In a recent discussion about developers running out of money (arch,
perl) and what can we do for helping them I can up with an idea. I'd
like to share it with anyone in the list even though it'll probably be
disregarded or flamed. The Kernel may well be nicely funded, because of
companies supported. But that's not always the case, and the schema just
fails in a lot of key areas of OSS.
Why post it here then? Because for it to work it must be supported by
at least some of the grand developements of OSS.
Here's the actual idea, which actually is a slashdot repost
Re:Isn't dual-licensing with the GPL perfect for t (Score:2)
by fferreres on Friday July 26, @11:01AM (#3958396)
(User #525414 Info | http://www.arrancar.com/)
Yes, but that underfunds the projects. You can see this clearly when
Microsoft can sell lots of buggy software and of the best OSS developers
can't earn a decent salary.
I'd love to see a new license, that could be called the fGPL. That would
be the "Funded GPL". To be able to use fGPLd programs you'll HAVE to
contribute some small amount of money to the fGPL foundation. You'll not
be required to pay for any individual fGPL software, just a plain simple
yearly $10 or $20 charge. And you will be able to distribute exactly
where that money goes, among all the different projects. If you can't
pay $20 a year it will be no problem, just a bit penalty: all fGPL
software would be free as in beer once the year passes (old releases).
The money paid to the developers would only cover salaries and some
expenses that are needing to continue developement. So if any proyect
gets over-funded, you'll be noticed that you must reasign some of your
credits.
It'd always be free as in freedom. We only need to bring some beer for
that to happen. It'll also kill the anti OSS argument that the system is
for comunists or anti-american. I know that is FUD, but do your
representatives know that? It will also kill most of the other FUD
targeted at OSS and will also bust developement to unknown levels.
What do we need for this to happen?
To have the Linux Kernel, the Red Hat distro, mplayer, X and gcc (for
example, could be others as well) adopting the fGPL for the next
releases. After that, we'll see most every GPLd program adopting the
fGPL. After that, you'll start to see how much sense it made to pay $20
a year. And even the ones that can't pay (if any) will be able to use
the software (though 1 year old, but their hardware si severla years old
for sure).
This is my opinion. I'd gladly pay the $20, as long as EVERYONE ELSE
pays their $20. That's why we don't see many donations now: because you
have this felling everyone else is just waiting for a fool like you to
contribute to project X in order to save it.
------------- end -------------
Thanks for everything and to everyone here!
Federico
fferreres (@) ojf com
pd: Please CC if you need my reply as I am not on the list
On 26 Jul 2002, Federico Ferreres wrote:
>
> What do we need for this to happen?
>
> To have the Linux Kernel, the Red Hat distro, mplayer, X and gcc (for
> example, could be others as well) adopting the fGPL for the next
> releases. After that, we'll see most every GPLd program adopting the
> fGPL. After that, you'll start to see how much sense it made to pay $20
> a year. And even the ones that can't pay (if any) will be able to use
> the software (though 1 year old, but their hardware si severla years old
> for sure).
>
So, if I have a cheap/buggy motherboard and its faulty behavior is fixed
with a new kernel release last week, I don't have those $20... I have to wait
a year for a fix? I don't think so...
Regards
--
Robinson Maureira Castillo
Asesor DAI
INACAP
On Fri, 2002-07-26 at 13:15, Robinson Maureira Castillo wrote:
> So, if I have a cheap/buggy motherboard and its faulty behavior is fixed
> with a new kernel release last week, I don't have those $20... I have to wait
> a year for a fix? I don't think so...
You can chose to assign your money _whenever_ you want (you have 1 year
flexibility).
Project could open a "feature requests" if they like. Some of these
could work in a "reward" fashion. A reward could look like this:
"whoever finishes a working driver for <insert card> gets <sum of funds
offered by members>".
Federico
http://www.fool.com/news/foth/2001/foth011108.htm
} Wasn't there a South Park episode making fun of the dot coms where they
} were starting a business and someone asked about the plan and they said
}
} Step 1: Idea
} Step 2: Err...
} Step 3: Make money
}
} Or something like that. Where ever it was, it was pretty funny, it was
} quite obvious that the vast majority of the failures could be traced
} to a lack of a step 2. Seems so simple, doesn't it?
}
} On Fri, Jul 26, 2002 at 11:08:26AM -0600, Cort Dougan wrote:
} > "Bat-Boy creates open-source project"
}
} --
} ---
} Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
Wasn't there a South Park episode making fun of the dot coms where they
were starting a business and someone asked about the plan and they said
Step 1: Idea
Step 2: Err...
Step 3: Make money
Or something like that. Where ever it was, it was pretty funny, it was
quite obvious that the vast majority of the failures could be traced
to a lack of a step 2. Seems so simple, doesn't it?
On Fri, Jul 26, 2002 at 11:08:26AM -0600, Cort Dougan wrote:
> "Bat-Boy creates open-source project"
--
---
Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
On Fri, 26 Jul 2002, Larry McVoy wrote:
> to Microsoft. I've thought about it a lot and my attempts have pretty
> much failed, so it's encouraging to see someone else thinking hard
> about this.
Larry, what the hell are you smoking? It's a repost from tabloid, for
fsck sake - clearly says so in the beginning. Thinking is, indeed, hard
for these guys, but what's encouraging about that?
"Bat-Boy creates open-source project"
After defeating the evil-doers of the world Bat-Boy has taken up another
cause - titanic software companies. He vows to tirelessly create
sourceforge.net projects one after another until the sheer weight of them
collapses onto, and crushes, Microsoft.
"What's our business model?", asked Bat-Boy through a translator. "I eat
vermine and insects so I don't need much money. When I do need money, I
sell photos of myself to tabloids".
Bat-Boy, you do us all proud!
} Even tabloids can have good content, though I agree it's rare.
} After watching all the dot com and open source companies take a
} dive when it became apparent that no business model == no business,
} it's somewhat nice to see someone besides myself trying to figure out
} an answer which is actually sustainable. It's not an enjoyable thing,
} everyone hates you if you don't work all day long on their problems
} for free, but it's a good topic for further thought.
}
} I think that the guy is on the right track, I've frequently described
} software sales as similar to insurance, noone pays what it actually costs
} to handle the problem, everyone pays a little and the cost is spread
} out over everyone. The more people who pay, the less each has to pay,
} and that seems to be his message. I agree with that. He's basically
} right in theory, the problem is putting it into practice looks hard
} or impossible. But maybe someone will figure out a way, so I'm
} trying to be encouraging.
}
} Consider this my obligatory, once a year, "that's a good idea" post.
} Intended to balance out the zillion other posts saying "that's braindead". :)
} --
} ---
} Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
} -
} To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
} the body of a message to [email protected]
} More majordomo info at http://vger.kernel.org/majordomo-info.html
} Please read the FAQ at http://www.tux.org/lkml/
On Fri, Jul 26, 2002 at 01:01:21PM -0400, Alexander Viro wrote:
> On Fri, 26 Jul 2002, Larry McVoy wrote:
> > to Microsoft. I've thought about it a lot and my attempts have pretty
> > much failed, so it's encouraging to see someone else thinking hard
> > about this.
>
> Larry, what the hell are you smoking? It's a repost from tabloid, for
> fsck sake - clearly says so in the beginning. Thinking is, indeed, hard
> for these guys, but what's encouraging about that?
Even tabloids can have good content, though I agree it's rare.
After watching all the dot com and open source companies take a
dive when it became apparent that no business model == no business,
it's somewhat nice to see someone besides myself trying to figure out
an answer which is actually sustainable. It's not an enjoyable thing,
everyone hates you if you don't work all day long on their problems
for free, but it's a good topic for further thought.
I think that the guy is on the right track, I've frequently described
software sales as similar to insurance, noone pays what it actually costs
to handle the problem, everyone pays a little and the cost is spread
out over everyone. The more people who pay, the less each has to pay,
and that seems to be his message. I agree with that. He's basically
right in theory, the problem is putting it into practice looks hard
or impossible. But maybe someone will figure out a way, so I'm
trying to be encouraging.
Consider this my obligatory, once a year, "that's a good idea" post.
Intended to balance out the zillion other posts saying "that's braindead". :)
--
---
Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
On Fri, Jul 26, 2002 at 01:09:32PM -0300, Federico Ferreres wrote:
> I'd love to see a new license, that could be called the fGPL. That would
> be the "Funded GPL". To be able to use fGPLd programs you'll HAVE to
> contribute some small amount of money to the fGPL foundation. You'll not
> be required to pay for any individual fGPL software, just a plain simple
> yearly $10 or $20 charge. And you will be able to distribute exactly
> where that money goes, among all the different projects. If you can't
> pay $20 a year it will be no problem, just a bit penalty: all fGPL
> software would be free as in beer once the year passes (old releases).
>
> This is my opinion. I'd gladly pay the $20, as long as EVERYONE ELSE
> pays their $20. That's why we don't see many donations now: because you
> have this felling everyone else is just waiting for a fool like you to
> contribute to project X in order to save it.
If this model could be enforced, i.e., everyone had to pay, then this
would indeed be a revolutionary change to how software is developed.
It would bring in more than enough money.
One problem I see is that you'd be talking a huge amount of money,
potentially money on Microsoft scale. Managing that money, making
it go to the right places, without it sticking to the fingers of
management, isn't likely to happen. You'd need a real corporate
structure to do this and I suspect it would fail because noone would
trust them to do the right thing. There are plenty of people who
don't trust the FSF now. Imagine what the feeling would be if
$2B/year were headed their way.
One possible answer is to make each program its own profit and loss
center or corporation. But now you have to send $20 to the kernel.com
people and $20 to the apache.com people and $20 to ...
Another problem is that GPLed software is essentially software in the
public domain. Many people in many parts of the world will not obey
the license and will just stop shipping the source. Yes, you can
catch and pressure some of them, but you'll not catch the majority
of them, just the dumb ones. We found this out with BitKeeper,
people downloaded the source and promptly removed the openlogging
feature, even checking in a changeset with comments like "Disable
that stupid openlogging feature". You couldn't find a more blatent
violation of our license if you tried, but that doesn't stop people
from doing it.
The problem looks pretty intractable to me. I'm glad you are thinking
about it, I'd much prefer a world that was closer to open source than
to Microsoft. I've thought about it a lot and my attempts have pretty
much failed, so it's encouraging to see someone else thinking hard
about this.
--
---
Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
On 26 Jul 2002, Federico Ferreres wrote:
>
> You can chose to assign your money _whenever_ you want (you have 1 year
> flexibility).
>
> Project could open a "feature requests" if they like. Some of these
> could work in a "reward" fashion. A reward could look like this:
> "whoever finishes a working driver for <insert card> gets <sum of funds
> offered by members>".
>
You didn't get it... I _do not_ have the money, but I _do_ need the fix,
why do I have to wait a whole year?
--
Robinson Maureira Castillo
Asesor DAI
INACAP
The problem of not having enough resources to achieve a given goal has
plagued mankind for years. Open-source, GPL or magic dust are not going
to solve that problem.
} You didn't get it... I _do not_ have the money, but I _do_ need the fix,
} why do I have to wait a whole year?
On Fri, 2002-07-26 at 14:01, Alexander Viro wrote:
> Larry, what the hell are you smoking? It's a repost from tabloid, for
> fsck sake - clearly says so in the beginning. Thinking is, indeed, hard
> for these guys, but what's encouraging about that?
Would it have made any difference if I posted it here first? Or if I was
a respected developer? Shame on me, I happened to have studied
economics.
Anyway, the OSS model is just fine, so you shouldn't care. If at any
point you change your mind and want/need my $20 bucks, I will be ready
to support you (as well as the hundred millions x $20 waiting out there
arround the globe).
Federico
On Fri, 26 Jul 2002 15:48:21 -0400 (CLT)
Robinson Maureira Castillo <[email protected]> wrote:
> On 26 Jul 2002, Federico Ferreres wrote:
>
> >
> > You can chose to assign your money _whenever_ you want (you have 1
> > year flexibility).
> >
> > Project could open a "feature requests" if they like. Some of these
> > could work in a "reward" fashion. A reward could look like this:
> > "whoever finishes a working driver for <insert card> gets <sum of
> > funds offered by members>".
> >
>
> You didn't get it... I _do not_ have the money, but I _do_ need the
> fix, why do I have to wait a whole year?
>
Because you do not have the money.
If you want to donate money to a project(s), I'm sure that a simple email
to the maintainer(s) of that project would get you an address to send such
a donation to. Myself, I feel that donation should be optional, besides
the obvious practical difficulties with something like this (e.g., the
task of keeping track of who has paid, not to mention the necessary
restrictions on downloading at various sites, not to mention the fact that
people will end up just getting source and posting it for everyone, not to
mention...)
On 26 Jul 2002, Federico Ferreres wrote:
> On Fri, 2002-07-26 at 14:01, Alexander Viro wrote:
> > Larry, what the hell are you smoking? It's a repost from tabloid, for
> > fsck sake - clearly says so in the beginning. Thinking is, indeed, hard
> > for these guys, but what's encouraging about that?
>
> Would it have made any difference if I posted it here first? Or if I was
> a respected developer? Shame on me, I happened to have studied
> economics.
>
> Anyway, the OSS model is just fine, so you shouldn't care. If at any
> point you change your mind and want/need my $20 bucks, I will be ready
> to support you (as well as the hundred millions x $20 waiting out there
> arround the globe).
>
> Federico
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
>
On Fri, 26 Jul 2002, Cort Dougan wrote:
> The problem of not having enough resources to achieve a given goal has
> plagued mankind for years. Open-source, GPL or magic dust are not going
> to solve that problem.
I Agree.
But I don't think that the kernel should adopt this new license, what stop
some developers to fork another tree under the GPL?
IMHO, kernel should stay free...
Best regards
--
Robinson Maureira Castillo
Asesor DAI
INACAP
On Fri, 2002-07-26 at 17:11, Federico Ferreres wrote:
> On Fri, 2002-07-26 at 16:48, Robinson Maureira Castillo wrote:
> > On 26 Jul 2002, Federico Ferreres wrote:
> >
> > >
> > > You can chose to assign your money _whenever_ you want (you have 1 year
> > > flexibility).
> > >
> > > Project could open a "feature requests" if they like. Some of these
> > > could work in a "reward" fashion. A reward could look like this:
> > > "whoever finishes a working driver for <insert card> gets <sum of funds
> > > offered by members>".
> > >
> >
> > You didn't get it... I _do not_ have the money, but I _do_ need the fix,
> > why do I have to wait a whole year?
> >
> >
> > --
> > Robinson Maureira Castillo
> > Asesor DAI
> > INACAP
Idea 1: You do have $10 and you are trying to free-ride (honest answer).
Idea 2: You don't have $10 or $20 a year to "spare"? Then probably you are using
(old and) already supported hardware. And because past years developements
are free as in beer. If you ARE a developer, you get a free permament
membership so you shouldn't care about all this.
Idea 3: Make bugfixes and hardware support are always GPL and not fGPL. The fGPL
will force you to distribute under GPL (or fGPL, at your choise).
Idea 4: Write your own drivers if nobody else would.
Idea 5: You are a student and as such are granted a free membership
until you finish your studies.
Federico
On Fri, 2002-07-26 at 17:29, Jon Portnoy wrote:
> If you want to donate money to a project(s), I'm sure that a simple email
> to the maintainer(s) of that project would get you an address to send such
> a donation to.
Some people don't want to be only ones donating, some others just will
just wait & see if somebody elses money saves the project. Also, your
money may not be enough to fund that project and you are left with no
improvements. Or you could be overfunding the project (never seen this
happen though). You have no warranties where and how the money would be
usued.
> Myself, I feel that donation should be optional, besides
> the obvious practical difficulties with something like this (e.g., the
> task of keeping track of who has paid
A fundation focusing on this issue alone would be the best solution for
solving this "fragmentation" problem.
> not to mention the necessary
> restrictions on downloading at various sites, not to mention the fact that
> people will end up just getting source and posting it for everyone, not to
> mention...)
No problem here really. No restricted downloads, not anti-piracy code.
No added burdens. No BSA. Just the legal obligation to pay $10 a year.
If you don't pay, nobody will come looking for you, but at least you'll
know you owe somebody real money.
Federico
> On 26 Jul 2002, Federico Ferreres wrote:
>
> > On Fri, 2002-07-26 at 14:01, Alexander Viro wrote:
> > > Larry, what the hell are you smoking? It's a repost from tabloid, for
> > > fsck sake - clearly says so in the beginning. Thinking is, indeed, hard
> > > for these guys, but what's encouraging about that?
> >
> > Would it have made any difference if I posted it here first? Or if I was
> > a respected developer? Shame on me, I happened to have studied
> > economics.
> >
> > Anyway, the OSS model is just fine, so you shouldn't care. If at any
> > point you change your mind and want/need my $20 bucks, I will be ready
> > to support you (as well as the hundred millions x $20 waiting out there
> > arround the globe).
> >
> > Federico
> >
> >
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to [email protected]
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> > Please read the FAQ at http://www.tux.org/lkml/
> >
> >
>
On Fri, 2002-07-26 at 17:07, Robinson Maureira Castillo wrote:
> On Fri, 26 Jul 2002, Cort Dougan wrote:
>
> > The problem of not having enough resources to achieve a given goal has
> > plagued mankind for years. Open-source, GPL or magic dust are not going
> > to solve that problem.
>
> I Agree.
>
> But I don't think that the kernel should adopt this new license, what stop
> some developers to fork another tree under the GPL?
Nothing is preventing them, and that's a good thing. As at is now,
everyone can fork a kernel. And everybody is inteligent enough to know
which one to pick.
If they want to switch the license to GPL, they have to start with a 1
year old source or stick to the fGPL.
Now, It may be true that some projects should not be fGPLd or have
exceptions, for one reason or another. It's up to the developers to
decide what's best for them.
> IMHO, kernel should stay free...
Free as in no money to the developers? It isn't really free. You need
Internet access to download the kernel (or to buy a CD), a computer and
electricity among other things. And the developers need to eat, to have
bandwith donate, hardware donated, etc.
Federico
> Best regards
> --
> Robinson Maureira Castillo
> Asesor DAI
> INACAP
>
On 26 Jul 2002, Federico Ferreres wrote:
[sigh... I shouldn't be doing that, but...]
> Idea 1: You do have $10 and you are trying to free-ride (honest answer).
>
> Idea 2: You don't have $10 or $20 a year to "spare"? Then probably you are using
> (old and) already supported hardware. And because past years developements
> are free as in beer. If you ARE a developer, you get a free permament
> membership so you shouldn't care about all this.
>
> Idea 3: Make bugfixes and hardware support are always GPL and not fGPL. The fGPL
> will force you to distribute under GPL (or fGPL, at your choise).
>
> Idea 4: Write your own drivers if nobody else would.
>
> Idea 5: You are a student and as such are granted a free membership
> until you finish your studies.
What you and the rest of armchair generals do not get is that "adding
features" is _not_ the hard part of work. Doing that in a way that
wouldn't be a permanent source of bugs afterwards and cleaning up the
existing sources of bugs _IS_. So is doing infrastructure work. So
is auditing code. So is removing crap code.
None of that is covered by your "model". 99:1 that your "working driver
for card" is going to contain a bunch of root holes. _And_ be unmaintainable.
I had seen quite a few vendor drivers. Every time I'm looking at one of
them, I'm reminded of MST3K. Yup, Mistery Science Theater 3000. With
guy being forced to watch crap selected to drive him mad.
And it's not just vendor drivers. Example: recently a new version of
cpio(1) had been released (after 6 years of inactivity). It still
contains idiotic holes reported (with fixes) years ago. Many times.
Some of these holes going back to 1993 (first report I'd been able to
find, followed by ~5 rediscoveries of the same bug). Who should be
paying to whom in cases like that?
The thing being, absolute majority of software is crap. That has nothing
to getting paid for it and everything with average quality of programmers.
For how many years did the (well-paid) rogering tosspots in SGI ship IRIX
with sendmail choke-full of known root holes? (not to mention that
it was configured as an open relay effectively hiding the IP of submitting
host - spammers' dream come true). For how many years did Sun ship
systems with mind-boggling default configuration? (NIS holes galore)
For how many years does Microsoft ship what they are shipping?
As long as crap software is considered acceptable and people who write
crap - employable, the things will be bad and job market - overcrowded.
It's that simple.
On Friday 26 July 2002 05:23 pm, Alexander Viro wrote:
> I had seen quite a few vendor drivers. Every time I'm looking at one of
> them, I'm reminded of MST3K. Yup, Mistery Science Theater 3000. With
> guy being forced to watch crap selected to drive him mad.
>
> And it's not just vendor drivers.
[cut]
> The thing being, absolute majority of software is crap.
http://www.tuxedo.org/~esr/jargon/html/entry/Sturgeon's-Law.html
Nothing special about programming, this applies to television programs, acts
of congress, the skill of most drivers on the highways, grocery shopping...
Fighting off Sturgeon's law is what maintainers DO. It's why 90% of all
patches are dropped...
It's also the core business of Red Hat, Google, Slashdot, Slashdot's
moderation system, KernelTrap and Kernel Traffic... (This isn't to say that
the end result is less than 50% crud, just that they try very hard to change
the percentages. With varying degrees of success...)
Rob
On Fri, 2002-07-26 at 18:23, Alexander Viro wrote:
>
>
> On 26 Jul 2002, Federico Ferreres wrote:
> What you and the rest of armchair generals do not get is that "adding
> features" is _not_ the hard part of work. Doing that in a way that
> wouldn't be a permanent source of bugs afterwards and cleaning up the
> existing sources of bugs _IS_. So is doing infrastructure work. So
> is auditing code. So is removing crap code.
Al, your point is 100% valid. That's why is said "Projects could open
"feature requests" if they like." I think the important thing is that
developers are the ones that should run the show.
The idea here is to offer a monetary compensation for the hard work of
developers directly from the users in such a way that the costs will be
spread among millions of people.
How to pump money into projects and at the same time allow developers to
do The Right Thing, is something that can be discussed and shaped
according to each developer needs. The kernel will not want to accept
code that came from a feature requests (maintainability or security or
for whatever the reason). And that's good! Only good code should make it
in.
An idea for funding the core developements would be to preassign a fixed
15% of all the funds at this developements (gcc, glibc, kernel, OpenSSH,
etc.)
The rest will go to userland as the members see fit, as most (home)
users think everything is about apps. We might even be able to fund
games and (uneeded but wanted) things like that, boosting Linux
adoption. We may be even able to buy some patents (like the SGI ones
that MS bought, before somebody elses limits us). Or we may be able to
open the code to certain apps to turn them into fGPL, if need be.
So in brief, the power will remain at the developers, because they are
the ones that know what they are doing. If some folks want an insecure
piece of code to make it into (for example) the kernel, they'll be so
out of luck as they are today.
The idea may be crap in many people minds. It's not perfect, it's not
without dangers. Maybe OSS doesn't have a need for those funds. But they
money is there. It just needs a way to flow from users to developers.
Right now it's flowing from the users to Microsoft. That's fine, but we
could do better, faster and bring freedom to users.
All we'll need is discussion of pro/cons and whoever likes the pros more
can fGPL. After 4 or 5 core apps fGPL, everyone will want to fGPL and
get some money to pay the expenses and hard work they are doing.
The choice is all yours ... you own this market, you control it and you
deserve to manage things as you see fit. My $20 will be waiting for a
way to contribute in an inteligent manner which also forces everyone
else to do the same.
Federico
On 26 Jul 2002, Federico Ferreres wrote:
> All we'll need is discussion of pro/cons and whoever likes the pros more
> can fGPL. After 4 or 5 core apps fGPL, everyone will want to fGPL and
> get some money to pay the expenses and hard work they are doing.
This isn't even possible for the kernel nor is it possible for any GPL
app. If you want this you will need to start from scratch since "fGPL"
would not be compatable with GPL you would need to get every last person
who submitted a patch to relicence.
I just don't see the point of trying to argue for what is essentially an
imposibillity and that's even if everyone here agrees to it.
Gerhard
PS There is something to be had here that is better than money.. it's a
chance to read and learn from the best. I'm not kernel level yet but I do
contribute to other projects. I learn a lot just by reading and one day I
hope to have code good enough to be slammed by Alan Cox, Al Viro or even
Linus himself ;)
--
Gerhard Mack
[email protected]
<>< As a computer I find your faith in technology amusing.
On Sat, 2002-07-27 at 00:02, Gerhard Mack wrote:
> you would need to get every last person who submitted a patch to
> relicence.
That's a real problem (and that's probably why Sun tries to be the
copyright holder in OO)
> I just don't see the point of trying to argue for what is essentially an
> imposibillity and that's even if everyone here agrees to it.
Maybe the kernel would be wrong place to start, besides the kernel is
already well funded. (It could provoke a demostration or cascading
effect).
> Gerhard
>
> PS There is something to be had here that is better than money.. it's a
> chance to read and learn from the best. I'm not kernel level yet but I do
> contribute to other projects. I learn a lot just by reading and one day I
> hope to have code good enough to be slammed by Alan Cox, Al Viro or even
> Linus himself ;)
Yes, that's clear. It's great. You learn, then you contribute back. But
that doesn't allow you to pay your bills, does it? The thing is massive
amounts of people _don't_ code, never will, are willing to contribute
but don't do it, because they are not beign asked for support in the
right way. You need them to know they'll put $10 and get $100.000.000
worth every year.
But if OSS finds a better way to fund developement, it certainly won't
hurt your learning experience. It may well improve it (more code, more
research, more accesibility). The point is to allow more people full
dedication to OSS.
Federico
PD: Maybe some day we'll find out the perfect solution.
> --
> Gerhard Mack
>
> [email protected]
>
> <>< As a computer I find your faith in technology amusing.
>
On 27 Jul 2002, Federico Ferreres wrote:
> Yes, that's clear. It's great. You learn, then you contribute back. But
> that doesn't allow you to pay your bills, does it? The thing is massive
> amounts of people _don't_ code, never will, are willing to contribute
> but don't do it, because they are not beign asked for support in the
> right way. You need them to know they'll put $10 and get $100.000.000
> worth every year.
Thank you for taking Economics 101. Your final score is F-.
Don't let it discourage you from taking the test next year.
Pay particular attention to the reasons why pyramid schemes
are not sustainable. We wish you a nice summer and hope
to see you next semester.
Al,
On Sat, 2002-07-27 at 02:34, Alexander Viro wrote:
> Thank you for taking Economics 101. Your final score is F-.
> Don't let it discourage you from taking the test next year.
> Pay particular attention to the reasons why pyramid schemes
> are not sustainable. We wish you a nice summer and hope
> to see you next semester.
>
Do you know the conditions for a pyramid scheme? A piramid scheme starts
when a guy can convince some (more than 1) other guys to pay him a
certain amount, with the promise that they will be able to do just the
same sale to other people. Of course, at some point the system stops
working as everyone knows (easy math, no 101 needed).
The fGPL scheme means you pay $10, and you get code in exchange. The
developer receiving the funds either buys food, a computer, a book, or
maybe even uses it for a nice trip with his/her family after their hard
work is done. Let me know when you find the pyramid!
You can be agressive all that you want, but it doesn't look good when
you are making a fool of youself. Millions of people like me are free
riding your hard work. Dump my idea, laugh at it, hate me for being so
stupid...we couldn't care less :)
Federico
PD: I do know you are one of the most talented people around, and I
couldn't thank you enough for the hard work you do everyday. But that
doesn't mean you are always right. Sorry.
On 27 Jul 2002, Federico Ferreres wrote:
> The fGPL scheme means you pay $10, and you get code in exchange. The
> developer receiving the funds either buys food, a computer, a book, or
> maybe even uses it for a nice trip with his/her family after their hard
> work is done. Let me know when you find the pyramid!
Sigh... That stops working at the same point where pyramid does - when
number of recepients becomes a sufficiently large fraction of all potential
participants. It is not sustainable.
Simple math: Debian got several thousand developers. Most of the packages
contain both contributions from said developers and stuff from upstream.
By very conservative estimates it's tens of thousands. If your $10 is
per package update - it's impossible to pay. _Really_ impossible - on a
reasonable system it will easily amount to $50 _daily_. If it's $10 per
year - count the number of installations and look how much it will give
for one developer.
It won't work for the same reason why pyramids are unsustainable - when
you have too many recepients, both go to hell. And there _is_ too many
for that to work.
It's $10 or $20 per user per year to be able to use fGPL code _alone_.
In addition to that, you'll have to pay for any extras (services or non
fGPL code) you may like purchasing: the nice Red Hat Network services,
your Debian CD and my Slackware CD (but I would preffer to just fund
Patrick though fGPL instead). The membership just allows you into "the
club".
Also, publishing something under the fGPL doesn't mean the developer
will get the funds, unless people choose to fund it or if it's a core
app (this last decided by a board of directors for example, lead by
Linus or whoever is most respected and wants to get involved and up to a
fixed % of the yearly fund raising).
This doesn't kill any other funding source you already have established.
IBM may not want to fGPL webalizer, Covalent or MySQL AB as well. And
that's fine. Copyright owners can offer them other licenses suitable for
them (X style, BSD, whatever). Also if IBM (or any other company) wants
to fund a different project because it makes sense for them, and users
and small companies do not want to pay for it, then they will fund it
themselves, as it is the case today.
Now some figures: if you have 50.000.000 (end users and companies) there
paying $20 (as an example), that's about $1B a year average as Larry
suggested. If you have 5.000.000 it's only $100M. You at least have
some extra funds available that go DIRECTLY to developers. And as all
projects progress faster, you will have widespread adoption of OSS: I
will just be the natural step money-wise.
If 1B a year is not enough, then some projects will have to wait, or the
users base must grow (and it will grow surelly).
The Linux Kernel and the core of OSS will have a fixed percentaje pumped
to them, so the ones that know what must be done are not bothered. Say
10% (or a different %). That's between $100M and $10M extra a every
year, and you can always count on it. If you don't need that much, then
you could resign the right. And if it's not enough, then it's much
better than nothing / "charity".
It works for Microsoft and they make crappy software which is also
propietary, closed, and expensive. Why wouldn't it work for OSS? You'll
be making everyone else in the world a favor. They collect most of their
revenues due to the Windows plataform. The kernel may provide that base
for all the GNU deck.
>From the end user perspective it would also make a lot of sense. Whoever
can't pay $20 year, doesn't know how to code, is not a student or a
charity organization may well thank everybody else's for they will be
able to use last year developement for free as in freedom and in beer.
If anyone is concerned about the impact $20 will have, as long as you
allow piracy (a la Microsoft), there's no problem. It may even help OSS.
Now the real concerns would be:
- Will the money be used wisely?
- Will money change the spirit of OSS and bring bad code with it?
I couldn't answer those. If the answer is negative, then OSS is better
off underfunded (as it is today). All for the love.
But what if it works out nice? You are indeed underestimating the
willing of non-coders to contribute. They will probably even be proud
members of the fGPL foundation.
Federico
On Sat, 2002-07-27 at 03:29, Alexander Viro wrote:
>
>
> On 27 Jul 2002, Federico Ferreres wrote:
>
> > The fGPL scheme means you pay $10, and you get code in exchange. The
> > developer receiving the funds either buys food, a computer, a book, or
> > maybe even uses it for a nice trip with his/her family after their hard
> > work is done. Let me know when you find the pyramid!
>
> Sigh... That stops working at the same point where pyramid does - when
> number of recepients becomes a sufficiently large fraction of all potential
> participants. It is not sustainable.
>
> Simple math: Debian got several thousand developers. Most of the packages
> contain both contributions from said developers and stuff from upstream.
> By very conservative estimates it's tens of thousands. If your $10 is
> per package update - it's impossible to pay. _Really_ impossible - on a
> reasonable system it will easily amount to $50 _daily_. If it's $10 per
> year - count the number of installations and look how much it will give
> for one developer.
>
> It won't work for the same reason why pyramids are unsustainable - when
> you have too many recepients, both go to hell. And there _is_ too many
> for that to work.
>
On Sat, 2002-07-27 at 03:29, Alexander Viro wrote:
> Sigh... That stops working at the same point where pyramid does - when
> number of recepients becomes a sufficiently large fraction of all potential
> participants. It is not sustainable.
I forgot to answer that one. It doesn't matter how many people want to
join a project. It's up to the copyright owners to decide how much stuff
they need, who to "work with", "how much folks are needed" and how is
the line of comand structured. At some point, contributors will have to
send free patches as they are doing now. That may or may not earn them
into the "core team".
Users will decide how much money to put in such projects. If not enough
funds are pointed in your direction (based on how much the developer
requested), you'll just have to do with less (which is better than
nothing). And the asked funds will probably make sense, because if they
don't there will be an incentive to branchthe project (except for the
core apps as already pointed out).
In the Debian scheme, people can't choose where the money goes, and
there's no strict limit to the number of developers, and it's not a big
enough user base (though big). As you point out, the ships sinks, as
it's a classical example of the Fishermen dylema: the more you capture,
the better of you are off, but you can't prevents others from doing the
same. After some time there's so much fishermen that nobody capture
enough to make a living from it.
And that's why fGPL should allow people to choose their funded projects
(except the core systems) and why lead developers should be able to
manage the funds as they see fit.
Federico
On 27 Jul 2002, Federico Ferreres wrote:
> It works for Microsoft and they make crappy software which is also
You want us to make crappy software with silly restrictions, too ?
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
On Sat, Jul 27, 2002 at 12:42:14PM -0300, Rik van Riel wrote:
> On 27 Jul 2002, Federico Ferreres wrote:
> > It works for Microsoft and they make crappy software which is also
>
> You want us to make crappy software with silly restrictions, too ?
Cute. How about we ease off on this guy a bit and ask ourselves if he
doesn't have a point? It's not clear that he has the right answer but
it is blindingly clear that pure open source business models don't work.
Every single one of them which has any success has found a way to derive
money from some other source to fund the open source development.
You could argue that everything is fine and you'd be wrong. Not all
applications can find a way to piggyback on some other sale. Consider a
Word replacement, how do you generate money for that? You can't tie it
to hardware sales, you can try the support route but that isn't going
to work, so what do you do? And there are a million other apps which
are the same.
It's easy to beat up on this guy but he's trying to find a reasonable
compromise which allows open source to be self sustaining for all
projects. That's a great goal. Right now, open source funding is
a joke. Microsoft wipes their butt with more money than all of the
open source revenue put together. Yeah, it's very macho to say that
you'll program circles around them, but it's also wrong. You may in
one or two areas but they have a zillion apps that lawyers and doctors
and other non-geeks use every day. Where is the money for those apps?
It's sure as hell not coming from "donations".
So how about we all cut the guy some slack, maybe he's misdirected, maybe
he has the wrong answer, but he's not working on the wrong problem in
my opinion. Try helping him, it might be in our best interest to do so.
--
---
Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
On Sat, 27 Jul 2002, Larry McVoy wrote:
> On Sat, Jul 27, 2002 at 12:42:14PM -0300, Rik van Riel wrote:
> > On 27 Jul 2002, Federico Ferreres wrote:
> > > It works for Microsoft and they make crappy software which is also
> >
> > You want us to make crappy software with silly restrictions, too ?
>
> Cute. How about we ease off on this guy a bit and ask ourselves if he
> doesn't have a point? It's not clear that he has the right answer but
> it is blindingly clear that pure open source business models don't work.
> ... but they have a zillion apps that lawyers and doctors and other
> non-geeks use every day. Where is the money for those apps? It's sure
> as hell not coming from "donations".
Agreed. What we need is funding for the development of those
things which aren't already taken care of by the community.
In the areas where open source software is doing well, there
already is funding by interested parties (companies, government,
universities, ...).
Some other areas won't ever get the funding through donations,
simply because people will freeload whenever they can and try
funding development as much as they can. We've seen that with
BitKeeper and you had to "tighten up" the license a bit in order
to make sure development stayed funded.
In short, I believe the voluntary donations aren't needed in
most areas people would donate to and won't make enough of an
impact in the areas where they are needed.
regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
On Sat, Jul 27, 2002 at 01:06:56PM -0300, Rik van Riel wrote:
> Some other areas won't ever get the funding through donations,
> simply because people will freeload whenever they can and try
> funding development as much as they can. We've seen that with
> BitKeeper and you had to "tighten up" the license a bit in order
> to make sure development stayed funded.
Yeah, and I wasn't thrilled about it, I can tell you that. I had the
same starry eyed idealistic idea that things would just work out and
they don't. So we're essentially in a proprietary model with a funky
dual free/pay licensing model. That's OK for us because the money
people value their privacy.
Actually, that's an interesting topic. Other applications could use
the BK model of "free if you're out in the open" and pay otherwise.
It's pretty effective. However, it doesn't work very well when the
community beats you to hell for not being GPLed. I had a thick enough
skin to deal with it, I doubt others would, they'd give up. It also
doesn't work when people refuse to obey the license because they
don't agree with it (we had plenty of that).
That leads to the real question: what is an acceptable model for the
community and the vendor for applications which don't work under the
standard GPL model?
> In short, I believe the voluntary donations aren't needed in
> most areas people would donate to and won't make enough of an
> impact in the areas where they are needed.
Agreed, but his thought was to make it non-voluntary. I have my doubts
about that working too, but there ought to be something which would work.
--
---
Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
On Saturday 27 July 2002 18.22, Larry McVoy wrote:
> On Sat, Jul 27, 2002 at 01:06:56PM -0300, Rik van Riel wrote:
> Actually, that's an interesting topic. Other applications could use
> the BK model of "free if you're out in the open" and pay otherwise.
> It's pretty effective. However, it doesn't work very well when the
> community beats you to hell for not being GPLed. I had a thick enough
> skin to deal with it, I doubt others would, they'd give up. It also
> doesn't work when people refuse to obey the license because they
> don't agree with it (we had plenty of that).
Trolltech does the same. And have taken the same amount of heat.
(Probably A LOT more... since their Qt is the base for KDE)
/RogerL
--
Roger Larsson
Skellefte?
Sweden
Replying to Alexander Viro:
> What you and the rest of armchair generals do not get is that "adding
> features" is _not_ the hard part of work. Doing that in a way that
> wouldn't be a permanent source of bugs afterwards and cleaning up the
> existing sources of bugs _IS_. So is doing infrastructure work. So
> is auditing code. So is removing crap code.
One remark is - Microsoft actually "conquered" the market by adding features
_only_.
--
Paul P 'Stingray' Komkoff 'Greatest' Jr /// (icq)23200764 /// (http)stingr.net
When you're invisible, the only one really watching you is you (my keychain)
On Sat, 2002-07-27 at 13:46, Roger Larsson wrote:
> On Saturday 27 July 2002 18.22, Larry McVoy wrote:
> > On Sat, Jul 27, 2002 at 01:06:56PM -0300, Rik van Riel wrote:
> > Actually, that's an interesting topic. Other applications could use
> > the BK model of "free if you're out in the open" and pay otherwise.
> > It's pretty effective. However, it doesn't work very well when the
> > community beats you to hell for not being GPLed. I had a thick enough
> > skin to deal with it, I doubt others would, they'd give up. It also
> > doesn't work when people refuse to obey the license because they
> > don't agree with it (we had plenty of that).
>
> Trolltech does the same. And have taken the same amount of heat.
> (Probably A LOT more... since their Qt is the base for KDE)
>
> /RogerL
The problem with the present type of donation systems is knowing your
donation is not going to be abused.
I though a model that may work is a United Way type of model ... have a
central non-profit act as a clearing house for donations to specific
open source projects. The non-profit would do the proper account
reporting of receipts, overhead, and funding to the various open source
projects (an effort to keep donation abuse down and allow open non-bias
account reporting). Allowing either a general donation or target
donation from the public to a particular project. Then a specific
project would register with the non-profit to receive funding help. In
order to qualify you would need to provide certain technical/financial
reporting of your project to the non-profit to demonstrate you need the
financial help and you are working on a needed project.
The real problem is getting something like this started.
On Sat, 2002-07-27 at 13:06, Rik van Riel wrote:
> Agreed. What we need is funding for the development of those
> things which aren't already taken care of by the community.
The problem is not the community, but how the comunity is structured in
certain areas. If big companies, educational institutions or goverments
are not interested in your project, the OSS model fails as you stated.
> In the areas where open source software is doing well, there
> already is funding by interested parties (companies, government,
> universities, ...).
100% true.
> In short, I believe the voluntary donations aren't needed in
> most areas people would donate to and won't make enough of an
> impact in the areas where they are needed.
Good point. I started at the wrong place. Anyway, if your funding is
based on final users payments you need to make payment a non-option. So
if vital parts of GNU don't help them a bit, it may just not work. If
Windows/DOS didn't help fund Office, would we be having the Office
compatibility problems now? Extend the list to all MS userland apps.
As long as payment is optional and fragmented, OSS will have a hard time
in userland core apps. Maybe an fGPL could work for them. Bad luck the
projects that could help are already well funded.
I'll get back and rethink the problem ...
Federico
PS: Larry, it's ok really. Harsh comments that make sense are much
better than silence. Though without your posts it'd have been mostly
flames. Thanks!
> regards,
>
> Rik
> --
> Bravely reimplemented by the knights who say "NIH".
>
> http://www.surriel.com/ http://distro.conectiva.com/
>
Larry McVoy wrote:
> On Sat, Jul 27, 2002 at 01:06:56PM -0300, Rik van Riel wrote:
> > Some other areas won't ever get the funding through donations,
> > simply because people will freeload whenever they can and try
> > funding development as much as they can. We've seen that with
> > BitKeeper and you had to "tighten up" the license a bit in order
> > to make sure development stayed funded.
>
> Yeah, and I wasn't thrilled about it, I can tell you that. I had the
> same starry eyed idealistic idea that things would just work out and
> they don't. So we're essentially in a proprietary model with a funky
> dual free/pay licensing model. That's OK for us because the money
> people value their privacy.
>
> Actually, that's an interesting topic. Other applications could use
> the BK model of "free if you're out in the open" and pay otherwise.
> It's pretty effective. However, it doesn't work very well when the
> community beats you to hell for not being GPLed. I had a thick enough
> skin to deal with it, I doubt others would, they'd give up. It also
> doesn't work when people refuse to obey the license because they
> don't agree with it (we had plenty of that).
>
> That leads to the real question: what is an acceptable model for the
> community and the vendor for applications which don't work under the
> standard GPL model?
>
> > In short, I believe the voluntary donations aren't needed in
> > most areas people would donate to and won't make enough of an
> > impact in the areas where they are needed.
>
> Agreed, but his thought was to make it non-voluntary. I have my doubts
> about that working too, but there ought to be something which would work.
> --
> ---
> Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
Hi to u all!
It's funny how some things seem to pop in to peoples minds
almost simultainously. I believe some people call this
phenomenon "A collective consciousness" I've had some
open source funding Ideas spinning in my head since the
beginning of May. My Ideas differ from Fredricos in some
crucial ways. But I believe that they might be connectable.
in a way that will form a nice symbiosis.
I'm willing to put considerable amount of time in trying to
resolve, design and develop both of them together with others,
at least in part into a reasonable functionality.
This will take coordinated discussions, transcripting, more
coordinated discussions A lot of book reading ( law practise ),
advocacy, contacting people at nearby Universities, forum creation,
advertizing on the internet, Writing forms, making inquiries,
contacting authorities, etc, etc, etc
It's not something that can be done just by one or two dudes.
It will probably not be achievable unless at least 50 persons
get deeply committed around the world for starters. and
perhaps 200 guys for additional support.
First I want to know If more people are interested in doing
the same. I have an Email adress that is not yet in use.
, so if you send a mail to
[email protected]
with "some cool subject line"
and a body saying something like:
======================
I'm In for now. I have some spare time
to do some tedious and boring stuff.
Let's see where this "fantasy" is heading.
======================
I will send the result as a reply to everyone that
showed interest in about three days from the time
this letter got posted. It will NOT go on LKML.
Oh yes, almost forgot about the math part:
$1000 000 000 / 100 000 hackers = $10 000 per hacker.
I would be quite happy if I got $10 000 a year just for getting
a little patch that I wrote in a day or two into some GPL:ed software.
Especially if I knew that someone sent me this money just because
this dude appreciates the software that I work with specifically.
( even just a $1000 which would finance two new boxes / year )
The basis of my own idea is built in part on creating an
Open Patent Funding service organization.
Finance is partly solved by administration of several
small charity foundations with identical purpose,
created by the same amount of SW-companies in order
to avoid tax, as well as being able to use new SW-methods
freely. I've googled for this without any success.
kind regards
/Daniel Mose
On Sun, 28 Jul 2002, Daniel Mose wrote:
> Hi to u all!
>
> It's funny how some things seem to pop in to peoples minds
> almost simultainously. I believe some people call this
> phenomenon "A collective consciousness" I've had some
> open source funding Ideas spinning in my head since the
> beginning of May. My Ideas differ from Fredricos in some
> crucial ways. But I believe that they might be connectable.
> in a way that will form a nice symbiosis.
... and leverage the synergy. Or something.
[more spew snipped]
Larry, do you really like the company? Slashdorks and pretentious
self-important wankers speaking PHBese... YMMV, but IMO "prose" above
is every bit as disgusting as usual GPL advocacy. Same depth of
thought and amount of clue behind it, same lack of taste...
On Sun, Jul 28, 2002 at 01:21:56AM -0400, Alexander Viro wrote:
> [more spew snipped]
>
> Larry, do you really like the company? Slashdorks and pretentious
> self-important wankers speaking PHBese... YMMV, but IMO "prose" above
> is every bit as disgusting as usual GPL advocacy. Same depth of
> thought and amount of clue behind it, same lack of taste...
I'm not a big fan of fanatics and I see your point. My perspective is
probably skewed from the whole BK experience. These guys may be a
waste of time but I don't see why they should be shut down. I happen
to agree that there is a problem which needs solving. If they want to
solve it, or better yet, if they simply act as a catalyst and someone
else solves it, that's cool by me.
--
---
Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
I stated a simple idea aimed at solving a real world issue. And you
haven't proved it wrong. It may not be what you or the kernel hackers
need/want (which is FINE). But it would solve ALL the funding problems
at least.
The name calling is ok, but only if it makes you feel smarter or
superior.
Regards,
Federico
On Sun, 2002-07-28 at 02:21, Alexander Viro wrote:
>
>
> On Sun, 28 Jul 2002, Daniel Mose wrote:
>
> > Hi to u all!
> >
> > It's funny how some things seem to pop in to peoples minds
> > almost simultainously. I believe some people call this
> > phenomenon "A collective consciousness" I've had some
> > open source funding Ideas spinning in my head since the
> > beginning of May. My Ideas differ from Fredricos in some
> > crucial ways. But I believe that they might be connectable.
> > in a way that will form a nice symbiosis.
>
> .. and leverage the synergy. Or something.
>
> [more spew snipped]
>
> Larry, do you really like the company? Slashdorks and pretentious
> self-important wankers speaking PHBese... YMMV, but IMO "prose" above
> is every bit as disgusting as usual GPL advocacy. Same depth of
> thought and amount of clue behind it, same lack of taste...
>
On 28 Jul 2002, Federico Ferreres wrote:
> I stated a simple idea aimed at solving a real world issue. And you
> haven't proved it wrong. It may not be what you or the kernel hackers
> need/want (which is FINE). But it would solve ALL the funding problems
> at least.
You don't get it. So far the only guy who had been charitable was Larry, who
felt that problem was real but had serious doubts about viability of your
idea. I don't feel charitable and I've no reason to hesitate telling that
you guys _are_ waste of time. No maybes about it. It's that simple...
On 28 Jul 2002, Federico Ferreres wrote:
> I stated a simple idea aimed at solving a real world issue. And you
> haven't proved it wrong.
He doesn't need to prove it wrong, new ideas need to be proven right.
regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
On Sun, 2002-07-28 at 07:35, Alexander Viro wrote:
> You don't get it. So far the only guy who had been charitable was Larry, who
> felt that problem was real but had serious doubts about viability of your
> idea.
Actually, this is what Larry said:
| One problem I see is that you'd be talking a huge amount of money,
| potentially money on Microsoft scale. Managing that money, making
| it go to the right places, without it sticking to the fingers of
| management, isn't likely to happen.
This is not a problem of fundraising, but a problem of fund management
and policy. You get the funds, how you use them is another concern.
| Another problem is that GPLed software is essentially software in the
| public domain.
But we don't want less people using (for example), the Linux kernel. We
don't even want people who can't pay to pay. As long as a large enough
group pays, you're really better off.
And here is what Rik said:
| In the areas where open source software is doing well, there
| already is funding by interested parties (companies, government,
| universities, ...).
Meaning the kernel and some other OSS areas are already well funded.
| Some other areas won't ever get the funding through donations,
| simply because people will freeload whenever they can and try
| funding development as much as they can. We've seen that with
| BitKeeper and you had to "tighten up" the license a bit in order
| to make sure development stayed funded.
Here he seems to be suggesting that certain areas of OSS that can't get
enough funds should either relicense, drop the towel or do it for the
love as a part time job. I am suggesting they could all relicense to a
single license that leverages each others work and centrally manages
fundraising (but not fundspendings).
| In short, I believe the voluntary donations aren't needed in
| most areas people would donate to and won't make enough of an
| impact in the areas where they are needed.
Now he is poining out something that I had stated in my first message:
On Fri, Jul 26, 2002 at 01:09:32PM -0300, Federico Ferreres wrote:
+ The Kernel may well be nicely funded ...
+ Why post it here then? Because for it to work it must be supported
+ by at least some of the grand developements of OSS
... and has nothing to do with the ability to raise the funds. The
Kernel may not need the money but that it could help in a lot of areas
of OSS where a kickstart is needed (after which I am sure they will be
able to contribute funds back to the pool).
Regarding your opinion on the pyramid nature of the system well that can
be forgiven :)
The real facts that I am aware are:
- the kernel doesn't need the money
- the idea of _enforcing_ payment may not be an option or what is wanted
- there are doubt's about how many people would end up paying
- it will not be easy to manage the funds and could have side effects
- it may be nearly imposible to relicense the kernel due to the large
number of people involved even if they wanted
But the last fact invalidates all others:
"The idea is discarded, nobody likes it"
Which is fine! And I certainly am at the wrong place, poluting a
developement list and wasting your time. So I'll stop replying to your
flames (even though I find them funny and creative :-)
Regards!
Federico
Larry McVoy wrote:
>
>
> However, it doesn't work very well when the
>community beats you to hell for not being GPLed.
>
If you made your source code available so that it could be customized
then I would stop beating you for your license. As it is, any new
features have to get written by you, and I have requested a trivial to
author feature from you that I need and you haven't written despite your
promising to do so before I made the switch to bitkeeper. I don't at
all blame you for being busy, but if we had the source it would be 5
minutes for us to add the quite valuable to us feature.
Make your license charge for privacy but have the software come with
source code, and the license is brilliant. Otherwise, it is a source of
frustration, and yet another example of how software copyright is being
abused and twisted away from its original intent so as to permit
information to be hidden rather than published. In the long term, it is
the accumulation and compounding of knowledge that matters, not limited
time monopolization for 14 years (original length of copyrights), and
secret source code frustrates that.
--
Hans
Alexander Viro wrote:
>On 28 Jul 2002, Federico Ferreres wrote:
>
>
>
>>I stated a simple idea aimed at solving a real world issue. And you
>>haven't proved it wrong. It may not be what you or the kernel hackers
>>need/want (which is FINE). But it would solve ALL the funding problems
>>at least.
>>
>>
>
>You don't get it. So far the only guy who had been charitable was Larry, who
>felt that problem was real but had serious doubts about viability of your
>idea. I don't feel charitable and I've no reason to hesitate telling that
>you guys _are_ waste of time. No maybes about it. It's that simple...
>
>
>
>
Viro is abusive to everyone (by email, he is likable in person oddly
enough), usually without understanding what he is talking about at a
level of depth any deeper than it is new therefor wrong (see devfs
thread where he rejects devfs on the basis of endless details without
understanding that the basic idea had any merit).
Your idea has some merit in my opinion. I think that my version of it
that I presented at linuxworld some years ago, which is called an "Open
Sale", has some advantages.
"Open Sales" are the best method for reducing the economic
distortion caused by non-zero marginal pricing of
products with zero marginal cost without losing the incentive to
produce. An open sale is when a user group agrees to
pay X% of their hardware expenditures to all of those who provide
them with the right to use their software, with the
users performing the allocation to the providers based on the
users' usage and perception of quality. The users
determine the value of X, setting it in accordance with their
interest in attracting providers. It is open in the
sense that any software producer may add their software to the list
whose usage and quality will be
evaluated. Automation of usage monitoring and sampling techniques
are expected refinements of the basic idea. It is
based on a belief that economic distortion is proportional to the
ratio between marginal price and marginal cost, not
just the dollar total of the gap, and that increasing the price of
hardware by X% is less of a distortion than
increasing the cost of software by an infinite ratio over its
production cost. An open sale retains the consumer/user
driven decentralized economic allocation of traditional sales, and
in the case of government agencies and other large
purchasers, it enhances it. Payment of the X% is required to
legally use the software.
My approach has the advantage that the fee scales with hardware costs,
and that it is set by users.
However, you should understand that an idea is not enough, you must have
sufficient sociological mass to pull it off. Neither you nor I are in
that position at this time. Most people will be hostile to you if you
propose an idea that you lack the sociological position to effectuate.....
--
Hans
On Mon, 2002-07-29 at 04:39, Hans Reiser wrote:
> Your idea has some merit in my opinion. I think that my version of it
> that I presented at linuxworld some years ago, which is called an "Open
> Sale", has some advantages.
Your description of the problem is perfect. The solution, though, is
difficult to enforce as there's not a single incentive for OEMs to fund
anyone else that MS (reinstalling Windows).
> My approach has the advantage that the fee scales with hardware costs,
> and that it is set by users.
That's really an improvement over my idea (in the corporate area mostly.
In the home user area it doesn't matter, as he will probably use 1
computer at a time).
I can clearly see that a combination of both schemes would be needed,
because they are opposite sides of the same coin (your idea could be
though as an x% tax on hardware to fund developement, mine as a direct
payment from the users).
So one natural way of getting the best of both worlds would be:
- Hardware EOM could pay an x% per machine to the fGPL foundation, and
that would grant the buyer of that hardware a permanent license to all
fGPLd software. Users could request the OEM (IBM, etc) to offer them
that non-expiring (per machine) license instead of a preinstaled MS
Windows. This would mean OSS remains totaly free as in freedom and
_beer_.
- Hardware manufacturers that do NOT want to offer their customers that
option will be penalized because the user will have to pay the regular 1
year fGPL I have already described or get their hardware somewhere else.
This would create a great incentive for OEM to offer the fGPL license
instead of a preinstaled MS Windows.
- If desired, all past hardware sales could be granted a free
non-expiring "per-machine" fGPL license (so you'll see funds starting
low and growing every year in size) so that there can be NO complains
from anyone. Maybe the grant (for old hardware) could be limited to
individuals, non-profit organizations, goverments and educational
institutions.
- Manufacturers will be granted the right to claim to be GNU, Linux or
OSS ready ONLY if they actively offer the fGPL license as choice to
customers (instead of the preinstaled Windows). They could also be
required to include a "GNU/Linux ready" sticker in their hardware.
I believe this can be made to work in reasonable time, with minimun
effort and minimun hassle. There are problems, but they could be worked
out if there's support.
Everything else remains the same (as Hans and I have said, the users
will have a limited ability to chose what they need. I say limited
because for an Office application to work the core must remain funded
even though the user may not notice it).
> However, you should understand that an idea is not enough, you must have
> sufficient sociological mass to pull it off. Neither you nor I are in
> that position at this time. Most people will be hostile to you if you
> propose an idea that you lack the sociological position to effectuate.....
Makes sense. But if the idea is good, the ones that can make a
difference will support it, assuming they analized what is being
proposed.
> --
> Hans
>
I will (silently) put some resources on polishing the hard edges of
idea. It believe it may be worth it.
Federico
PS: I probably never reach the needed sociological position in OSS, so
at some point someone will have to help (if they believe it's a good
thing).
Most of you, people, didn`t ever thought about the fact that the money
scheme
doesn`t simply work at all. That`s it.
It`s being a _very_ fundamental issue. Look at the history:
Money were invented as a human work equivalent. What it
became now - this exercise is for the reader.
Let me prove my point (and by proving i mean pointing out so called worst
cases):
1. you are born in a poor family. you can work very hard, but have nothing.
2. you are employed as a miner. you indeed
work hard. look at a manager
doing brainfarting 24 x 7 x 365.
So the life shows that we cannot rely on the
money as a human work equivalent.
The worst thing is that money corrupt evrything they touch.
Thats why i`m doing my OSS developement for
free, and i don`t plan to change that.
Let the freedom be with you. And remember - freedom never comes outside of
you.
---
cheers,
Samium Gromoff
Hi,
On 26 Jul 2002, Federico Ferreres wrote:
> Idea 5: You are a student and as such are granted a free membership
> until you finish your studies.
...will certainly kick most of the payers.
Thunder
Federico Ferreres wrote:
>
>
>>My approach has the advantage that the fee scales with hardware costs,
>>and that it is set by users.
>>
>>
>
>That's really an improvement over my idea (in the corporate area mostly.
>In the home user area it doesn't matter, as he will probably use 1
>computer at a time).
>
>I can clearly see that a combination of both schemes would be needed,
>because they are opposite sides of the same coin (your idea could be
>though as an x% tax on hardware to fund developement, mine as a direct
>payment from the users).
>
Mine is a direct fee payment from users with allocation of fee
instructions accompanying the fee. The fee is less for buyers of cheap
computers since the hardware cost is low, so folks in Brazil on average
pay less than folks in Silicon Valley. The more you spend in hardware
the more you can afford to pay/contribute for software.
>
>Everything else remains the same (as Hans and I have said, the users
>will have a limited ability to chose what they need. I say limited
>because for an Office application to work the core must remain funded
>even though the user may not notice it).
>
Limited? What limit? If you pay the X% of hardware cost fee you can
use all the software in the pool.
>
>
>
>>However, you should understand that an idea is not enough, you must have
>>sufficient sociological mass to pull it off. Neither you nor I are in
>>that position at this time. Most people will be hostile to you if you
>>propose an idea that you lack the sociological position to effectuate.....
>>
>>
>
>Makes sense. But if the idea is good, the ones that can make a
>difference will support it, assuming they analized what is being
>proposed.
>
You are assuming they think. They don't. They sense herd movement.
This is wise of them, because unless a large portion of the herd moves
to it, it has no value. Thus it has no value. Thus the herd does not move.
--
Hans
On Mon, 2002-07-29 at 10:11, Hans Reiser wrote:
> Mine is a direct fee payment from users with allocation of fee
> instructions accompanying the fee.
Does it matter who collects the money? It should only be based on
convenience. And it's much easier/safer/efficient to have the OEM
collect the funds than any other party.
Imagine going to a store and asking "Two Sony Vaio laptops please, and
make them OSS, NOT Windows" and beign charged the computer's quoted
price (or maybe even less, but never more). Imagine not having to
research on how to get a refund for that Windows license that came
bundled and that you don't need. Imagine not having to _require_ the OEM
to preinstall any OSS on the machine for it to "OSS ready" (unless you
ask for redhat/suse/etc), yet beign able to use any OSS without having
to pay any extra dime.
There's this ilusion in normal people's minds that Windows is free
because it comes with the computer. We should take advantage of that
fact. And anyone that claims OSS is free is having an ilusion for OSS
costs money to develop and to be able to use OSS you need to spend a lot
of money (even if you don't pay anything back to the OSS developers).
Small note: funding OSS is not about the money. The money here is a
medium (funds) for an end (free software). Only businesses invert the
relation and put money as the objective objective and turn work into a
medium.
> >Everything else remains the same (as Hans and I have said, the users
> >will have a limited ability to chose what they need. I say limited
> >because for an Office application to work the core must remain funded
> >even though the user may not notice it).
> >
> Limited? What limit? If you pay the X% of hardware cost fee you can
> use all the software in the pool.
Here's a clear version of what I meant: Limited ability to "allocate the
money paid" based on what they'd like to be improved.
> You are assuming they think. They don't. They sense herd movement.
> This is wise of them, because unless a large portion of the herd
moves
> to it, it has no value. Thus it has no value. Thus the herd does not
move.
Whatever they decide is fine for me as it's their show and not mine. You
and Larry can think differently, as you are involved in developement.
> --
> Hans
>
Federico
On Mon, 2002-07-29 at 10:11, Hans Reiser wrote:
> Mine is a direct fee payment from users with allocation of fee
> instructions accompanying the fee.
Does it matter who collects the money? It should only be based on
convenience. And it's much easier/safer/efficient to have the OEM
collect the funds than any other party.
Imagine going to a store and asking "Two Sony Vaio laptops please, and
make them OSS, NOT Windows" and beign charged the computer's quoted
price (or maybe even less, but never more). Imagine not having to
research on how to get a refund for that Windows license that came
bundled and that you don't need. Imagine not having to _require_ the OEM
to preinstall any OSS on the machine for it to "OSS ready" (unless you
ask for redhat/suse/etc), yet beign able to use any OSS without having
to pay any extra dime.
There's this ilusion in normal people's minds that Windows is free
because it comes with the computer. We should take advantage of that
fact. And anyone that claims OSS is free is having an ilusion for OSS
costs money to develop and to be able to use OSS you need to spend a lot
of money (even if you don't pay anything back to the OSS developers).
Small note: funding OSS is not about the money. The money here is a
medium (funds) for an end (free software). Only businesses invert the
relation and put money as the objective objective and turn work into a
medium.
> >Everything else remains the same (as Hans and I have said, the users
> >will have a limited ability to chose what they need. I say limited
> >because for an Office application to work the core must remain funded
> >even though the user may not notice it).
> >
> Limited? What limit? If you pay the X% of hardware cost fee you can
> use all the software in the pool.
Here's a clear version of what I meant: Limited ability to "allocate the
money paid" based on what they'd like to be improved.
> You are assuming they think. They don't. They sense herd movement.
> This is wise of them, because unless a large portion of the herd
moves
> to it, it has no value. Thus it has no value. Thus the herd does not
move.
Whatever they decide is fine for me as it's their show and not mine. You
and Larry can think differently, as you are involved in developement.
> --
> Hans
>
Federico
"Samium Gromoff" <[email protected]> writes:
> Most of you, people, didn`t ever thought about the fact that the money
> scheme
> doesn`t simply work at all. That`s it.
>
> It`s being a _very_ fundamental issue. Look at the history:
> Money were invented as a human work equivalent. What it
> became now - this exercise is for the reader.
Money is not a human work equivalent. Money is a commodity that
everyone will barter with. Making bartering more efficient.
With pure barter you have to do something like:
work -> stuff1 -> stuff2 ->stuff3 -> stuff you want.
With money this usually becomes
work -> money -> stuff you want.
> So the life shows that we cannot rely on the
> money as a human work equivalent.
This makes the assumption that all work is equally valuable. But you
already made the assertion that a manager sitting around is not as
a miner.
Eric
Hans Reiser <[email protected]> writes:
> Alexander Viro wrote:
>
> >On 28 Jul 2002, Federico Ferreres wrote:
> >
> >
> >>I stated a simple idea aimed at solving a real world issue. And you
> >>haven't proved it wrong. It may not be what you or the kernel hackers
> >>need/want (which is FINE). But it would solve ALL the funding problems
> >>at least.
> >>
> >
> >You don't get it. So far the only guy who had been charitable was Larry, who
> >felt that problem was real but had serious doubts about viability of your
> >idea. I don't feel charitable and I've no reason to hesitate telling that
> >you guys _are_ waste of time. No maybes about it. It's that simple...
> >
> >
> >
> Viro is abusive to everyone (by email, he is likable in person oddly enough),
> usually without understanding what he is talking about at a level of depth any
> deeper than it is new therefor wrong (see devfs thread where he rejects devfs on
>
> the basis of endless details without understanding that the basic idea had any
> merit).
There are a couple of interesting points raised in this thread.
1) That a substantial part of the work of software is not building it, but
is maintaining it.
2) That several people feel that it is hard to make a business plan with
open source software.
3) Many successful software companies, have made money with a hardware tax.
And open source could feasibly get part of that action, all you have to do
is to demand that Linux be pre-installed. And the distributions should
get involved to help the hardware manufacturers.
4) Raising money is conceptually very simple. But practically hard.
5) How society perceives the solution is a very important part of any solution.
A lot of economics is modeling and finding a system where the greedy
algorithm, can be used to optimize the system. As all decisions are local
with the greedy algorithm it scales very well. Most forms of central
planning, distribution whatever fall down because they are not built
on algorithms that scale.
The basic economic model of supply and demand with competition works
fairly well. But it has problems when you have either an external
cost (Pollution) or an external benefit (Free Software). With some
small amount of government regulation it is theoretically possible to
introduce these external costs, into the direct costs normally dealt
with. The best scheme I have seen proposed is to sell permits to
pollute, with the maximum number of permits limited by the maximum
amount of pollution you wish to allow. It might be possible adapt
this to free-software by allowing having permits to sell non-free
software. But that requires a large shift in how this are
accomplished.
But the assertion made by Al Viro that software maintenance is where
the bulk of the work is, is interesting. Long term this is trivially
true because the all of the code has been written, and there is no new
development to do. Services like distributions and device driver
writers, and kernel maintainers appear to be in the area where
maintenance is important. Maintenance can be handled by
maintenance/support contracts, making the economic model with closed
source and open source the same, except with open source it is easier
for multiple maintainers to cooperate. And in the areas where the
work is primarily maintenance is where open source has been observed
to be well funded so this appears to work in practice.
Given that software maintenance is the primary problem, it is only
the creators of innovative open source programs whose costs are
external to the economic model, that making business plans harder to
deal with. So the question becomes how in the open source community
do we encourage true innovation, while not encouraging it so much we
fail to weed out the dumb ideas. Innovation always has a large share
of external benefit so the problem of how to encourage and compensate
innovators is not new, but the open source landscape is.
The only idea that spring to my mind to encourage true innovation are
obligating the distributors to pay the innovators something for the
programs they distribute. Or having universities and other research
institutions pay the salaries of the researchers. Off the top of my
head I cannot think of anything that takes advantage of the potential
to bypass an old guard of maintainers, and universities and go
directly to the consumer.
Eric
Eric W. Biederman wrote:
>But the assertion made by Al Viro that software maintenance is where
>the bulk of the work is, is interesting.
>
Another large part of the work is in pursuing innovations in software
that fail. Note that failed software is generally not maintained.
What I don't understand about Viro's argument is why he thinks that
maintainers won't also get paid under the schemes discussed. There is
nothing that prevents paying for maintenance in either scheme.
--
Hans
>> Most of you, people, didn`t ever thought about the fact that the money
>> scheme
>> doesn`t simply work at all. That`s it.
>>
>> It`s being a _very_ fundamental issue. Look at the history:
>> Money were invented as a human work equivalent. What it
>> became now - this exercise is for the reader.
>
> Money is not a human work equivalent. Money is a commodity that
>everyone will barter with. Making bartering more efficient.
>
> With pure barter you have to do something like:
>work -> stuff1 -> stuff2 ->stuff3 -> stuff you want.
>With money this usually becomes
work -> money -> stuff you want.
>> So the life shows that we cannot rely on the
>> money as a human work equivalent.
>This makes the assumption that all work is equally valuable. But you
>already made the assertion that a manager sitting around is not as
>a miner.
I`m sorry but you should forget that crap.
Humans are born for cooperation, otherwise
they seem not to survive.
And the said "work valuability" doesn`t seem to be a mental excuse
for the extreme success of selected individuals in society, does it?
We don`t want to reinvent the broken wheel, because all known
problems will arise then. Again.
>Eric
---
cheers,
Samium Gromoff
On Mon, 2002-07-29 at 22:47, Eric W. Biederman wrote:
> But the assertion made by Al Viro that software maintenance is where
> the bulk of the work is, is interesting. Long term this is trivially
> true because the all of the code has been written, and there is no new
> development to do. Services like distributions and device driver
> writers, and kernel maintainers appear to be in the area where
> maintenance is important. Maintenance can be handled by
> maintenance/support contracts, making the economic model with closed
> source and open source the same, except with open source it is easier
> for multiple maintainers to cooperate. And in the areas where the
> work is primarily maintenance is where open source has been observed
> to be well funded so this appears to work in practice.
>
> Given that software maintenance is the primary problem, it is only
> the creators of innovative open source programs whose costs are
> external to the economic model, that making business plans harder to
> deal with. So the question becomes how in the open source community
> do we encourage true innovation, while not encouraging it so much we
> fail to weed out the dumb ideas. Innovation always has a large share
> of external benefit so the problem of how to encourage and compensate
> innovators is not new, but the open source landscape is.
OK. How about this?
Let us observe that the fundemtal 'problem' here is that closed source
software companies get funding by levraging it's control on the
inelectual property it generates to get money - 'pay us or live with out
it'.
The challange is therefore to find a way, a 'business plan' if you will,
to get money *without* execersizing control on generated intellectual
property. To do this, let's note that there are already equivelent
fields of business with some very close charateristics to software (that
is they involve brain work, not digging coal) where for various reasons,
control of intelectual property is not how you get paid.
Let's take a specific example: lawyers cannot control the intelectual
property they generate because it's part of the law system that once a
lawyer thought up a new idea anyone can use it. In fact, AFAIK
precedence is one of the basic ideas of the modern law systems.
Are lawyers out of a job? Are there less and less lawyers around because
they can't make money? Are they perhaps not innovative? I don't think
so. As a matter of fact some would even argue that they maybe too much
lawyers around but let's not go there...
Ok. So how do they do this? simple, the way they organise their business
is suitable for the job: AFAIK, a typical law firm has a couple of
senior lawyers who are the 'partners'. The rest of the staff, including
the less senior lawyers, are hired help.
A couple of interesting to note:
1. The top brass are lawyers, a lot of the time practicing lawyers, not
'managment'.
2. No sales, marketing etc. etc..
Where these are not true is usually in the really big (and therefore we
assume succfull) companies. That is it begins to maybe become slightly
different only AFTER you succeed.
To paraphrase Robert Heinlin, in a typical law firm senior managment
'everyone drops'.
And that's it. (I think :-) the problem with current Open Source
operations which don't do so well is that they are trying to build a
company structured around how closed source software companies are built
- senior manamenbt that don't program. CEOs from that other (closed
source!) company that worked. Peopole who do PR. Sales people. People
who do not 'drop'. Can you say 'Overhead'?
In short and plain words, what I think the lawyers know and we need to
learn is that to make a business built on something else then
intelectual property control, you need low overhead. You need a firm
where (almost) everyone drops. Building bloated companies only works if
you grab someone in the balls (the customer) and make him maintain the
overhead. When you can't grab someone in the balls you've got to be mean
and lean. When you're big enough then you can efford some of the bloat
and by then maybe you even need it. Maybe.
Imagine for a second a software company built according to 'law firm
principles' - we have the 'Senior Hackers'. They all write code in
addition to taking care of the other sides of business. Sure, they might
hire an accountant and other employees to take some of the work etc. but
there is no bloated 'managment' that can't hack. Intresting concept if
nothing else...
I'm sure by know you are all smiling and thinking "OK, I want to have
some of what he's been smoking...". Fine, but the lawyers are doing
something like this for a long time. It works somehow. Maybe we can
learn.
Just my 2EUROs,
Gilad.
--
Gilad Ben-Yossef <[email protected]>
"You got an EMP device in the server room? That is so cool."
-- from a hackers-il thread on paranoia
On 29 Jul 2002, Eric W. Biederman wrote:
> But the assertion made by Al Viro that software maintenance is where
> the bulk of the work is, is interesting. Long term this is trivially
> true because the all of the code has been written, and there is no new
> development to do. Services like distributions and device driver
> writers, and kernel maintainers appear to be in the area where
> maintenance is important. Maintenance can be handled by
> maintenance/support contracts, making the economic model with closed
> source and open source the same, except with open source it is easier
> for multiple maintainers to cooperate. And in the areas where the
> work is primarily maintenance is where open source has been observed
> to be well funded so this appears to work in practice.
That's _not_ the assertion I've made. It's not about innovation vs.
maintenance and frankly, I'm insulted by that interpretation (which
is quite an achievement - I've got really thick skin).
Actual split is not into innovation and maintenance - it's into
new visible features and everything else. Which happens to include
* writing maintainable code.
* designing sane interfaces.
* writing code in a way that allows audits.
* _doing_ audits of your code.
* getting already existing code into form that allows aforementioned
new features - and if they are really new it's 99% of the work; the only
way to avoid it is by creating, er, private copies of existing subsystems
and kludging them up into something you can use. Effects of such strategy
are left as an exercise to readers. Or, if your new features happen to
be similar enough to something not new you get to reuse the work already
done back when those features were implemented.
* etc., etc., including the things traditionally considered as
maintenance.
All mechanisms I've seen mentioned in that thread encourage only one thing -
new features as fast as possible at whatever cost for future work. Calling
that innovation and relegating everything else to support contracts... I'd
beg to differ.
THAT is the problem - mechanism that gives incentive to being quick, dirty and
sloppy (and encourages featuritis, while we are at it - inventing features
that are not new by any stretch of imagination and selling them as great
innovations is the oldest trick in the book). Folks, it's not a good thing.
There's already more than enough pressure in that direction and it had
already given us dungpiles of unsupportable, exploit-ridden software.
Commercial and free alike. And guess what? After a while you _can't_
do anything really new on a platform, since it's already choke-full of kludges
from "quick innovators" - to the point when touching it in any non-trivial way
breaks something and nobody actually understands the thing enough to fix
the breakage, so more and more kludges are layered. Sheesh...
Gilad Ben-Yossef <[email protected]>:
> The challange is therefore to find a way, a 'business plan' if you will,
> to get money *without* execersizing control on generated intellectual
> property. To do this, let's note that there are already equivelent
> fields of business with some very close charateristics to software (that
> is they involve brain work, not digging coal) where for various reasons,
> control of intelectual property is not how you get paid.
>
> Let's take a specific example: lawyers cannot control the intelectual
> property they generate because it's part of the law system that once a
> lawyer thought up a new idea anyone can use it. In fact, AFAIK
> precedence is one of the basic ideas of the modern law systems.
Ummm. nope. Wrong (or at least very incomplete).
Lawyers CAN own the new idea - they can copyright their presentations
preventing others from using them without first paying for the right
to use all/part of the presentation in other cases.
Lawyers are closer to high profile consultants.
> Are lawyers out of a job? Are there less and less lawyers around because
> they can't make money? Are they perhaps not innovative? I don't think
> so. As a matter of fact some would even argue that they maybe too much
> lawyers around but let's not go there...
The same can be said for consultants :-)
> Ok. So how do they do this? simple, the way they organise their business
> is suitable for the job: AFAIK, a typical law firm has a couple of
> senior lawyers who are the 'partners'. The rest of the staff, including
> the less senior lawyers, are hired help.
>
> A couple of interesting to note:
>
> 1. The top brass are lawyers, a lot of the time practicing lawyers, not
> 'managment'.
Actually, they are the management. These practicing lawyers only handle the
most exclusive clients (and for the most $$$$ possible). All others contribute
a cut to the firm (and charge less $$$$ to the customer).
Thats the same way a consulting company works.
> 2. No sales, marketing etc. etc..
>
> Where these are not true is usually in the really big (and therefore we
> assume succfull) companies. That is it begins to maybe become slightly
> different only AFTER you succeed.
>
> To paraphrase Robert Heinlin, in a typical law firm senior managment
> 'everyone drops'.
>
> And that's it. (I think :-) the problem with current Open Source
> operations which don't do so well is that they are trying to build a
> company structured around how closed source software companies are built
> - senior manamenbt that don't program. CEOs from that other (closed
> source!) company that worked. Peopole who do PR. Sales people. People
> who do not 'drop'. Can you say 'Overhead'?
>
...
> Imagine for a second a software company built according to 'law firm
> principles' - we have the 'Senior Hackers'. They all write code in
> addition to taking care of the other sides of business. Sure, they might
> hire an accountant and other employees to take some of the work etc. but
> there is no bloated 'managment' that can't hack. Intresting concept if
> nothing else...
And who is charged for the time?
Who gets the direct benifit of the work?
You must have a customer/client that is identified for all billing.
Personally, I think Red Hat is doing the best job of this:
1. provide consulting to clients paying for help.
2. providing custom software to clients paying for it.
3. providing short debug/fix/patches cycles to clients/customers paying for it.
4. advertising their capabilities to attract more (paying) customers/clients.
5. funding developement out of their charges to customers.
6. funding debug/fix/patch cycles for applications that do not have paying
clients/customers. Probably charging this to "overhead" or
"marketing/advertisment" to expand their list of capabilities.
7. Being good enought to pass those developements/fix/patches to the rest of
the world via GPL/LGPL licenses.
IBM is doing a similar thing, though I suspect they count the expense of the
open source development/debug people into the charge passed to the customer
purchases of their hardware (ie. bundled into hardware cost) or
overhead/marketing overhead, or in consulting fees.
Not being an employee of either RH or IBM, the above statements are, of,
course, my opinion of how they are looking at things.
The other side of coin is "what do the customers get"
1. someone to call when they have problems
2. customized software aimed at their environment
3. a lower expense than the M$ tax
4. a better (ie. more reliable/secure) product than M$ can produce
What does the world get out of it:
1. an alternative to the M$ tax
2. a better (ie. more reliable/secure) product than M$ can produce
3. the ability to modify the software, learn how things work, limitations
of the software
4. a better personal knowlege environment allowing the potential to be better
employed by other busineses.
5. an ever expanding knowlege of how/why/what software works
-------------------------------------------------------------------------
Jesse I Pollard, II
Email: [email protected]
Any opinions expressed are solely my own.
Viro, do I understand you right that you feel that what is visible to
users will get more funding from users, and that because there will be
far more suppliers the users will be more inclined to not pay for the
unseen benefits than when buying cars?
My answer to this is that users should be given a sample of, say, five
providers, each of whom gets to make their case for what they have done
for the user.
We seek not perfection, but improvement.
Hans
PS
Some of you may be inclined to suggest that such a sampling scheme would
tend to benefit the small contributors..... and to this I say that more
sophisticated mathematical models which reduce the variance between the
importance of those in each sample so as to increase user precision of
assessment can come later after the basic idea is started up.
On Tue, 30 Jul 2002, Hans Reiser wrote:
> Viro, do I understand you right that you feel that what is visible to
> users will get more funding from users, and that because there will be
> far more suppliers the users will be more inclined to not pay for the
> unseen benefits than when buying cars?
>
> My answer to this is that users should be given a sample of, say, five
> providers, each of whom gets to make their case for what they have done
> for the user.
>
Does the average user know about code clarity and maintainability? I
think I see his point. I've seen what happens when non coders make
decisions as to what features need to go in. Don't think it happens in
the open source world? Go look at what most irc nets use for code.
It's why I actively avoid programming for a living.
Gerhard
--
Gerhard Mack
[email protected]
<>< As a computer I find your faith in technology amusing.
On Tue, 30 Jul 2002, Gerhard Mack wrote:
> Does the average user know about code clarity and maintainability? I
> think I see his point. I've seen what happens when non coders make
> decisions as to what features need to go in. Don't think it happens in
That's part of the problem, indeed, but nowhere near the entire picture.
> the open source world? Go look at what most irc nets use for code.
Or, easier yet, go to prep.ai.mit.edu and try to RTFS. Use of safety
barfbags is mandatory.
There is nothing in the funding mechanism that causes features to go in,
merely to cause features to get paid for.
Do you understand that capitalism has this problem in general, that when
consumers/users decide who gets money they make a lot of mistakes. The
only thing we know is that when the experts of production decide things
instead the results are usually worse (communist systems have things
determined by producers not consumers, though it must be said that there
are some things that communists designed better than the west).
Not seeking perfection, just improvement....
Hans
Alexander Viro wrote:
>On Tue, 30 Jul 2002, Gerhard Mack wrote:
>
>
>
>>Does the average user know about code clarity and maintainability? I
>>think I see his point. I've seen what happens when non coders make
>>decisions as to what features need to go in. Don't think it happens in
>>
>>
>
>That's part of the problem, indeed, but nowhere near the entire picture.
>
>
>
>>the open source world? Go look at what most irc nets use for code.
>>
>>
>
>Or, easier yet, go to prep.ai.mit.edu and try to RTFS. Use of safety
>barfbags is mandatory.
>
>
>
>
>
--
Hans
On Tue, 30 Jul 2002, Hans Reiser wrote:
> There is nothing in the funding mechanism that causes features to go in,
> merely to cause features to get paid for.
>
> Do you understand that capitalism has this problem in general, that when
> consumers/users decide who gets money they make a lot of mistakes. The
> only thing we know is that when the experts of production decide things
> instead the results are usually worse (communist systems have things
> determined by producers not consumers, though it must be said that there
> are some things that communists designed better than the west).
A little research will show that the governments have little to
do with design. Governments just tend to determine what will
be designed. Most designers, Engineers, Architects, etc., design
in spite of governments influence, not because of it. Both governments
and consumers generate markets for products. When consumers make
mistakes, people lose their jobs. When governments make mistakes people
lose their lives. Government is both necessary and dangerous, like oxygen.
Cheers,
Dick Johnson
Penguin : Linux version 2.4.18 on an i686 machine (797.90 BogoMips).
The US military has given us many words, FUBAR, SNAFU, now ENRON.
Yes, top management were graduates of West Point and Annapolis.
On Fri, 26 Jul 2002, Alexander Viro wrote:
> What you and the rest of armchair generals do not get is that "adding
> features" is _not_ the hard part of work. Doing that in a way that
> wouldn't be a permanent source of bugs afterwards and cleaning up the
> existing sources of bugs _IS_. So is doing infrastructure work. So
> is auditing code. So is removing crap code.
>
> None of that is covered by your "model". 99:1 that your "working driver
> for card" is going to contain a bunch of root holes. _And_ be unmaintainable.
This is totally unrelated to the ecconomic model, we have many proofs that
code quality is unrelated to financial compensation. People write crap
code for both fun and profit. So what you say is totally true, but has
zero to do with why the author wrote the code. You have to QA any code
before using it, why the developer wrote it is irrelevant.
That said, let me suggest another possible model for funding free
software. It's in two parts. It would be helpful perhaps to discuss "is
this a good thing to do" first, before "how could you do that," and I do
know some similar things have been proposed and tried in the past.
First the user driven part:
1. User wants functionality X. Defines a functional spec. Also defines
the open source license to be used for the finished work. GPL, LGPL, BSD
or similar.
2. Concept approval. If the functionality is a library change, driver,
kernel hack, or similar, the entity in charge commits to accept the
functionality *if and only if* it meets some standard of fitness,
reliability, and maintainability. Otherwise some neutral party (user
group, FSF, individual) is selected as the acceptor.
3. Bidding. At that point the proposal goes up {somewhere} with a price (I
will pay $Y for functionality X). Developers may offer to do the work, and
most importantly other users may add to the offer. Not a fixed $20, but
whatever it's worth to them. After some time either an acceptable
developer is found or the offer expires.
4. Acceptance. Developer tests the software, publishes the result. The
acceptor tests the software and either accepts it or rejects it for a
objective reason related to not meeting the functional specs.
5. Deployment. Developer gets paid, software released under open source
license, if it's part of a package it's queued for the next release.
Developer driven part (only changes from user shown):
1. Definition. Developer states s/he wants to develop X, states functional
spec, license, and price of the work.
3. Bidding. If the stated desired price is not met after some time, other
developers may accept the current pledge (with user approval, of course).
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
On Wed, 31 Jul 2002 16:03:59 -0400 (EDT), Bill Davidsen wrote:
>This is totally unrelated to the ecconomic model, we have many proofs that
>code quality is unrelated to financial compensation. People write crap
>code for both fun and profit. So what you say is totally true, but has
>zero to do with why the author wrote the code. You have to QA any code
>before using it, why the developer wrote it is irrelevant.
No matter how many proofs you have or how good they are, I won't believe it
because this fails the giggle test. Here's a simple counter-proof. I want to
write an SQL server from scratch. I create two teams, one with $50,000 and
one with $5,000,000. You can honestly tell me that it's equally like that
either team will produce a higher quality SQL server?
This reminds me of the proofs that supposedly showed that locking up
convicted criminals for longer didn't lower the crime rate. Are we honestly
supposed to believe that otherwise honest people commit more crimes to make
up the difference?
DS
On Thu, 2002-08-01 at 10:32, David Schwartz wrote:
> No matter how many proofs you have or how good they are, I won't believe it
> because this fails the giggle test. Here's a simple counter-proof. I want to
> write an SQL server from scratch. I create two teams, one with $50,000 and
> one with $5,000,000. You can honestly tell me that it's equally like that
> either team will produce a higher quality SQL server?
Intuition is misleading. It mostly depends which of the teams has the
good engineers and smart management. The .com bust is just one example
of that.
> This reminds me of the proofs that supposedly showed that locking up
> convicted criminals for longer didn't lower the crime rate. Are we honestly
> supposed to believe that otherwise honest people commit more crimes to make
> up the difference?
In some cases yes. There are a whole variety of well understood reasons
why this occurs - probability of capture not length of capture is the
deterrent. Most criminals won't reoffend after a short or long sentence
(and there is little evidence that length of sentence decreases the
probability of reoffence. In addition there is a category of crime (that
involving a dispute between two parties) where one person being sent to
jail causes their entire family and relations to become 'criminals'.
I don't however see the relationship between the two, other than both
being demonstrations that you must do the actual science and statistics
before you rely on intuition.
On 1 Aug 2002, Alan Cox wrote:
> On Thu, 2002-08-01 at 10:32, David Schwartz wrote:
> > No matter how many proofs you have or how good they are, I won't believe it
> > because this fails the giggle test. Here's a simple counter-proof. I want to
> > write an SQL server from scratch. I create two teams, one with $50,000 and
> > one with $5,000,000. You can honestly tell me that it's equally like that
> > either team will produce a higher quality SQL server?
>
> Intuition is misleading. It mostly depends which of the teams has the
> good engineers and smart management. The .com bust is just one example
> of that.
I don't get it.. he should know full well there are programmres out
there who will not provide good code no mater how much you pay them.
In fact, I can name at least one example from the past employees of
the company providing his email address.
Also see "Brook's Law" from here:
http://info.astrian.net/jargon/terms/b/Brooks_s_Law.html
Gerhard
--
Gerhard Mack
[email protected]
<>< As a computer I find your faith in technology amusing.
On Thu, 1 Aug 2002, David Schwartz wrote:
>
>
> On Wed, 31 Jul 2002 16:03:59 -0400 (EDT), Bill Davidsen wrote:
>
> >This is totally unrelated to the ecconomic model, we have many proofs that
> >code quality is unrelated to financial compensation. People write crap
> >code for both fun and profit. So what you say is totally true, but has
> >zero to do with why the author wrote the code. You have to QA any code
> >before using it, why the developer wrote it is irrelevant.
>
> No matter how many proofs you have or how good they are, I won't believe it
> because this fails the giggle test. Here's a simple counter-proof. I want to
> write an SQL server from scratch. I create two teams, one with $50,000 and
> one with $5,000,000. You can honestly tell me that it's equally like that
> either team will produce a higher quality SQL server?
First, we were talking about written for free vs. written to make money.
Second, the quality of the output depends on the quality of the process,
not how much you pay for it. Equally likely isn't what I said, either.
One last time: commercial software is not a guaranty of quality nor is
being free an indication of being shoddy. Clearly if you underpay people
for any work you are likely to get poor work, but that doesn't apply to
someone who is being paid in satisfaction and recognition, and who has a
real motivation to do it to the best of her/his ability.
> This reminds me of the proofs that supposedly showed that locking up
> convicted criminals for longer didn't lower the crime rate. Are we honestly
> supposed to believe that otherwise honest people commit more crimes to make
> up the difference?
Glad it reminds you, I sure as hell don't see the point... and I never saw
any such thing. Studies show that locking people up longer doesn't make
*that person* less likely to commit a crime, which is not at all the same
thing as the crime rate in crimes per unit time by all persons.
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
First:
I have received a rather harsh complaint email that told
me that it was considered to be RUDE to post to Subject:
"Funding GPL projects..(etc)" in context of LKML.
I agree in that Subject Sentence is rather [OT], but it
seems like SUBJECT is of interest to air for many LK
developers (and others). Therefore I propose the above change
in subject line to narrow the subject down to specifically
mean kernel related development projects. I hope that no one
is offended by this remark of mine. Most postings so far in
subject does in practise touch funding of kernel related
development in one way or the other.
Bill Davidsen wrote:
> On Fri, 26 Jul 2002, Alexander Viro wrote:
>>
[ snipped ]
>
> That said, let me suggest another possible model for funding free
> software. It's in two parts. It would be helpful perhaps to discuss "is
> this a good thing to do" first, before "how could you do that," and I do
> know some similar things have been proposed and tried in the past.
>
> First the user driven part:
>
> 1. User wants functionality X. Defines a functional spec. Also defines
> the open source license to be used for the finished work. GPL, LGPL, BSD
> or similar.
>
> 2. Concept approval. If the functionality is a library change, driver,
> kernel hack, or similar, the entity in charge commits to accept the
> functionality *if and only if* it meets some standard of fitness,
> reliability, and maintainability. Otherwise some neutral party (user
> group, FSF, individual) is selected as the acceptor.
>
> 3. Bidding. At that point the proposal goes up {somewhere} with a price (I
> will pay $Y for functionality X). Developers may offer to do the work, and
> most importantly other users may add to the offer. Not a fixed $20, but
> whatever it's worth to them. After some time either an acceptable
> developer is found or the offer expires.
>
> 4. Acceptance. Developer tests the software, publishes the result. The
> acceptor tests the software and either accepts it or rejects it for a
> objective reason related to not meeting the functional specs.
>
> 5. Deployment. Developer gets paid, software released under open source
> license, if it's part of a package it's queued for the next release.
>
> Developer driven part (only changes from user shown):
>
> 1. Definition. Developer states s/he wants to develop X, states functional
> spec, license, and price of the work.
>
> 3. Bidding. If the stated desired price is not met after some time, other
> developers may accept the current pledge (with user approval, of course).
> --
> bill davidsen <[email protected]>
> CTO, TMR Associates, Inc
> Doing interesting things with little computers since 1979.
I believe that a lot of kernel development funding is
already ruled somewhat (unofficially of course ) by similar
structures as above. Only: The User in point 1. is a larger
company such as f ex IBM. By making the scheeme above available
to anybody would probably be a nice improvement. Some question
that remains to be solved is however:
A. Who will want to make the nescessary work that is
involved to achieve this? I do believe that most of
the Linux kernel development team are a bit too buzy
making linux patches, while working on ordinary jobs
on the side.
B. How should the work be organized in order to achieve
project credibility and fair progress ?
There are numerous court examples of funding abuse.
C. The scheeme above might actually turn out to "need" funding
for it self. There is even a big "riscue" that it will it
self eat up all the funding that it receives. ( Old bad own
experiences. )
Kind regards
/Daniel Mose
P.S.
If any body actually find my posting in subject above
to be rude I would of course be happy to know this by mail.
Being rude is certainly NOT my intention. Of course any
kind of feedback besides this is always welcome
D.S.
Hi,
On 29 Jul 2002, Federico Ferreres wrote:
> There's this ilusion in normal people's minds that Windows is free
> because it comes with the computer.
Not to mention all the people who think that Windows is the computer,
indivisibly, and never heard of the term "Operating System". Oh, well...
Thunder
--
.-../../-./..-/-..- .-./..-/.-.././.../.-.-.-
> seems like SUBJECT is of interest to air for many LK
> developers (and others). Therefore I propose the above change
egads! the price of cheese is of interest to many LKML readers,
but that DOES NOT MAKE IT a legit topic. please, please direct
any followups to alt.linux.kernel.cheese.
On Thu, 1 Aug 2002 12:51:16 -0400 (EDT), Gerhard Mack wrote:
>I don't get it.. he should know full well there are programmres out
>there who will not provide good code no mater how much you pay them.
Sure, and there are people who can't write a symphony no matter how much you
pay them. But, believe it or not, the people who can write symphonies are
more likely to get paid to write them than people who aren't. Or, in other
words, there will tend to be a strong correlation between how much a person
is paid to write symphonies and how well he or she writes them. This is
because in general the people who control the flow of money tend to pay the
people who write the symphonies that meet their needs.
Now if you do a study where you randomly choose 100 people and pay them an
extra $500 to write symphonies, you are very unlikely to get improved
symphonies. But this hardly shows that more money won't make better
symphonies.
For an individual given programmer, in most cases code quality won't
correspond to financial compensation. You can't pay a randomly selected
person more and more money and find them writing better and better
symphonies. However, if I had to get a symphony written by any means at my
disposal, the more money I had, the better the symphony would be.
DS
>First, we were talking about written for free vs. written to make money.
>Second, the quality of the output depends on the quality of the process,
>not how much you pay for it. Equally likely isn't what I said, either.
I give one person $2,000 and one person $50,000 to buy a car. Would you
argue that they are equally likely to come back with quality cars?
>One last time: commercial software is not a guaranty of quality nor is
>being free an indication of being shoddy.
Now you're changing the argument on me. First we were talking about code
quailty and financial compensation. Now we're talking about code quality and
cost. The two are really not related. A person can produce a product that is
free either with or without being paid.
>Clearly if you underpay people
>for any work you are likely to get poor work, but that doesn't apply to
>someone who is being paid in satisfaction and recognition, and who has a
>real motivation to do it to the best of her/his ability.
Yes, but who will that person be and how many of them will there be? That
will depend upon how much money is available. Some people will do wonderful
work for free, and if you have no money, those are the only people you can
use and you get as much time as they can spare or afford to give at best. If
you have money, you can still use those people, but you can also use people
who need money.
>> This reminds me of the proofs that supposedly showed that locking up
>>convicted criminals for longer didn't lower the crime rate. Are we honestly
>>supposed to believe that otherwise honest people commit more crimes to make
>>up the difference?
>Glad it reminds you, I sure as hell don't see the point... and I never saw
>any such thing. Studies show that locking people up longer doesn't make
>*that person* less likely to commit a crime, which is not at all the same
>thing as the crime rate in crimes per unit time by all persons.
This is exactly the point. A given programmer may not create better code
with greater compensation. However, a programming project can create better
code with greater compensation. (Assuming we don't randomly pick projects and
dump money on them, of course. Assuming the money is at least placed by the
people who went to the trouble of earning it choosing where they think it
will do the most good.)
DS
On Thu, 1 Aug 2002, David Schwartz wrote:
>
> >First, we were talking about written for free vs. written to make money.
> >Second, the quality of the output depends on the quality of the process,
> >not how much you pay for it. Equally likely isn't what I said, either.
>
> I give one person $2,000 and one person $50,000 to buy a car. Would you
> argue that they are equally likely to come back with quality cars?
If you send both to area famous for seedy dealers? Yeah - both will
come back with painted pieces of shit, unless they'll have uncommon
luck to find something better. And that's a *BIG* luck.
Face it - in a lot of categories the market of software is market
of snake oil. When all alternatives on the market are utter crap and that
situation hadn't been changed in decades... And I mean all - be they
commercial, free, for Linux, Solaris, Windows, VMS, HP-UX, yaddda, yadda;
it doesn't matter.
It's nice to pretend that this is not the case. Just don't forget
to pray when you view a PDF or PostScript document from some site. Or
listen to mp3. Or view a picture in one of $BIGNUM formats.
And don't give me that crap about Sturgeon's Law - for many categories
we are talking about 100%, not 90%. Sheesh...
>I don't however see the relationship between the two, other than both
>being demonstrations that you must do the actual science and statistics
>before you rely on intuition.
They're both example of how the properties of individuals differ from the
properties of groups of those same individuals, and I believe the same error
is being made in both cases.
In the programming case, a programmer working on a project will not be
likely to produce better code simply because his salary goes up. Generally,
people will produce the best code they are capable of producing. Similarly,
longer lengths of incarceration won't decrease the probability that a
particular person will re-offend. In both cases, there just isn't a cause and
effect relationship with the individual.
However, this does not mean that a group will behave the same way. For
example, if we lock everybody who commits a violent crime up for twice as
long, crime in the group as a whole will go down simply because repeat
offenders will spend less time out of prison.
Similarly for programmers. With more money, you can employ more and/or
better programmers. You can still employ cheaper programmers, it doesn't stop
you.
How many people currently working on the Linux kernel would devote more time
to it if they received regular anonymous donations? How many new talented
programmers who can't find employment (or don't want/need to) would take more
interest in working on Linux?
I'm not saying this guarantees better code. For example, it's possible that
the greater volume of code produced might overload choke points in the
development that can't be expanded due to key individuals already working at
their limits. However, money brings new options.
As you (I think?) pointed out, most free software is crap and most
commercial software is crap. The hard part is to find the good software and
the good people and then give them an incentive to produce the code that you
really want. Money is a tool for doing that.
DS
I'd just like to add that I respect Larry's problems regarding:
* small number of seats means large number of dollars per seat is
necessary to justify the business investment
* people modifying the source to turn off open logging and thus
defeating his original business model (this would really piss me off also)
He is also an all around nice guy who has given us some nice tools, and
tries hard to be generous while keeping his business alive.
I would like to see him try making his license such that if open logging
is not used, and he is not paid a license fee, then the user has agreed
that the software is GPL'd and any potential user may sue to claim that
software. I think this would scare managers into paying their fees, and
I think it would work, but I also understand and respect that it is not
my business that would go under if I was wrong, and it is not I who
would have to explain to his workers why payroll couldn't be met.
--
Hans
I have already donated $30 last year (but to the EFF as I think freedom
is also important). But how many individuals donate though? I'd say 99%
of the people just free ride. That means resources are very scarce and
not enough OSS is produced.
But something is wrong with this. If I donate then it's no longer free.
And only if you never donate then you can call it free. Efectively, that
means that to get to a certain level of funding donators have pay 99
times the amount that would otherwise be needed if everyone contributed.
This reminds me of my countrie's idea of "social justice": the rich go
to the tax funded, free universities while the poor grow corn and work
18 hours, 7 days a week and also pay the taxes (40%). Of course these
poor can't pay for housing and food while studing. But every time
someone suggests charging a small fee to univ. students (offering state
loans for the ones that can't really pay that would cover food, housing
and the fee) to partially fund those universities, and in order to lower
the consumer tax (poors tax), they are dismised because "free access to
the university is there so that the poor can study".
OSS is not much different either, but it's very difficult for most
people to understand it for some unknown reason.
Federico
On Tue, 2002-07-30 at 19:20, Pavel Machek wrote:
> Hi!
>
> > The choice is all yours ... you own this market, you control it and you
> > deserve to manage things as you see fit. My $20 will be waiting for a
> > way to contribute in an inteligent manner which also forces everyone
> > else to do the same.
>
> If you own to donate $20, give it to FSF. Or buy RedHat box. Or buy
> SuSE box. Or give them to me. If you want to force everyone else to
> give $20, you are out of luck.
> Pavel
> --
> Worst form of spam? Adding advertisment signatures ala sourceforge.net.
> What goes next? Inserting advertisment *into* email?
>
Hi!
> I have already donated $30 last year (but to the EFF as I think freedom
> is also important). But how many individuals donate though? I'd say 99%
> of the people just free ride. That means resources are very scarce and
There's no such thing as a free ride. Programs are buggy, all of them,
and each "free rider" helps testing. Programs are not
feature-complete, and each "free rider" is pretty likely to add
feature they need. This means no more money once all bugs are killed
and once all features are done, but thats okay.
Pavel
--
Casualities in World Trade Center: ~3k dead inside the building,
cryptography in U.S.A. and free speech in Czech Republic.
[email protected] (David Schwartz) wrote on 01.08.02 in <[email protected]@whenever>:
> On Thu, 1 Aug 2002 12:51:16 -0400 (EDT), Gerhard Mack wrote:
>
> >I don't get it.. he should know full well there are programmres out
> >there who will not provide good code no mater how much you pay them.
>
> Sure, and there are people who can't write a symphony no matter how much
> you pay them. But, believe it or not, the people who can write symphonies
> are more likely to get paid to write them than people who aren't. Or, in
Unfortunately, the computing field is full of counter examples to that.
They tend to follow a fairly simple logic: If I hire an unskilled worker,
I can pay him less; the problems with that approach only surface in one or
two years. Then I simply replace one monkey with another monkey - assuming
it's even still my problem and I haven't moved to a different job
meantime.
Thus, as Alan said, the .com bubble.
Thus, Microsoft.
> other words, there will tend to be a strong correlation between how much a
> person is paid to write symphonies and how well he or she writes them. This
Unfortunately, reality does not seem to bear that out.
> is because in general the people who control the flow of money tend to pay
> the people who write the symphonies that meet their needs.
In general, that's exactly what they don't do. In fact, this not happening
in reality is exactly why the Berne convention was written to reserve
rights for half a century after the death of the creator.
> For an individual given programmer, in most cases code quality won't
> correspond to financial compensation. You can't pay a randomly selected
> person more and more money and find them writing better and better
> symphonies. However, if I had to get a symphony written by any means at my
> disposal, the more money I had, the better the symphony would be.
I really, really doubt that, *unless* you *yourself* already understand
(and I mean really *understand*) symphonies to start with.
Unless you can judge the quality of composers yourself, the money is
rather unlikely to buy you a good one.
MfG Kai
>But something is wrong with this. If I donate then it's no longer free.
>And only if you never donate then you can call it free. Efectively, that
>means that to get to a certain level of funding donators have pay 99
>times the amount that would otherwise be needed if everyone contributed.
The GPL explains your misunderstanding:
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it
if you want it, that you can change the software or use pieces of it
in new free programs; and that you know you can do these things.
Free software is about freedom, not about pricing.
DS
>> For an individual given programmer, in most cases code quality won't
>>correspond to financial compensation. You can't pay a randomly selected
>>person more and more money and find them writing better and better
>>symphonies. However, if I had to get a symphony written by any means at my
>>disposal, the more money I had, the better the symphony would be.
>
>I really, really doubt that, *unless* you *yourself* already understand
>(and I mean really *understand*) symphonies to start with.
>
>Unless you can judge the quality of composers yourself, the money is
>rather unlikely to buy you a good one.
>
>MfG Kai
First you say that nobody knows what a good programmer is and it's virtually
impossible to tell the difference. Then you assert that I won't find one.
Well, if your first argument is correct, then there's no way to know whether
I've found one or not, and it becomes impossible for you to assert that I
won't find one.
To put it another way, if nobody can tell the difference between good and
bad programmers, then it doesn't matter whether you have a good or bad
programmer. If it matters, it must be possible to tell the difference.
Even if the only way you can tell is to wait until you're done and then see
whether you have a good or bad program, with enough money, you can keep
trying over and over.
Suppose one in a hundred programmers are really good and they all charge the
same but nobody can tell the difference. The more money I have, the more
independent programming teams I can hire, and then I can choose the best
result. Unless you want to argue that I can't even tell the results apart,
but then you're back to their being literally no difference between good and
bad programmers.
Unless you want to make an argument that more money actively causes harm,
you can't do anything about the obvious fact that more money means more
options.
The original argument that money doesn't produce better code quality was
based upon programmers getting more money, not projects getting more money.
With this I agree, in general, paying a person more won't make them produce
better code. But it doesn't follow that this applies to projects or even to
groups of programmers.
DS
> Our General Public Licenses are designed to make sure that you
> have the freedom to distribute copies of free software (and charge for
> this service if you wish), that you receive source code or can get it
> if you want it, that you can change the software or use pieces of it
> in new free programs; and that you know you can do these things.
That is fantastic, unless what you need is not beign developed and
supported. In these cases, the GPL effectively prevents the software
from beign developed and funded.
Suppose Adobe wants to GPL Photoshop. They need funds to keep improving
it, patenting stuff so that other don't abuse their ideas (imagine those
patents under a General Public Patent License). The designers want to
have a Photoshop. But Adobe will have no way to charge for it, except by
redistributing it. But who want's to pay for a Photoshop CD or download
from Adobe when you'll have it bundled in your RedHat (or favorite)
distro? So Adobe doesn't want to GPL it, GIMP doesn't get enough funds
to improve it faster and the designed end up having to use a lower
quality product (GIMP) or having to pay for Windows and Photoshop.
The GPL effectively help the distribution of software but not the
developement of software. The fGPL helps developement of software and
doesn't harm redistribution.
But I do not expect any developer that's getting paid by a distribution,
the goverments or a big company to agree with me. Any other developer
that's _trully_ independant (from distributors, universities, tax and
big corps money), did a great contribution for OSS and can't get the
funds to keep on working, should agree.
Federico