GPL not encouraging new technology

[Niall Douglas]

On 29 Nov 2002 at 16:18, Marcus Brinkmann wrote:

> > To prove this last probably contentious point, look at GNU/Linux. I
> > personally cannot see anywhere in the entire system any completely
> > unique technology. It's merely an improved version of existing know-
> > how. There's no real innovation in there AFAICS, not say like Plan 9
> > or EROS is a reconception from the ground up.
> 
> Uhm, but EROS is GPL and LGPL. 

And its development was exclusively funded by a university (ie; a 
government) as far as I understand it.

> So is the GNU Hurd, which is another
> innovative operating system.

You're right it is, but it was much more radical back in ~1990 when 
it was conceived. My understanding of it is that it's quite similar 
to the NT kernel.

However GNU Hird proves my point exactly. Work has gone extremely 
slowly on a badly needed kernel which would in one fell swoop address 
many of the technical problems with GNU/Linux. Because it was and is 
considered so radical, and mostly I suspect because the much more 
retrograde GNU/Linux kernel felt more comfortable to more developers, 
it has been sidelined.

> > So, hopefully you don't think me a troll. I am genuinely interested
> > in what all your thoughts are.
> 
> I think that programmers of all kind should come up with innovative
> programs, and that managers should come up with innovative business
> models for free software ;)  Also, most progress is probably
> incremental.  There are steep pragmatic barriers when it comes to the
> question of establishing innovative software.

True, but one of the most major is funding because without that it's 
very hard to attract enough programmers to some whacky unproven idea.

> I also think that the availability of free software among pupils and
> students can spur some innovative new free software.  Studying real
> programs is a very important aspect of learning to program, and the
> GNU software base has some very good examples (I often look into the
> GNU C library, for example).

There's no reason why commercial software can't achieve the same 
thing, it's just we've all grown used to not receiving the source 
with our software.

Nick Mailer wrote:

> Stop there for a moment. You see, you already frame the GPL's "big"
> problem in economic terms. In other words, if some philosophy is not
> "economically worthy", it is a "big problem". This is highly
> contentious, for a start.

I never said that.

> > *except* 
> > when the product is already well-established and mature.
> 
> I'm not sure there's a particular problem with this? *Most* software
> projects, propriatory or Free, do not make money until they are
> well-established and mature.

No, but money arrives to GPL projects much later (and less of it)  
than to commercial so therefore proving my point.

> Other software projects learn the value of the GPL through a trial of
> fire: like Trolltech's QT and MySQL. That is not to say they have
> abandoned propriatory licenses (through a dual-license) strategy, but
> that they have realised the value in allowing those who would benefit
> from a GPL'd product access thereto. This seems to be an inreasingly
> sustainable model.

MySQL had a mature product when they looked to go commercial. Qt was 
less so at the time, but it was still relatively mature.

Both are better implementations of existing technology, there isn't 
much radically different in them.

You're not addressing my point at all.

> Why should you assume the right to make a living out of such a thing?
> In a sense, the world offers you no living at all - you EARN one. If
> your talents are sufficiently useful to someone, you will be paid to
> work therewith. A case in point: we employ a Debian GNU/Linux package
> maintainer who, at least 1/5 of the time, must continue to work
> maintaining this package for the Debian project. It's worth it for my
> company.

Ever heard of the entrepreneurial spirit? ;)

Why should I want to work for someone else? Why should I care if I am 
of value to someone else? Engineering, like art, has intrinsic value 
with no relation to whether someone wants to pay for it - however, 
unlike most art, engineering is useful.

Unfortunately we live in a world where work in itself is not 
rewarded. It must be of value, and because of this artists and 
engineers alike must bend their creativity towards making profit.

Hence yes, I do assume the right to make money out of writing 
software, just as much as painting a wall or anything else others 
consider of value.

> As for your comments about programmers - I think that's patronising.
> Take a look at the Kernel mailing list to dozens of programmers
> thinking "out of the box" every day. This contradicts some of the
> comments you make below.

cf. my comments about re Hurd.

> > Until that point, you'd probably have to work to 
> > support yourself and do the radical project in your own limited free
> > time.
> 
> No you won't. Many people work on Free Software projects in their
> spare time or as part of other employed work. Take a look at the Exim
> mail-server as an example. Lots of innovative technology has emerged
> this way.

Emm, you've just repeated what I said!

> > Hence, I would feel that the GPL is bad for blue-sky technology
> > startups. The GPL is excellent for developing a better version of
> > already existing technology which cannot be stolen by others, but no
> > use for creating new technology.
> 
> Do you know anything about the history of the Free Software
> Foundation? Do you know about the creation of GCC? Indeed, of the
> whole GNU project? This seems to contradict what you say: certainly,
> it was designed to look and feel like Unix, because this is what
> people were used to. But, in virtually every case in point, the GNU
> tools out-performed and out-innovated anything available for
> propriatory Unix. Indeed, even today, propriatory Unixes like Solaris
> pretty much recommend the downloading of the GNU toolset to get
> anything like a comfortably useable operating experience!

Again you're repeating exactly what I just said! And if you do 
examine the history of GNU, their attempts at their own kernel were 
trounced by Linux which was a very retrograde kernel when it was 
first created. Therefore, supporting my original argument.

Don't get me wrong. Given twenty years of spare time by enough 
people, you'll eventually get there. But vastly more people will 
volunteer their time on established paradigms that radical step 
changes, hence the GPL is not good for blue-sky projects.

> There are very few examples of "completely unique [sic]" (something is
> either unique or it isn't, by the way - it's one of those words ;-) IT
> technologies around at the moment.

I think there are more that many would believe. Acorn RISC-OS had 
quite a number, as did NextSTEP.

> Certainly, Microsoft haven't
> created many.

Absolutely agreed! In fact, I'd go further and say everytime they've 
tried something new and innovative it's failed dismally.

> I would argue, strongly, however, that the Free Software
> movement is innovating far more frequently and interestingly than
> propriatory software. Take a look at some of the fascinating things
> that are going to be included in the 2.6 kernel. Take a look at the
> Reiser4 file system.

I like Reiser, him and I think similarly. I wholeheatedly support 
Reiser4 but I don't think it has anything unique in it. Reiser6 will 
be unique though.

> > Now I personally am not a free software advocate, never have been
> > and probably never will (I come from an Acorn background) 
> 
> So do I, but I don't see how this precludes one from a belief in Free
> Software. Indeed, much of the benefit of the Acorn world was in the
> home-brew programming mentality it spawned.

Precisely, but free software is libre. The typical Acorn mentality 
was to distribute the software with source but for a cost if you 
could get away with levying one.

> To be honest, if Acorn's dying movements had been to Free RiscOS, it'd
> be in a far, far better state than it is now. And it would have any
> number of wonderfully innovative features.

Possibly, but I'd doubt it. Software needs hardware to run and RISC-
OS is totally dependent on ARM because it's all assembler. Arm don't 
want to produce performance processors, so RISC-OS is effectively 
screwed.

MJ Ray wrote:
> > I'll firstly offer my own position on this: the big problem I see
> > with the GPL is that it does not make any money for vendors *except*
> > when the product is already well-established and mature. [...]
> 
> I do not think that the GPL is particularly better or worse in this
> regard than most software licences.  I suspect that selling GPL
> software is easier because there is a near-guarantee of not being able
> to screw the users for more money in exchange for no new work.  I am
> testing this suspicion.

You're right with the standard business model regarding software, and 
it's also true that GPL software saves much of the repetition which 
wastes so many closed-source resources.

> To me, your business model seems short-sighted and hemmed in by the
> thinking of proprietary software producers.  *shrug*

However just because IBM and Microsoft said things should be this 
way, it doesn't mean it's a strict black and white choice between 
free software and existing business models. I personally support a 
model whereby you pay a small amount for the binaries (users), more 
for binaries and source (students or developer evaluation) and there 
is a sliding percentage royalty for reuse of the software in other 
commercial products (resale in a derivative work). This business 
model benefits engineering, the software, the users and has most of 
the benefits of GPL software but it can *make* *money*!

Something like the QPL goes half way there, but its problem is it 
isn't scalable - to reuse a derivation off Qt requires a one time 
payment of the lump sum by each developer and this reduces the 
nesting level of reuse because the lump sums accrue with each 
derivation. With a sliding scale of royalties, the more derived you 
are the less percentage you get, which keeps overall royalties low 
but ensures derivations which are 95% combination of other pieces of 
software sending most of their earnings to the people who wrote the 
software.

In all of this, there is no reason why royalties cannot be waived for 
non-commercial use. There would need to be some limiting to prevent 
someone distributing a slightly repackaged pay-for program for free 
in order to undermine them.

Nick Mailer wrote:
> > To me, your business model seems short-sighted and hemmed in by the
> > thinking of proprietary software producers.  *shrug*
> 
> Quite. Let's examine the issue at hand:
> a. An individual wishes to produce software
> b. An individual wishes to earn money
> 
> The fallacy is in assuming that there is only one sphere of causality
> where both of these desires are fulfilled. Notice that once you remove
> all the illusory connective tissue: "I wish to be paid to produce
> software, whereafter I wish to receive further royalties from every
> copy of the software sold; and to deprive the right of any (even
> paying) user access to my work except in a very specific and
> delineated way" you are left with simply a. and b.

I never said any of that. You assumed that's what I thought. cf. 
above.

> The *truly* creative programmer, who can GENUINELY "think out of the
> box", will provide innovative connective tissue that does not rely on
> the effective monopolistic extortion and strongarming that a
> fully-operative proprietary realm demands.

No, you as someone also from an Acorn background should realise the 
commonly practicised software business model is one invented mostly 
by US multinationals. It is not the only business model.

> There are many models available at the moment where the GPL fits very
> comfortably. It is only someone extraordinarily lazy or greedy who
> cannot conceive such.

To encourage innovation, society must make it worth while to the 
inventors. Currently, we do this with money. I don't feel I am lazy 
nor greedy to feel that the ten hours I put into something per day 
which will benefit all of mankind should be adequately rewarded.

The ideal people will produce for the benefit of mankind for no 
reward may work for some people, but they are a tiny minority. People 
are primarily centered around self-interest (look at any four year 
old child), it's simply the way we are.

Someday our culture may evolve to create a majority less egocentric. 
But that is a long way off, and until then this is the way it is. 
Hoping or dreaming for something not possible is at best a waste of 
time, and at worst dangerous.

Cheers,
Niall