" both C and C++, and the extensions have for the most part had good consequences. So you could use the words "embrace and extend", but I'd contest a claim that this is necessarily a bad thing. Hence my comments about "dark mumblings" (sorry if that sounded harsh)

It is my opinion that any non-open approach to defining standards is a very negative thing for the community. The process is as important as the result, and the open, collaborative process here is being abandoned and supplanted by a coercive, closed process that is designed to benefit one or two entities, under the guise of serving the community.

Again, I disagree. The claim I'd contest is that this is being sold as a replacement for LSB and other standards. Standards codify existing practice, committees aren't very good at designing and innovating, but they are good for formalising and clarifying details about how things should be done. If United Linux was attempting to replace LSB, I'd be against it, but at this stage, I don't see evidence that supports such a claim.

Re:United Linux trying to reduce choice?

[FreeUser]

Again, I disagree.

I think we'll have to agree to disagree on this.

f United Linux was attempting to replace LSB, I'd be against it, but at this stage, I don't see evidence that supports such a claim.

The issues are more subtle than outright replacement (as embrace and extend imply), but the results can and generally are just as bad when dealing with the subversion of open standards through a non-open process. I do take some issue with the word claim ... I voiced a concern, not a claim, though the more reading I do on the subject the greater that concern grows and the more uneasy I grow with this entire process.

Fortunately the GNU/Linux community is by and large a pretty bright bunch of people, so I doubt this effort at hijacking the standards process, so to speak (if in fact that turns out to be what is happening) will have much success.

Debian Graphical Installer

[MADCOWbeserk]

Progeny has made a great graphical installer available for Debian woody. Sorry I can't remember the link. But you need to do the install over the net, AFAIK.

Re:Debian Graphical Installer

[gol64738]

..and how is this on topic?

Re:For anyone..

[skotte]

you know. i hate the fFact that you started your comment in the subject. thats awfully irritating. BUT NEVERMIND THAT. :)

you really have a good point.

I am remninded of Golf. I hate golf. golf is a stupid stupid game in which old men walk around and swing a stck at the dirt fFor hours on end and come home grumpy.

But golf was not *meant* fFor me. It's meant fFor old men who like to walk around all day, and would come home grumpy regardless.

lesson: fFor linux to be successfull in the Real World, it must comply with a fFew things which the Real World expects software to be like. my boss used to say he would never use linux because if something goes wrong, who do you sue? Not that he would sue, or even would be ABLE to sue. (ever try suing microsoft?) But it's the idea that out there somewhere is someone who speaks the business language and is standing behind a product and saying "This will work."

You don't like UnitedLinux? well, you're reading slashdot. so probably, UL was not meant fFor you.

Red Hat already does what UL is going to do

[d3xt3r]

Does everyone who is claiming that United Linux is bad and won't contribute to Linux development realize that Red Hat already does what United Linux is planning to do?

In late April, Red Hat released Advanced Server - a distribution that will be guaranteed to be supported for an 18-month period with minimal change. If you go to ftp.redhat.org and poke around for this distro, you will find that Red Hat has a nice collection of SRPMS (source RPMs) for the distro, yet no public binary release. Any complaints?

Do people not realize that this is going on, or is it not a problem because it's Red Hat?

Personally, I think it would be great if all Linux companies made all of their distros freely available in binary form on ISOs, but sadly this is not the case, not even with Red Hat.

So what's it going to be? If everyone is going to get up in arms over United Linux, they better put Red Hat in the same boat (IMHO).

Insight into Linux economics

[God! Awful]

What I read from the subtext of this article is that the major Linux distribution companies have lost confidence in the GPL. He doesn't come right out and say it, but it's clear that they are fighting the license. Without violating the GPL, they are trying to make it inconvenient for users to redistribute the software and avoid the per-seat licensing. He also says that he not does believe in a Linux model that requires ongoing charity to survive. This sounds like a dig at Mandrake.

He is right, though. Charity is a terrible business model. The only time it works well is when you campaign for some worthy cause (e.g. feeding starving children in Africa) and then spend 90% of the money on "administration". Here's an interesting tidbit from the other side of the political spectrum. Did you know that Ayn Rand (a laissez-faire capitalist and staunch opponent of taxes) believed that if income tax was abolished then the rich citizens of America would voluntarily donate money to establish an army? Can you imagine a world in which armies were established and paid for solely by robber barons? I mean, they'd probably be sent off to fight any country that threatened to mess with our oil supply. Oh wait, that already happens.

-a

Re:Having the source lets you create the binaries

[Junta]

A bit offtopic, but I was wondering if something could be a loophole here. While the source should match the output, does the GPL make any requirements about the build process? I mean, would it be any less GPL because it includes Visual Studio Build information as opposed to automake/autoconf stuff? What about no build information at all and raw source? That last option could really make a codebase so not worth putting together that a customer would rather buy than try to piece it together. Of course, someone could release a GPLed builder of the product and that would go away. But on the other hand, what if the company devises and implements a really different and proprietary build system with which they build the product, that can't easily be reverse engineered and leave the source mostly unusable?

I guess the ultimate example of this would be a company developing a proprietary language for internal development. Release things in this language, but make no compilers or platform documentation available outside the company. This extreme example is not plausible (company would lose more than it gains), but it still raises an interesting question.

Also, where is the line drawn between source and binary? One could make the argument that binary is source, if you can just grok machine code. A simple translation from assembly code, and many gpl projects at least include assembly, and I guarantee you there is at least one entirely assembly based project under the GPL. In those circumstances, whether you have the source or the binary doesn't matter too much (well, depending on the linking). Same as having Java class files, Java class files easily decompile to readable, modifiable code.

Problem with 2/3s of the linux community...

[Sheeplet]

...is that they have no clue about how enterprise computing works. You can't just be like, "oooh, here's a problem, throw a linux box at it", that would never happen. Companies that are worth millions and billions in business every year move very carefully. Every piece must either work perfectly or have channels of responsibility, which in the case of 3rd party products (like Operating Systems) often falls back to the vendor.

Why doesn't Linux get deployed at the enterprise level (5000+ employees, not to mention 170,000 employees like the company I work at)? Because there are no support channels, and no, "ask the local LUG or post a message to a mail list" is not a valid support channel at that level.

UnitedLinux may be on the right path, why not support them for what they're trying to do, bring linux onto the enterprise stage, and not make up things like they're trying to take away your rights or make linux proprietary or some other nonsense. UnitedLinux, IMHO, is trying to innovate, they might get some things wrong here and there, but very few people have tried to bring Linux into the enterprise space and they deserve a little support just for trying that.

-Chris
My $.10

Sorry, I agree with him on that one

[Anonymous Coward]

What? GNU/Linux was created for business deployment? Where do you get that?

Posts like this reinforce my belief that most Slashdot posters are computer-crazed teens with no business experience. They can't see the big corporate picture. But in truth, businesses don't want to have to learn every detail about Linux. They want to get on with their operations. Computers and operating systems are a means to an end.

I think Ransom acknowleged the importance of software generated by the open-source community. But until packaged in a business-ready (read: dumbed-down, simple, hassle-free, supported, whatever) distribution, the hundreds of open-source projects we're talking about really don't do much for business.

Obviously there are other distributions out there, some might cost less. Ransom feels that there's a market for a business distribution. He's probably right, that's where the money is.

Re:Business model....

[Roblimo]

I fixed the formatting error. Thanks.

As for straightupness, remember that we're talking about it in corporate CEO context here, not about a normal person. For a CEO, Ransom is pretty okay. He uses buzzwords because it's part of his job, but the use here is light compared to most.

(An aside: Peter Wayner and I have registered BuzzwordInstitute.org so we can issue "buzzword compliance certificates" and such; it's a joke, one we'll get going as soon as we have time to actually make the site. We can use some help if you're up for it; we're both very busy. We'll give you some sort of impressive title, even. "Executive Director in Charge of ______" or some such. Peter pointed out that inflated titles are part of buzzwordness, necessary "for executive retention purposes.")

Anyway, I'll admit that I prefer interviews with people like Moshe Bar, who simply speak their minds, but some of the corporate activities are important and it's worth knowing what the Linux corpses are up to even if you have an RMS shrine in your dining room before which you genuflect 5 times daily.

- Robin

An actual corporate user chimes in

[funkyaintit]

Many of the people in this discussion are talking in broad generalities about how you *think* corporations are going to react to UnitedLinux; let me tell you how *I*, a sysadmin for a company that uses both Linux and Windows 2000 servers, gauge this interview.

First of all, Mr. Love is entirely correct about what corporations want, which is a standard, consistent server platform with regular upgrades. We currently use Caldera OpenServer for file sharing and DNS, and we love it. However, we're nervous about utilizing this platform for more than these two functions.

Example: we want to move to an LDAP mail and message server, akin to Exchange. So, I hit the boards and forums, check SourceForge (of course), and come across 8 - 10 server products that *sound* like they fit our needs. My problem is this: how can I tell that this is going to work as advertised on our systems, and how do I know that it's not going to bork everything else we're running? Sure, it's all *supposed* to run properly and play nice with the other boys in the garden, but everyone knows that server platforms, regardless of manufacturer, have interop glitches. Yes, it's true that, if I had all the time in the world and was being paid 200% more, I could download each of these products, set up a staging environment, and try to hammer out the bugs myself.

Or, I could install Exchange and be up and running with a high confidence level in about 3 hours. I know, I know- you all think Exchange is a steaming pile; but the reality is that it's quick and easy to install and administrate. Why? Because Microsoft has farms full of paid developers making sure that it is.

If UL can provide my company with the peace of mind Linux does not currently provide, it will have made a new customer. In turn, the rest of the community will benefit because the installed base will expand, and perhaps another developer gets a job. There's a whole lot of sweetness to be passed around!

One more thing: I would like to remind all die-hards here that corporate profits are the *only* thing supporting Linux development right now. Every developer who works on any open source project gets his beer/rent money from a job someplace; even though he doesn't get paid (most of the time) for the time he spends on a project, he only has the luxury of spending spare time because he has a job to cover the expenses. Why not support UL, so that maybe a few more Linux geeks can have a good job doing something they love?

Re:Ah yes the LWYAWMYA disease.....

[rice_burners_suck]

Yeah, man, don't you get it? Eschew duplicated efforts by... duplicating efforts!!! Seriously, make yet another [insert type of program here], so that everyone will have to maintain 100 different programs that ls in color, rather than just one program that does the same thing without the 100 times duplication! Oooooooooooh well.

Re:binaries not free?

[KjetilK]

IANAL, but AFAICS, yes. But you wouldn't really be competing with UL anyway, as what they really sell is the promise that "this will work".

Bogus Logic Alert - GPL Violation Ahead!

[ibis]

The binaries that are certified by the major ISVs and OEMs will not be made freely available for distribution by anyone.

If he is referring to binaries of GPL'ed programs, this is a violation of the GPL. The GPL specifically states:

"3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form..."

and

"6. ... You may not impose any further restrictions on the recipients' exercise
of the rights granted herein. ..."

So anyone who receives copies may redistribute.

'nuff said. Flame Mr. Love at will!

Re:Bogus Logic Alert - GPL Violation Ahead!

[Anonymous Coward]

Anyone may distribute the binaries if they want, but neither Caldera, or anyone else, is under any obligation to do so.

Re:Bogus Logic Alert - GPL Violation Ahead!

[Software]

I think you're misinterpreting his answer. When he writes:

The binaries that are certified by the major ISVs and OEMs will not be made freely available for distribution by anyone.

what he means is that the United Linux vendors will not be making the binaries available for free. This is perfectly legal under the GPL. People who get the source and compile it can distribute the binaries. But United Linux isn't going to help them.

Re:Bogus Logic Alert - GPL Violation Ahead!

[Linux_ho]

So anyone who receives copies may redistribute..

One person mentioned that recipients of the binaries can redistribute as long as they don't use the name UnitedLinux. I'm not even sure that's true (but IANAL). It would be illegal to redistribute a UnitedLinux binary ISO if you changed a single byte on the CD. But if it's the exact same thing they're distributing, I can't see how it's a violation of trademark. If I buy a Ford car, I can turn around and sell it to my neighbor as a Ford car, even though I have no association with Ford. When the BSA goes after somebody printing thousands of pirate MS Office CD's, they don't prosecute for trademark violation, they prosecute for violating the copyright. You can only call it a trademark violation if you change the product, or if the product isn't actually something that was produced by the trademark holder. RedHat got away with threatening Cheapbytes on the basis that the CDs they produced might not have the same quality control as RedHat CDs. Even if RedHat didn't encourage .iso mirroring, there would be nothing stopping anyone from distributing an iso electronically and calling it RedHat 7.3 or whatever, because it would be EXACTLY the same as the product RedHat produces.

Since they don't hold the copyright for the vast majority of software they're distributing, they probably can't use that to prevent binary redistribution. The first guy who buys a UL distro will get the binaries for UL all nicely packaged on one CD, and I can't see any legal reason why he couldn't just post the .iso on a server somewhere. The fact that it's going to contain one single line of GPL'd software on the CD means that they can't put a restrictive license agreement on the package as a whole. The only scenario I can see that would work the way Ransom described it is like so:

They might put some proprietary software on the CD along with all the open stuff, and then they could restrict binary distribution of the entire CD, although there would be nothing stopping anyone from removing the proprietary code and redistributing the rest of the binaries. Of course, then they would not be able to use the UnitedLinux name, because they changed the content of the CD.

Personally, I'm sticking with RedHat for my company's servers. I'm happy with their support and support of their product by the community. I'm delighted that they can cultivate a positive relationship with that community while at the same time making money and paying the salaries of free software developers. They understand the community, and they get its support; it's as simple as that.

Re:Bogus Logic Alert - GPL Violation Ahead!

[FreeUser]

what he means is that the United Linux vendors will not be making the binaries available for free. This is perfectly legal under the GPL. People who get the source and compile it can distribute the binaries. But United Linux isn't going to help them.

Quite right.

However, anyone purchasing a copy of the certified binaries may in fact distribute those binaries, for free, to as many people or institutions as they like. If United Linux were to require people to sign contracts agreeing not to redistribute these certified, but nevertheless still GPLed, binaries, then they would most certainly be in violation of the GPL, having added additional restraints to the GPLed product in direct violation of that license.

So, in other words, United Linux cannot go after Cheap Bytes when they release a $2 CD of UL's $800 product.

Re:Bogus Logic Alert - GPL Violation Ahead!

[elflord]

One person mentioned that recipients of the binaries can redistribute as long as they don't use the name UnitedLinux. I'm not even sure that's true (but IANAL). It would be illegal to redistribute a UnitedLinux binary ISO if you changed a single byte on the CD. But if it's the exact same thing they're distributing, I can't see how it's a violation of trademark.

I think it's a violation of copyright. One can still copyright an ISO image. I'm not fully sure of the trademark implications, but if the United Linux CD carries any promise of support, then you obviously can't redistribute it without creating the false impression that you're also transferring the support. Your car analogy is not analogous, because you're not making unsupported clones of factory models.

Re:Bogus Logic Alert - GPL Violation Ahead!

[Bodrius]

I would think they would only care about someone misrepresenting the name UnitedLinux; that is, pretending to be a distributor that provides/transfers the year-of-service contract along with the copy.

Anyone will be able to take the CD, copy it, and give it as a "copy of the UnitedLinux distribution", but they would not be able to call it "the UnitedLinux product" because what they are selling is not the software per se, it's the software + support.

I guess it will be similar to OEM generic electronics sold by part-number and retail versions. You can resell them and trade them, but putting them in a nice box with some cables and selling them as the "ATC GraphicBoom 8080" or the "Crux PC2100 RAM with eternal warranty" would be fraud and probably a trademark violation. You will have to sell them as generic or provide a lot of information that differentiates them from the retail product.

Re:Bogus Logic Alert - GPL Violation Ahead!

[iCEBaLM]

One person mentioned that recipients of the binaries can redistribute as long as they don't use the name UnitedLinux. I'm not even sure that's true (but IANAL). It would be illegal to redistribute a UnitedLinux binary ISO if you changed a single byte on the CD. But if it's the exact same thing they're distributing, I can't see how it's a violation of trademark. If I buy a Ford car, I can turn around and sell it to my neighbor as a Ford car, even though I have no association with Ford.

You can change the tires, install a new audio deck, modify the engine and still sell it as a Ford car.

You can do the same with UnitedLinux, the GPL provides for that right even.

-- iCEBaLM

Re:Bogus Logic Alert - GPL Violation Ahead!

[Linux_ho]

I think it's a violation of copyright. One can still copyright an ISO image. I'm not fully sure of the trademark implications, but if the United Linux CD carries any promise of support, then you obviously can't redistribute it without creating the false impression that you're also transferring the support. Your car analogy is not analogous, because you're not making unsupported clones of factory models.

Claiming copyright on the iso image might give them a legal means to limit distribution -- you might be right there. But it's legally pretty murky, because a copyright must be a significantly original creation: "sweat of the brow alone is not the 'creative spark' which is the sine qua non of originality." (*).

No CD or iso image carries a promise of support. Hell, there is a disclaimer of no warranty and no support in nearly every software package that will be on the UnitedLinux CD. Caldera, RedHat, Microsoft, and most everyone else don't offer support unless you have a valid, unique key code or other proof that you paid for the support. Why would people's expectations change for UnitedLinux?

With the car analogy, I was just making the point that using someone else's trademark to sell a product isn't against the law unless you're misrepresenting the product you're selling as something it's not.

(*) Bridgeman Art Library v. Corel Corp, 36 F. Supp. 2d 191 (S.D.N.Y. 1999)

Re:Bogus Logic Alert - GPL Violation Ahead!

[Linux_ho]

You can do the same with UnitedLinux, the GPL provides for that right even.

Whoops, I should have been clearer. When I said, "It would be illegal to redistribute a UnitedLinux binary ISO if you changed a single byte on the CD," I really meant, "It would be illegal to redistribute a UnitedLinux binary ISO and still use the name UnitedLinux if you changed a single byte on the CD."

The arrangement and choice of software packages are what defines a Linux distribution. If you change anything, it's not really a UnitedLinux distribution anymore, and you're violating the trademark if you say that it is.

Re:Bogus Logic Alert - GPL Violation Ahead!

[quantaman]

You're of course assuming that all the software included with the CD is GPL'ed. Perhaps they just take a few BSD programs, work them heavily into the system and relicence them. You can't redistribute this program and it's heavily integrated into the OS (maybe even like IE) and while it acn be removed it is such a pain and a legal problem seperating the GPL'ed components that it cannot be easily redistributed.

Re:Bogus Logic Alert - GPL Violation Ahead!

[elflord]

No CD or iso image carries a promise of support. Hell, there is a disclaimer of no warranty and no support in nearly every software package that will be on the UnitedLinux CD.

That might be true, but the fact remains that customers might expect a certain quality of service associated with a certain brand name. That in itself would probably give them some kind of case.

With the car analogy, I was just making the point that using someone else's trademark to sell a product isn't against the law unless you're misrepresenting the product you're selling as something it's not.

Right -- so when you say you're selling "United Linux", when united linux is actually a product that is supplemented by support, then you are misrepresenting the trademark.

Re:binaries not free?

[ComputerSlicer23]

Yes you can. However, you can't call it UnitedLinux, that is an infringement on their Trademark (I'm assuming they have that name trademarked). If they are very clever, they will pull the OpenBSD trick of copyrighting the ISO Images and directory layout.

Suse has had nearly this policy for a while. I know they haven't given out ISO's for a while, I'm not sure on binary packages. I'm going to guess that some of the companies are going to give out free ISO images that are based on the UnitedLinux source, but aren't "UnitedLinux". So you can't claim it is certified.

People will still want the free desktop OS for workstations and whoever continues to provide that has a huge marketing/PR advantage over everyone else. However, that advantage can be quite costly to provide. So it might all come out in the wash.

RedHat does essentially this in the Advanced Server Platform. Go try and find those ISO's. They aren't anywhere on RedHat's site. I've hunted high and low for them. It is very similar to the Personal/Professional release execpt for some of the kernel patches they apply, and a few 3rd party tools.

The next step after this is that RedHat isn't bothering to get Oracle certified under all of the Personal/Professional Releases, only on the Advanced server. If isn't certified, Oracle shouldn't support you. I would imagine with all big proprietary products will be the same. It isn't cost effective to re-certify RedHat for all the big name products every release. So from a purely business perspective it makes a lot of sense.

So RedHat can give away free workstation copies but get you to buy a copy of Advanced server for all of you're server machines. I'd imagine that UnitedLinux will end up doing something similar somehow.

Kirby

GPL is a SOURCE CODE license

[Uggy]

I always wondered why Redhat and Mandrake and the ilk always made binaries available for download for free. People would sometimes scream and cry about SUSE not offering ISO's at EXACTLY the same time as the packaged distro for purchase. People would scream bloody murder but forget that the GPL is a SOURCE CODE license, not a binary license. The binaries are yours to do with what you want. If you want to sell your brand of Linux for 1 million dollars per seat, you can. However, you cannot deny anyone access to the gpl'd source code. If they wish to recompile the entire distro and sell it for $1 or give it away, there's nothing Caldera can do to stop them. And there's no reason for Caldera to worry about that.

They don't care. They want to do business with people that want to exchange money. Companies forge relationships by the exchange of money. That's how business is done.

They like it.

Open Source coders build relationships by exchanging code.

They like that.

The GPL allows for both and encourages both. It's really pretty simple and elegant... any engineer can appreciate that can't they?

Re:GPL is a SOURCE CODE license

[Tomun]

That's not entirely correct. You can deny "anyone" access to the source code. Only those who have received the software (in binary or source form) are entitled to the source code.

You can sell binaries with source on cd under the gpl and refuse to give the source or binaries to anyone else.

Of course those who have it are entitled to redistrubute but they bear the responsibility of source distribution, not you. They cant put that burden on you if you have already made the source available to those that you distributed to.

My god you are all pricks!

[shren]

Here goes my karma.

A CEO takes time out from his busy day, and the highest ranked comments accuse him of being a liar or against the 'cause'.

Moby says that just maybe P2P cost him a dime or two, and now he's a no-clue traitor to the cause who sucks now anyway.

You suck. You all suck. Nobody should agree to be interviewed by you. Other sites would be better off refusing HTTP referals from slashdot because too many people here are incapable of being civilized.

The embarassment to the OSS movement and the DRM debate caused by Slashdot posters has probably set both respective causes back 6 months.

Now I understand the trolls.

Now I agree with the trolls!

Please people, learn how to be decent human beings!

Re:My god you are all pricks!

[Hard_Code]

"My god you are all pricks! (Score:3, Troll)"

Haha...it's marked Troll. Priceless! The circle is complete.

Re:My god you are all pricks!

[_Sprocket_]

Here goes my karma.

Any post that starts with a comment about karma is immediately suspect. If you truely have conviction in your words, don't try to paint yourself the martyr.

A CEO takes time out from his busy day, and the highest ranked comments accuse him of being a liar or against the 'cause'.

Moby says that just maybe P2P cost him a dime or two, and now he's a no-clue traitor to the cause who sucks now anyway.

Both Ransom Love and Moby CAN, in fact, be wrong. And when people perceive that they are wrong, they will voice those observations in forums like this one. Its less about "the cause" and more about differing opinions, perceptions, and fact.

Now I understand the trolls.

Now I agree with the trolls!

Trolls either enjoy the sport of creating chaos or are simply unable to fathom differing opinion and possibly lack the ability to properly frame their own argument. In the end, they are either simply noise or the exact foaming-at-the-mouth zealot they like to claim to be exposing.

Please people, learn how to be decent human beings!

Right. And use phrases such as "My god you are all pricks" and "You suck. You all suck" perhapse? Good point.

The purpose of an environment such as Slashdot is to discuss issues. This discussion will include opinion and nobody will agree on every opinion expressed. But this is the stuff of good debate. Join in. But don't take it personally if someone find fault with your thoughts.

Re:Gee....

[photon317]

Whether he succeeds or not is irrelevant. His *intention* is to to do as I stated.

Clarification re: Red Hat

[jeremy_hogan]

[snip from question #6]
Now you appear to be in competition with Red Hat (on server) and Mandrake (on desktop) who both give their software away. Red Hat makes it's money from service contracts and Mandrake from special software for paying customers. I guess my question is how can you compete against them, when they are just as good and give it away for free or cheaper?
[/snip]

[snip from response]
Ransom:

It should be noted, first off, that Red Hat has moved to a model on advanced server where they are not giving away the binaries and they are charging around $800+ for their advanced server product.
[/snip]

The wording here is very misleading. No fault of Mr. Love, he has not the insight to make statements about our positioning or products, so here's a bit of clarification of what Advanced Server is and isn't:

1) The charge is for the packaging, updates and support subscriptions, not for the software in and of itself.

2) The code is still available. Binaries, Red Hat Network, ISO's/CD's are available for those who pay for the whole package. The source will be available publically for all to build their own.

The intent is *not* a per boxed set price of $800+. The intent is *not* to keep any errata or other updates exclusive to paying AS users, but to keep the convenience and associated services exclusive to those that pay for them. Which is very fair.

There was also a question about patents and a comment as to whether we could/would be held to uphold "Our Promise".

Our patent policy is issued as an estoppel statement which binds us as well as subsequent "owners" of such patents to the spirit of our intent. That is, any party relying on Our Promise has an absolute defense to an assertion of patent infringement by Red Hat (and subsequent owners of any such patents), subject to the limited obligation that they not sue us for patent infringement with respect to software we have produced.

http://www.redhat.com/legal/patent_policy.html

I'm no lawyer, but in layman terms, the agreement is much more than a press release.

--jeremy

Re:Clarification re: Red Hat

[tlambert]

"The source will be available publically for all to build their own."

Yes. But if someone compiles it on their own, are they permitted to call it "RedHat Advnaced Server"?

The answer is "no".

This is the same branding play that UnitedLinux is trying to sell. IMO, it's a direct response to the effective RedHat monopoly as a third party Linux applications platform.

-- Terry

Re:Clarification re: Red Hat

[Zigg]

... If you've got binaries out, you have to give up the source.

Do you have the binaries? No? Then they don't have to give you squat.

I can take something GPL'd, modify it, and sell the compiled result for $1,000,000. I only have to give the source for my mods to the people who bought the binaries from me. They, of course, are free to redistribute once they've got it.

In related news...

[zurab]

Mr. Love has charged /. and is planning to charge all readers a "per question fee" for this interview. This is to limit the support liability for these companies and to ensure a high quality, consistent answers. These fees will not be per seat as previously feared. Instead, the restriction is per server.

On a more serious note, it seems like UnitedLinux will be trying to feed corporate customers to what they are used to - being ripped off. The general attitude at an average corporate IT consumer, most will agree, is and has always been "you get what you pay for". It's certainly so for the management who makes decisions. At the same time the same attitude is being smashed and ridiculed by most in the community when it comes to software (I am not talking about support!). UnitedLinux will be landing 2 different impacts with this move - make corporate friends, and alienate an average developer to develop for UnitedLinux. They will try to work this from top to bottom, not from bottom up.

But by pushing only in corporate server market, in my opinion, will not put UnitedLinux over the hurdle; you need to have the word of mouth and average developer support. Again, in my opinion, UnitedLinux selling back to developers what developers have given to UL for free will not sit right with many. That was I think one question that should have been asked - Mr Love, to accomplish UL goals do you believe today that you need developer support? What do you think his answer would be?

Italics quoted from interview.

Re:Elaborate please

[photon317]

Oh sure, that would be a good point, if it weren't so fatally flawed. Please elaborate on your use of the word "so". In what context does it really apply here?

Multiple is better!

[Steveftoth]

what happens when you hose one script in linux with multiple scripts....

that service doesn't start.

what happens when you hose your one master script file?

the computer doesn't start?

I agree that simpler is better but sometimes you need to be modular and multiple files solves that problem better then one large modular, specially formatted file.

example: xinetd vs inetd, inetd uses one script for all services, it got really really messy. Format couldn't be changed when new options came in... etc.
xinetd has one files per service, the files are extendable format. If they decide to change the format of the file in the future, they can without breaking old files.

Two conflicting UL target markets?

[rakslice]

"The restriction on binaries is to ensure product quality and consistency of the brand for hardware and software vendors and for the quality of support within the business organization. "

Heh... Well, that could be an argument against the distribution of _modified_ binaries not labelled as such, but not one against the distribution of _unaltered_ binaries. The only reason I can see to restrict the distribution of unaltered binaries is to obtain more money.

Now, keep in mind that all GPLed bits of UL will be freely available to the public in short order. (After all, the UL members can't enjoin their customers from distributing the GPLed bits, so even if the UL members won't do it directly, the sharing that is the basis of the linux community will ensure that they sneak out somehow...)

Anyway, the UL base seems to have a good chance of reaching the segment of the enterprise market looking for well-supported linux server distributions (that Red Hat is trying to reach with their Advanced Server product, for example). They tend to buy distros that contain a lot of non-GPLed value-add anyway, and will spend big chunks of cash on wider (more "end-to-end"-integrated) products.

But, it's kind of odd to try to go after them with the "standardization" angle, isn't it? I mean, they don't care very much about standardisation as a tool to integrate products from different vendors, because they don't want to spend any time on that anyway. Reaching a minimal cost to fit the requirements isn't really very important to most enterprise customers, when it comes right down to it, as long as the cost is below a subjective "too high" level. So wouldn't it be better for UL to pump out as much distribution lock-in as the customers can handle?

So why not a free (beer) UL-based distro...

[weave]

This certainly isn't my idea, I read it in previous UL thread, but I haven't seen it mentioned here yet.

If the UL distro is required to distribute binaries, why can't someone just compile them all and distrubyte/release a UL-based binary and source distro with no strings attached under some different name? FUL anyone? (Free United Linux that is... :)

Re:So why not a free (beer) UL-based distro...

[weave]

Sigh...

s/distribute binaries/distribute source/

s/distrubyte/distribute/

:-(

Re:So why not a free (beer) UL-based distro...

[spitzak]

This question was asked several times here, and from the responses, nothing stops a person from doing this.

However they will not be able to call it "United Linux" and their binary will not have the "certification" from Oracle and other companies that their software works on it. Thus this will have no effect on United Linux's business plans because their market will be uninterested in this free version.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Having the source lets you create the binaries

[Junta]

Seems valid, but I wonder where that sits if it came to court. How is the "preferred form" defined? How is "normally distributed" defined, and it specifies "the" operating system on which the executable runs, but operating system could be vague too. For example, say Company X released a version of it's software for Company's X version of BSD with Linux Emulation. They say it is GPLed and release source but requires tools available only in their special version of BSD though the sources run under vanilla linux. They also withhold their version of BSD from external use. So now, there is source as modified by the company, and it just needs something normally distributed with the operating system on which it runs.

This is all too troublesome to actually happen, but I have to worry about the shaky ground of the GPL at times. Time to flush the karma down the drain for that last statement.

Robilmo is right

[haggar]

Usually I would disagree with Robilmo, if nothing else then just for the heck of it ;o) but this time I can't but agree completely: this interview really did provde me with straight answers. A very useful and as a consequence, rather envolving.

Now I wish more than 10 questions could have been allowed....

As an aside; I was a bit surprised R. Love hasn't mentioned Caldera's contribution to Linux's TCP/IP stack. If you ask me, that's more useful than RPM.

Mr. Love is not United Linux

[HiThere]

Unfortunately Mr. Love is the most outspoken (at least in English) member of the United Linux group, because he is not someone that a sensible party would choose. I *do* wonder what Caldera brought to the party, as practically any other English Linux distribution would appear to have more reasonable spokesmen.

OTOH, he is making sure that they get publicity. Negative, perhaps, but publicity.

Re:Mr. Love is not United Linux

[ivan256]

As far as I can tell, the only reason any of these companies package distributions is for control. They're out to make a buck on someone else's labor, and they'd have a hard time selling something that they have no control over. Once guys like Love realized that they couldn't make money selling something that was free, they started this scramble to keep their heads above water instead of slowing down and rethinking their plan. There is money to be made in linux, but they aren't paying attention to where the open wallets are. One thing is for certain, they're not going to get themselves any further by telling their volunteer labor staf that we 'owe them' or something.

It's amazing how high the percentage is of paying linux software customers that run Debian instead of one of the hyped up corporate distributions...

Re:Mr. Love is not United Linux

[elflord]

As far as I can tell, the only reason any of these companies package distributions is for control. They're out to make a buck on someone else's labor, and they'd have a hard time selling something that they have no control over.

Packaging distributions takes a lot of time and effort. They're not "making a buck on someone else's labor", what they're doing is taking care of the packaging and integration work that most programmers don't have the time to deal with.

nce guys like Love realized that they couldn't make money selling something that was free, they started this scramble to keep their heads above water instead of slowing down and rethinking their plan.

Packaging a distribution is not "free", it takes time and effort, which has to be paid for (commercial distros) or donated by volunteers (Debian).

It's amazing how high the percentage is of paying linux software customers that run Debian i

What, it's up to 1 already ? (-; Seriously, what is it ?

Re:Mr. Love is not United Linux

[ivan256]

Packaging a distribution is not "free", it takes time and effort, which has to be paid for (commercial distros) or donated by volunteers (Debian).

Yes, but that cost is a small fraction of what they would have payed had they developed the software themselves. Yet they still expect to command the same fees as someone like Microsoft. Why do they think they can charge premium support fees for software they only packaged? Customers don't want 3rd party support for full price.

...paying linux software customers that run Debian

What, it's up to 1 already ? (-; Seriously, what is it ?

I can't pull the numbers together quickly enough to give you an exact figure, but on new purchases it's around 50%. It's also used by every OEM that integrates our products (by their choice).

(The previous comment is in no way an endorsement by my employer of any particular distribution)

Re:Mr. Love is not United Linux

[elflord]

Yes, but that cost is a small fraction of what they would have payed had they developed the software themselves. Yet they still expect to command the same fees as someone like Microsoft.

They don't sell as many boxes as Microsoft either. And Microsoft don't do a comparable job with packaging. They only package their operating system, they certainly don't package hundreds of third party applications.

Customers don't want 3rd party support for full price.

I'm not clear on what you mean by "full price", or "third party support".

I can't pull the numbers together quickly enough to give you an exact figure, but on new purchases it's around 50%

I don't know where you get that figure from, but I don't find it plausible.

Re:Mr. Love is not United Linux

[elflord]

It's very simple. They don't write the software, so they are a third party. Full price is what you'd expect to pay to a first party OS vendor like Sun or Microsoft.

Which is a bogus way of computing price, because it makes the obviously false assumption that (a) the products are comparable (they are not. Microsoft do their own packaging, build management and integration, and the products they do that with are less numerous than a Linux distribution), and (b) there are no other factors which can/should impact price.

Re:Mr. Love is not United Linux

[elflord]

It doesn't matter what's bogus, the only thing that matters is the thing that gets someone to write you a check. If you go out of business because nobody buys your product, you can't be crying about how bogus the marketplace is and make it all better.

If you look at it that way, corporate customers are not like cheapskate slashdot lusers who whine about how something is "too expensive" on the basis of some ill-founded claim that they think the product "should be" cheaper. They're only interested in value for money. If a commercial Linux distribution provides better value to the customer than competiting alternatives (like a free distribution, or a free distribution with separately purchased support, or Microsoft or Sun), then they will go for it. Corporations are not very interested in nonsensical moral arguments about how much someone "should" charge,

Heh, he'll have the last laugh.

[S1mon_Jester]

People forget that OSS is a development model, not a business model. Linux has proven the effectiveness of the OSS development model. (So the OSS stuff is just snow, okay?)

Now...I think Love has correctly interpeted the business market and recognizes that unless some radical changes are soon had Caldera (and a number of the other small distributions) are going to be out of the market. Right now, RedHat has the mind share and Mandrake the market share. (Debian, much as I like it, is a non-issue for most businesses and is, for the most part, a speck on the business horizon. --- which is, imo, a good thing.)

So, Love proposes (to other soon to be out-of-business distributions) commodizing the OS. This isn't a bad thing (especially for Love if he can get them to agree). This allow other distibutions to differentiate on other things ... sorta like how Dell, Gateway, and Micron all support Microsoft. Now, the question is how much of a hold will Love keep on this? If he keeps too loose a control, the distributions will fragment (losing effectiveness), if he keeps too much control, the distributions will quit and won't join him.

It'll be interesting to see if he can pull it off.

But then again, considering his alternatives, he didn't have much of a choice.

Re:Number of days before someone clones their dist

[the-banker]

Ok, I think you completely missed the point on most of his answers. The fact that the binaries can be assmebled onto an ISO with an install script is nothing. Go ahead and do it. Licensing is on your side, and it isn't what they are selling.

Go to a large enterprise (their target), and give them a choice:

1. This distibution, built form a consortium of companies CERTIFIED to run with Oracle, DB2, Checkpoint-1, etc etc... for some price (say $2,500)

2. This free distribution which has all the same components, but is backed by some volunteers over the internet. It mirrors (1) in all regards but support and certification.

Most medium to large companies will go with (1) in a heartbeat. So go ahead. Make a distro. The linux community will shun it because it is essentially UnitedLinux, businesses will shun it because it isn't what they asked for.

I am no Ransom Lover (had to get that in), or a fan of SuSE, but they are finally hearing what commercial enterprises have been demanding - reassurance. Red Hat has moved even further down this path with training certs. I don't think United Linux will make a huge impact. I don't ever see myself considering it. But then again, my annual revenues are markedly lower than the annual revenues of the people they are targetting this thing for.

Re:Having the source lets you create the binaries

[Hard_Code]

For example, here is the entire codebase of my free Enterprise Application Server:

/* this code covered by GPL, see COPYING */
a

Of course, you can just pay $GAZILLIONBUCKS to me to save yourself the hassle of compiling it. :)

Re:*Because*.... You as a corporate 'admin'

[HiThere]

I do have experiences that back your assertion as to how they will figure things. I may feel that they are lamentably ignorant of history, and foolishly trusting of salesmen's promises, but they do frequently figure that way.

This doesn't keep them from being wrong, of course. But I can understand trying to sell in that market. There are a lot of buyers they. Of course they are mainly committed to MS, with a faith that even the recent license changes won't shake, but there are a lot of them...

Re:Great interview

[bolthole]

Debian only includes GNU-certified Free Software.

Incorrect. Debian only has IN MAIN, software that complie with the Debian Free Software Guidelines. That includes GNU, AND other licenses. And then there's a buncha stuff in 'contrib', that isnt strictly 'opensource', but IS 'freely distributable'.

Re:For anyone..

[laserjet]

Exactly what the hell is wrong with you kKeybBoard? Or this is an attempt to be cool?

I still dont understand the total lack of binary

[nurb432]

If all the 'distros' will be based off a common 'universal' cd ( the first disk he spoke of ) and then VAR cds added to the set to refect each companies slant..

Why isnt the *first* disk to be available if its
not got any 'special' stuff that must be supported, as he says is the reason for not releasing binaries..

Ever hear of 'unsupported' download versions?

Sure they need to make money too, but what if i dont care about the VAR parts, but would like to be compatible with this 'standard'...

Does that mean i can download the source to recreate the 'base' system totally?

Re:Having the source lets you create the binaries

[HiThere]

Sure there are loopholes, of that variety.

But to use that loophole they would need to train their programmers to write in that language. And the language would need to be different enough from all of the standard languages that it would be difficult to figure out what it was doing (or it would just get translated). And then they would need to have those programmers develop, in that language, something that was so useful that it would entice people to use their software...

Somehow this doesn't strike me as a good bet. Legal, but not practical.

I suspect that the main problem is that we are being presented with this one particular spokesman, who seems to specialize in antagonizing Linux advocates. I do wonder what the people in Germany and Japan are hearing. I'm relatively sure that the spokesmen choosen by SuSE and TurboLinux would not be so ... abrasive.

Contributions? really

[Eversor]

Ransom gave this link for contributions from caldera: http://www.caldera.com/developers/community/contri b/ Lets take a look at these. AIM benchmarks: Well if you are not satisfied with cat /proc/cpuinfo now you can clock your system with benchmarking software that caldera has provided for you. Caldera Open Administrations System (COAS): these are supposed to be contributions to the entire linux community not a small crappy GPL'ed program they use to promote their own bastardized distro? CScope: it is amazing how they take credit for this, as SCO (Santa Cruz Operation) released this openly before Caldera had anything to do with them Java: Now you are taking credit for Java?!?!? Linux Kernel: and I quote "Caldera has contributed several Linux kernel enhancements, including Windows support, IPX support, NFS, and more". Well I know that I have excellent "Windows" support in my linux kernel, how about you. It looks as if the ipx support is genuine, but NFS has been around longer than caldera. Lizard: Yet another installation system. but this one seems to be very locked down, with not mention of OS licensing, or any available downloads. Again isn't this supposed to be contributions to the Linux community? Netscape: Well sure you saved netscape! OS is the only thing that has saved netscape to this point. NCPS for linux: more netware stuff. Well being that ransom did have a lot of dealings with novell, I guess contributing to this dying networking technology seemed like the right thing to do. NKFS: yet more dying netware support. OpenSLP: Well another BSD style license, for I program I'm sure everyone uses everyday. RPM: quote: "Working with Red Hat, we developed the first package manager." BS! The first and best package management system was the good old tarball. RPM has cause me more trouble than it has avoided me. Webmin: yet another BSD style licensed project "started" by caldera. WordPerfect 6.0: Well thanks for paying for the copy of WP for linux that I also paid for. (someone made a lot of money off this deal). WP for linux was so bad I had to revert back to staroffice after using wp for a week. some things are best not shown as a badge of honor. UDI: vaporware. looks like it was started back in 99, and 3 years later, it is still as un-heard of. So what we have here, is some loose attachments to BSD style licenses, and a bunch of programs written for the dying netware protocols. Ransom, look at what the linux community has actually DONE for you, such as what the foundation for your entire company is. and your ticked!! This guy dosn't stop, so he must be stopped.

Hmm.

[mindstrm]

Wasn't Ransom Love a character out of some old sci-fi from the 60's or 70's? I'm sure I have the book....

You seem to misunderstand.

[mindstrm]

They can charge whatever they want for selling those copies of GPL binaries. They are not in any way restricted in this.

What you mean is they are not charging you for a license to the binaries, but for the copy itself.
That's the GPL.

You seem to be implying that they can't charge much because it's only a copying fee; that isn't so; I can take GPL stuff, modify it, and charge enormous sums of money for copies.

the bottom line...

[gol64738]

basically, ransom love is rallying behind this concept:

windows admins can now be linux admins without the learning curve of compiling, libraries and linking.
binaries are guarenteed to work so any dummy should be able to install anything.

lovely...

Re:Having the source lets you create the binaries

[msouth]

well, to make it more interesting you want to have them create an internal tool that translates c into language X, and then have them translate the linux kernel and GNU tools into X, then make some really spiffy must-have ("now butters toast!") modifications to it (in language X), and then sell it, distributing the modified code in language X.

If I had an infinite amount of money and time this might be on the list of the things I would do just to see what happens.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Bogus Logic Alert - GPL Not Violated!

[Linux_ho]

Eewwww. Good point. I was thinking they would add proprietary software to the CD, but then I didn't think SuSE or TurboLinux would want them to do that.... I think you have hit the nail on the head, smiff.

Still, I expect to see "Untied" Linux available on the net as an iso with all proprietary items removed within about a week of the first UnitedLinux release. Ultimately, I don't care what they do, but I think all this negative publicity they're stirring up by not releasing the binaries isn't worth whatever they think it would have cost them to release the binaries.

As a corporate sysadmin I'm still going to recommend my company continues to subscribe to RedHat Network, because they manage to make money and pay free software developers while maintaining the support and goodwill of the free development community at the same time. How DO they do it? Perhaps Ransom will one day discover their secret methods.

Where can I get one?

[Roy Ward]

> This is the equivalent of Apple making computers without sound cards or high-end video, and Jobs saying that they think it's more "professional" that way (but ignoring the majority of Mac users who do audio/video work!)

I'm a Mac user who'd be more than happy not to have sound card or high-end video - as long as this was reflected in the cost of the machine. The Mac 'community' (and the Linux 'community' for that matter to bring this a little back on topic) are more diverse than some people realise.

Re:Number of days before someone clones their dist

[God! Awful]

Go to a large enterprise (their target), and give them a choice:

1. This distibution, built form a consortium of companies CERTIFIED to run with Oracle, DB2, Checkpoint-1, etc etc... for some price (say $2,500)

2. This free distribution which has all the same components, but is backed by some volunteers over the internet. It mirrors (1) in all regards but support and certification.

Most medium to large companies will go with (1) in a heartbeat.

This assumes that the purchasing managers at large companies are idiots, and will continue to act like idiots. Let's see how long they are willing to flush money down the toilet when the economic slump continues for another 3 years. Unless you are an experienced con-artist, any business plan that depends on the gullibility of the consumer is probably flawed.

Seriously, why would you pay $2,500 to get something when you can get the *EXACT* same thing for free. We're not talking about a cheap knock-off here. It's the *same damn thing*. So what if it's backed by a consortium of companies? If a bug in Linux causes me to lose data or allows an intruder to hack into my network, I guess I can sue Caldera, then? Why would I bother suing a financially strapped company with no discernable revenue source; why wouldn't I just use the free version and buy some insurance from a bank?

The way you describe it, it really does sound like the UnitedLinux companies are just selling insurance. Sure, they may contribute the odd bug fix here and there, but it reminds me of that insurance commercial where the car insurance company describes how they fund research into eliminating dangerous intersections "because ensuring your safety saves us money."

-a

Going forward

[Chris Johnson]

"Going forward, there will only be two platforms certified by the major hardware and software vendors, Red Hat and UnitedLinux"

I fail to see how this, for the open source community, qualifies as 'going forward'. It sounds like the seizing of power, instead.

Is this a blithe prediction of what will happen in the absence of any activity to the contrary? Or is it a statement that UnitedLinux and/or Red Hat will try to BLOCK 'certification' of other distributions? (cue 'we're just trying to turn a profit' meme here)

Somehow this just seems to be a rejection of something... or at the least a refusal to work towards something... why EXACTLY would JoeBlowLinux not deserve 'certification'? Is it because Joe isn't getting help in meeting the needs of other vendors? ...if it's that, then why not?

It's all about appearances.

[eddy the lip]

(In a similar context, Peter Wayner and I have acquired the sought-after BuzzwordInstinute.org to enable our issuance of "Buzzword Compliance Certificates" and related offerings. We will capitalize on the humour potential inherent in such a product immediately after the completion of an extensive Research and Development phase. We would be interested in exploring the potential for a mutually beneficial partnership with you. We can discuss a position on the Board if you feel, as do we, that this would enstrengthen our business relationship. Peter has brought to my attention that suitably descriptive titles are an integral part of Buzzword Compliance, and necessary for "Executive Retention Purposes").

In other words,

[Kashif Shaikh]

Who's the sucker? The company "stealing" 5000-man-hours code created by volunteers or the volunteers not understanding what the GPL is?

If you, the author of your code, don't like "Big Companies" stealing code, then change your license. Attach a license that more or less says, that no one can use your code commercially(for profit).

IF you don't realize this or don't care(and think GPL is the holy grail of "Open Source"(tm)), then I don't give a flying fuck who's using your GPL'ed code as long as they comply with the license.

As an aside, don't cry about UL not releasing binaries either; as they are adding a "cost-to-use" factor. For example, not many people would install linux, if you had to compile the kernel, compile 100s of packages, configure and write scripts, etc. all on your own.

Business-speak? Business model!

[einhverfr]

and a valid business model for Linux (Supportable business quality product through limited binary distribution and 12 month maintenance agreement with every license sold).

Hmmm.... IMO, this has been Caldera's problem from Day 1. It is impossible, IMO, to make money be decommoditizing Linux, and it looks like Caldera is going to try this one. Caldera has a worse track record in terms of businesses than the other members of their team, so I am unsure why exactly SuSE, TurboLinux, etc. are so quick to follow suit.

My prediction is that if Red Hat, SuSE, etc. are to survive for more than another few years, that they will have to come up with a free, unified distribution and stop trying to sell it in competition with eachother-- instead selling the services that are their life-blood, they are pumping large quantities of money into distributions which are costly to produce and are loss leaders.

Instead, a free universal distribution would help the consultants and OEMs reduce prices, and be better able to compete with UNIX and Windows. I had hoped that UnitedLinux would be it, but it looks like it will be really good for RedHat.

Again, they are trying to sell Linux to the wrong crowd. Too bad...

?? Re:Graphical Installers are bad.

[einhverfr]

For the Joe Sixpack user, I think the Packman game in the Caldera install is pretty useful ;)

I think some other distros offer this sort of thing (and Windows doesn't).

I stopped earlier

[Erris]

when ransom said:

I do not believe in a Linux model that requires ongoing charity to survive.

While I can't speak for them, I don't think Linux Torvalds or Richard Stallman believe in such a model either. Both Torvalds and Stallman are earning decent livings with their skills. A "limited binary" distribution is not a viable business model, service is. It's not the shiny disk it's the ability to use it that people value. Obfuscating the works by binary distro will make United worthless. How does the fsf say it, something about having the choice of who to give up your rights to.

OK, I lied, I did not stop there, I read most of his answers. He wants people to pay him a fee to develop his code? This is better than the current free code how? He's not going to say he will be using patents but defends their use? He's leaving he QC for "value added components" to his fee paying member companies?

He says many of the right things but his approach is exactly opposite of succesful coding:

For Linux to move from the peripheral of the business network into mainstream application server market, businesses must be assured that their platform is certified and will work with other applications and hardware solution in their environment. What the UnitedLinux customer is paying for is 1) the assurance that his applications will work together, and 2) the ongoing maintenance and support of that certified platform. The restriction on binaries is to ensure product quality and consistency of the brand for hardware and software vendors and for the quality of support within the business organization. I believe that Red Hat is moving to a similar model with business customers. The majority of the value will be in product assurance and maintenance. Both of these are of tremendous value to the business customer.

The only way to insure this is to use free code and nothing but free code. Certifying configurations of that free code will make plenty of money. Trying to rebuild everything yourself and trying to tack on propriatory junk that no one else can sync with will never work.

As for you, what's this all about? When there are applications you need that run on linux, you get linux. I see it every day as the purchase orders come in. No distribution provider out there looking to make a buck understands yet.

What do I need that runs on any other platform that does not have a substitute on Linux? I'm not aware of it, that's why I own one of those white boxes like 45% of all other computer buyers last year. These white boxes are going to come with Linux, Lindows, even BSD and restrictive software companies are goint to fade away into some foggy nightmare.

OK, I can stop now.

Re:I stopped earlier

[ivan256]

What do I need that runs on any other platform that does not have a substitute on Linux?

That's not what I mean. I mean that you go and get a linux based platform when there is software for linux that you can't get anywhere else. A question I hear probably once a week is "You're stuff looks great, does it run on Solaris?" When I say that it doesn't, probably 25% of the time they'll come back and say "Ok, so what do I have to do to get Linux?"

UL not so different from RHAS

[_|()|\|]

Red Hat has moved to a model on advanced server where ... they are charging around $800+ for their advanced server product. [Love]

Binaries, Red Hat Network, ISO's/CD's are available for those who pay for the whole package. The source will be available publically for all to build their own. [Hogan]

The availability of binaries and ISOs has obviously contributed to Red Hat's success. I find it obnoxious that some companies stick a proprietary license on the packaging tool to enforce, in effect, a per-seat license on the whole distro.

The proof will be in the pudding, but UnitedLinux sounds like it will be very similar to Red Hat Advanced Server. The core will be free software (with the possible exception of Acrobat Reader). The lack of binaries, ISOs, and a trademark license discourage use of the product without support. Support will cost much more than the regular products, and is, of course, per server.

This business of certifying a specific combination of Dell, EMC, RHAS, and Oracle is what really strikes me as exclusive. I wonder whether you could get an Oracle support contract on a built-from-source version of RHAS, without Red Hat support. (Given the relative price of Oracle and RHAS, the question is academic.)

What I find curious is that RHAS and UL are positioned as increasing developer support. As a developer, I'm going to find barriers to entry (such as lack of ISOs) annoying, considering that it may be but one of a dozen potential platforms. For that matter, are these really platforms to bet on? The increased cost may increase the O/S vendor's profits, but probably over fewer customers. If I'm Oracle, I may be able to require an "enterprise" Linux distro. Otherwise, I'd rather make it easy to use my product.

Verification and testing

[_Sprocket_]

Example: we want to move to an LDAP mail and message server, akin to Exchange. So, I hit the boards and forums, check SourceForge (of course), and come across 8 - 10 server products that *sound* like they fit our needs. My problem is this: how can I tell that this is going to work as advertised on our systems, and how do I know that it's not going to bork everything else we're running? Sure, it's all *supposed* to run properly and play nice with the other boys in the garden, but everyone knows that server platforms, regardless of manufacturer, have interop glitches. Yes, it's true that, if I had all the time in the world and was being paid 200% more, I could download each of these products, set up a staging environment, and try to hammer out the bugs myself.

Or, I could install Exchange and be up and running with a high confidence level in about 3 hours. I know, I know- you all think Exchange is a steaming pile; but the reality is that it's quick and easy to install and administrate. Why? Because Microsoft has farms full of paid developers making sure that it is.

I'm kind of curious as to where the leap happened between

"Sure, it's all *supposed* to run properly and play nice with the other boys in the garden, but everyone knows that server platforms, regardless of manufacturer, have interop glitches"

and

"...I could install Exchange and be up and running with a high confidence level in about 3 hours."

I would assume at some point, your organization took Exchange and ran some tests against it to ensure it would manage to deliver what it claimed. Perhapse you even talked to Microsoft and got some information as to what to expect out of their product and good implementation strategies. And you would have sought out reviews and opinions from industry news sources and technical discussion forums.

Many of the same strategies apply to investigating Open Source applications for the Enterprise. One advantage to Open Source is that its community and developers are rather exposed to the public - discussions over implementation, scalability, bugs, and other technical issues are often a google search away. Furthermore, these projects usually offer forums (web forums, mailing lists, usenet groups, etc) where one can seek out informed answers to direct questions. Sometimes these projects are the product of a group who consult on implementing the software - engage them. Or seek out some of the other organizations involved in providing business support for Linux and associated products (such as RedHat, SuSE, or Caldera). There may be enough free information available - or a small enough fee that is easily absorbed as a "cost of business" when compared to future licensing fees (or lack thereof).

The final step is simply testing. After you've narrowed down on a few possible products based on your desired needs (you DO know what your user base NEEDS, right?), underlying architecture, etc... implement it. Set up a testbed. Stress test it. Look for odd performance kinks and usability or administration issues.

Of course, this should be old hat. After all, we know not to trust the glossy brochure. The proof is in the performance. And that holds true no matter if the product comes from a Mega-corporation or a modest listing on Sourceforge.

UL does not change any of this. Granted - it does provide another vendor listing. But this is not something new. Linux vendors, contractors, and independant contractors have been serving business interests and concerns for years.

Jumping on the "brand"-wagon

[tlambert]

The reason for no binaries is obviously to control the brand "UnitedLinux".

Say I wanted a preinstalled "UnitedLinux" system, or I wanted a prepackaged "UnitedLinux" distribution.

What this does is prevent "Bob Schmoe's UnitedLinux distribution", since Bob is not able to legally use the trademark on a binary release compiled from their sources. Only binaries from the participating companies can use the trademark. Bob would have to "buy in" to obtain the same rights (Ransome said as much).

This puts UnitedLinux in relative the same boat as the UNIX trademark, where use of the trademark is controlled by licensing fees and/or buy-in to the club. The main difference appears to be that the base system is, in fact, compatible -- at least until you load on the vendor "value add" or try and load software from a third party that depends on a particular vendors value add (this is the same problem all non-RedHat Linux distributions already face today, since RedHat has used the same embrace-and-extend technique).

The consumer benefit to this is that if a software package that depends only on the UnitedLinux base system is shipped by a software company, they are guaranteed that the software will run on any UnitedLinux system.

The LSB fails to provide the extended features of UnitedLinux, and therefore there is a barrier to entry for LSB compliant systems, and even systems derived from the UnitedLinux code base, in the corporate market.

Or, in one sentence: it is a marketing play.

I actually rather doubt that there is a liability issue involved for UnitedLinux vendors, unless they (some day) pool their technical support services. The only liability left is on the part of software vendors, who want to guarantee that their applications will "run on UnitedLinux". If each vendor guarantees that, then a third party hacked up version could potentially fail to run the tvendor's application. So it's also a minor amount of risk mitigation for the vendors they hope to attract to get them to port software to the combined platform.

Most likely, what will happen instead, is that each vendor will have it's own support services for their product, and each vendor will embrace-and-extend UnitedLinux -- the same way that the UnitedLinux vendors have done the same thing in regard to the LSB.

Unfortunately, "standard, plus extensions" is, by definition, non-standard.

I don't think this is really a reesclation of "the UNIX wars"; GNU already lit that fuse when they started embracing and extending command like "tar" and "gprof", etc., with options that are not present in POSIX, and with more than single character options, and with expression reevaluation in GNU Make, etc..

What this works out to is a shot across the RedHat bow. It says that RedHat is not going to be permitted to own the distribution standards for Linux.

Frankly, someone needs to shoot at FreeBSD the same way (IMO).

-- Terry

patches

[_|()|\|]

How can one Linux distribution be 'more stable' than another? It has the same FSCKING kernel! The SAME drivers, THE SAME libraries all written by THE SAME people!

No, it doesn't. Linux from Scratch or Gentoo may use the latest and greatest, but most major distributions patch (i.e., fork) the major components. They back port bug fixes to a relatively stable version. That, and testing, is why Red Hat 7.2 stayed on Linux 2.4.9.

In fact, the situation is even crazier with Red Hat Advanced Server 2.1. It has asynchronous I/O, more scheduler patches, and lots of goodies you won't find in 2.4.18, let alone 2.4.9.

I actually find the RHAS scenario kind of scary. Red Hat 7.3 was unusable on our VA Linux box until the ext3 update (2.4.18-4) and the Mylex update (2.4.18-5). Do I really trust them with more advanced features and an even smaller (albeit, higher paying) user base?

Re:Having the source lets you create the binaries

[sydneyfong]

Yes but if you base your source code off a GPL'd codebase, then if you moe the source to the build code, you also have to GPL the build code, thus still releasing the code. If you didn't base your code off a GPL'd codebase, then why bother to GPL anything if you don't want to?

MacOS 1.0 didn't have a graphical install.

[Per Abrahamsen]

It was pre-installed on the Mac, and it didn't run on older machines (well, it kind of ran under emulation on the Lisa). It may even have been stored in ROM on the Mac.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Business model....

[Enigma2175]

an RMS shrine in your dining room before which you genuflect 5 times daily.

Don't you mean GNUflect?

You have no idea what the GPL is.

[jotaeleemeese]

So please, don't even try to say who the enemies of the OS movement are.

As long as a company respects the GPL they are very welcome to use GPLed software to make as much money as they possibly can.

Widespread use of OS software makes everybody's life easier and better, so all the power to companies trying to make a buck from GPLed software.

And why do you care?

[jotaeleemeese]

What matters is:

1.- That Caldera respects the GPL (or any licensing for that matter, but most people here have an interest in OS licenses).

2.-Caldera can convince clients that it is worth paying for the "marketspeak gobbledygook" as you put it.

What should you or anybody else take issue at them as long as they respect the GPL?

Re:And why do you care?

[wfrp01]

There's more to life than the GPL. ;)

The problem with "marketspeak gobbledygook" occurs when people pay for something they aren't getting. My question is, what does UL's "certification" really mean? If it means someone will stand behind the certification and take a hit if they don't fulfill their promises, that's worth something. If not, why should anyone prefer UL over any other brand? I see brand building, but so far, that's all I see.

And that violates GPL how?

[jotaeleemeese]

Please enlighten us.

Get a real job and come back.

[jotaeleemeese]

In a corporate environment there is not enough time and resources (people) to compile your stuff for every single utility or program you need.

When your priority is to keep systems running the last thing you need is to "learn" to link and compile, these activities are completely superfluos to the objectives of your work.

In case you did not notice (illiteracy is high everywhere), the source code is fully available (as it should be, this is GPL stuff we are talking about) so if you find a problem you can still contribute to solve it. Or you can try to suit some code to your needs. You still have the certified binaries but can play with code to adpat it to your needs if so are you inclined.

There is need for some standarization.

[jotaeleemeese]

Big companies require standards, otherwise the cost of support is too high, and also some support form an entity entirely devoted to provide it is highly desirable.

A bank or an oil company are not in the business of testing and releasing operating systems, that is why they turn to other companies to do the work.

Companies like Caldera an RedHat can sparehead the introduction of Linux in big companies, thus bringing money into Linux development and convincing the skeptics about the quality of OS software.

People still have the other distros, and failing everything else, you can still choose to get the source from Caldera;s distribution and compile and link, activities you seem to be very fond off.

Re:Flawed logic?

[photon317]

Ok, I'll clear it up for you:

UL is repackaging a distro. It's not all that hard work, relatively speaking. The only things they'll contribute in the way of code will be things on the order of graphical installers. They'll employ a few hackers, but they won't have the budget to do much on that front in the long run.

IBM does make huge, significant contributions to Linux. They employ tons of linux people. They employ kernel hackers. They contribute significant code in all areas, both at the kernel and application level. They back linux publicly on their website, giving it corporate legitimacy. They back it with their basketball/linux commercials, which I'm sure cost a pretty penny to produce.

The bottom line is that IBM is doing right by the community, and by the spirit that made all of this software to come to be. They are giving back, and they are making the Linux model work. They are making their money off of hardware and service, and dumping money into linux (both the kernel and the larger "os" sense).

UL is doing wrong by the community and the spirit of things. They're taking a huge body of work performed by tons of people in the spirit of OSS, and they're repackaging it with a few frills and compatibility gaurantees. That alone is fine, that's basically what RedHat and everyone else does today. The difference is UL's domineering attitude, and their willingness to sacrifice the spirit of the open software just to attempt to make some cash. They may be complying with the written letter of the GPL and other licenses (I would hope), but you can bet that a large percentage of the code authors never envisioned their code being twisted into something like this, and that it makes them feel uneasy at best.