fsuk-manchester
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Fsuk-manchester] RMS on Swedish Pirate Party vs Free Software


From: Andrew John Hughes
Subject: Re: [Fsuk-manchester] RMS on Swedish Pirate Party vs Free Software
Date: Mon, 3 Aug 2009 02:50:32 +0100

2009/8/2 Pater Mann <address@hidden>:
> Andrew John Hughes wrote:
>> I disagree that proprietary software
>> needs to or should be able to exist though.  Such an argument is based
>> on the idea that the source code is somehow special and unique.
>
> It may be! I wrote a library that implements B-trees in a unique way
> that has advantages over conventional implementations. As far as I am
> aware, no-one else implements B-trees in that way. (I must confess
> that I haven't looked very hard though!) I have used my skill and
> expertise to create this library and, although I currently have no
> plans to sell it, I don't see why I should give the details away to
> anyone and everyone without any recompense.

It really depends here what you mean by 'recompense'.  Many such
algorithms usually receive their recompense not by making profit
through sale, but by academic prowess through publishing important
results.  I guess that brings me back to the benefit of computer
science in general rather than a single company.

There probably is a market for a proprietary library that provides a
(presumably better performing) B-tree algorithm.  But I don't imagine
it's that big because of the tradeoffs.  Any developer wanting to use
this has to consider all the legal implications of using it and
distributing it with their software, and be suitably content with its
API.  They also have to be happy by the maintenance service provided
by the original developer, because they have to rely on this developer
for that future maintenance by virtue of it being proprietary.

The history of the LZW algorithm is relevant here.  It was patented by
Unisys, which put it in a worse position than your B-tree algorithm
because there is at least nothing to stop an independent
implementation of your algorithm being produced (if someone could work
out how it behaved).  In the LZW case, the algorithm was published but
unusable and developers had to actively avoid treading on the landmine
created by that patent.  This increased development but not in the way
Unisys would want; alternate and better compression algorithms were
developed for developers to use without having to pay license fees to
Unisys.

> Even if I did decide to
> try and sell the software, you don't have to buy it but, if you did, I
> would also give you the source code so that you were not dependant on
> me for fixes and extensions.

That helps, though it's less of a normal case.  What you get there is
a community of 'honoured developers' who have the source as well as
the original developer and can improve it.  In the long run, that
probably means every person who buys your software ends up maintaining
their own fork.  This is very like how Sun handled Java for the last
fifteen years and Microsoft have had similar 'shared source' or
'research license' schemes.  They don't tend to succeed too well. Plus
if you give away the source code, it's not going to remain secret for
long.  It only takes one developer to ascertain the idea from that
source and publish an article describing it.  Then someone else can do
a cleanroom implementation.

 It would not be FOSS because you would
> have to buy it to get the source

As Simon mentioned, that's misinterpreting FOSS.   The GPL states that
if the binary is distributed, then the source must be available for 'a
nominal fee'.  Prior to the widespread use of the Internet, this would
mean sticking it on disc or cassette and mailing it to them.  Sure,
someone else can then redistribute it but the only difference between
that and the pirating of proprietary software is the legality.  In my
experience, that doesn't tend to stop a lot of people.

and it would be tied up with legal
> agreements that would prevent you from passing it on to others but at
> least you would not be locked in.
>

You're still locked in in a sense.  You may be able to develop it
yourself, but that assumes you can.  It's still a secret which has to
remain confidential to you or the company.

>> Keeping the source code
>> secret is born of fear that something dreadful will happen to it.
>
> Not necessarily. I have no fear that something dreadful will happen to
> the source code to my library. I may well open-source it at some point
> in the future - I just choose not to do so at the moment.
>
> However, if by "something dreadful" you mean that someone with little
> or no expertise themselves could simply take advantage of the time,
> money and effort that I have poured into developing the software
> without giving anything back whatsoever, then I agree that would be a
> big worry for a software company. If they cannot make money from the
> software that they develop, what incentive is there to write it in the
> first place? Even worse, if someone else can make money from it
> without having contributed to it in any way. It seems to me that could
> potentially stifle innovation rather than encourage it, at least as
> far as software companies are concerned.
>

The 'something dreadful' was meant to be suitably ambiguous because it
varies from developer to developer and company to company.  If they
haven't worked with FOSS before, they have this fear that it will
somehow eat their children or something.  I agree with your second
point that those outside the community shouldn't benefit from those
within it (which is basically what you seem to be saying).  It's why I
wouldn't license my own work under a BSD-style license.  I'm very much
a Free Software person with a belief in working together to some
greater goal, as idealistic as that may be.

>> the developer is already providing a good service in producing and
>> supporting the software, FOSS developers aren't going to attack and
>> fork the project, just as you point out below.  I just really don't
>> see how keeping source code proprietary helps anything.
>>
>> This wasn't an anti-Microsoft argument.  Feel free to replace them
>> with Apple, Adobe, etc. if you want.
>
> I was merely using your example. My argument also applies to many
> other companies such as those you mention.
>

Indeed, it just sounded like you were trying to add some
anti-Microsoft slant to what I said.  Personally, I treat all
proprietary companies the same.  Were Microsoft to release the source
code for Windows under the GPL tomorrow, I'd treat this with the same
interest and cautiousness with which I greeted the release of OpenJDK
by Sun.

>> But I'm going off the actual point which was simply that the actual
>> development cost is overshadowed by the amount of profit made because
>> there is no continuous outlay to produce more copies.  It's not like
>> making sofas where you have to make back the cost of producing each
>> one.
>
> Apart from the obvious fact that a company has to make more money than
> it costs to produce (and market) the product, in general the price
> that you pay for something is rarely directly related to the
> production cost. If a product is in short supply or is available from
> a limited number of sources, it will generally carry a high price
> regardless. If the product is freely available from a variety of
> sources then demand comes into play. High demand generally leads to
> low prices as companies compete for business; conversely, low demand
> generally leads to a high price because companies need to make more
> money from each sale. The final element affecting price is value to
> the customer. For example, if I create something that saves you a
> thousand pounds per year then I could argue that I am perfectly
> justified in charging you say two hundred pounds for it even if it
> costs only pennies to make. You are still saving 800 pounds per year
> and I get a good sum of money to develop the product further and/or
> create new products whilst still having enough for a decent holiday!
> [grin]
>

My suggestion wasn't that companies shouldn't make a decent profit.
My point was that the cost is a one-time fixed cost followed by
nominal costs for duplication of the software, whereas tangible goods
have to be produced from scratch on a per-sale basis.  The same
applies to other intangible items such as music, movies, etc.

> One other thought: If I put tap water into bottles and then sell them
> at 5 pounds each, providing that I make it quite clear that it is
> simply tap water both on the label and in the marketing, am I doing
> anything wrong? If so, why? It is not fraud because I am not
> pretending it is something that it isn't and the people who buy it (if
> there are any foolish enough to do so) know exactly what they are
> getting.
>

Your argument is flawed in that at some point you still have to pay
the water company for this water you're using in your bottles.  You
don't have an inexhaustible supply of tap water.  Even if you pay a
fixed price for your water supply, I'd hope any decent water company
would wonder why you are using so much and hike the price! :)

>> You can still satisfy people and make money without having to pretend
>> the software source code is some kind of secret sauce that only your
>> company can produce.
>
> I may not be only one who *can* produce the software but I may be the
> only who actually does so...
>

If no one else does so, that may also mean there is as yet no demand.
If your software then creates demand, then others will follow.

>>  Competition should be based on the actual value
>> of the product to the user.
>
> And people often do not value things that they do not pay for. If they
> can get it for free, they tend not to consider it valuable. A company
> who I used to work for had a client who acted as a good sales
> reference for them. Because of this, the client was rarely charged for
> work done and when they were it was only a small amount. Consequently
> the client became more and more demanding and treated the company very
> badly. One day, the company decided that enough was enough and
> informed the client that in future they would have to pay the going
> rate for any further work. Immediately the client's attitude changed
> and the company and client ended up with a very good business
> relationship. The problem was that while the work was free, it did not
> have any perceived value even though it was actually saving the client
> considerable sums of money. Although the value to the client's
> business had not changed, once they had to pay for the software, it
> suddenly gained a tangible value in their minds. Of course, they did
> complain at first but it wasn't long before they came to appreciate
> the change.
>
> Another old joke: Give a man a fish and he will eat for a day; give a
> man a rod and teach him to fish and he will complain about the loss of
> his free meal-ticket! [grin]
>

Indeed this is a general issue with intangible goods.  On the other
side, a lot of people don't see why a CD they bought is worth hundreds
of pounds either, which is why piracy of software, music and movies is
rife.

>> That's true of business generally but sadly I don't see a lot of
>> competition or innovation in the software field.  The only reason some
>> still exists is because of Free Software forcing proprietary vendors
>> to rethink their ideas.  A lot of the most prominent pieces of
>> software continue to exist by charging their users for new versions
>> that add fairly minor features and fix bugs.
>
> Sadly, you are right in many cases. The problem is that the majority
> of commercial software is bought by businesses and these tend to have
> a lot of inertia. To change something as fundamental as the office
> software used - or, even worse, the operating system - can be a
> support nightmare for a company so the tendency is to stay with the
> status quo.
>

Not only that, but most of the smaller businesses in today's
proprietary software industry have been driven out or bought by larger
vendors which just increases inertia.

> The same applies to individual users albeit to a lesser extent. That
> is why so many people are still using IE for their browser in Windows.
> It works (up to a point) and it takes effort to find and install an
> alternative. Not a lot, I know, but unless there is a compelling
> reason to change, non-techies simply will not bother.
>

That's because normal users just want things to work.  Microsoft have
done a fantastic job with Windows from a business point of view.  It
comes with every computer you buy and because the user buys the
computer with Windows from someone like Dell or PC World rather than
Microsoft directly, it's the computer as a whole that generally gets
the blame when things go wrong and not Microsoft's product.  It's lead
to a world where crashes are the norm, where viruses are the norm,
etc.

> Cheers,
> patermann
>
>
> _______________________________________________
> Fsuk-manchester mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/fsuk-manchester
>

Cheers,
-- 
Andrew :-)

Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

Support Free Java!
Contribute to GNU Classpath and the OpenJDK
http://www.gnu.org/software/classpath
http://openjdk.java.net

PGP Key: 94EFD9D8 (http://subkeys.pgp.net)
Fingerprint: F8EF F1EA 401E 2E60 15FA  7927 142C 2591 94EF D9D8




reply via email to

[Prev in Thread] Current Thread [Next in Thread]