|
|
| Linux is Not "About to Fork" like Unix |
 |
|
Sunday, November 21 2004 @ 02:46 PM EST
|
You may have seen
Paul Krill's concern that the Linux kernel may be about to "fork" and go the way of Unix. He attended the SDForum conference recently, which covered the following topics, Open Source in the Enterprise: Practical Considerations, Vendor Perspectives, Success Stories, Open Source Legal Issues, and Investing in Open Source. One of the keynote addresses was by Andrew Morton, maintainer of the Linux 2.6 kernel, and something he said worried Krill. Never fear, Mr. Krill. The kernel is safe, as you will see. Here's what Krill wrote:
"Linux could be about to fork. In a worrying parallel to the issue that stopped Unix becoming a mass-market product in the 1980s - leaving the field clear for Microsoft - a recent open source conference saw a leading Linux kernel developer predict that there could soon be two versions of the Linux kernel.
"Each version of the kernel requires applications to be compiled specifically for it. Today, only one Linux kernel is current but Andrew Morton, lead maintainer of the Linux kernel for Open Source Development Labs (OSDL), said that he expects the 2.7 version of the platform to fork to accommodate large patch sets.
"Commenting on the planned 2.7 release of the Linux kernel, Morton said OSDL expects a few individuals with big patch sets will want to include them in the kernel. But there will be no place to put them - presumably because functionality for major kernel changes won't be applicable to all or even most users. One example might be between desktops and servers. So at some point, Linux founder Linus Torvalds will fork off version 2.7 to accommodate the changes, Morton said at the SDForum open source conference." Not to worry. I contacted Mr. Morton and asked for clarification, and you'll find his response reassuring.
Here's what he wrote: "Paul has misinterpreted the word 'fork'. I was referring to the
software
engineering process of branching off a stable release of your product
so
that development can continue against the tip-of-tree codebase.
"We did this for the 2.0 kernel series, the 2.2 series and the 2.4
series.
One day we'll do it for the 2.6 series." Linux is developed like this, so rather than being worrying, it
is good news of pending Linux progress. That's OK. We can't all be techies, but everyone can go back to a relaxing Sunday, safe in the assurance that there is no doom on the horizon for Linux. Of course, a little research on Mr. Krill's earlier articles makes for a fun Sunday in itself, I must say. Here's a catchy title:
"Maybe SCO has a point". In that article, written back in January when SCO sent its letter to Congress about free and open source software allegedly endangering the economy, Krill articulated his concern about software developers: "However, if the trend of giving away software continues to gather momentum, how do developers and software companies put bread on the table? Work a second job?
"This question is something I've pondered before, and now SCO seems to be backing me up.
"The capitalist economy is based on selling products and services for the top dollar that the market demands. If the user community begins to expect its software free of charge, what happens to the innovation and incentive to improve software, or to even build it at all?" Of course, that was back in January. I'm not so sure anyone would want to associate themselves with SCO today. But in the earlier article he refers to, titled, "How can open source fly?" he expressed this worry: "Pardon me, but am I missing something here? Do people make money in a capitalist system by giving things away?" And in a November 17, 2004 blog entry, "Can't say I didn't tell you so," he writes: "Previously, I have asked what happens to the software industry if enterprises expect to acquire their software for free via open source.
"Well, it looks like this expectation is coming to fruition. A Wells Fargo executive at the SDForum's 'Open Source Entering the Mainstream' conference this week said the tide has switched from companies being suspect of open source to now openly seeking it out as an alternative to commercial products.
"Which leads us back to the question of what happens to innovation if at some point there is no money being made on the actual selling of software. . . . How long can open source be sustained if developers have to work a separate job to pay the bills and deal with open source as a hobby? I guess that depends on the devotion of the developers, whose dedication to their craft is certainly admirable and even enviable. . . . "I have to wonder, though, whether the days are numbered for software companies to make billions of dollars a year by selling software." As you can see, he gets the big picture. Microsoft's artificially-high-priced software business model is doomed. But is that bad? Won't we all, businesses included, then have money to spend and invest in other ways? Is that not good for the economy? And what about the question about programmers paying their mortgages? More and more, FOSS programmers work for vendors like Red Hat and Novell. Andrew Morton gets paid. He is sponsored by OSDL and works full-time on Linux kernel development. So does Linus. And so forth. Many companies, not just vendors, see it as in their corporate interests to assign some of the coders to working on the kernel, so they can get incorporated the things they need for their business purposes. I believe IBM is oriented toward making money in a capitalist structure. Morton writes to corporations on how to get what they want incorporated into the kernel: "I'd like to encourage corporate developers to become more involved and to contribute more to Linux. There are some cultural and even legal problems, as well as some common pitfalls for corporations working with the kernel and the open source development process.
"Developers should always work with their management and in-house counsel to be clear about IP issues and internal processes. Each company needs to individually address these issues.
"Companies should try to avoid what I call the 'SourceForge Syndrome.' A company will set up a big project and then beaver away at it independently for months. Suddenly a 50,000-line patch appears on a mailing list and no one understands what it does or why it does it. Such large patches usually have significant architectural problems and even duplication of other efforts. By this stage the originating development team is deeply invested in their current code and may even have run out of budget for rework, but the code may be unacceptable. It's generally a disaster.
"Companies must understand that there are significant advantages for them to merge their code into the mainstream kernel. This reduces their maintenance effort and costs, increases their tester and reviewer base, and encourages other developers to contribute additional feature work. Other people will magically fix your bugs for you. . . .
"Rather than setting up an external SourceForge project, you should aim to get a small core of your feature into the base kernel and develop against that, introducing new features on a frequent basis. . . ..
Companies may want to nominate a lead individual as their contact point with the rest of the kernel development team." Many do write it just for fun, though. It may seem hard to accept that folks would write software without charge, but by now, after more than a decade, is it not an established fact that they will and they do? It's comparable to novelists, in a way. You can't stop them from writing, even if there may not be a publisher, because it's a creative outlet and they enjoy doing it. Working on the Linux kernel is satisfying because there is a use for your work, probably some functionality you have coded that you personally want, on top of the creative fun. And you know the entire world will benefit from your labor. So, let's just posit that folks will do it, based on the scientific basis that we see that they have and they do. If you are curious about how it all works, I suggest you might enjoy reading Keeping the Kernel", by Andrew Morton. He explains how the process works, and it's interesting to see an example of how the openness of the work benefits the kernel, in a box, "Building a New Scheduler":
"Inspired by a paper on an anticipatory scheduler by Peter Druschel and Sitaram Iyer (http://www.cs.rice.edu/~ssiyer/r/antsched), a young Australian developer named Nick Piggin came out of the woodwork to ask Jens Axboe [another seasoned kernel developer] and Andrew Morton questions about the existing I/O scheduler. Per Morton's suggestion and with Morton's encouragement, Piggin implemented the scheduler on Linux. Six months and 140 patches later (90 percent of the contributions from Piggin), the new scheduler was stabilized and merged into the mainline kernel."
It's not a free-for-all, though, where anybody can just do whatever they want with the kernel. Morton provides clear instructions on how to become a contributor and how long it takes to get accepted into the process usually: "Approximately 1,000 individuals contributed to the 2.5 and 2.6 kernels. For the Linux kernel the '20/80' rule very much applies. About 20 people did 80 percent of the work. Or maybe it was '10/90.'
"The kernel has become a lot more complex than it was in the 2.0 days, and the learning curve is longer and steeper. A committed developer who's competent in C can begin to contribute usefully after as little as a few weeks of study. (For an example, see the sidebar 'Building a New Scheduler.') In general, it probably takes six to twelve months of full-time work to reach the level of a mainstream kernel contributor." I think we can see from this that while a few work at this full-time, many do not, yet they are still able to contribute meaningfully. That's how it works. The best way to get started, Morton writes, is the way he got started, fixing bugs: "Of course, new developers are encouraged and are always welcome. We always need help with drivers, bug fixing, tuning, and testing. A new developer may choose to go through the kernel Bugzilla database and identify any problems that appear to be unresolved. See if you can reproduce the problem, work it with the originator, and then develop a fix. Similarly, the various development mailing lists are a good source of current bug reports." Here's how Morton says he got familiar with the code, "I spent two years working partly on network drivers, but mainly on bug fixing. It didn't matter where in the kernel the bug was, I'd try to fix it. That was a great way to learn how the kernel worked." So, while there are adjustments in the software industry to be made, you can see from the way it is already playing out that innovation and advances are coming from a merging of two groups: programmers who love to program and corporations who need things incorporated into the kernel for their business purposes. And what's wrong with that?
|
|
|
|
| Authored by: Anonymous on Sunday, November 21 2004 @ 02:56 PM EST |
As ever, grateful for you holding the torch here.
These people should realsie that they serve their own
misguided aims ill by speaking like this. They just
force(?) OSS in to an ever more documented safe place. Do
they not understand thy're not creating FUD?
It's true irony, really. [ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 03:14 PM EST |
1 - If Linux actually did fork, it wouldn't be the end of the world. Part of
the strength of open source is that it can fork.
2 - Anyone who insists that software is written mostly for immediate economic
gain is seriously out of touch with important aspects of modern economic
theory.
One book is must reading for anyone who wants to understand what's going on:
"The Success of Open Source" by Steven Weber. He takes a scholarly
look at what has happened so far and presents that information in a very
understandable manner. He makes it clear that, at this point in history, open
source is going to be much better for the economy generally as well as for most
of the 'computer' industry. The alternative is to mess with the law and kill
open source by regulation. That would of course create grave harm to the
economy.[ Reply to This | # ]
|
| |
| Authored by: archonix on Sunday, November 21 2004 @ 03:18 PM EST |
"Pardon me, but am I missing something here? Do people make money
in a capitalist system by giving things away?"
It's a
funny thing, but... yes. People do make money by giving things away, or selling
as such a low price as makes no difference anyway. It's called loss-leading.
Supermarkets do it all the time.
Then there's marketing offers such as "free
drink with every Hindz Beenz tin purchased!" or some such rubbish.
ANd then
there's the concept of "value add," which is the alternative to trying to sell
more by lowering prices. Say you have a restaurant and you aren't making as much
money as you expect. The conventional wisdom is to drop prices in order to
attract more custom, but that means you hve to sell a *lot* more for less gian
than you might anticipate. The alternative? Spruce the place up a bit, add some
live music and make people want to pay your prices beause of the
atmosphere. Give away free drinks, free bread and free "stuff".
Inf act, I'd go
so far as to say that giving things away is one of the cornerstones of a
thriving market economy.
But I'm just a some-time graphic artist and computer
service bloke. What do I know?---
disclaimer: I'm human. I make mistakes
too, but I try to fix them if you ask nice.
Graham Dawson - Gentoo Masochist and another Scared Cat Owner [ Reply to This | # ]
|
| |
| Authored by: earthforce_1 on Sunday, November 21 2004 @ 03:22 PM EST |
I think he shares this unfortunate revelation with Pamela. (And Linus with his
patent comments)
Anything you say or do can and will be taken out of context or twisted and used
against you. Kind of like the negative ads in the last US election. The higher
your public profile, the greater the chance of this happening.
[ Reply to This | # ]
|
| |
| Authored by: paul_cooke on Sunday, November 21 2004 @ 03:32 PM EST |
unbelievably clueless... or else deliberate... Of course
Linux is about to fork... it's perfectly natural... 2.7
will be branched off from 2.6 and we all know that 2.7 is
what is known as the development kernel... where the
exciting new things get thrashed out until they have no
major showstoppers... then when they're nice and steady
enough for enterprise /normal use, the freeze goes in and
everything gets nailed down solid enough to be called the
new stable branch. We have a few release candidates where
those who're brave get involved in wider bug fixing, and
then it gets declared stable enough for normal use and it
will be the 2.8 final whenever Linus is satisfied with it.
We're in the envious position of having several different
stable kernels, 2.2 for those who absolutely have to have
stability... 2.4 for those who can't run stuff yet on 2.6,
2.6 for most of us now where we're getting down and dirty
with mass testing it for stability, and soon, there will
be the bleeding edge for those who really really enjoy
living dangerously. Which will be me using a spare box for
"hacking" around on.
---
Use Linux - Computer power for the people: Down with cybercrud...[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 03:32 PM EST |
I disagree - i've been saying for a while now that I feel
Linux is starting to fork. The 2.6 kernels are still
unstable, and i'm not the only one who thinks that.
Nvidia drivers have been broken by the kernel developers
now 3 or 4 times since the release of 2.6. This doesn't
inspire confidence in hardware manufacturers. Add to that
the attitude of the kernel developers of "well, Nvidia are
contributing binary only drivers, we can't see what
they're doing, so they can just fix it [the problem] can't
they" Sounds like the kernel developers didn't really
appear to talk to Nvidia about it before making the
changes to the kernel.
CD burning anyone? i've just went back from 2.6.9 to
2.6.7 and will not be using anything later because CD
burning is broken unless you are root. The excuse?
Someone might crack my system and wipe the firmware on my
cd burner. Puhlease! A poor enough excuse if ever I
heard one.
Add to that all the other minor instability issues that
i've seen/heard/read about, add to that the fact that it
doesn't really offer any improved performance whatsoever
(system - Athlon 1ghz, 768mb pc 100 sdram - period 4
months). I noticed no noticeable difference in
performance.
Add Andrew Mortons comments that the public kernel
repository at www.kernel.org shouldn't be used by users,
but they should expect to get their kernel sources fixed
by their vendors! The kernel src at www.kernel.org may
not be stable! Gee...do that and you'll get various
kernels for various distributions and ergo forking. And
it'll kill most of the smaller distros who can't afford to
be writing kernel patches or testing them.
Dave [ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 04:22 PM EST |
Ever since 2.6.8(.1), I've been having horrible processor stalls. Didn't the
developers decide to not necessarily make 2.6 a stable branch?[ Reply to This | # ]
|
| |
| Authored by: hauva on Sunday, November 21 2004 @ 04:22 PM EST |
Just look how different the kernel of Red Hat Enterprise Linux 3.0 is from the
vanilla 2.4 kernel. Even the scheduler is a different one. It's a fork. It seems
that the forking of the code is not a problem but please call a fork a fork.
---
Ari Makela, Helsinki, Finland
My name is Ari and I am a grokholic.[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 05:00 PM EST |
If all software is free, how will software companies make money?
Over the years I can recall Micro$oft giving away many free products. There
were disk compression programs and tcp/ip stacks and browsers and media
players... lots of stuff. I doubt that the M$ developers are wondering where
their next meal is coming from. Of course M$ is teaching the world a valuable
lesson in monopoly economics and political power (read thuggery). I certainly
can't see where free software hurt good old M$.
</sarcasm>
-- Alma[ Reply to This | # ]
|
| |
| Authored by: John Hasler on Sunday, November 21 2004 @ 05:20 PM EST |
> "Each version of the kernel requires applications to be
> compiled specifically for it."
This is false.[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 05:56 PM EST |
Mr. Krill clearly does not understand the difference between capitalism and
commercialism.
Regarding Linux and open source destroying the programming community, Mr Krill
might try the following experiment - place a stack of Linux distribution CD's
(any new release will suffice) on top of his computer and then watch to see if
it loads itself, and starts to automatically program and run all of his
applications.
No doubt, he will soon find it necessary to call in a couple of programmers to
get the job done and maintain the system - and by the way , since it is open
source, the programmers might even optimize the code to his specific needs to
improve the efficiency of his operation and hence his all-important bottom line.
Who knows, maybe Mr. Krill will be so tickled, that he will release the
improvements under a GPL license so others can contribute improvements to his
bottom line also.
[ Reply to This | # ]
|
| |
| Authored by: Tony on Sunday, November 21 2004 @ 05:56 PM EST |
I think our friend here has subscribed to one of the 'big myths' of the
software sales industry. The 'software alliances' would like us to believe that
programmers need consumers walking into stores (or their employers signing
expensive licenses) to earn their living. Just ain't so, folks.
The
percentage of programmers who earn their living writing software for 'boxed'
sale is incredibly low. It has been a while since I've seen estimates and I
tried to find some now without sucess but I do remember numbers below 1%
recently.
Most programmers are like me. I spent 15 years as a
professional programmer and the entire time I was working on custom applications
for particular businesses. The only difference the higher availability of open
source software made to me was to lower the cost to my customer and
increase my income through better productivity.
Tony
Williams
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 06:07 PM EST |
I have heard this MANY MANY MANY times before.
Hasn't happen yet...
This is getting old. Just shut down Microsoft now and staple Steve Ballmers
mouth shut![ Reply to This | # ]
|
| |
| Authored by: ssavitzky on Sunday, November 21 2004 @ 06:28 PM EST |
A couple of years ago, some people in the embedded systems industry took the
Linux kernel and ripped the memory management out of it so they could use it on
chips that didn't have an MMU, like the 68000-based processor that was in the
early Palm Pilots. The result was called uCLinux (for "microComputer
Linux").
As of 2.6, the uCLinux changes have been folded back into the mainline kernel.
So Linux has forked (in the Unix project sense), and unforked. Similar things
have happened to gcc. Happens all the time.
---
The SCO method: open mouth, insert foot, pull trigger.[ Reply to This | # ]
|
| |
| Authored by: Maciarc on Sunday, November 21 2004 @ 06:35 PM EST |
However, if the trend of giving away software continues to gather
momentum, how do developers and software companies put bread on the table? Work
a second job?
Where does Mr. Krill get the idea that these
developers and software companies are owed a living from us? It's real simple,
when the world chamges, change with it or become extinct. Just because they've
made a living selling a product, it doesn't mean that they desreve to continue
making that living selling the same product when the world has moved on. I don't
recall any abacus companies whining about calculators when they first came
out.---
IANAL and I don't play one on TV, I'm just an "anti-SCO
philippic." [ Reply to This | # ]
|
| |
| Authored by: arch_dude on Sunday, November 21 2004 @ 06:44 PM EST |
This is a major red herring, and is exactly analagous to the situation with the
RIAA and musicians. A propriatary software company pays approximately 15 percent
of its revenue to programmers. The rest of the revenue goes to overhead and
profit.
With FOSS, the major losers are the non-programmers at propriatery SW companies.
The programmers will in general go somewhere else and keep programming.
In the music industry, the RIAA claims that illegal copying takes money from
artists. This is true, but misleading. Since only a tiny percentage of revenue
for recordings goes to the artists, the major losers are the non-artists in the
industry. My response to the RIAA is to NOT purchase or listen to any new RIAA
music. I will listen to Indy music and old stuff I already have.
Just as most artists have very little sympathy for the parasitic non-artists in
the recording industry, most programmers have very little sympathy for the
parasitic non-programmers in the software industry.
I am a programming professional with more than 35 years of experience. I am
telling you what my collegues and I think. This is not a theoretical analysis.
There are jobs for every good programmer that wants one. The propriatary SW
companies agree: they are strenuously lobbying for more H1-B visas on the
grounds of a serious programmer shortage, and they are out-sourcing to foreign
countries for the same reason. I am happy to have collegues from India, either
here in the US or in India. I am not happy to pay a toll to non-productive
non-programming drones at Microsoft.[ Reply to This | # ]
|
| |
| Authored by: Mark Levitt on Sunday, November 21 2004 @ 06:53 PM EST |
The problem is that people don't get that forking, in itself, is not a bad
thing.
The problem for UNIX was that each company who had a license to UNIX could
create there own fork *and keep it to themselves*.
If each UNIX vendor had been able to use the best ideas from all the other
vendors, UNIX would still have forked, but it wouldn't have been such a big
problem.
Forks in FOSS are usually a good thing. Two people/groups disagree about the
direction of the project. The fork the project. In the short term, the team may
be split and less work gets done. But, in the long term, they each develop cool
new features and start to cross-pollinate.
Eventually, they two "forks" get merged together or the each take on
an idependant, but viable life of their own. I've seen this over and over again
with projects like Samba, KDA, the kernel, emacs, etc. Mandrake started out as a
"fork" of redhat. Many distros are forks of debian. Etc, etc.
Think of it as evolution in action.
[ Reply to This | # ]
|
| |
| Authored by: jkondis on Sunday, November 21 2004 @ 06:53 PM EST |
"Pardon me, but am I missing something here? Do people make money in
a capitalist system by giving things away?" - Krill
I'm sorry but
this kind of sentiment is bad for the economy and just plain wrong. People make
money by *creating value* for the economy. In the case of open source, a lot of
value is created either cheaply or freely. Operating systems, instead of being
an industry that provides entitlement to software programming graduates, becomes
an inexpensive resource for all to take advantage of. There - the value of the
OS has been created, now spend your programming skills customizing that resource
for people and you will make money. That same money has now provided for the OS
and its customization; that is a great stimulus to the economy, not a
drag.
Since the major software companies are tripping over each other to see
who can hire the most programmers in India and lay off the local labor force,
one might ask, what's the difference between cheap labor (India) and free labor
(Open Source)? Local (American) companies are laying the programmers off
anyway.
The difference is that the shareholders and executives of the large
corporations such as Microsoft *take* more of *your* money in the end for the
same or even less functionality. If *you* weren't paying that money, in the
form of OEM license fees, large enterprise server contracts, higher costs for
doing online transactions, etc., then you can do other things with that money,
like buy better hardware, a widescreen TV instead of a small one, a fluffier
couch, etc.
Oxygen in the atmosphere is quite free. It is one of the most
free-as-in-beer things we have, yet provides incredible value. I would argue
that it is a very good thing for the economy that is free, not the other way
around. This is because it is a resource that anyone can and does take
advantage of. We are not losing out on something economically because it isn't
owned by some corporation, to be dished out to folks for an annual
fee!
"The capitalist economy is based on selling products and
services for the top dollar that the market demands." - Krill
I
would *love* to see a reference for that assertion. *NEVER* has selling things
for "top dollar" been the basis for a capitalist economy. The assertion is
simply wrong. Otherwise, someone please explain to me why I can fly to Florida
from California for less than $100, or why I can download the next version
of IE for free, or why I can buy a walkman-style CD player with MP3 playback for
like $30 at the local department store.
Here is a better statement on
the basis of capitalism: You are free to choose what kind of good or service
you'd like to sell on the marketplace, and free to charge what you like. The
consumer is free to take or leave your offerings at the price you chose, or
negotiate a better price with you. And of course, you are free to not adapt to
the changing market conditions and be a financial failure because you failed to
discover or figure out what it was that people are willing to pay for.
Think
about that statement "willing to pay for". If people are "willing to pay for"
something, then it follows that something is providing value to people.
However, if people are "compelled to pay for" something, then it's not providing
real value for society.
Won't we all, businesses included, then
have money to spend and invest in other ways? Is that not good for the
economy?
PJ your sentiment is accurate. Real economists don't buy
this "everything has to be non-free and expensive to be good for a capitalist
society" bull crud.---
Don't steal. Microsoft hates competition. [ Reply to This | # ]
|
| |
| Authored by: PolR on Sunday, November 21 2004 @ 07:23 PM EST |
Guys, we should keep creatings the standard threads. We need a correction thread
and a OT thread fo every article. The task of creating those belong to the first
posters.
[ Reply to This | # ]
|
| |
| Authored by: PolR on Sunday, November 21 2004 @ 07:25 PM EST |
So PJ can easily find them.
This is a standard thread. See the OT thread for details.
[ Reply to This | # ]
|
| |
| Authored by: rweiler on Sunday, November 21 2004 @ 07:53 PM EST |
In the article, Mr. Krill refers to Kim Polese as 'he'. Ms. Polese is, of
course, a woman, and was probably the best known woman in the computer industry
6 years ago as Sun's marketing voice for Java and later as CEO of Marimba. If
Mr. Krill can't even get the gender of the speakers correct, the rest of his
story is highly suspect.
---
Sometimes the measured use of force is the only thing that keeps the world from
being ruled by force. -- G. W. Bush
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 07:58 PM EST |
A nice piece of journalism.
Let's see - if PJ can utilize her fame/clout to just pick up the phone (or send
an email) and correct the errors of other journalists on the Linux beat by going
directly to the source - wow!
Yet another service to Linux - in addition to all the legal stuff - brought to
you courtesy of PJ.
Nice work,
Epaminondas[ Reply to This | # ]
|
| |
| Authored by: unsubtle on Sunday, November 21 2004 @ 08:49 PM EST |
How long can open source be sustained if developers have to work
a separate job to pay the bills and deal with open source as a
hobby?
PJ pointed out that increasing numbers of FOSS
developers are being paid by vendors. but remember that this is an extra. GNU
and Linux started out with just volunteers, and they wrote an OS and a
lot of applications, having started from nothing. now there are still
volunteers, and indeed many more of them, since the success of FOSS has
attracted more people, and there are also paid developers.
if mr
krill had made this kind of argument 20 years ago, when the GNU project was
starting, he might have had a point. the project might have failed, and huge
credit is due to the early volunteers who made sure that it didn't. but now,
when we know (with hindsight) that a few volunteers succeeded, we're being told
that many volunteers and paid developers are likely to fail! this does
not compute.
i've noticed a few different people using this flawed argument
about sustainability. i don't know if they're copying one another, or if it's
organized FUD, or if it's just a mistake that people who don't know much about
FOSS tend to make. [ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 08:57 PM EST |
| "Many do write it just for fun, though. It may seem hard to accept that folks
would write software without charge, but by now, after more than a decade, is it
not an established fact that they will and they do? It's comparable to
novelists, in a way. You can't stop them from writing, even if there may not be
a publisher, because it's a creative outlet and they enjoy doing it.
"
The comparison between software writers and traditional writers such
as novelists is a good one in my opinion. Impossible to do the subject justice
in a note here but I'd like to raise a few points for contemplation with my
apologies in advance for the resulting oversimplification.
Software, like
the traditional written word, comes in many flavours. The bitter taste perhaps
is the necessary evil of device drivers, units of software such as network
drivers, disk drivers and so on. These form an essential underpinning of a
working computer system and have an affinity with various branches of
engineering. For those working on them, there is a certain satisfaction of being
an unsung hero with perhaps tens of millions of users relying on but unaware of
your work. I have been there myself. At the other end of the spectrum are the
programs whose features and content are very immediate to the user, the design
of interactive metaphor, ease of use, relevance to purpose, directly exposed to
criticism (and occasionally, delight!). I have also worked in this, the sweeter
side of software.
Invisible software has never been sold in its own right,
it has always been seen as free as in beer, as a part of a hardware or software
solution whose revenue has paid for the work. FOSS has had it easy here (Linux
kernel most obviously). However most consumer/education software that people
voluntarily spend their own money on falls at the other end, games being only
the most obvious example. Much of the high cost corporate/government style
projects fall somewhere in-between.
If Linux is to escape from the ghetto of
corporate/government it needs to reach out to the sweet side. Reading earlier
comments, I think there is far too much complacency about the current
effectiveness of contemporary FOSS structures. There are many changes and
challenges ahead IMO and repeating the usual stuff about paying the mortgage by
working for OSS in big multinationals and government to me seems
dangerous.
Back to the novel and the comment about writing for fun. Quite
true as far as it goes but ask yourself how many novels you have read recently
which failed to find a willing print publisher? Personally, I find writing user
and subject-oriented software extremely similar to writing a book. However,
currently, FOSS gives very few opportunities for individuals or small
independent groups to make a living in this very area that is crucial if Linux
is to seriously enter the Windows domain. I would, more controversially, go
further and say this is the key area Linux needs to lose its history of simply
catching up and look to taking the lead in new approaches to subject matter by
enabling the independents with good ideas so side-step the established
corporations.
What is needed is good new ideas of how we can do this and
remain solvent. Think of software writers as novelists, not engineers.
[ Reply to This | # ]
|
| |
| Authored by: inode_buddha on Sunday, November 21 2004 @ 09:08 PM EST |
| I had read Morton's comments sone weeks ago, and the odd thought struck
me: What we are doing at Groklaw, PubPat, etc. is "debugging" the legal
system. Hopefully. ---
"When we speak of free software, we are referring
to freedom, not price." -- Richard M. Stallman [ Reply to This | # ]
|
- An odd metaphor - Authored by: Anonymous on Sunday, November 21 2004 @ 09:21 PM EST
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 09:19 PM EST |
I dunno what kind of crack you guys are on, I'm using the latest kernels and
have no issues burning cds (or dvds for that matter) as any old user - with no
root privileges at all!
I dunno what yer issue is dude, maybe you have a lame config or something. I'd
bet dollars to donuts tho that it's been discussed ad nauseum on the lists/etc
for a long time now.
solution - config your system properly and maybe get a decent burner, install
properly and patch right. problem solved.[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, November 21 2004 @ 11:20 PM EST |
If one says they are forking a project, that really means forking. It isn't a
word which is normally used for branching because it means something so
different.
He really should have said branching, as that is a normal development process
for any non-trivial project.
Although to be honest, the whole 'for 2.7' bit totally gives it away as 'he
meant to say branching' anyway.
[ Reply to This | # ]
|
| |
| Authored by: cerebus on Sunday, November 21 2004 @ 11:30 PM EST |
There was some discussion previously on a 'MS-Linux'. Maybe as part of
extend-embrace-extinguish they will attempt a fork if they do move that way.
Image MS-Linux - New! Improved! Special features only available from MS!!
MS-Linux-Office can only be guaranteed to work with MS-Linux auto-update(R)
feature!
Maybe they'll attempt to move the code base on their fork faster than others can
assimilate the changes.[ Reply to This | # ]
|
| |
| Authored by: eamacnaghten on Sunday, November 21 2004 @ 11:49 PM EST |
| I must confess, when I first read Paul Krill's article it went in one eye and
out the other. I immediately dismissed it as a "wannabe open techie writer"
publishing an article about a subject he knew nothing about. Of cource
the kernel will "fork" to 2.7 at some stage. Many say it should have happened
earlier. I did not think it warrantied a comment. More the fool I, I did not
think of the FUD factor.
As for making money on free software - the fact is
thet the IP value of software is decreasing at an alarming rate. As little as
five years ago a Software House would make as much as 80% of it's revenue on
software license sales. Today you would be lucky to reach as high as 50%, the
rest being made up with services. In five years time you will probably find
sales from licenses to be in single figures (this is based on my personal
experience, I have no formal survey figures about this). That is for software
houses too! Most software technicians do not work for those, most work for "end
user" companies where the software has zero sales value. FLOSS simply takes
this concept to the ultimate conclusion. The revenue for a FLOSS company comes
entirely from the services you sell, and you cut down enormously on the
development investments by employing file Open Source model.
Web Sig: Eddy Currents.
[ Reply to This | # ]
|
| |
| Authored by: k12linux on Monday, November 22 2004 @ 12:42 AM EST |
And what about the question about programmers paying their
mortgages? More and more, FOSS programmers work for vendors like Red Hat and
Novell. Andrew Morton gets paid.
As others have said most
of the programers and innovators in software developement do NOT get their
paychecks by selling software. Most are "in house" working on programming which
will only be used within their own company/organization.
Imagine if a
company is able to save $250,000/year in licensing fees alone. That easily
can translate into both salaries for additional programming positions and more
profit for the company... or more money for R&D perhaps. What you won't see
is companies saying, "Hey, we can get this stuff for free, lets fire our
programmers!" The main job of the "in house" programmers isn't to develope
software. It is to customize software to the company's needs or to develope
applications specific to the company.
As far as Krill's concern about the
economy it seems like creating hundreds of thousands of new "in house"
programming jobs might be a boon for any nation's economy. Especially if there
is currently a lot of outsourcing to other nations going on by the proprietary
software companies doing business there. ---
- SCO is trying to save a
sinking ship by drilling holes in it. -- k12linux [ Reply to This | # ]
|
| |
| Authored by: Anonymous on Monday, November 22 2004 @ 03:09 AM EST |
Also saw that the website it was on was covered with M$ adverts, links to
"Get the Facts", and offering a "free"
"independent" study comparing Linux to Windows.
Mr. Krill is just another hired FUDster.[ Reply to This | # ]
|
| |
| Authored by: bliss on Monday, November 22 2004 @ 04:23 AM EST |
There has been a lot of noise about the coming 2.6/2.7
Linux kernel fork from the spin doctors. This is not at
all like the fragmentation the proprietary Unix operating
systems suffered in the eighties and nineties, it is a
normal part of the development process.
Linux kernels are numbered like so:
<major version>.<minor version>.<subversion>
The major version number changes when there are very
significant architectural changes - I believe 1.x -> 2.x
happened due to an a.out to ELF binary format change, but
this was a long, long time ago and I'm just a system
administrator, not a developer, so I've forgotten the
details.
The minor version number indicates how old the kernel is
*AND* if it is a production kernel or an R&D kernel.
Production kernels are 2.0, 2.2, 2.4, and 2.6. Development
kernels were 2.1, 2.3, currently 2.5, and soon 2.7 will
begin. The subversion number indicates more information
about the age, with x.x.1 being the first and numbers
after that rising sequentially.
Vendors will typically take a kernel, 2.6.4 in the case
of the SuSe 9.1 system I am using to write this, then
they'll append their own identifier to that number. The
full version reported by this system is 2.6.4-52-default,
with the 52 being some sort of internal reference for SuSe
support.
This sort of naming schema as a part of revision control
exists in other operating systems as well. The FreeBSD
project maintains two major kernel branches, currently
4.10 and 5.3, and within each branch there is a -STABLE
and -CURRENT, with -STABLE being trustworthy and -CURRENT
being likely to not compile on any given day.
The impending fork between 2.6 and 2.7 means what this
has always meant; 2.6 is becoming a mature body of code
and there are exciting new things coming that will get
hashed out in 2.7 prior to the release of the 2.8 kernel.
-bliss
---
Information becomes fragmented, knowledge does not. What causes fragmentation in
information is scholasticism - Ramitani[ Reply to This | # ]
|
| |
| Authored by: gbl on Monday, November 22 2004 @ 07:32 AM EST |
Code forks are only a problem when the code is closed or mostly closed. In the
Linux GPL world, even when a code base forks, neither fork is denied the new
code developed for the other.
The old Unix forks resulted in multiple super-sekret-source trees where similar
functionality was implemented in a way that was almost but not quite totally
different to all the others. This really annoyed the application coders who had
to deal with each Unix seperately which multiplied costs.
It is VERY important that Linux provide stable interfaces. Interfaces should
always be extended rather than changed. This allows long term stability and
reduces 3rd party development costs.
---
If you love some code, set it free.[ Reply to This | # ]
|
| |
| Authored by: phrostie on Monday, November 22 2004 @ 07:51 AM EST |
here we go again.
I've lost track of the number of people i've had to explain it to.
the difference between Linux and Unix is the GPL.
all forks find their way back home.
---
=====
phrostie
Oh I have slipped the surly bonds of DOS
and danced the skies on Linux silvered wings.
http://www.freelists.org/webpage/snafuu[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Monday, November 22 2004 @ 11:44 AM EST |
This is an old analogy, but most mathematics
development is done by passionate individuals
for little or no pay. Many others are paid to
apply the concepts to solve real-world problems.
The Balmers and Sontags of the world would have us
believe this won't work for software.
[ Reply to This | # ]
|
| |
| Authored by: Jaywalk on Monday, November 22 2004 @ 01:22 PM EST |
Do people make money in a capitalist system by giving things
away? Umm, yeah. All the time. "Loss leaders" are a popular
means in retail of getting folks in the store by giving something away (or
selling it cheaply). But the big winner in making money by giving something
away is Gillette. "Give away the razors and sell the blades" has become an
axiom for how to make money by giving things away. You buy a razor once, but
you'll continue to buy blades for the rest of your life.Compare this to IBM
which gives away a free operating system and sells consulting services. The
amount that IBM puts into Linux development pales in comparison to the amount of
money they make integrating and developing for Linux systems. Not to mention
hardware. Give away the OS -- which, incidently, IBM got for free -- and sell
the services. Sounds like a workable business plan to me. ---
=====
Murphy's Law is recursive. ===== [ Reply to This | # ]
|
| |
| Authored by: Chugiak on Monday, November 22 2004 @ 02:24 PM EST |
I believe you can still get hand-made shoes, but any made in the western world
will likely cost $500 or more. Yet you can go to the nearest retail outlet and
buy shoes for $30. Even before the onset of cheap overseas labor the cobbler
was a vanishing breed. Changes in manufacturing were making his product too
expensive. take a trip back in time if you wil: Imagine you're the head of a
household with ten children (read CTO) and are faced with either buying
hand-made shoes from the cobbler or off-the-shelf shoes at Woolworth's. New
hand-made shoes for the family will cost $100 and come in leather, leather, and
leather. Shoes from Woolworth's will cost $40 and come in leather, vinyl
(ewwww) and canvas. The canvas ones have rubber soles which provide good
traction for various athletic activities. By spending less on shoes you can
afford piano lessons for the eldest child. What will your choice be?
When making this decision, the farthest thing from your mind will be whether the
Footwear Industry Artists Association members are maintaining the high profits
they've enjoyed for years following the old business model.
Times change, and industries change to meet the new needs and survive in the new
environment. [ Reply to This | # ]
|
| |
| Authored by: sjf on Monday, November 22 2004 @ 02:57 PM EST |
Linux forks all the time. There are numerous forks for all sorts of reasons.
1)Supporting different hardware example:
-PPC and other non x86 hardware
(Yes I know in theory the mainline linux supports PPC, but having compiled a
kernel for my ibook. I know better than to go to kernel.org rather than
ppckernel.org.)
-untold numbers of embedded releases
2)Developer test trees:
-MM Morton
-AC Alan Cox's tree
3)Comerial requirements:
-Suse/Novell
-Red Hat
-Yellow Dog
Forking is good in linux. It allows people to do things that are right for
them, and not worry about how it might be bad for others. Due to the gpl
everyone can cherry pick the things that worked from everyone else's trees. This
survival of the fittest code benefits everyone and results in a better overall
kernel.[ Reply to This | # ]
|
| |
| Authored by: valdis on Monday, November 22 2004 @ 03:32 PM EST |
Microsoft warning about "forks" in the Linux kernel is quite amusing,
given that Microsoft has forked *their* code.
What did you *think* Win/NT, Win/2000, Win/XP, and Longhorn are? By their
definition, they're forks of Windows....[ Reply to This | # ]
|
|
|
|
|
