octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: overloaded function handles


From: Jaroslav Hajek
Subject: Re: overloaded function handles
Date: Mon, 27 Jul 2009 19:49:52 +0200

On Mon, Jul 27, 2009 at 7:33 PM, Robert T.
Short<address@hidden> wrote:
> John W. Eaton wrote:
>>
>> On 27-Jul-2009, John Swensen wrote:
>>
>> | I think patent number 6,857,118 is the more applicable one: "Function  |
>> values in computer programming languages having dynamic types and  |
>> overloading."
>>
>> Yes, sorry.  I pasted the wrong number in.
>>
>> |  That being said, I find it ridiculous that the ability  | to overload a
>> function by string name is actually patentable.  There  | has got to be gobs
>> of languages that did the same thing before this  | patent was issued,
>> right?
>>
>> Yes,  I agree that the claims for using '@' and 'feval' seem to go too
>> far, and I don't see how they could hold up, but they could.
>>
>> However, if I understand it correctly, that's not the major point of
>> this patent.  Instead, the point is that you can create a function
>> handle in one context, and then use that handle in another context to
>> call a function, and the dispatch will be determined by the arguments
>> given and information about which functions are available at the point
>> where the funtion handle was created, not where it is used.  Do you
>> know of prior art for that?  Is there a way to implement this feature
>> in a way that does not use the method described in the patent?  The
>> language of claim 1 seems fairly broad.
>>
>> jwe
>>
>>
>>
>
> Just glancing through the patent, I don't see how it is possible to avoid
> infringing.  I also don't see how a patent like that ever issued, but that
> is really beside the point.

I second that. While I haven't read the patent, what John sketched
seems simply (and obviously) necessary for any reasonable
implementation of the mere idea of taking a handle from an overloaded
function.

> Since the patent has issued it has already been
> evaluated for obviousness and prior art.  Even if you know of prior art,
> challenging an issued patent is not likely to be a pleasant or cheap
> endeavor.
>
> I am not a member of any patent bar so this certainly can't be taken as
> advice.  I have, however,  spent an absurd amount of time in the patent
> infringement world.  I will take a look.  It will take some weeks though.  I
> am seriously over committed at the moment.  For the time being, I would say
> not to worry about it, but we do need to recognize that all the Mathworks
> has to do is file a lawsuit and we will have to react - probably just by
> immediately pulling contested features out of the language (or maybe pulling
> octave from all U.S. distribution sites?) while we examine the problem.
>  Since there is no profit involved, there won't be any royalty issues, or at
> least I don't think so.
>

Although I agree that some action would be desirable in the case of
infringement threat, probably reverting the relevant changes, I would
also plead for a way to keep the useful feature working for those of
us not (yet, at least) affected by this patent. The idea of forking a
"patent-safe" branch of Octave is not very nice, though. Hmm - does
the mere fact that I uploaded those changes create problems for anyone
in USA? What if they use Octave, but don't use overloaded handles?
What if there's a global flag to disable it, i.e. turn the infringing
code into dead code? Is that acceptable? Better consult a lawyer, I
guess.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz



reply via email to

[Prev in Thread] Current Thread [Next in Thread]