Fellow monks,
May focus you attention on an issue that might trouble at least all Europeans that work with the web and cgi.
The European Patent Office has granted some very strange patents, examples of which are collected at the
European Software Patent Horror Gallery.
With the permission of
the people there I quote two examples:
- a) dynamic webpages
Mainclaim:
A service agent for fulfilling requests of a web browser client coupled to a network;
comprising: a control program agent tangibly embodying a program of instructions
executable by a supporting machine environment for performing method steps for receiving
a request initiated at said web browser client and fulfilling the request by a providing
a result, said method steps comprising:
displaying an HTML document to said web browser; invoking a control program agent;
receiving data entered by the user from the HTML document
and passing said user entered data to said control program agent upon
invocation as input parameters to said control program agent that were
returned from said HTML document; using an API set for invoking executable
command files and programs accessible via a associated command file object.
Which means:
Creating dynamic webpages by invoking a
script: This seems to cover any webserver
that processes HTML forms and invokes a
program via a common gateway interface,
such that this program returns a webpage.
- database queries
Mainclaim:
An information retrieval system for retrieving information from a database, comprising:
a parser for parsing a natural language query into its constituent phrases to produce
a syntax analysis result; virtual table for converting phrases
of the natural language query to retrieval keys possessed by the database, said
virtual table accounting for particles that modify the phrases; a collating unit for
preparing a database retrieval formula from the syntax analysis result by
selecting a virtual table that is used to convert the phrases of the natural
language query to keys possessed by the database; and a retrieval execution
unit for retrieving data from the database on the basis of said database retrieval
formula.
Which means:
Database retrieval system for responding to
natural language queries with corresponding
tables: translating natural language questions
into database queries by using undisclosed
parsers and virtual tables.
note: The links leading to the European Patent Office
may or may not lead to the information.
Their website is slow and I got some errors.
Many things we can do with Perl are affected, as is telnet and others.
If you want to do something, you can sign the petition against software patents. Any comments are welcome.
neo - horri - phyte
Update: Malkavian is right, you can do more than just sign a petition, and that is what those people really want to do. It is the purpose of their collection and workgroup activities. But they also realize that many people do agree but are not willing to spare the time to do more than filling in a form. This is the reason, why they asked me to provide a link to the petition. This is the reason why I did just that. Otherwise I fully agree with Malk |
Re (tilly) 1: European Software Patents vs. Perl?
by tilly (Archbishop) on Nov 21, 2000 at 19:29 UTC
|
Those of us in the US should not just dismiss this as not affecting them. Multinationals have long understood and used the strategy of trying to get whatever countries they can to pass legislation, getting a treaty signed codifying the standard, and then forcing legislation in the rest based on "international standards". The goal being to fight all of the difficult battles out of sight in the easiest place to fight them, then be able to fight them in the hard jurisdictions with the "weight of international opinion" on their side.
In this case I believe that the pattern is that software algorithms first were justified in the US based on patenting machines that implemented said algorithms. They then moved on to algorithms in general. Then there was pressure to have patents on software in Europe obstensibly to create a uniform environment since the US already had them. Strangely enough, though, in Europe the standard is aiming substnatially lower than the (already absurd) US limit. Finally the goal would be to have the lower standard pushed on the US as well.
This type of strategy has been pursued successfully in many areas. For instance it is why in the US the terms of copyrights well beyond anything dreamed of in the US Constitution... | [reply] |
Re: European Software Patents vs. Perl?
by Malkavian (Friar) on Nov 21, 2000 at 18:56 UTC
|
I'd entirely agree that those kind of frivolous patents are nothing but a large scale pain in the derriere to those of us who just want to get the job done.
That, of course, doesn't go against those patents that are actually for things worthwhile, just those frivolous ones.
Take, for example that CGI patent. How quickly can you come up with some examples of this, prior to the publishing date of 1996?
Just a few seconds, I guess, which completely blows it out of the water on the prior art clauses of patents.
As for the database queries in natural language, published in 1993.. I can draw a handful of examples of this being done in university departments way prior to this.
Thus, the idea is not unique to these people, as prior art exists, and as such, the patent is invalid.
I often find that petitions don't quite cut the mustard. Lots of people sign, you hand it to the powers that be, who look at it, and say "Well, we already knew you didn't like it... What's new??".. Which, I admit, is just my view of it..
Perhaps, a better method is for each person who would sign a signature on that form to find an example of a spurious, and overwide patent in the patent office database, and decry it, proving prior art.
This way, if submitted as a "Petition with rationale", the EPO is forced to admit that there are glaring errors in the system, by the proof submitted.
And, of course, once the evidence is put in that format, in a public forum, there's no way for the powers that be to shrug it off as just a group of zealots with no backing.
Anyhow, that's just my tuppence worth,
Malk. | [reply] |
|
|
Unfortunately, whether a patent is frivolous, obvious, or
just plain nefarious does not shield one from letters written
on the letterhead of a law firm. I rather suspect that anyone
who receives such a letter is in for a fairly significant
cash outlay.
If either of these patents hold, and are enforcable on
this side of the Atlantic, then we might as well shut down
our web site. We have had a www server since 1993, and have
been using CGI and query forms (and Perl, btw) since early
1994 to query a relational database and retrieve specified
results.
In fact, when was the Common Gateway Interface Specification
first proposed? It had to have been before 1994, if we were
already using it.
-----
"Computeri non cogitant, ergo non sunt"
| [reply] |
|
|
At which point, you explain you've been running it since 1993 (webserver with CGI), which blows the prior art right up. No law company would really follow that in a European court, 'cos they know they'd lose, and have to shell out. It's non-enforcable.
Also, for the database query one, it refers to parsing of natural language. However, using SQL, or a strict computational language is not covered by the above patent.
It's meant to be able to extract the meaning of a user with no batabase training who says something like: "Give me a list of all the beers which are dark please, computer."
And that's what I was referring to in my previous post of seeing that around in many places prior to the patent publication date.
Your regular database queries are safe.
I'll wholeheartedly agree that nothing shields you from that dreaded letterhead, but, just because a good many lawyers are vultures, doesn't mean they're stupid.
If you give them that bit of prior art, which they know will completely blow their case out of the water in any court, they won't persue it.
It's an annoyance, but, not insurmountable.
Cheers,
Malk
| [reply] |
|
|
