Re: [linux-audio-dev] Re: [linux-audio-user] Re: Free Software vs. Open Source: Where do *you* stand?

From: Hannu Savolainen <hannu@email-addr-hidden>
Date: Tue Feb 21 2006 - 14:56:01 EET

On Mon, 20 Feb 2006, David Kastrup wrote:

> Well, then they might have some expectation to be able to use it, no?
> Without the ability to adapt the software to different devices or
> applications, or fix errors (or pay someone to do that), the software
> is crippled in its usefulness.
Usually software companies take care of fixing the errors themselves. If
they fail to fix them then you have the right to get your money back. When
you buy some software you certainly know it's only going tor un under some
given operating system(s). If you expect anything else it's clearly your
fault.
 
> When buying electronic appliances, at one time you could rely on the
> schematics being in the inside. That meant you could make full use of
> the appliance, adapt it to different problems (using a radio as a
> guitar amplifier), repair it and keep it in working order, and you
> could take it to service men of your choice to have it adapted or
> fixed.
The point is that having full schemantics for some hardware device doesn't
make you capable to make free copies of the device. It's just a document
that makes the device itself more usefull. So to make the schematic
usefull you have to buy the actual product.

However software source code is not a document. It's the software itself.
Having the sources makes you able to improve the code. It also makes it
possible to produce competing products very easily. So if you give your
source code you very much kiss goodbye to your rights.

> That's basically what workmanship is about: offering the best to the
> customer to make use of.
>
> Just 20 years ago, it was customary to provide computer purchasers or
> service people with schematics, BIOS listings and similar stuff
> (partly on request and for payment). Now it is trade secret this,
> closed source that, not for your eyes this.
I think the problem is pretty much the size of the software. In 80's
typical BIOS was just tens of pages of paper. Listings of current BIOSes
are likely to be thousands of pages.

I quess you don't want the BIOS listing but the full BIOS (source) code
itself. However these are very different things.
 
> It is annoying. If I want a Porsche engine in a VW bug, I can buy the
> parts and all relevant service manuals and plans, and put a mechanic
> to work.
When you buy any serious software (other than some el cheapo consumer
stuff) you will get full documentation (for some parts of it you will
have to buy the SDK). Practically every major software package has all
kind of setup files, scripts, plugin/DLL interfaces, built in custom
programming/scripting languages and things like that. These
docs/tools/interfaces will make it possible to do things you can't even
dream to do with a VW bug.

Yes. If you don't want to pay anything for software your only choice is to
get some morally clean free sofware with no other documentation than the
source code. This works well if you are a programmer and have enough
spare time to find out how the source code works.

However for large majority of computer users this free model doesn't work
at all. They simply don't have any capability to modify or even understand
the source code. In fact they don't even know what source code means. So
they find it easier and cheaper to use evil commercial/proprietary software.

It may be morally wrong to produce this kind of software but in practice
it's the only way to fund the extra work required to make the software
easier to use, to produce the additional documentation and to handle the
packaging/marketing/logistic required by this model. This is about 90%
of the total effort while writing the software code means just 10%.

Best regards,

Hannu
-----
Hannu Savolainen (hannu@email-addr-hidden)
http://www.opensound.com (Open Sound System (OSS))
http://www.compusonic.fi (Finnish OSS pages)
OH2GLH QTH: Karkkila, Finland LOC: KP20CM
Received on Sun Feb 26 20:17:45 2006

This archive was generated by hypermail 2.1.8 : Sun Feb 26 2006 - 20:17:45 EET