Re: [dvd-discuss] Gedanken Experiment -Unix and Norton

[Joshua Stratton <cpt(at)gryphon.auspice.net>]

On Thu, 10 Jul 2003 microlenz@earthlink.net wrote:

> Look at this analogy. 
> 
> Book Manuscript = Trade Secret
> Published Book = Copyright.

Yes, I've seen that before. It's wrong.

Assuming that software is properly the subject of copyright anyway, 
copyrighting software as we do it presently is worthless. That is, the 
public fails to benefit from it to the same degree that they benefit from 
the publication of a book, or a movie, or music.

Compiled software is too difficult to read. It's entrance into the public 
domain doesn't satisfy various public goals such as being able to create 
derivative works, or considering the short lifespan of software, being 
able to adapt it so that it continues to run. It is possible, but it is 
far too difficult.

If software is going to EARN copyrights, it needs to include additional 
disclosures. Source is NOT a manuscript. Rather, it, plus information on 
the method of compilation, is merely a translation of the exact same work. 
It is not a rough draft; it is more along the lines of a handwritten MS as 
opposed to a typeset book. With the book example there is no significant 
difference in information conveyed to a reader. There is not anything of 
note that isn't just as easily available later. This isn't true for 
software, not even a little bit. The whole Open Source movement wouldn't 
have even started if compliled software was a substitute for source code.

We've not shied away from things like deposit requirements before. And 
it's important to note that when a deposit is required, a best edition is 
typically what has to be deposited. The concept of deposit isn't just to 
help keep track of what is copyrighted when, it also aims towards 
preservation.

So it is perfectly in line with these noble aims of copyright --
meaningful preservation (i.e. enough info to get software to run in the
future), and the creation of derivative works -- to make a copyright for
software conditional on the deposit of the complete and well-commented (to
LoC specifications) source.

If that invalidates trade secrets, it's no problem of mine. It's just the 
cost of getting a copyright. Happens in the patent field all the time. 
It's what you pay to get better protection.

Note also that although perhaps the source need not be seriously 
published, (not effecting its term length) it should be publicly available 
to inspect. Authors can read one another's works to gain insights; it's a 
good thing. We might as well encourage the same with software. This might 
include studying algorithims, but it could surely extend to other areas as 
well -- interoperability seems like one, and there's always good old 
fashioned criticism.

This is my position anyway. Think of it as "disclosed source." Can't copy 
it literally, can copy ideas or facts, just as with other copyrighted 
materials, and you do get to look at it.

It has been pointed out that there may be other types of works that 
similarly are not sufficiently useful without additional disclosure. Let's 
do the same with them, then.

An MS I'm not too worried about. Complex musical compositions only 
released as phonographs -- sure. Maybe even additional material with 
films, e.g. scripts, computer models, etc.

Whatever it takes to make that grant of copyright really pay off to the 
public. I seriously doubt that it will harm authors of any type even a 
little bit. After all, the thing about trade secrets is, you never know 
when someone else has figured it out too -- and that means it's no secret.