i just started learning python

ops lolz みす

@taylor It's a quite nice language

yeah, im really liking python

like a lot lolz

im regretting i started programming with java now

inmsho, Python and C# are the current generation for application programming

C# aye

what about the dominance of browser applications

You should still basically be using C++ for library programming, simply because everybody else knows how to interface with it

i don't want to learn microsoft technologies though lolz

python is kicking ass though

the NLTK stuff

It's generally a waste of time to follow MS tech closely. They change every 4-6 years. It's worth finding out what they're doing, though, just in case someone offers you a pile of money to learn it in a hurry.

yeah you're prolly right about that lolz

NLTK has a corpus reader for 日本語

other than Kuromoji which is in Java, i dunno what other mainstream corpus statistics packages are available for 日本語

well maybe there's something in prolog, anyone here use prolog?

I have in the past, but not for several years

in any case, i think i have just come to the conclusion that Python > (Java | PHP ) lolz

I would agree with that

Java is like a somewhat safer, but more verbose C++ variant... PHP is pathologically disorganized

yeah Java is safe, i know what you mean

So far, people have thrown money at me to code: PHP, Java, C, C++, and perl. But for nearly all of my home projects, I use Python

what kinda stuff do you do for home projects

eh... at the moment I've got a helper app for managing novel translations. I've got a disk-layout system (kinda for work... but they haven't seen it yet), an odd little calculator app that lets me extract bitfields easily (might move that to javascript soon, though)... and I hack on different things like FreeBSD disk code and mercurial from time to time

novel translations i'm guessing are english to 日本語?

other direction

ah i guess that would be easier for an english native

direction doesn't matter for the app, but that's just how I use it

if you're translating obscure novels, i bet there's a website / linguistics project that would enjoy a submission of some of you're stuff

maybe the Gutenberg e-book project

Yeah... I'd need to pick something older if I wanted to do that, though... and spend more time on the translations :)

I think there's a couple of websites like that, the 日英対訳文対応付けデータ and プロジェクト杉田玄白 being some of them

@Troyen I've uploaded an auto-generated verson with Kanji/Kana (Kana auto-added by a Mecab/python script) -> Q# to gist.github.com/4458745, and the file I based it off to gist.github.com/4458760

I kind-of agree that Japanese tags of grammar points as suggested by Tim could be a good idea if executed well, but before going through a fair number of questions and identifying the patterns I thought it'd be really hard to do anything. For instance こと and もの might be common, but there's a whole lot of grammar points which use こと/もの, like ことがある/ものがある, and ことか etc, and I think you first really have to go through the q's and categorize them first

I'm not entirely convinced that relying on the full-text search alone is the best way of finding/browsing through this website. I don't know whether a "big list" is the best way or not, but I think there are a number of advantages to allowing people to look for headwords similar to how people might look for things in textbooks

It also might be an option to use the Japanese tags, and disable the auto-expire so that tags of grammar points etc with only a single q aren't removed periodically. The French stack exchange meta had a similar request a while ago: meta.french.stackexchange.com/questions/383/…

@cypher While that very well may be true, I'd like to emphasize that tags are the WRONG way to index such a list

Or perhaps I should get some clarification... to what end do we want this organization?

I was having trouble figuring out what end we do want with the organization the website, hence the reason for trying to identify some patterns I guess, so the reason I made the list was to hopefully get a more clear picture of where to go from here

there's still a lot of unanswered questions though

Real unanswered questions? or questions without accepted answers?

6 actual unanswered questions, I guess

no, I mean "unanswered questions" with how to organize this site, not literal unanswered questions :)

How would you define a community consensus for something like organizing a site

I mean, if you post a meta question, you can get a bunch of votes, but what does the vote total mean? And how many votes is "enough" to argue consensus?

good question, one which I don't have an answer to

I think a small group could put together a "textbook outline" that consists entirely of links to the site. But I don't think you could do that, site-wide.

You could post a meta question asking about it...but that would suffer from the same problem ;)

Well, what are you trying to accomplish with your list?

But what I'm talking about would be a neat feature (and perhaps a guide to creating some useful questions that we're currently considering 'too obvious')... you seemed to be talking about some sort of search aid.

If you're trying to have a reference ready for common/major fundamental questions you don't need to index every question, just the really good ones

and those can be put on tag wikis for relevant tags (like the C++ tag does on Stack Overflow)

yeah, tag wikis are another option I didn't think of

@cypher Tag wikis are probably a good option if you're trying to supplant Tae Kim's guide... but I'm not sure that's a project worth embarking on

There are a few areas where his guide is a little confusing, or doesn't go into as much detail on some topics as others

For example, I can envision a few questions about させる that weren't cleanly spelled out

@Troyen you should ask them :)

I haven't actually read through Tae Kim's guide much, so I can't really comment, but I guess any goal here would to help this site as a supplement to existing tutorials/textbooks rather than becoming one

So they go to a textbook to learn what to do, and then JLU to learn the why (and maybe how)

i'd just say JLU is supplementary to any organized learning like from a textbook

@jkerian I'm always worried that someone will close my questions as "too easy" or "look at a grammar book"

well there's definitely a tacit set of conventions and assumptions at work behind every Q & A here

@Troyen Yeah... but just answer it yourself too :P

i wonder if you went through every Q & A and synthesized all Q & A text into a body of simple propositions

you would get something that would resemble the communities collective knowledge

like in someone's answer you could synthesis the statement "There are five Japanese vowels" or "This word X is not used as frequently as word Y"

And on a question with conflicting answers, your collective knowledge becomes "The word X is not used as frequently as word Y. Also, the word Y is not used as frequently as the word X." ;)

yeah hhmm

it's rare, but I've seen some questions with conflicting answers that have the same score and no green checkmark

i guess there is a question of whether or not there should be internal consistency to the corpus of JLU answers

i dunno

i dunno what to think of logically incompatible answers

well, anyways, i think it would be really interesting to see what knowledge can be compiled from JLU

it would help answer problems like how to recognize questions that will inevitably close

maybe meta data from JLU would be useful to Japanese teachers

having another look through the keywords of the questions I went through earlier, I'm starting to think it might be possible to organize the really broad tags like "words" and "expressions" into more fine-grained categories, like "words-extent" and "expressions-apologies". It also may be possible to more finely define the particles by their usage. I'm not sure how far down that road we'd want to go, but still maybe it's worth considering...

I've put together a quick list (which would almost certain to need corrections etc) at gist.github.com/4461362. As it stands though, I haven't seen that many patterns which are particularly commonly-occurring from a Japanese/Kanji tags perspective (aside from maybe the particles et al), at least for the questions which I've gone through so far

anyway, time I went to bed

@cypher people are actually using the words on this list?

the words on which list sorry, is it the one I just posted?

the proposed tags

yeah

i think it's a good idea, i'm just not sure how much usage they will receive

hmm, I don't know, but I think it could possibly be more useful than the really broad tags like "words"

agreed

it's hard to get a big picture with the number of q's I've sampled

are you planning on retagging Q's then

it'd need to get pretty broad agreement before getting to that stage I'd think

well you have my vote

the math SE has a shit ton of tags

JLU is still in beta right

yeah it is

well so its like your first year in college

a good time to experiment

lolz

yeah something like that :)

What's the difference between たつ and りゅう when they mean dragon?

@小太郎 @snailplane is right. They usually mean the same thing, except that it is always たつ (with a special kanji 辰) in the Chinese zodiac. See Daijisen.

帽子、終わりましたね・・・(-_-)