1) Open up File Manager and click on 'Removable memory card' and create a folder here called 'stardict' . Then within this folder create another folder called 'dic'. [folder names have to named exactly as stated]
StarDict automatically reads dictionaries that are in:
"media/mmc1/stardict/dic" [as above]
"media/mmc2/stardict/dic"
"~/.stardict/dic"
"usr/share/stardict/dic"


2) Download Collins, Oxford English dictionaries from here:
http://stardict.sourceforge.net/Dict...w.dict.org.php

You want the 'tarball' version which when you download it will be called xxxxxx.tar.bz2 or such.

You can extract this compressed file as follows:
a) If you downloaded it to your Windows machine & have 7-zip installed for example :
- right-click on xxxxxx.tar.bz2 and then click on 'extract here'; this will give you a file called xxxxxx.tar
- right-click on xxxxxx.tar and then click on 'extract here'; this will give you a folder called xxxxxx(containing 3 files).
-Transfer this folder from your Windows machine into your recently created 'dic' folder on your NIT
DONE

b) If you downloaded it to your Linux machine:
- extract it as follows: tar -xjvf xxxxxx.tar.bz2
-Transfer this folder from your Linux machine into your recently created 'dic' folder on your NIT

c) If you download it directly to your NIT into the 'dic' folder just created.
-[could not find a way to extract them in Xterm using 'tar xv xxxxxx.tar.bz2' or such ][anybody know??]
- The easiest way I found to extract them, strangely enough, is to install another dictionary program called WhiteStork
- Once installed, start WhiteStork and from the menu click on 'Dictionaries' and then on 'Load dictionary'. Here select xxxxxx.tar.bz2 that you previously downloaded into your 'dic' folder. WhiteStork will ask you whether you want to 'use it immediately' and whether you want to 'optimize' it: choose 'yes' for both[As a side benefit you now have 2 dictionary programs to access your dictionaries]

I know that the initial set-up is quite a bit of work, however I find that StarDict's ability to (optionally) automatically throw up a definition whenever you highlight something in a program running on the NIT (browser, FBReader etc) is worth the trouble.

I have attached a couple of screenshots of StarDict , with auto 'scan' option turned on, interacting with FBReader and Google News.