Eisspeedway

User talk:Likebox: Difference between revisions

Content deleted Content added
Line 39: Line 39:


It's far more likely we don't have stack, despite the fact that an elegant representation of grammar suggests one. The solution, according to people like [[Charles J. Fillmore]], is that people don't have an elegant representation of grammar in their heads. It's something else. ---- [[User:CharlesGillingham|CharlesGillingham]] ([[User talk:CharlesGillingham|talk]]) 15:57, 1 June 2009 (UTC)
It's far more likely we don't have stack, despite the fact that an elegant representation of grammar suggests one. The solution, according to people like [[Charles J. Fillmore]], is that people don't have an elegant representation of grammar in their heads. It's something else. ---- [[User:CharlesGillingham|CharlesGillingham]] ([[User talk:CharlesGillingham|talk]]) 15:57, 1 June 2009 (UTC)

:Your examples can indeed be parsed, but you have to agree some effort is required. To parse these sentences, I think we are not just using our unconscious linguistic competence. We're using all of our cognitive facilities to understand them (i.e., we're solving them the way we would solve a puzzle.) Our "parsing" unit (i.e. [[Wernicke's area]]) can't finish the job alone, otherwise these sentences would parse as effortlessly as any other sentence. I think there's a back and forth between [[semantic]] clues, grammatical constraints and general intelligence. ---- [[User:CharlesGillingham|CharlesGillingham]] ([[User talk:CharlesGillingham|talk]]) 19:53, 1 June 2009 (UTC)

Revision as of 19:53, 1 June 2009

Edit-warring

It is very unwise to immediately resume edit-warring when coming off a 48-hour block. Guettarda (talk) 04:34, 8 April 2009 (UTC)[reply]

I was not edit warring. I just don't know any other way to save the text, which is a pain to type up. Now it is in the talk page.Likebox (talk) 04:39, 8 April 2009 (UTC)[reply]
You were blocked for repeated reverts. You have returned to the same sort of behaviour that got you blocked. The point of a 3RR block isn't to teach you how to skirt the 3-revert rule, it's to get you to stop the behaviour that caused the problems in the first place. Discuss, don't revert. Guettarda (talk) 04:43, 8 April 2009 (UTC)[reply]
I am not skirting anything! I reverted for the purpose of moving the deleted text to the talk page. There is no rule breaking going on. If you want, I'll undo myself.Likebox (talk) 04:51, 8 April 2009 (UTC)[reply]

Thank you. Guettarda (talk) 13:58, 8 April 2009 (UTC)[reply]

Mass–energy equivalence

Hi Likebox:

Thanks for the note. I confess to feeling you weren't receptive to change or discussion, but I've changed my mind. I think the article is better than it was. Brews ohare (talk) 23:44, 14 April 2009 (UTC)[reply]

Wikipedia:Meetup/NYC and green cheese

Hey, it was great to meet you last week. I got to start something on one of those minority-scientific theories, The Moon is made of green cheese :)--Pharos (talk) 19:12, 24 May 2009 (UTC)[reply]

intro article on QM radically altered

Hi,

A new editor has unilaterally many drastic changes to the article Introduction_to_quantum_mechanics to which you have made contributions. I do not think that the changes are desirable. I do not want to start an edit war. Could you please have a look at it? Thanks. P0M (talk)

Stacks inside brains

I thought it was interesting that you specifically mentioned "stack (computer science)". (I.E. The argument, based on Chomsky, is that "brains must implement stacks" because the formal structure of English grammar suggests that any natural language parser must use a stack.) Robert Wilensky had a great counter example to this. Consider this sentence:

  • *"The gun the man the police arrested used was lost."

This sentence can't be parsed by most English speakers, but they can parse all these sentences.

  • "The police arrested the man."
  • "The man the police arrested used a gun."
  • "The gun was lost."
  • "The gun the man used was lost."

The sentence we can't parse has a subordinate clause in a subordinate clause. It requires that the parser use a stack with a depth of at least two. This suggests (as Wilensky quipped in a natural language processing course I took from him) that "human stack depth is between one and two." If we have a stack, we devote about a byte (out of our quadrillions of bytes of memory) to it.

It's far more likely we don't have stack, despite the fact that an elegant representation of grammar suggests one. The solution, according to people like Charles J. Fillmore, is that people don't have an elegant representation of grammar in their heads. It's something else. ---- CharlesGillingham (talk) 15:57, 1 June 2009 (UTC)[reply]

Your examples can indeed be parsed, but you have to agree some effort is required. To parse these sentences, I think we are not just using our unconscious linguistic competence. We're using all of our cognitive facilities to understand them (i.e., we're solving them the way we would solve a puzzle.) Our "parsing" unit (i.e. Wernicke's area) can't finish the job alone, otherwise these sentences would parse as effortlessly as any other sentence. I think there's a back and forth between semantic clues, grammatical constraints and general intelligence. ---- CharlesGillingham (talk) 19:53, 1 June 2009 (UTC)[reply]