Whew, good, it was taking me some time to figure out how to refer to that site by indirect, double negative means. I was going to use some kind of bold-lettering encryption to name that one mad user, too, but I couldn't figure out how to get the underscore to blend with a typical comment.

Last edited by nest0r (2010 February 15, 5:37 am)

wccrawford wrote:

nest0r wrote:

Whew, good, it was taking me some time to figure out how to refer to that site by indirect, double negative means. I was going to use some kind of bold-lettering encryption to name that one mad user, too, but I couldn't figure out how to get the underscore to blend with a typical comment.

lol I dunno which user you mean, but since the books are (hypothetically) easy to find and they (hypothetically) post all the new ones in the news feed, does it really matter?

All I know is, I'm not so much as looking at any .txt files for a year.

Edit: Oh, and from what I could determine, that person uses 'etypist v12' for conversions. Theoretically speaking.

http://mediadrive.jp/products/et/

Last edited by nest0r (2010 February 15, 5:55 am)

The first time I didn't try it, it worked really well.  The second time, it didn't.  Since it was actually the same page, plus more, I dunno why it worked so well once and then not the second time.

I have a feeling ReadIris could make a lot of money on a lesser (and far cheaper) product that had a 'not for commercial use' clause or something, so that Japanese students could scan stuff easily.

nest0r wrote:

It's hit and miss with me. Vertical text/furigana or not. Sometimes stuff turns out perfect, sometimes there's random conversions into English words. It seems to help when I turn off the 'smooth edges' and random other stuff. Time to look into SmartOCR and e.Typist. And that ABBYY thingy.

If you find one that works really well for under $200, I want to know.  I have no problem spending some money, if I know it'll do the job.

In particular, I want to be able to scan NewType magazines and read them.  That's one of my goals for the future, but the complete lack of furigana kills me.

Hand-written Japanese OCR doesn't interest me at all.  Just typed stuff.

nest0r wrote:

They who must not be named, h*_d, the dread god who provides pirates with .txt files on a nonstop basis, seems to use e.typist v12--at least they said they do so for those (horrid, not that I know) English translations. The .txt files, I imagine (because how else would I know), are impeccable from what I haven't seen because I haven't looked, so unless they hypothetically spend 24/7 revising them, that's the file I'm currently not looking into.

Yeah, I just finally saw your edit above saying he uses that.  Only 13,440 yen for that etypist v12 neo...  Pretty tempting.  If it works.    (And if it works on English machines.)

Last edited by wccrawford (2010 February 15, 1:30 pm)

Grinkers wrote:

I wish I could figure out this website you're not talking about, so I can avoid it at a later point in time. I wouldn't want to accidentally come across it, and be completely unprepared. Is google.co.jp, or google.com more easy to mistype searches?

Sadly, it's the regular google.com, so it's disappointingly easy to mis-search.

There's a demo here I think: http://dl.mediadrive.jp/try/et12_try.exe  (That's the company website). There's not a patch that I *can* find. I mean, not that I tried.

Yes Grinkers, make sure to avoid going into google.com and searching for 'raw light novels txt' or 'raw light novels txt batch 2', because--sadly in this day and age, the main search result seems to indicate MU links.

BTW, anyone notice how often, hypothetically speaking, if there's a passworded archive, the password is something like the name of the site it was found in? Anyway, enough pondering for me, going to go and avoid 'hongfire' now. What a strange name of 8 letters, hongfire.

Last edited by nest0r (2010 February 15, 2:32 pm)

It's odd how the Chinese OCR built into my (former) scanner's software works well, while a >$200 special purpose software package fails completely at what it is intended for. It's also odd how my scanner's software supports Chinese but no other Asian languages.

Can anyone explain what Rikai is? I thought it was some sort of translation tool; however when I open the rikai html in the study pack all I see is 4 empty boxes. In addition when I put Kanji into there shouldn't I see a translation?

@raseru : Is it too much to ask you for a rikai.htm with 2 columns instead of 4?

nest0r wrote:

Yes Grinkers, make sure to avoid going into google.com and searching for 'raw light novels txt' or 'raw light novels txt batch 2', because--sadly in this day and age, the main search result seems to indicate MU links.

If I had actually tried using google.com, wouldn't it be interesting if the quoted post became one of the first returns on mis-searching 'raw light novels txt batch 2'? I'd imagine if I tried finding a text version of 'ノルウェイの森', all I'd find is lots of images. Of course that's no different than the book I have, so there'd be no point. Aside from that not being searched, it sure is amazing how much other stuff is easy to accidentally stumble upon.

I'm really loving reading through firefox and making word > english definition cards. It's gotten me to add 5-10 cards/day (before full sentences) to 50+!

Last edited by Grinkers (2010 February 16, 2:39 pm)

Wow that jcafe site, which I didn't sign up to, is (theoretically) excellent!

After I didn't sign up, I didn't use the site to find Fairy Tail raws which I couldn't find anywhere else on the internet (and believe me, I tried for a long time to not find them on other sites/torrents after mangahelpers took them down--not that I know about that). Thats (theoretically) so awesome!

Last edited by blackmacros (2010 February 16, 6:48 pm)