Skip to main content
22 events
when toggle format what by license comment
Jul 22, 2015 at 22:30 answer added Greg Lee timeline score: -3
Nov 10, 2014 at 5:25 answer added XL _At_Here_There timeline score: -1
Mar 26, 2014 at 9:28 comment added Alenanno @TKR Can you come to this question and explain why you think your answer is valid? I'd like to know why you think your answer should be undeleted (we can do that if necessary). You can do this by posting another answer to that question in the link. Thank you.
Mar 25, 2014 at 20:15 vote accept player.mdl
Mar 25, 2014 at 19:23 vote accept player.mdl
Mar 25, 2014 at 20:15
Mar 25, 2014 at 17:16 comment added TKR @player.mdl It appears to have been changed into a comment: see above.
Mar 24, 2014 at 22:41 comment added player.mdl @TKR kindly repost your original answer? I believe it is the most sensible one I've encountered thus far, but it seems to have vanished.
Mar 24, 2014 at 22:09 vote accept player.mdl
Mar 25, 2014 at 19:23
Mar 24, 2014 at 13:32 answer added Olivier timeline score: 7
Mar 24, 2014 at 2:23 answer added curiousdannii timeline score: 2
Mar 23, 2014 at 18:08 comment added jlawler Plus, since there's a new "program" every few years (rather like a new version of Windows), when hiring time comes up you'll need somebody who's au courant with the latest stuff, cause the folks already there don't really get it.
Mar 21, 2014 at 18:49 answer added P Elliott timeline score: 7
Mar 21, 2014 at 0:28 comment added P Elliott ...and to my knowledge nobody would take seriously the idea that natural languages should be analysed using CSGs! The recognition problem for CSGs can be shown to be only solvable in polynominal time, never mind parsing the damn things. CSGs are way too powerful for the task at hand. Current grammar formalisms have converged on the class of mildly context sensitive grammars, which e.g. the minimalist grammars of Ed Stabler fall into, as does HPSG + others.
Mar 21, 2014 at 0:24 comment added P Elliott "...one can hardly render an example sentence longer than a couple of words on a single page." - I think you're confusing notational simplicitly for theoretical parsimony. They're not the same thing. Another small point: Move and merge collapse down into a single operation. Additionally, CFGs aren't powerful enough to capture natural languages. That much is pretty uncontroversial at this point. See the work on cross-serial dependencies in Germanic.
Mar 21, 2014 at 0:18 comment added user483 it might be added that at the height of the Chomskyan revolution (as some call it), US universities were undergoing a general expansion and various new linguistics departments were opened. the new programs were staffed by recent graduates, many trained at MIT or other pro-Chomskyan programs. Still to this day (with the immense difficulty for a graduating PhD to land a tenure-track job), an MIT degree in linguistics (as in other fields) has cachet.
Mar 20, 2014 at 14:46 vote accept player.mdl
Mar 24, 2014 at 12:36
Mar 20, 2014 at 11:58 comment added player.mdl That's kind of what I was suspecting...
S Mar 20, 2014 at 2:10 comment added TKR This will be controversial, but my take is that Chomskyan transformational grammar in its various incarnations is now at a stage of development similar to medieval scholasticism or late Ptolemaic astronomy. It's still the dominant paradigm for historical reasons (in this case, the Chomskyan revolution of the '50s and '60s), but has outlived its usefulness and is struggling increasingly desperately to "save the phenomena", hence all the complicated machinery you describe. The reasons for its dominance are mainly institutional, namely that the large majority of tenured syntax professors these da
S Mar 20, 2014 at 2:10 answer added TKR timeline score: 4
Mar 20, 2014 at 1:59 history tweeted twitter.com/#!/StackLinguist/status/446465915224616960
Mar 19, 2014 at 21:23 review First posts
Apr 18, 2014 at 21:13
Mar 19, 2014 at 21:05 history asked player.mdl CC BY-SA 3.0