iLiad Does anybody planning to develop a dictionary software?

[wagnerian]

I'd really like if I can use dictionary on my iLiad.
The link below is about dictionaries for Zaurus:
http://www.users.on.net/%7Ehluc/myZa...l#dictionaries

[scotty1024]

The trick to a dictionary is you usually want to pop it up while reading, check a word, then pop it back down and continue reading.

I know how to do the pop up trick, I use it for my HP48. But it involves modifying /home/root/start.sh. If something goes wrong editing that file: bricked iLiad.

How badly do you want that dictionary?

[wagnerian]

Originally Posted by scotty1024

The trick to a dictionary is you usually want to pop it up while reading, check a word, then pop it back down and continue reading.

I know how to do the pop up trick, I use it for my HP48. But it involves modifying /home/root/start.sh. If something goes wrong editing that file: bricked iLiad.

How badly do you want that dictionary?

Popping up is not so important to me and it's OK to execute it manually.

You know, currently I'm using a dictionary on iLiad, but it's pdf files converted and divided alphabetically from a single dictionary file. So I can browse around them.

And I prefer to take out my Zaurus in my bag...

[k2r]

Originally Posted by scotty1024

The trick to a dictionary is you usually want to pop it up while reading, check a word, then pop it back down and continue reading.

...in an ideal world this would be handled by an extension of mutant-bionic-ipdf. Highlight a word (if it's not a bitmapped pdf, but that would not be ideal, of course.) and send it to the (dictionary) helper application (which in a very very ideal world could to be defined per file in the manifest or by some kind of language tags and a matching table. I have English, German and French documents on my iLiad and sometimes a document has multiple languages).

Maybe in an ideal OS there was somehow a way to integrate stuff like this in some kind of framework that could be called "service" and would be available in every application :-)))

Another thought:
If I understand a little how differently pdf's are organized maybe it's easier to let the user draw a box and find all the letters in it than finding all lines and words in advance and let the user use a highlighter?

And just another Sci-Fi thought:
In a paradise-and-heaven-like world bitmapped pdf would be no problem of course but would be fed to tesseract :-)))

Well, it's christmas soon, maybe if iRex, Santa and the Tooth-Fairy get together to work for us?
#!chris

[tribble]

I would like to see a dictionary icon in the page bar. Turn it on, and then click a word.
The pagebar hides and shows the translation until you click another word or flip the page or click the pagebar area.

[scotty1024]

For phase one a button to pop the dictionary, type in the word and then pop it back down again would I think be better than no dictionary.

Even "text" PDF's need something close to a tesseract to figure out what text is where.

I invite anyone whom wants to, to examine the code to handle hyper links in PDF, and those are designed to be clicked on.

[tribble]

Originally Posted by scotty1024

For phase one a button to pop the dictionary, type in the word and then pop it back down again would I think be better than no dictionary.

Even "text" PDF's need something close to a tesseract to figure out what text is where.

I invite anyone whom wants to, to examine the code to handle hyper links in PDF, and those are designed to be clicked on.

Sounds like a good start (can it be done, that the dictionary is somehow always present, like a small window ontop of PDF, so you daont have to close the currently read file?)

And where do i have to start looking? in the poppler libraries?

[k2r]

Originally Posted by scotty1024

Even "text" PDF's need something close to a tesseract to figure out what text is where.

PDF sometimes seems to be synonymous with "fear the worst".

Preview.app somehow does this job, however. But Apple has legions of programmers to implement something like that and doesn't give us the sourcecode...

I hoped that it was relatively "easy" to figure out which letters are located in a given rectangle and to just order them up-to-down-left-to-right and then send the resulting word to whatever helper application.

edit: I found some very nice links that explain basic why-PDF-can-be-difficult-to-handle-stuff:

http://www.dclab.com/converting_from_pdf.asp
http://www.dclab.com/converting_from_pdf2.asp
(have a look at the links on the right in the second article)

#!chris

[scotty1024]

I have one tesseract I use in my Sony BBeBook software but it definitely doesn't handle every PDF out there.

It can be done, as you say, Preview.app handles it.

And yes, by pop open, lookup, pop closed I mean a persistent state. I use the same trick for x48 on my iLiad. A little pop up button down in the toolbar and the x48 is left running the entire time once its popped open the first time.

Maybe I can figure out how to do a application loaded pop up button to avoid the hacking of /home/root/start.sh...

[yvanleterrible]

I'm looking forward to the day a portable, downloadable Wiki Comes around and be usable on a reader...You just update it every other week, for a fee, and if you could use it while reading some other text would be great!

[CommanderROR]

But I would not try that in PDF...^^

maybe once the openreader, HTML viewer and other reader software are available for the Iliad...

[yvanleterrible]

For any reader as well. Adobe's Digital Edition would be great as a native format on a reader...

[k2r]

> I'm looking forward to the day a portable, downloadable Wiki Comes

I don't see a reason why mysql / apache / php / mediawiki shouldn't possibly run on a beefy Linux-Box with 64MB RAM and many gigabytes of diskspace...

Don't panic!

[tweq112]

I really want something like this as well...
I see eink readers being really useful and popular for
people reading in a language other than their native one,
especially in Asia where a lot of the population studies
English as a second language, I've actually been working
on something like this for J2ME-enabled cellphones over the
past year but cellphone screens are soo small... Still hoping
that maybe one day there'll be a SDK for one of these
devices...(V2 is taking forever to come out;;; ).

[ericshliao]

I think the very first question for a dictionary software on iLiad should be: which format should it comform to? It would be smart to adopt a generally accpeted format, such as EPWing (mainly for Japanese), MDX (mainly for Chinese), and of course, StarDict.