Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Red Hat annual Report

By vectro in News
Thu Jul 12, 2001 at 09:02:03 PM EST
Tags: Software (all tags)
Software

Below, I reproduce a section of the FY2001 Red Hat form 10K on the risks of open source.

I think this is an excellent collection of the problems associated with an open source business model. I'm interested to see the community's response.


Risks Related to our Linux-based Open Source Business Model

Our open source software business model is unproven

 We have not demonstrated the success of our open source business model, which gives our customers the right to freely copy and distribute our software. No other company has built a successful open source business. Few open source software products have gained widespread commercial acceptance partly due to the lack of viable open source industry participants to offer adequate service and support on a long term basis. In addition, open source vendors are not able to provide industry standard warranties and indemnities for their products, since these products have been developed largely by independent parties over whom open source vendors exercise no control or supervision. If open source software should fail to gain widespread commercial acceptance, we would not be able to sustain our revenue growth and our business could fail.

We depend on the support of Linux developers not employed by the us [sic] to release major product upgrades and maintain market share.

 We may not be able to release major product upgrades of Official Red Hat Linux on a timely basis because the heart of Official Red Hat Linux, the Linux kernel, is maintained by third parties. Linus Torvalds, the original developer of the Linux kernel, and a small group of independent engineers are primarily responsible for the development and evolution of the Linux kernel. If this group of developers fails to further develop the Linux kernel or if Mr. Torvalds or other prominent Linux developers, such as Alan Cox, David Miller or Stephen Tweedie, were to join one of our competitors or no longer work on the Linux kernel, we would have to either rely on another party to further develop the kernel or develop it ourselves. We cannot predict whether enhancements to the kernel would be available from reliable alternative sources. We could be forced to rely to a greater extent on our own development efforts, which would increase our development expenses and may delay our product release and upgrade schedules. In addition, any failure on the part of the kernel developers to further develop and enhance the kernel could stifle the development of additional Linux-based applications.

We may not be able to effectively assemble and test our software because it consists largely of code developed by independent third parties over whom we exercise no control, which could result in unreliable products and damage to our reputation.

 Official Red Hat Linux, in compressed form, consists of approximately 1.1 gigabytes of code. Of that total, in excess of 1,000 megabytes have been developed by independent third parties, including approximately 10 megabytes of code contained in the Linux kernel. Included within the 1.1 gigabytes of code are more than 800 distinct software components developed by thousands of individual programmers which we must assemble and test before we can release a new version of Official Red Hat Linux. If these components are not reliable, Official Red Hat Linux could fail, resulting in serious damage to our reputation and potential litigation. Although we attempt to assemble only the best available components, we cannot be sure that we will be able to identify the highest quality and most reliable components or successfully assemble and test them. In addition, if these components were no longer available, we would have to develop them ourselves, which would significantly increase our development expenses.

The scarcity of software applications for Linux-based operating systems could prevent commercial adoption of our products.

 Our products will not gain widespread commercial adoption until there are more third-party software applications designed to operate on Linux-based operating systems. These applications include word processors, databases, accounting packages, spreadsheets, e-mail programs, Internet browsers, presentation and graphics software and personal productivity applications. We intend to encourage the development of additional applications that operate on Linux-based operating systems by attracting third-party developers to the Linux platform, by providing open source tools to create these applications and by maintaining our existing developer relationships through marketing and technical support for third-party developers. If we are not successful in achieving these goals, however, our products will not gain widespread commercial acceptance and we will not be able to maintain our product sales growth.

We may not be able to generate revenue from sales of Official Red Hat Linux if users can more quickly download it from the Internet.

 Anyone can download a free copy of official [sic] Red Hat Linux from the Internet. However, because this download can take up to 36 hours using a standard telephone connection, many of our users choose to buy the shrink-wrapped version of Official Red Hat Linux. If hardware and data transmission technology advances in the future to the point where increased bandwidth allows users to more quickly download our products from the Internet, users may no longer choose to purchase Official Red Hat Linux. This could lead to a significant loss of product revenue.

We may not succeed in shifting our business focus from traditional shrink-wrapped software sales to offering subscription-based product and service offerings.

 We are focusing our sales and marketing efforts on providing subscription-based products and services as opposed to relying on sales of shrink-wrapped software. This change has required us to expend significant financial and managerial resources and may ultimately prove unsuccessful. The failure to successfully implement this transition of our sales model could materially adversely affect our operating results.

Our customers may find it difficult to install and implement Official Red Hat Linux, which could lead to customer dissatisfaction and damage our reputation.

 Installation and implementation of Official Red Hat Linux often involves a significant commitment of resources, financial and otherwise, by our customers. This process can be lengthy due to the size and complexity of our products and the need to purchase and install new applications. The failure by us to attract and retain services personnel to support our customers, the failure of companies with which we have strategic alliances to commit sufficient resources towards the installation and implementation of our products, or a delay in implementation for any other reason could result in dissatisfied customers. This could damage our reputation and the Red Hat brand and result in decreased revenue.

We may be unable to predict the future course of open source technology development, which could reduce the market appeal of our products and damage our reputation.

 We do not exercise control over many aspects of the development of open source technology. Historically, different groups of open source software programmers have competed with each other to develop new technology. Typically one of these groups develops the technology that becomes more widely used that that developed by others. If we adopt new technology and incorporate it into our products, and competing technology becomes more widely used, the market appeal of our products may be reduced, which could harm our reputation, diminish the Red Hat brand and result in decreased revenue.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Poll
Biggest risk
o Unproven business model 21%
o Depend on developers 11%
o Difficult to test 0%
o Scarcity of applications 19%
o Internet downloads 14%
o Failure of subscription model 21%
o Difficult to install 4%
o Failure to predict future of development 7%

Votes: 42
Results | Other Polls

Related Links
o FY2001 Red Hat form 10K
o Also by vectro


Display: Sort:
Red Hat annual Report | 40 comments (34 topical, 6 editorial, 0 hidden)
wow..redhat said it for me (2.00 / 7) (#1)
by rebelcool on Wed Jul 11, 2001 at 04:12:10 PM EST

all very good reasons why you just cant make money off of it.

COG. Build your own community. Free, easy, powerful. Demo site

Hmm. (4.16 / 6) (#2)
by Mawbid on Wed Jul 11, 2001 at 04:26:04 PM EST

No, these are all very good reasons why you might not be able to make money off of it. That's not to say I think there's real money to be made selling a Linux distribution, just that you're confusing "might not" and "can't".

[ Parent ]
Gawd. (2.66 / 3) (#11)
by regeya on Wed Jul 11, 2001 at 08:30:02 PM EST

It's all so simple for you, isn't it?

I envy you.


[ yokelpunk | kuro5hin diary ]
[ Parent ]

What is so complicated? (none / 0) (#32)
by darthaya on Thu Jul 12, 2001 at 10:05:11 AM EST

If you want to look at simple things in a over-sophisticated way, that is your choice.

[ Parent ]
Sure. (none / 0) (#40)
by regeya on Sun Jul 15, 2001 at 06:22:59 PM EST

And if you are willfully being a simpleton about it, that's your perogative.

[ yokelpunk | kuro5hin diary ]
[ Parent ]

Ya, and? (4.14 / 7) (#5)
by drivers on Wed Jul 11, 2001 at 04:43:32 PM EST

Companies put this kind of stuff in their reports all the time. It's supposed to prevent lawsuits, not that it helps much if the stock goes down. For a humorous example, read Cryptonomicon by Neal Stephenson.

Cryptonomicon (2.50 / 2) (#16)
by delmoi on Wed Jul 11, 2001 at 09:03:50 PM EST

I did read cryptonomicon... but it's a pretty damn big book. Could you be more spesific?

But I agree, this is basicaly the same as the notes on software that say, basicaly, that you shouldn't even expect the software to do anything, that it dosnt' have "fittness for any purpose".
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
Cryptonomicon (none / 0) (#39)
by scorchio on Fri Jul 13, 2001 at 09:41:46 AM EST

I did read cryptonomicon... but it's a pretty damn big book. Could you be more spesific?

The Epiphyte(II) Business Plan. Testament, holy book, inspirational text. Slickly and groovily word-processed by Avi, printed on paper handmade by isolationist Buddhist monks. Try opening the book randomly, you may find the page. Serendipity, and a' that.

[ Parent ]

The booklet was super-cool tho. (4.25 / 4) (#6)
by jayfoo2 on Wed Jul 11, 2001 at 05:08:04 PM EST

I must say I really enjoyed the little booklet they sent out with the annual report . Lots of cool propaganda and some neat graphs.

Ok for the serious part of this post: I remember reading Red Hat's annual report last year and being shocked by how many of these disclaimers there were (much more than most companies). I don't want to commit to this and I don't have the time to do the analysis today, but my very strong feeling is that this year's report has far less CYA items than last years.

The purpose of these type of statements is to help defuse shareholder class action lawsuits. You see them in every annual report and prospectus, even for big companies. Some of these are absolute boilerplate for technology companies (such as the one about how the internet isn't theirs).

The one piece of analysis that I might try to dig into this weekend unless someone does it before then is a comparison of RH's caveats and those of some big compaines (MS, Oracle, Sun, Intel, etc.) at the same phase.



"Open Source" not the real issue (4.00 / 7) (#7)
by Arkady on Wed Jul 11, 2001 at 05:19:16 PM EST

The document you're quoting here, though a good description of Red Hat's business issues, has very little to do with Open Source in general and everything to do with the particular parasitical style of business Red Hat was designed to pursue.

As the doc points out, Red Hat themselves produce basically none of their core "product", which is actually produced by "thousands of individual programmers". Instead, Red Hat takes their work, bundles it up into a couple of CDs, and sells it. Thus it is only reasonable that they're completely dependant on the continuing operation and goodwill (or at least tolerance) of these thousands.

Red Hat's specific problems are due to the fact that they themselves are not producers, but rather consolidators, in an environment in which such middlemen aren't actually necessary. Any unnecessary middleman is in a precarious business position. Red Hat is at no more risk because they're re-distributing Open Source than they would be were they redistributing proprietary software which could also be downloaded directly from the producer. The status of the source code is completely irrelevant.

-robin

Turning and turning in the widening gyre
The falcon cannot hear the falconer;
Things fall apart; the centre cannot hold;
Mere Anarchy is loosed upon the world.


Re: "Open Source" not the real issue (4.66 / 6) (#9)
by bigbird on Wed Jul 11, 2001 at 06:36:50 PM EST

They depend neither on goodwil nor tolerance, as most of the software they distribute is GPL / BSD (or similar) licensed. Microsoft has no goodwill in the community, and could package the same software as Redhat, under the same conditions.

"Parasitical" is certainly not a word I would apply to Redhat. They contribute to a wide range of projects, from Gnome to the kernel. I am not aware of much (any?) software they produce which is proprietary, unlike Caldera. Everything goes back into the community, and other successful distro's have built on Redhat, including Mandrake and Suse.

As for middlemen not being necessary, try installing Debian with a 2.4 kernel. With Reiserfs. With working sound (although Debian includes a sndconfig package, which is a Redhat configuration script. However, I still needed to compile a custom kernel to get my sound card and Reiserfs working in Debian). Redhat remains more user friendly. In terms of marketing, look at the four P's - product, place, promotion, price. They have a product which is better than non-middleman alternatives such as Debian for most markets. Place - they distribute boxed sets everywhere from software stores to Wal Mart, and not everyone gets 200 kB/s from their local mirror. Promotion - their marketing budget and public exposure are high enough that Redhat == GNU / Linux for most people. Price - boxed sets are far cheaper than any proprietary software, and their mirrors appear to remain in operation to maintain the size of their user base.

Middlemen will remain around for some time, as long as they add value. Redhat adds considerable value, even if their product is not unique. People will often buy a boxed set just for the manuals,and, outside of /. and k5, the rest of the world has no interest in ./configure, make, make install. The value they add is convenience, and support services (because corporations are not about to depend upon support from a nebulous "community"). Most people are less than keen on having some 15 year old tell them to RTFM in a Usenet group.

IIRC, they also employ Alan Cox and several other prominent GNU/Linux developers, support multiple projects, and employed some of the Gnome team back when the project started.

bigbird

For I am not ashamed of the gospel of Christ: for it is the power of God unto salvation to every one that believeth; to the Jew first, and also to the Greek. Rom 1:16
[ Parent ]

Re: "Open Source" not the real issue (4.00 / 3) (#10)
by Arkady on Wed Jul 11, 2001 at 07:52:52 PM EST

Despite the fact that they're dealing with GPL/BSD/etc licensed code, they're still dependant on the goodwill of the developers. It's true that code, once released under a license, cannot be revoked, but there's nothing in the license that commits the developer to releasing the next version (or patches) under the same license. If Red Hat pissed of, say Icaza, then he could release the next version of GNOME under the GPL for everyone except Red Hat, who could be forbidden from using it. That one version has been released under one license doesn't commit the developer to continue that way.

Perhaps parasite is strong for Red Hat, who have contributed some to the projects off which they make their money, but I would argue that their contributions have been too low to warrant calling them a symbiot. They cannot, in fact, raise themselves to that level, now that they're public. They're now committed legally to doing only those things that maximize the financial returns to their shareholders (just as Microsoft is ...).

I wouldn't know about Debian, but I use Slackware and have definitely had the frustration of doing things by hand, but at the same time I've enjoyed the challenge, so I can do without Red Hat.

And I didn't say that the middlemen like Red Hat were useless, though I for one certainly prefer to put in some extra effort to avoid them, I said they were unnecessary, which they are. Red Hat, because the product they distribute is available directly from its producers, is in the same position that BMG and the other record distributors are: they provide an unnecessary service and, as you pointed out, Red Hat has tried to convince the world that they are Linux, in order to keep people from noticing this in numbers large enough to hurt their cash flow.

By the way, what's "IIRC" mean?

-robin

Turning and turning in the widening gyre
The falcon cannot hear the falconer;
Things fall apart; the centre cannot hold;
Mere Anarchy is loosed upon the world.


[ Parent ]
Very unlikely (3.33 / 3) (#15)
by delmoi on Wed Jul 11, 2001 at 09:00:30 PM EST

If Red Hat pissed of, say Icaza, then he could release the next version of GNOME under the GPL for everyone except Red Hat, who could be forbidden from using it.

I'm not sure you can use the GPL that way. In any case, if he were to do that, GNOME would no longer be 'open source'

What would actualy happen is people would just stop buying their stuff...
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
GPL (3.00 / 1) (#23)
by vectro on Thu Jul 12, 2001 at 01:15:42 AM EST

Icaza could in fact license it under the GPL to anyone but Red Hat. But it is a provision of the GPL that you can distribute it to anyone you choose, so anyone he distributed it to could then redistribute it to Red Hat.

I suppose you could write a new license that was like the GPL, but forbade distributing to Red Hat. Such a license would not be free software, and you couldn't use the GPL as a base - the license is copyrighted by the FSF, and modifications are now allowed.

Of course, the point is moot since, as has been pointed out previously, there are so many contributors to the GNOME project that relicensing would be impractical.

“The problem with that definition is just that it's bullshit.” -- localroger
[ Parent ]
Response (4.66 / 6) (#17)
by sigwinch on Wed Jul 11, 2001 at 09:13:39 PM EST

Despite the fact that they're dealing with GPL/BSD/etc licensed code, they're still dependant on the goodwill of the developers.
True, but BFD. It isn't as if all the free software developers are suddenly gonna disappear or something.
It's true that code, once released under a license, cannot be revoked, but there's nothing in the license that commits the developer to releasing the next version (or patches) under the same license.
That's true in theory, but in practice nearly every project includes contributions from random third parties that are under the original license. To relicense, they core developers would have to get licenses/assignments for every patch that was ever applied to the code. On a large project that is simply impossible.
If Red Hat pissed of, say Icaza, then he could release the next version of GNOME under the GPL for everyone except Red Hat, who could be forbidden from using it.
Even if he could do that, it would instantly decimate de Icaza's market share. Most people will not go to the trouble of finding, downloading, and installing New Gnome. They'll either use the Old Gnome that ships with Red Hat, or they'll use KDE. It's a matter of laziness.

Having your software ship with the Red Hat standard distribution is considered a major accomplishment for a project. Red Hat would have to fuck up big-time to turn it into a liability that projects would avoid. IMHO, something sufficiently heinous to drive away projects would drive the customers away first, so it doesn't really matter.

Perhaps parasite is strong for Red Hat, who have contributed some to the projects off which they make their money, but I would argue that their contributions have been too low to warrant calling them a symbiot.
Some things Red Hat has done:
  1. gcc. They bought Cygnus and are now the gcc maintainers. (Cygnus, incidentally, was profitable.)
  2. Kernel enhancements. They funded much of the zero-copy networking patches that enable Linux to trounce Windows 2000 at static file serving.
  3. Integration. Red Hat makes hundreds of packages coexist peacefully on the same machine. More on this below.
  4. Variety. The basic distribution includes enough software packages to do just about anything you want. Windows still has a 5-10 year headstart on truly excellent packages, but free software is catching up.
  5. Support. If you get stuck with a silly little problem, they'll help you.
  6. Distribution channels. Thanks to Red Hat, the average person in a small city in the U.S. can easily buy CDs to play with Linux. More on this below.
More on systems integration: you may not mind spending a few hours fighting to install a package, but it's a *killer* for the average person. A test: for every package included with RH Linux (including Powertools CD), obtain the nearest equivalent package for Windows, and install them one-at-a-time onto a clean Windows system. This will probably render the Windows system nonfunctional, and no one will consider this remarkable at all. Windows machines are expected to crap out when you install 1000 packages. RH Linux will probably be fully functional for equivalent packages, and this fact will be considered completely unremarkable.

More on distribution channels: somebody already mentioned this, but you can go to Wal-Mart and buy Red Hat Linux (or at least you could; Wal-Mart changes stock all the time). In my mind, this is a watershed event for free software -- it's no longer something that geeks have to put together from the hidden recesses of the Internet. Anybody with enough money to play with commercial software can simply buy a CD from the local store.

I wouldn't know about Debian, but I use Slackware and have definitely had the frustration of doing things by hand, but at the same time I've enjoyed the challenge, so I can do without Red Hat.
This is like a mathematical savant swearing off calculators and calling TI and HP parasites. The average person considers it a privilege to pay $40 for an electronic calculator, and is glad to find that calculator for sale at the local store, and glad to have a number to call if the calculator goes all screwy.
Red Hat, because the product they distribute is available directly from its producers, is in the same position that BMG and the other record distributors are:
Unlike BMG, Red Hat's product is more than just entertainment, it is a tool. A complex, sophisticated, powerful tool that many people are building their livelihoods on. Sure, you can buy the pieces and put the tool together yourself, but that's not cost effective for most people. If you give the average businessman a choice between coal+ore+smelter+foundry and a premanufactured pair of pliers, he will glady lay down money for the pliers.
...they provide an unnecessary service...
I can replace or upgrade anything on my car. Not theoretically, either. In the past, I've replaced my transmission, I've helped rebuild engines, and I've helped build race cars.

Does that make mechanics unnecessary? No. They're the most cost-effective solution for many people, who have neither the skill nor the capital to do the work themselves.

By the way, what's "IIRC" mean?
If I Remember Correctly.

(Apologies for any typos. I'm late for supper at home and have to run.)

--
I don't want the world, I just want your half.
[ Parent ]

big response there ;-) (none / 0) (#26)
by Arkady on Thu Jul 12, 2001 at 02:34:28 AM EST

I responded to dsome of the things you're saying in my response to another comment, so I won't type it in again.

I just want to point out that I'm not saying that a revolution against Red Hat's attempt to hijack control of Linux is underway (though from the way I worded that, I obviously think it'd be good ... ;-).

What I'm saying is that Red Hat is fragile, in that they both depend on others' production for their business and they perfom a service which is unnecessary. It may be desirable, and it may be useful, and they may contribute some resources back to those producers, but they are dependant on both the goodwill of those producers and on the generally low level of computer interest and ability in the population. A change in either could be disastrous for their ability to make a buck at their chosen business.

It may be unlikely, but it is possible; and it's got nothing at all to do with "Open Source". It has to do with the particular style of business they run. Consider that Nike, for example, is also dependant (on the goodwill of the governments of the countries in which they run their sweatshops) in the same way that Red Hat is. It's not uncommon in the modern business world.

-robin

Turning and turning in the widening gyre
The falcon cannot hear the falconer;
Things fall apart; the centre cannot hold;
Mere Anarchy is loosed upon the world.


[ Parent ]
dependency (none / 0) (#37)
by kubalaa on Thu Jul 12, 2001 at 08:49:23 PM EST

You made me realize something; a business is inherently dependent, if not on its supply chain then obviously on its customers. Business are all about creating mutually-beneficial dependencies.

[ Parent ]
you forgot to add (5.00 / 1) (#33)
by mami on Thu Jul 12, 2001 at 12:27:17 PM EST

Some things Red Hat has done:

1.gcc. They bought Cygnus and are now the gcc maintainers. (Cygnus, incidentally, was profitable.)
2.Kernel enhancements. They funded much of the zero-copy networking patches that enable Linux to trounce Windows 2000 at static file serving.
3.Integration. Red Hat makes hundreds of packages coexist peacefully on the same machine. More on this below.
4.Variety. The basic distribution includes enough software packages to do just about anything you want.
Windows still has a 5-10 year headstart on truly excellent packages, but free software is catching up.
5.Support. If you get stuck with a silly little problem, they'll help you.
6.Distribution channels. Thanks to Red Hat, the average person in a small city in the U.S. can easily buy CDs to play with Linux. More on this below.

You have forgotten to add:

7. E-commerce : They bought two companies, MiniVend and Tallyman, which became Akopia, hired their developers and build their own RedHat Interchange e-commerce platform on the basis of that code. Perfect example that RedHat is doing a lot for the open source code programmers AND for end users. They might have a long way to go to set up services for all those non unix people, who discover less expensive and more flexible ways to build ecommerce sites, but hopefully RH won't behave like the primadonna "we'll change the world" snobs and CAN handle to technically service and support any uncle Joe, who is willing to pay for it. I don't see anything wrong with that. Contrary, I wait for that to happen and have no problem whatsoever to buy my boxed RedHat Servers and eventually RH services, if I would need them.

8. Databases: People have criticized RedHat on /. for being parasitic, putting out their *own* "Red Hat Database" based on PostgreSQL. So, why is that bad ? What are you all coding for if not for the code being used by people who can't code ?

rant on:
If a company tries to make the application packages easier to use, better integrated in their Linux Distro and offers technical support, backup, ugrade and security services, you start whining. Why ? What's so terribly wrong with servicing people, who are NOT hackers ? SuSE has tried to do this, putting a lot of work in documentation of their distro. It seemed to work all the way for mere mortal Germans. Can you tell me why a similar approach by RedHat, even much more loyal to the purist GPL coders than SuSE and definitely more American in its "flair", is all wrong ?

Just come down to earth and get real. If rusty et al. has to beg for $ 5.00 a month to the hacking community for getting enough cash in to run this site independently, I would be more than happy to pay some $30.00 a month or so for having my whole system serviced, upgraded and secured by RH professionals.

Just get real about what kind of service should cost how much. Reputation is built on code, service and reliability, not talk.

Having watched the general attitude on /. against RedHat makes me just wonder. To me it seems the biggest threat to put open/free source code an mere mortal's servers is the mental attitude of the hacking community itself. Very strange indeed.
/rant off

[ Parent ]

Re: "Open Source" not the real issue (4.50 / 4) (#18)
by bigbird on Wed Jul 11, 2001 at 09:19:36 PM EST

If Red Hat pissed of, say Icaza, then he could release the next version of GNOME under the GPL for everyone except Red Hat, who could be forbidden from using it. That one version has been released under one license doesn't commit the developer to continue that way.
I belive this issue has come up before either here or on Slashdot. He could only change the license if he gained the consent of every person who previously contributed code to the project while it was licensed under the GPL. And then, because many free software developers write free software on principle, any non-free version would inspire a fork from the last free version, and the non-free version would likely die quickly. It would immediately be placed into non-free in Debian. I cannot think of a single license in use which excludes a specific company. And then, of course, it is easy to comply with the letter of an exclusive license - Red Hat could engineer a reverse takeover by a company which is not excluded from the license, set up a shell company to produce the distributions, or use any number of workarounds.

Red Hat, because the product they distribute is available directly from its producers, is in the same position that BMG and the other record distributors are: they provide an unnecessary service
But it is necessary. If I need a new pickup truck, I will not buy a turbo-diesel engine from Cummins, some big tires from Michelin, a windshield from PPG, paint from BASF, ABS brakes from Bosch, headlamps from Hella, seats from Recaro, carpet from DuPont, some sheet metal, and mount it all on a custom-designed frame. I get better value buying a Dodge 3/4 ton extracab, and avoid having to devote my life to assembling my truck.

Same for my computer - every now and then I like to use it, not just tinker with it. Your argument works for computer enthusiasts, as they will obtain the software from the producer. It will not work for the other 98% of the population.

IIRC = If I recall correctly

bigbird

For I am not ashamed of the gospel of Christ: for it is the power of God unto salvation to every one that believeth; to the Jew first, and also to the Greek. Rom 1:16
[ Parent ]

Re: "Open Source" not the real issue (4.00 / 1) (#25)
by Arkady on Thu Jul 12, 2001 at 02:25:40 AM EST

"He could only change the license if he gained the consent of every person who previously contributed code to the project while it was licensed under the GPL."

Good point, and it definitely demonstrates why GNOME (as such a massive project) probably wasn't the best example since it would require the agreement on many people (quite possibly too many) to have the effect. The point, though, is that it's definitely possible. A smaller project, or even one as large as GNOME in response to a particularly egregious action by Red Hat, could definitely force themselves out of Red Hat's distro. And it doesn't do any good to form weird corporate structures if the prohibition is along the lines of "may not be used or distributed to any entity associated in any way with Red Hat Inc.". ;-)

Unlike Debian and Slackware, Red Hat is a publicly traded corporation and its operators are legally obligated to put the stockholder's investment return as their highest priority. This makes Red Hat much more likely to get into a situation which would piss off the Open Source developers on which they depend.

"Same for my computer - every now and then I like to use it, not just tinker with it. Your argument works for computer enthusiasts, as they will obtain the software from the producer. It will not work for the other 98% of the population."

I never said that they weren't _useful_, I just said they weren't necessary. Your example is similar to the other commenter's statement about a mechanic: yes, many of us can replace the transmissions in our own vehicles, but it can be more cost-effective (or prefereable for other reasons) to have a mechanic do it. That doesn't make mechanics _necessary_ (though with the vast majority of drivers having no clue, they might still be), though they are still useful. The same applies to most computer consultants and all consolidators (like Red Hat). They are useful, yes, and desirable to many folks, but they certainly aren't necessary.

Both of these seem to be fine lines, and easy to miss or disregard, but they're both important distinctions. Though it may be unlikely that GNOME (or KDE, or Apache, or any other major project) would get pissed off enough at Red Hat to forbid their use, it's possible and it would be devastating for their business. Similarly devastating would be a general rise in interest and ability with computers, which would reduce the potential market for Red Hat's particular distro (considering the prices they charge) since Red Hat, unlike Microsoft or Apple, has no real control over the actual contents of their system.

They're in a fragile position and, though i may seem unlikely that these particular vulnerabilities would get hit, it's certainly possible.

-robin

Turning and turning in the widening gyre
The falcon cannot hear the falconer;
Things fall apart; the centre cannot hold;
Mere Anarchy is loosed upon the world.


[ Parent ]
Oddly, no (none / 0) (#35)
by bigbird on Thu Jul 12, 2001 at 01:01:35 PM EST

A large part of what you have said is based on your definition of necessity. As the only real neccessities are things like food, air, water, shelter, and so on, that point merits no further discussion. "They are useful, yes, and desirable to many folks" - well thats about all any corporation can hope for now, isn't it? As well, you underestimate the creativity of MBAs and lawyers - I doubt a workable and effective "anyone but Red Hat" license could be worded.
Red Hat is a publicly traded corporation and its operators are legally obligated to put the stockholder's investment return as their highest priority. This makes Red Hat much more likely to get into a situation which would piss off the Open Source developers on which they depend.
Anything could happen. Hasn't yet. Not only that, but given that Red Hat recognises their dependance upon external developers, their obligation to provide shareholder returns requires them to maintain a good relationship with the developer community. Messing up community relations would be like killing the goose who lays golden eggs. If Red Hat was managed in a manner similar to my workplace, they would have to close their doors within the year.

Red Hat could still disappear, even without screwing up. They could fail to grow fast enough, they could be overtaken by Suse or Mandrake (but not Caldera, thank goodness). They could also surprise everyone. For the foreseeable future, someone will have a role like Red Hat's, writing tailored docs, and completing systems integration, to ensure that the different packages play nicely with each other, as pointed out by sigwinch. Boxed sets will not disappear until people give up dead tree manuals, which won't be anytime soon. And, in spite of the work of the FSF, software corporations face little risk of disappearing anytime soon.

bigbird

For I am not ashamed of the gospel of Christ: for it is the power of God unto salvation to every one that believeth; to the Jew first, and also to the Greek. Rom 1:16
[ Parent ]

Very true (mami in ranting mode) (none / 0) (#36)
by mami on Thu Jul 12, 2001 at 03:55:02 PM EST

Redhat adds considerable value, even if their product is not unique. People will often buy a boxed set just for the manuals,and, outside of /. and k5, the rest of the world has no interest in ./configure, make, make install.

Yeah, but the best of all is that even if you had no interest in ./configure today, you might have that interest tomorrow and then "mami dummy" can learn it. So, just get it, convenient services plus having the option anytime to dig intro learning from the sources, is the deal.

That's why we want documentation, not "intuitive" GUIs (not that we wouldn't take them - but that's not the real important thing) and that's why we want convenient services, because we want the freedom to stay "mami dummy" or become " mami smarty". And for the real female comprehending impaired geek, listen carefully, the choice to make is ours and NOT yours. Go preach to yourself to when to RTFM and not to your potential customers.

The value they add is convenience, and support services (because corporations are not about to depend upon support from a nebulous "community"). Most people are less than keen on having some 15 year old tell them to RTFM in a Usenet group.

Amen. Cluttered archives of mailing lists (which haven't been self-cleaned by the commentators, who just would love to eliminate superfluous questions, wrong answers and other flame-oriented trash talk they might have engaged in... hint, hint - I love self-editing empowerment of the forum and mailing list endusers) are OUT. Especially not, if the skript kiddie-man preaches religion, guns, sex, morals and politics between each and every bit of code they have not (re)-invented.
/rant off

Hmm, now I feel guilty of being impolite. See, it's not good for business to make your user-turned-customer feel so bad, that they start ranting and on top of that asking them to pay for it. Just thought it might be good to demonstrate this.

[ Parent ]

Huh? (3.75 / 4) (#14)
by delmoi on Wed Jul 11, 2001 at 08:57:12 PM EST

Redhat does produce a lot of their 'core products'. They hire many of the open-soucre hackers, such as Alan Cox, they bought Cygwin, etc.

If redhat did go down, it would be a pretty big loss for Linux. I think.
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
This is why FreeBSD is the superior model. (3.60 / 5) (#20)
by eLuddite on Wed Jul 11, 2001 at 11:19:15 PM EST

The *organization* writes almost all the code they distribute in the core OS, userland and kernel. Even the GPL stuff like gcc is maintained as a branch rather than simply imported. Almost everything in Linux, on the other hand, is individually pursued by selfish egoists. The author of ls, for example, has no constraints and is free to rewrite ls according to the demands of the Invisible (but vocally vociferous) Linux Hand. Well, good for him or her. Why should this mean that he or she must also, as a necessary consequence, be allowed to *hijack* attempts at building a coherent, unified, and socially equitable OS? Spontaneous order out of chaos is a ridiculous ESR myth.

"Thanks for ls, we'll be maintaining it now; although your future input is welcome, your actual commit privileges are confined to your personal machine alone, just as if it were Linux."

Makes sense to me. The author of ls can subsequently find Microsoft religion for all the effect such an obvious eventuality will have on the FreeBSD code base. Of course, FreeBSD doesnt pretend to be a capital 'C' corporation, either, so it spends all of its organizational effort at engineering -- forcefully assigning tasks and duties -- instead of marketing to rational agents of capitalism.

FreeBSD, the statist OS.

---
God hates human rights.
[ Parent ]

I don't get it (4.00 / 1) (#27)
by thejeff on Thu Jul 12, 2001 at 07:34:46 AM EST

The *organization* writes almost all the code they distribute in the core OS, userland and kernel. Even the GPL stuff like gcc is maintained as a branch rather than simply imported. Almost everything in Linux, on the other hand, is individually pursued by selfish egoists. The author of ls, for example has no constraints and is free to rewrite ls according to the demands of the Invisible (but vocally vociferous) Linux Hand. Well, good for him or her. Why should this mean that he or she must also, as a necessary consequence, be allowed to *hijack* attempts at building a coherent, unified, and socially equitable OS?

If the author of ls decided to rewite it in some ridiculous direction, RedHat (and the other Linux distributions) would be free to decide not to include the latest version in their releases. If the change was extreme enough the project would fork, and someone else, possibly even paid by one of the distributions, would maintain it. How could he hijack the OS?

And how is this really different from BSD? Except that since the current maintainer is doing a good job, there is no need to duplicate effort by having someone else be the Redhat ls maintainer.

thejeff

[ Parent ]

fork, frko, fkro, fork (3.00 / 1) (#28)
by eLuddite on Thu Jul 12, 2001 at 08:12:13 AM EST

If the change was extreme enough the project would fork, and someone else,

There you go again, different hijacker, same as the old on. This must be what you all mean by the open source revolution -- source code going in circles.

Madness.

---
God hates human rights.
[ Parent ]

hijack? (4.00 / 1) (#30)
by thejeff on Thu Jul 12, 2001 at 09:26:48 AM EST

I'm not sure I understand what you mean by hijack. Could you give an example?

So each BSD maintains its own version of all the core utilities just on the off chance that the author could decide to go in a direection they didn't like, but that they could just ignore or take over when needed. Seems efficient to me.

There may be other, perfectly good reasons, to use a separate internal code base for core utilities, but I can't see 'hijacking' as a major one. thejeff

[ Parent ]

HIJACK! (2.00 / 2) (#31)
by eLuddite on Thu Jul 12, 2001 at 10:02:49 AM EST

I'm not sure I understand what you mean by hijack. Could you give an example?

color_ls

What's next, frilly fru-fru vi? Too late, vim. You must stop giving in to the demands of code hijackers and adopt a singular vision of quality before trinkets.

but that they could just ignore or take over when needed

You could do that if you are content to leave the motley crowd set the pace and flow of development; and you will be left with... Linux.

FreeBSD, the Sparta of operating systems. Because computing is at war with people who lead interesting lives.

---
God hates human rights.
[ Parent ]

Open source primadonnas and hypocrites (none / 0) (#21)
by mami on Wed Jul 11, 2001 at 11:57:33 PM EST

are the real problem and nothing much else. To call RedHat parasitic speaks for yourself. RedHat is not dependent on good will, they hire programmers and pay them. Isn't that nice for a change ? Do you have founded a pure open source software company, who is capable of doing the same in these days ?

[ Parent ]
Well, good for them. (4.00 / 3) (#13)
by regeya on Wed Jul 11, 2001 at 08:39:48 PM EST

So they're realists. Good for them.

One of the nice things about an executive summary like that is that it gives direction to a company--it doesn't just mean, "Oh look, this is bad, bad, bad, throw up your hands and give up, our competition will whip us no matter what."

*sigh* guess it's time for me to read the full document to see what conclusions their marketing, advertising, R&D and so on departments drew from this. Unfortunately, the author of the article didn't feel the need to link, instead reproducing only the parts that easily support the same "Closed Source Isn't Commercially Viable" mantra people have been chanting since the bottom dropped out of the market, affecting the tech sector (not just the part that's involved in Open Source projects.)

But eh, what do I know. I should just wise up and decide that RH and others will never make money, like all the smart people have done. (No, no sarcasm here!)

[ yokelpunk | kuro5hin diary ]

Not a strategy document (none / 0) (#22)
by vectro on Thu Jul 12, 2001 at 01:09:53 AM EST

Keep in mind this is not intended to document corporate strategy - the entire purpose of the form 10K is to document the risks involved in investing in RedHat. So no conclusions are drawn from what I posted.

Finally, I want to point out that I submitted this not because I thought open source can be unprofitable, or because I thought Red Hat's business models are flawed - On the contrary, I'm an investor in Red Hat.

I just thought that this part of the form raised some interesting points and wanted to see how people address them. Thus far, the response has generally been "Gee, it's a prospectus, and they don't mean it." Which is immensely unsatisfying, as well as a logical fallacy.

“The problem with that definition is just that it's bullshit.” -- localroger
[ Parent ]
If you had told people that (none / 0) (#38)
by Wah on Thu Jul 12, 2001 at 10:29:22 PM EST

it probably would have helped. IMHO, K5'ers aren't too good at making that extra step. Call it troll protection. If you want those types of opinions, you usually have to ask specifically for them in the article.

I work for a company that is trying to make money on Linux, mainly by selling administrative services and bandwidth. The distro's are largely irellevant, at least they will be once a standard OEM class killer comes along. If it does. I think Red Hat is on the right road and will find a way to make money eventually. M$'s FUD campaign shows how much they are hurting. Netscape showed you can't compete with a free product, did their prospectus have any golden edges with that theme in mind?
--
Some things, bandwidth can't buy. For everything else, there's Real Life | SSP
[ Parent ]

Distros are inherently nonprofitable (4.50 / 2) (#19)
by psicE on Wed Jul 11, 2001 at 11:13:17 PM EST

Red Hat will obviously die unless they make money somehow. Due to the nature of the market, especially the existence of Debian, Red Hat's only choice is to provide value-added features that people are willing to pay for. Some of these include paying $50 (approx, I don't know the current prices) for a box set with CDs, manual, Partition Magic and other proprietary software, and 6 mos. of support and premier updates, and paying some larger amount of money for Red Hat's "database solution" software.

There's two major problems with Red Hat's strategies, for the most part: #1, all of their value-added features, aside from non-GPL Cygwin products, are available for free or cheaper from other sources, and #2, it's hard being a for-profit when your main product is 90% designed by third parties. This isn't a problem for Debian, because they don't need to be profitable, and AFAIK they don't have much in the way of expenses (no advertising for example). I wouldn't be surprised if Red Hat, and almost all commercial distro makers, concede to Debian and refocus on commercial-level products where they can make, and charge money for, proprietary extensions to Linux that are useful for servers but not end-users (such as Red Hat's database).

RedHat makes money elsewhere (none / 0) (#34)
by snowlion on Thu Jul 12, 2001 at 12:41:27 PM EST

RedHat's primary money maker isn't shelve distro sales; I believe they are making most of their money by working with huge companies and setting up their networks and installs for them.

Yes, RedHat has positive cash flow now.

They are also doing well in the embedded realm; I believe they write custom kernel's for these people.

Forget the boxes in the stores; that's not the real deal.

As for their certification and training sales; I don't know how those are doing.

I believe they do consulting as well.

RedHat network automatically applies patches over the net. That has a monthly fee attached to it.

Don't make the mistake of believing that they're just selling boxes in stores.


--
Map Your Thoughts
[ Parent ]
Red Hat annual Report | 40 comments (34 topical, 6 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!