Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!samsung!mips!apple!voder!pyramid!athertn!hemlock!mcgregor
From: mcgregor@hemlock.Atherton.COM (Scott McGregor)
Newsgroups: comp.software-eng
Subject: Re: Software Copyright
Message-ID: <35221@athertn.Atherton.COM>
Date: 13 May 91 20:38:26 GMT
References: <3287@legs.UUCP>
Sender: news@athertn.Atherton.COM
Reply-To: mcgregor@hemlock.Atherton.COM (Scott McGregor)
Organization: Atherton Technology -- Sunnyvale, CA
Lines: 154

Richard Kaczmarek writes:
> The good I've heard is that it protects intellectual property from theft.
> The bad is that it inhibits the advancement of software engineering in
> some form or another.  What are your thoughts?

This is always a heated issue, and many of the past discussions have
generated mainly heat, and not much light.  There are some very eloquent
advocates for each side of the issue.  Stallman, is a notable opponent
of copyright protection of look and feel. Ben Schneiderman has made several
thoughtful defenses of copyright protection including the Viewpoint article
in the April 1991 Communications of the ACM. Both sides make good points
I think.  I imagine that what is really important is the details of
what is and is not protected, and how licensing does or does not take place.
Unfortunately, those are primarily questions about our legal processes,
and our business processes, and not about technical processes which have
been the central focus of expertise of people who participate in this forum.

While it is probably impossible to resolve this issue here, it may be
possible to clarify some of the important considerations. 

First of all, consider exactly what is being copyrighted.  There seems
to have been only limited opposition to software copyright as it refered
to strict copying of source code, i.e. where the program's instructions
were what was being compared.  People who put their time and effort into
THAT particular representation felt that they should be protected from
someone who would just COPY OUTRIGHT their code.  This is basically just
against the grossest form
of plagerism.

When that code wasn't copied exactly, but rather translated to another
language, there might be some more divergance of opinion: some people
claim that at least the translation provides added value and so for that
reason it should not be regarded as something that copyright protection
would prevent.

When you get to copying user interfaces, the issues seem to be even
cloudier.  One of the problems is that there has been difficulty
apportioning protection to software between disparate forms of
protection such as copyright, patent and trademark.  Few people object
to the right for the producer of a distinctive logo (say Apple's "bit
into apple" motif) who uses that mark on their screen  to protect their
mark from use by others (i.e. tradmark protection).  Or if
they have written an on-line "help" panel, many people would agree that
perhaps the specific wording or illustration that they include might
have some artistic merit worthy of protection under copyright. Patent
protection, since protecting fundamental algorithms is specifically
forbidden, has been a bit more dicey.
Should a chip that does a special form of rendering qualify for patent
protection?

But beyond these what of a particular rendition of say a trashcan
rendered with lighting from the up-right area? Should not it not have
copyright protection of its artistic content?  What of a clear ripoff
that merely changes the light source to the opposite corner? Is that not
a derivative work, covered by copyright law?  But now what about the use
of ANY trashcan motif? What about size and shape and color and where it
is placed on the desktop?

U.S. intellectual property law has always held that ideas are not protectable,
only objects which instantiate them.  In software, the line between what
instantiates the idea, and the idea itself is much less clear than say
in the case of a manufacturing machine. There is considerable value in
the precise specification of a language, or the precise configuration of
a user interface, precisely because some specifications and
configurations are easier to use than others.  It is certainly desirable
to encourage such improvements.

On the other hand, there is often value in standardization.   The ability to
switch between car steering/accelerator/brake mechanisms, or to use
different brands of keyboards using LARGELY the same interface is good
for usability.
But note that even these do not copy EXACTLY.  People may prefer a given
keyboard bad upon where the escape key is placed, how many function keys
it has, the springyness of the keys, and the shape of the key caps.  There is
still variety within the uniformity.  The same is true about automobiles.

It is desirable for the law to walk this narrow line encouraging variety
even but also allowing some uniformity.  The debates over software
copyright (particularly user interface "look and feel"  copyright) seem
to hinge upon which side of the uniformity/variety line the law should err.

Richard Kaczmarek wrote:

> The good I've heard is that it protects intellectual property from theft.

This is only good in that by protecting it from theft, it makes it valuable
to the person who has it.  This encourages the creation of more of it,
i.e. fosters variety. He also wrote:

> The bad is that it inhibits the advancement of software engineering in
> some form or another.  

The question is how does it inhibit the advancement of software engineering?
If it does so (and many claim it does), then it is only in that it inhibits
uniformity.

Neither enforces variety  (no one may care enough to invent another
variation), nor prevents uniformity  (it is always possible to disclaim
ownership, or to
license others).

So how then will business be affected under narrow vs. wide application
of copyright?  This will be determined primarily by what is economically
advantageous and secondarily (because intellectual property law is only
enforced when suit is filed--and economics determine that).

In the absence of protectable value in new user interfaces, will people
stop having great new ideas of how to improve things?  I think not. 
Evidence is
that many people are naturally creative, regardless of laws.   But the
question of whether their ideas will ever come to the marketplace adding
variety is less obvious.   Many people I have talked to have provided
anecdotal  evidence that many companies will not support the development
of new improvements where an industry (de facto) standard has been
created.   Outside of the software field, product improvements have
often been primarily limited to patentable areas.
So, as Schneiderman points out, there may be little financial/business
structural support for ongoing usability improvements if no protection is
allowed.  Small companies might not be able to get cash to fund their
new ideas, and large companies might have little need to change to hold
on to their positions.

On the other hand, many, such as Stallman, have warned of what will
happen if good user interface techniques are handcuffed in the form of
proprietary advantage from one company alone. Apple's and Lotus's suits
have often been used to point out the possible risks, of strong UI
protection--won't they
squash everyone else out?  If it is possible for them to do so, it is clearly
because there is value in the UIs they have built.  So perhaps they do
deserve some benefits--but clearly  if this goes to the point of
oppressive total monopoly it goes too far.  But there is remedy (in the
form of anti-trust laws)
if that is reached.  There are also marketplace remedies, for instance
when Adobe's licensing was regarded as oppressive, others such as Apple
moved to come up with an alternative standard.  Similar events lead to
OSF's creation.
Even in other industries, patent holders such as Charles Goodyear and
the Wright Brothers tried to hold their patents tightly and lost out as
a result.
There can be down sides as welll as upsides to not licensing patents at low
prices.

Lastly, protection only works once you go to the trouble of finding the
violaters and also of filing suit against them.  Small unseen companies can
actually "get away" with a lot until they earn enough to be noticable to
the intellectual property owner.  Some countries have laws that specifically
exempt copyright and patent protection in the case of work done for
research purposes.  Sales of improvements still require licensing of the
underlying property, but sales of the improvement also requires
licensing from the original owner.  It is no surprise that free
cross-licensing is often a result, which favors everyone.   This might
also be good in the U.S.

Scott McGregor