built-in Grammar checker

Posted Feb 15, 2012 0:19 UTC (Wed) by tialaramex (subscriber, #21167)
Parent article: LibreOffice 3.5 released

I hope this was an outside project rather than something the foundation itself supported. The grammar of most natural languages cannot usefully be assessed with the rudimentary machine "checking" that's popular and somewhat effective for spelling. Grammar checkers mostly serve to give people who aren't very confident in their use of language a bunch of superstitious beliefs that make their writing awkward and leave them frustrated (e.g. it's common for such checkers to forbid preposition stranding in English even though stranding is sometimes the right choice).

Of course (I assume) this is off by default, but it's one of those features that will sucker in users who'd be better off without it. Those most capable of rejecting bad rules from such a "checker" won't be foolish enough to enable it in the first place.

built-in Grammar checker

Posted Feb 15, 2012 3:10 UTC (Wed) by proski (subscriber, #104) [Link]

I often mistype short words, especially it/is/if/in, you/your and then/than. A grammar checker should be able to catch it. I'm not sure if the grammar checker is a separate project, but I don't see anything wrong with it being an integral part of LibreOffice. I consider it a grammar-aware extension to the spell checker.

built-in Grammar checker

Posted Feb 15, 2012 7:53 UTC (Wed) by macson_g (guest, #12717) [Link] (3 responses)

Just read on the associated blog entry:

http://libreoffice.hu/2011/12/08/grammar-checking-in-libr...

The implementation is based around detecting suspicious patterns. Looks like useful feature to have.

built-in Grammar checker

Posted Feb 15, 2012 9:11 UTC (Wed) by tialaramex (subscriber, #21167) [Link] (2 responses)

Well, on the upside the author has at least done enough research to discover that people who ought to know (such as G. K. Pullum) recommend against even trying.

On the downside they're trying anyway, confident that if their rules are just vague enough they're bound to help. I notice that they haven't tried the benchmark Professor Pullum implicitly offers, typing several pages of a major literary work (say, Moby Dick, or Pride and Prejudice) into this software with everything enabled and verifying that it flags none of the excellent prose as incorrect. I think that building a collection of such inputs would have been simultaneously a good practical test of the software and a disheartening lesson on the difficulty of the general problem.

Several of the rules cited in that link seem harmless, but aren't grammar rules at all. Choosing to highlight violations of style such as double spacing may or may not help people, but it has nothing to do with grammar.

Grammar checker unit tests appreciated.

Posted Feb 15, 2012 9:28 UTC (Wed) by mmeeks (subscriber, #56090) [Link] (1 responses)

> On the downside they're trying anyway, confident that if their rules are
> just vague enough they're bound to help. I notice that they haven't tried
> the benchmark Professor Pullum implicitly offers, typing several pages of
> a major literary work (say, Moby Dick, or Pride and Prejudice) into this
> software with everything enabled and verifying that it flags none of the
> excellent prose as incorrect.

So lightproof is designed to be minimal and give ~zero false positives. But your idea is a good-one :-) What would be awesome, would be if you could get several of these smallish but representative classic texts, and create some unit tests in the lightproof module, such that we can ensure that not only are there no false positives now, but there will be none in future :-) patches most welcome (this is a volunteer project). The lightproof git repo is here:

git clone git://anongit.freedesktop.org/libreoffice/lightproof

Thanks for checking this out though ! :-)

Grammar checker unit tests appreciated.

Posted Feb 15, 2012 12:10 UTC (Wed) by tialaramex (subscriber, #21167) [Link]

I don't see any test scaffolding in that repository. My python is mediocre, so I'm definitely not going to attempt to build one from scratch.

I will say that the "ying and yang" -> "yin and yang" suggestion, although it's not itself very useful because "ying" is flagged as a spelling mistake already [at least on this system] does show where "zero false positives" is somewhat practical for units larger than a single English word. Of course you would need a lot of work to establish which things are "always" errors.

For example "ad homonym" and "all intensive purposes" are almost always going to be errors, but for every few times you find "tow the line" used when "toe the line" was meant, you'll stumble over a case where an actual rope was being towed.

built-in Grammar checker

Posted Feb 15, 2012 10:25 UTC (Wed) by mpr22 (subscriber, #60784) [Link]

In response to the notion that grammar checkers for English are dangerous boondoggles: Isle great you're cheese, than.

built-in Grammar checker

Posted Feb 15, 2012 19:44 UTC (Wed) by boog (subscriber, #30882) [Link]

In some European languages at least, there are quite useful and simple checks possible that are called "grammar checking" (they involve more than one word and all words exist in the dictionary). A typical example would be to check that adjectival endings in French respect masculine/feminine gender and singular/plural. These checks have been available in Word for ages and I have missed them since getting rid of it.

built-in Grammar checker

Posted Feb 17, 2012 23:54 UTC (Fri) by csawtell (guest, #986) [Link]

From another linguistically aware blog.

Grammar: The difference between knowing your shit, and knowing you're shit.

Well does it?