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

ASP loophole - where is the line



The problem that is fundamental (for me, at least) about the "ASP 
loophole" is where to draw the line.  I'm currently of the opinion that 
distribution is a good line and any other is fuzzy, but I'd kind of like 
to be convinced otherwise.

Here's the continuum I see:

a) Joe opens a business "Joe's Typesetting Service".  His customer brings
in handrwitten pages, Joe uses GPL software (some of which he's modified
to fit his work patterns) to enter, typeset, and print the documents.  
The printed documents are given to the customer.

b) His customer brings in floppies with text files, and recieves printed
documents.

c) Customer prefers Joe to give him a floppy with Postscript output 
on it instead of printed documents.

d) Joe automates this conversion such that he can put a floppy into a 
drive and the textfile on the floppy will be replaced by a postscript 
file.  This saves him a lot of time.

e) To save further time, he allows his customer to put the floppy into 
the drive rather than interrupting Joe.

f) The customer e-mails Joe a textfile and joe e-mails the output.

g) Joe adds to his program the ability to e-mail the resulting postscript 
file automatically.

h) Joe sets up a webpage which uses a mailto: form to send him the file.

i) Joe sets up procmail to run his program when the customer sends 
e-mail.

j) His webpage is still up, now much more responsive.

k) Joe converts his process to run as a CGI, accepting form posts and
returning the Postscript immediately

l) Joe adds an XML-RPC interface to his page.

The challenges:

In which cases should Joe be forced to give his program to his customer?

Without mention of a specific protocol, please try to find a definition of 
"user" for a theoretical closure of the ASP loophole which would require 
Joe to give his program (including source, he wants to follow the GPL) to 
his customer.
--
Mark Rafn    dagon@dagon.net    <http://www.dagon.net/>  



Reply to: