GPL not encouraging new technology

Niall Douglas ned at nedprod.com
Sun Dec 1 18:54:47 UTC 2002


On 30 Nov 2002 at 22:48, Alex Hudson wrote:

> On Sat, 2002-11-30 at 21:06, Niall Douglas wrote:
> > Again, you're restricting yourself to the current business model
> > mindset. It doesn't have to be that way. If a law existed which said
> > that all COTS had to come with source and could be reused by anyone
> > so long as they paid a percentage of the sales cost (including zero
> > if they sold it for free) ie; in other words, if mandatory reuse
> > were legistlated for, then this whole situation would change.
> 
> If a law existed that mandated everyone had to buy my software, my
> whole financial situation would change :o)

Not what I said, but yeah, me too!

> Basing business models on forced laws is a bad idea. I don't think my
> unacceptance of that is a 'restricted mindset'. If you that 'I work
> with what I have' as 'restricted', then I suppose so...

No I must disagree. All law can be said to be the grease between the 
cogs of society and it is no more so in commercial enterprise. If you 
compare the illegal drug black market (which is an example of 
partially unregulated capitalism) with the sanitised version we're 
legally allowed practice, then it is clear laws define a marketplace.

Hence just because there is a law saying no one may make a copy of 
some software for a friend doesn't make a similar law guaranteeing 
right of distribution of derived works any less possible or valid. 
Regulation gives bias - it's the old carrot & stick scenario.

> "Distributed for profit" is also a fairly meaningless phrase, in this
> context. We're talking about marketing - but you're talking about
> marketing Free Software being more difficult that non-free. Saying I'm
> talking about marketing as being a non-profit activity is merely
> begging the question on your part - yes, if you assume that Free
> Software cannot be effectively distributed for profit, then when I
> talk about marketing I'm talking about a non-profit activity. It's
> illogical, though.

Never said any of that! Really! This list seems to constantly misread 
and misquote my arguments! I will assume it is because I am not being 
clear enough, nevertheless no other list I'm on does this so it seems 
odd.

Of course free software can be sold, and indeed is so. Here's my 
original point:
> > No, but money arrives to GPL projects much later (and less of it)
> > than to commercial so therefore proving my point.
> 
> I don't think that's necessarily the case. Proprietary software
> tends to come to market fully-formed (i.e., developed already);
> Free Software isn't usually written and then marketed.

Now I'll repeat myself, this time with hopefully more clarity. 
Proprietary software tends to get money much earlier in its 
development cycle because the workers need paying. Free software 
tends to get money only once it's mostly complete and mature because 
you can't sell something which isn't written yet. The only source of 
substantial funding for infant free software is government grants, 
and those really are bespoke.

And that's my problem with free software - it is not long-term self-
sustainable because it cannot encourage entrepreneurs except those 
making money off the back of selling other people's work. This is bad 
(in the long-term) for software and hence society.

> > Except that all bespoke projects I've ever worked on tend to reuse a
> > lot of COTS
> 
> I doubt that very much. You probably 'customise', or 'extend', rather
> than reuse - if software is reuseable, it doesn't fit the COTS
> definition. COTS software comes 'as is' - you cannot change it, you
> only get to use it.

In this you would be right. My apologies for using "reuse" 
incorrectly.

> > Much bespoke software as well uses customised GPL 
> > software because the GPL permits not releasing source if you don't
> > distribute the software. So, I think it's the opposite of what you
> > say.
> 
> How does the above further your argument? If it's bespoke, it's not
> COTS, and hence nothing to do with the COTS market we were talking
> about. 

No my point was much bespoke software customises or extends one or 
more COTS. Therefore bespoke software is intricately linked with COTS 
and will increasingly become so with time.

Therefore when we speak of COTS, it implies consequences for bespoke 
and indeed almost all software everywhere due to this intrinsic link.

> Anyhow, it's arguable whether the above is the case anyway: I suspect
> a contractor re-using GPL on a bespoke, proprietary, development is a
> licence violation, pure and simple. I can't see otherwise.

Depends entirely on what is called "distribution". It can be argued 
that a privately funded venture to produce some bespoke solution 
involves no distribution.

I've never worked on a project where we did this, but I had to fight 
hard to stop it happening ("If project X did this, why can't yours?" -
 answer: "Because I'm working on this project!" :) ) with a copy of 
MySQL.

> The two basic tests are a) originality, and b) non-obviousness. Test a
> is passed if there is no previously published work including the
> invention; this is a fairly easy test to pass. Test b is more a matter
> of opinion, since only someone 'familiar in the art' could judge it,
> yet they are not the ones who are asked. 

This is all true and I completely agree the patent system is not 
perfect. However if you consider the world before patents, the 
overall situation was worse.

Anyway this is besides the point. I referred to the *theory* not the 
implementation and I think if you do have someone familiar in the 
art, then they are a pretty good judge of how non-obvious it is. 
Obviously implicit in the definition of "innovation" is a strong 
subjective quality.

> Proprietary software has wham-factor; Free Software tends not to.

That would depend. Free software tends to be much more powerful in 
non-obvious ways. I go "wow" much more often with free software than 
with mainstream commercial.

> > I don't quite understand you here. Bespoke software is custom
> > usually once-written software to perform usually one strictly
> > defined task. So therefore surely the customer is exactly buying the
> > software?
> 
> Hmm, not really the case. Certainly, customers very rarely own the
> software (in terms of copyright), even if it is bespoke. More often,
> they buy the software, and then are tied into a support contract
> (since only the writer of the software can maintain it). 

In my experience all rights get transferred to the purchaser who can 
theoretically take out a support contract with someone else. They 
never do, but they can.

> Modern companies see software they right as their "intellectual
> property"; and tend to licence it, not sell it. Much like in the COTS
> world, you very seldom buy the software.

In the contracts I've worked upon, it's explicitly written that all 
rights become the purchaser's. Could be a military thing though.

> Your argument isn't flawed when someone gives an example of innovative
> free software, because you can always say "that is not innovative".
> Hence, no-one will argue with you because it's an unwinnable argument.
> You produce your criteria - possibly with examples of proprietary
> software that fulfils the criteria? - and maybe someone will knock
> them down for you. 

That is entirely possible, and of course I know the proprietary range 
much better than the free one. However I came at this point from the 
basis of logical theory and then looked for evidence to disprove it. 
So far, I have not found any.

> Personally, I don't see any proprietary replacement for Mosix, for
> example, I guess one probably exists somewhere, but unlikely to be as
> well developed.

Surely Cray have something?

> I personally think the Hurd is innovative, and I don't
> see your reference to the NT kernel whatsoever (if it is like
> anything, it is like OS/X in that it [currently] runs on Mach - it is
> *very much* not like any other OS in terms of the daemonized
> architecture). 

Daemons no but plugins yes. The NT kernel is quite extensible eg; it 
uses a unified namespace of which parts are provided by plugins eg; 
pipe manager, file system etc. Technically one could do much of what 
GNU Hurd can in NT - it's just Microsoft have chosen not to and 
indeed seem to actively have prevented anyone knowing much about it.

Very few people know that file points in NTFS can hold multiple 
streams or run through a translator eg; reparse points which work 
like symbolic links, or zip files appear as directories etc. I agree 
that GNU Hurd is somewhat innovative, but it's been overtaken by time 
and history and the fault for that, in my opinion, lies squarely with 
the psychological consequences of free software.

> Even Apache - with the 2.0 multiprotocol support - is doing things
> found no-where else in the industry. I disagree with the cloning
> thing. If Free Software were only cloning, proprietary software could
> keep ahead by innovating. That concept has been killed dead by the
> fact that Free Software competes, betters and stays far out ahead (see
> Apache, for example). This isn't about being energetic, as you said -
> it's about being innovative.

Apache, like KDE is coming from a strong existing base. Well-known 
and respected programmers within each project then posit their views 
on best future direction and democracy chooses the best path.

Unfortunately, it will tend to choose a path most comfortable to the 
most developers. This will tend to be conservative and not overly 
radical. Thus returning me to my original point of not encouraging 
innovation.

> However, I have no doubt you will knock over every example here for
> some reason - as noted, it's an unwinnable argument.

You'd be right :). It's probably impossible to reach a conclusive 
position as well. However, the chances are that if something looks 
funny and smells funny, it probably is funny. What I want to guard 
against is zealotry. The free software ideology is 99% good but I can 
also see some serious concerns about its long-term sustainability and 
trampling down commercial alternatives is fine and good until they no 
longer exist and are no longer bringing vast sums of new money into 
the profession.

> I will only note
> that 'unsubsidised' is a rubbish criteria: this has nothing to do with
> our argument.

It has everything to do with it. Government subsidies are AFAICS the 
only method free software has of creating real innovation. We spent 
hundreds of years getting away from dependence on government to allow 
private enterprise, and I think it a bad idea to return to that so 
long as governments are structured the way they are.

> Look at Microsoft: everything they produce, apart from
> Windows and Office, is *hugely* subsidised. It's in the figures, you
> can go see them.

I know, I used to be a shareholder. And I agree a company as large as 
Microsoft isn't good for software (neither was IBM). But this is a 
completely different kind of subsidy - free software isn't subsidised 
for pure motives, it's really publicly-owned bespoke solutions eg; 
DARPA's funding of ReiserFS. There is next to no chance of pure blue-
sky research in free software, whereas if nothing else Microsoft's 
many failed blue-sky projects show some effort is put that way with 
the hope of first-to-market.

> Ditto the quality argument: Free Software does not
> prescribe love from developers, more peer review, or any of that.
> These are not the differentiating factors.

Oh come on now! You telling me free software isn't on average of much 
higher quality than proprietary? I'm sorry, I just don't believe 
that.

Quality in software has a direct correlation with user satisfaction. 
If my absolutely critical Windows 2000 server crashes twice a year, I 
will be much less happy than with a crash once every two years. Same 
even goes with reading email, or browsing the web or indeed anything.

> > And it's precisely that reason that they still haven't succeeded ie;
> > because blue-sky stuff just doesn't sit well with free software. If
> > they'd set out to write the perfect functional clone (which is what
> > they have ended up doing anyway), their goal and ideology would have
> > been compatible.
> 
> How is the Hurd a functional clone? I don't see that they have ended
> up doing that at all.

Read again: I said that's what they *should* have done and if they 
had, they'd have succeeded like Linux has.

> I also don't see that your argument as presented
> the case that the Hurd has not "succeeded" because it's blue sky. Hurd
> development has been slow because there are not enough hackers working
> on it: it's pretty simple. 

And why aren't there enough programmers on this project? Answer me 
this one question straight, no mucking around. If you can answer this 
without fufilling my logic regarding free software offputting 
volunteers for radical ideas, well, we'll have made progress!

> > You should have read my words more closely. I said the Hurd was
> > innovative *in* *its* *time* but not really any more. RISC-OS was
> > extremely innovative in 1988, but it sure isn't today. Innovation is
> > 100% relative to the context it's in.
> 
> That wasn't my argument. You pointed at RiscOS as being an example of
> something innovative (at whatever point in it's product life-cycle).
> Clearly, you are not holding it to the same "strongly growing ..
> unsubsidised .." etc. standards that you hold Free Software to -

No you're mixing and matching my arguments about totally separate 
things to make it look like I'm saying something else - which isn't 
productive. The strongly growing unsubsidised qualifications 
exclusively referred to free software innovation. Obviously, every 
commercial venture will subsidise any new programming project, it has 
to and whether it's strongly growing is a matter of the speed of 
development of the code, not how it sells.

> RiscOS was never strongly growing, never hugely profitable, and in
> fact gained most of it's market share on the back of the British
> Broadcasting Corporation monopoly on computing in schools.

All true, but it was still innovative in many ways because it did 
things no one did at the time and indeed still can do quite a few 
things hard to do today. I still use it occasionally in the form of 
an emulator - just the find & replace tools in Zap are enough reason.

> Your point
> that use of the GPL "stifles blue-sky innovation" is clearly wrong, so
> you then use this "non-subsidised" device to make the argument
> unwinnable. All blue-sky innovation is subsidised: if you cannot
> afford the development fail, as a business person you must not do it.
> The risk with blue-sky is that it will probably fail, so you should
> expect to lose your money. 

I see what you're saying but I'm afraid I have not been clear enough 
in my arguments. I'm talking bigger pictures, over the space of 
years. Quite simply, I'm asking the question: who will produce the 
most radical innovation over twenty years? Free software or 
proprietary?

My own conclusion is proprietary will tend to win, but for all the 
wrong reasons. A third way could maximise it better than either.

> You seem to be interested in some fabled 'third way', which is very
> modern, but seems to me to be meaningless. Software is either free or
> not - in this sense, it's a very black and white issue. What you are
> talking about is trying to sugar-coat non-free software (or, in
> business-speak, 'increase it's value proposition').

No I'm talking about reversing the mistakes made in the conception of 
software as IP. There is a huge difference between design and 
implementation and the US multinationals chose it to be design. I say 
I should be implementation, and leave the design alone and free.

Thus under my view software patents are boneheadedly stupid. Also 
licensing software should be illegal - you should get what you pay 
for, and that includes source. You should also get the right to 
modify and redistribute with appropriate royalty payments.

> That's not
> something I suspect many people in this group would find particularly
> interesting, nor do I think it is innovative (cf. shared source,
> Borland's Delphi system, and numerous other software which have been
> distributed in the way you seem to propose).

I've not seen any guaranteeing right to produce derivative works.

> And ultimately, I don't
> see it being successful, because non-free software ultimately cannot
> compete with free software. Development practice, funding, etc. is all
> mostly irrelevant: when free software and non-free software compete,
> the free software tends to win. Third-way software depending on "free
> software development practice", "open source", etc., is no different
> to any other type of non-free software.

They cannot compete because free software is superior to the current 
proprietary model. But no one can say what happens between a 
/different/ 
proprietary model and free software.

In the end, IMHO free software gets 98% of the way there but it 
causes me grave concern about long-term viability. There is a better 
way, but we'll need (a) free software to become an extremely serious 
threat to proprietary (b) a leading step-change product endorsing a 
third way and (c) support from politicians in order to bring it 
about.

Cheers,
Niall




More information about the Discussion mailing list