[TAG] [Apertium-stuff] Google patent

[Jimmy O'Regan]

2008/8/17 Ren? Pfeiffer <lynx at luchs.at>:
> On Aug 17, 2008 at 1601 +0100, Jimmy O'Regan appeared and said:
>> 2008/8/17 Sergio Ortiz Rojas <sergio.ortiz at gmail.com>:
>> > Not in EU.  For the moment.
>>
>> <spectie> Answer: The problem is that software patents exist in some
>> ways in the EU. The power of patent governance is split between a
>> legislature, an executive, and a judiciary.
>> <spectie> The legislature (the European Patent Convention) says that
>> software ideas are not patentable.
>> <spectie> The executive (the European Patent Office) ignores this and
>> approves software patent applications.
>> <spectie> The judiciary (the national courts) usually declares the
>> EPO's software patents to be invalid whenever there is a court case.
>
> Yes, this is true. Above all, some national patent offices (including
> the Austrian one) give hints in workhops how to patent certain things
> properly (i.e. how to write the patent application). This leads to tests
> in court that require some amount of effort to kill these senseless
> patents:
> http://press.ffii.org/Press_releases/Amazon_patent_fully_revoked%3A_skirmish_victory_for_FFII
>

I have to admit to finding it difficult to follow just *what* exactly
they're patenting; does it cover *all* of the mechanisms described
within, or the whole?

"[0031]If the cursor remains over the selected text for a
predetermined period of time, the feedback window component 312 causes
the presentation of a feedback window"

Umm... Overlib is a javascript library that has provided that kind of
effect for years...

'Sticky' popups are described in the version released 2002-11-01
(http://www.bosrup.com/web/overlib/?Changes)

"[0036]According to the MTS, the alternate translation has a low
probability of occurrence in the target language; or [0037]The
alternate translation contains redundant words. "

Looks very similar to the sort of filtering done by
apertium-transfer-tools; the first
description I know of is here:

Felipe S?nchez-Mart?nez, Juan Antonio P?rez-Ortiz, Mikel L. Forcada.
Integrating corpus-based and rule-based approaches in an open-source
machine translation system. In Proceedings of METIS-II Workshop: New
Approaches to Machine Translation, a workshop at CLIN 17 -
Computational Linguistics in the Netherlands , p. 73-82, January 11,
2007, Leuven, Belgium.
http://www.dlsi.ua.es/~fsanchez/pub/pdf/sanchez07a.pdf

which predates Google's 'suggestion' stuff:
http://googleblog.blogspot.com/2007/03/suggest-better-translation.html

That paper builds on:

Felipe S?nchez-Mart?nez, Hermann Ney. Using alignment templates to
infer shallow-transfer machine translation rules. In Lecture Notes in
Computer Science 4139 (Advances in Natural Language Processing,
Proceedings of FinTAL 2006, 5th International Conference on Natural
Language Processing), p. 756-767, August 23-25, 2006, Turku, Finland.
http://www.dlsi.ua.es/~fsanchez/pub/pdf/sanchez06a.pdf

which builds on:

F. J. Och and H. Ney. 2004. The alignment template
approach to statistical machine translation. Computational
Linguistics, 30(4):417?449.

Franz Och is now a research scientist at Google
(http://googleresearch.blogspot.com/2006/04/statistical-machine-translation-live.html).
Wow. One degree of separation.