🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: gnu.misc.discuss
Article #75247

Re: Patents (was Re: Copyrights)

#75247
From: Stavros Macrakis
Date: Wed, 29 Aug 1990 08:00
55 lines
3411 bytes
(sorry for the belated followup; we've had some problems with our news server)

In article <41668@think.Think.COM> barmar@think.com (Barry Margolin) writes:
	...
   The economics of software are different.

   Suppose you are developing a piece of software, and the going rate for
   software of this class is $500/copy.  And suppose it contains ten licensed
   algorithms, at an average $20K/license.  This is a built-in overhead of
   $200K for the product .....

I too have reservations about patenting algorithms, but making up a
specific pricing structure won't prove your point.

Very substantial pieces of (copyrighted) software are licensed for
much less than the figures you cite.  Take a look at the pricing
schedule for OSF Motif (tm) or OSF/1 (tm), for instance, each of which
contains licensed software from a variety of sources.

   On the other hand, manufacturers of tangibles generally have large
   overheads to begin with; they have factories, factory workers, purchased
   and leased equipment, etc.  In order to become a manufacturer it is
   necessary to raise quite a bit of capital, and licensing fees are slipped
   in there relatively easily......

... or perhaps the licensing fees are a reasonable part of the cost
simply because they're constrained by competitive pressures....  You
only license innovations when the benefit outweighs the cost.

...and, by the way, making and selling software <<does>> require
non-negligeable capital investment: when you include all the
overheads, one engineer-year costs over $100k, and R&D is only a small
part of the overall cost of bringing out a product.  Of course, in
many startups, much of the capital is in the form of "sweat equity",
but that doesn't make it any less real.  Insufficient capital is the
most frequent cause of small business failure....

   If you're manufacturing hardware that is simple enough that it can
   be made in your basement rather than a factory then it probably
   doesn't have many patented parts; the same cannot be said for
   software that is developed at home.

Again, this is simply not true.  The "Screwpull" corkscrew and those
ubiquitous plastic snaps (that you push on both sides to open) are
both (I believe) still protected by patent, and would be very easy to
produce in your basement.  Note that they are not extraordinarily
expensive, for the simple reason that there are other ways of doing
the same thing.

In summary, the economics of software are indeed different, and
current intellectual property protection may be far from the ideal
solution, but you've not shown that there should be NO such property
rights.

	-s
Technical Details
Message-ID: <MACRAKIS.90Aug29090046@diderot.osf.fr>
Path: arc1-ca.pugleaf.net!gmdzi!unido!mcsun!i2unix!inria!imag!mururoa!carnot!macrakis
References: <66187@sgi.sgi.com> <58801@bbn.BBN.COM> <58830@bbn.BBN.COM> <41668@think.Think.COM>

Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 1025b4,a3783b857baac617
X-Google-Attributes: gid1025b4,public
X-Google-ArrivalTime: 1990-08-29 21:37:42 PST
Path: arc1-ca.pugleaf.net!gmdzi!unido!mcsun!i2unix!inria!imag!mururoa!carnot!macrakis
From: macrakis@osf.fr (Stavros Macrakis)
Newsgroups: gnu.misc.discuss
Subject: Re: Patents (was Re: Copyrights)
Message-ID: <MACRAKIS.90Aug29090046@diderot.osf.fr>
Date: 29 Aug 90 08:00:46 GMT
References: <66187@sgi.sgi.com> <58801@bbn.BBN.COM> <58830@bbn.BBN.COM>
 <41668@think.Think.COM>
Sender: news@imag.fr
Organization: OSF Research Institute--Grenoble
Lines: 55
Posted: Wed Aug 29 09:00:46 1990
In-reply-to: barmar@think.com's message of 18 Aug 90 00:49:52 GMT