Subj : Re: Software Patents
To   : comp.programming
From : CBFalconer
Date : Sat Jul 09 2005 07:04 am

websnarf@gmail.com wrote:
> Paul Dietz wrote:
>> CBFalconer wrote:
>
>>> The only effect of the patent was to stop all development cold.
>>> Other techniques were developed, and by now have far outperformed
>>> LZW. [1]
>>
>> This is actually an argument used by patent advocates.  The idea
>> is that a patent forces competitors to explore workarounds they
>> otherwise would have ignored.
> 
> Well, that works for LZW, but what about Arithmetic Encoding?  IBM has
> a patent on that one (or at least the most reasonably practically known
> implementation of it).  See, for its purpose, AE is known to be
> optimal.  No work arounds, which involve choosing a different
> algorithm, will help.  IBM has managed to patent a method which is
> mathematically proven optimal -- even though the patent office doesn't
> patent mathematics (Benoit Mandelbrot tried to patent the Mandelbrot
> set  and was told that "Mathematics cannot be patented").

I understand that that is simply IBMs self protection against the
Software Patent disease, and that they are perfectly willing to
issue (or ignore) licenses.  Of course they do not have to remain
benign.  The proper cure is to end software patents.

BTW AE is only optimal in terms of bit ratios in the same sense
that Huffman is optimal in terms of byte ratios.

-- 
"If you want to post a followup via groups.google.com, don't use
 the broken "Reply" link at the bottom of the article.  Click on 
 "show options" at the top of the article, then click on the 
 "Reply" at the bottom of the article headers." - Keith Thompson

.