[Project-ideas] [Anubad] Glossary Tool to index available strings

[Sankarshan Mukhopadhyay]

On Thu, Mar 29, 2012 at 2:38 AM, Arani Bhattacharya <arani89 at gmail.com> wrote:

> Thank you for your feedback. After some introspection and discussion,
> I felt that we can find a solution in the following way. First, we
> will have an index of words on some server. The user, while entering
> some text, can copy some word to the clipboard, and then by clicking
> some icon can check if the spelling is correct and, if wrong, then
> obtain suggestions (which will be obtained from central server). In
> case the user is convinced that his/her spelling is correct, but it is
> not present on the central server, the user can add this word to the
> central server's dictionary. On the server, a group of approved people
> can, whenever convenient, check what words have been added, and either
> approve or reject them one by one.

This is something that has promise if you opt to think along further
about the possible combinations of usage. And, whether the review of
the words can be made easier. A glossary is not merely about the
correct spellings. Here's a small example - consider the phrase "log
in". A somewhat ubiquitous and well worn phrase that everyone is
familiar with. And yet, various pieces of software could have
different ways of translations and, newer pieces could have more
different ways. For example, it could be different on GNOME, KDE,
Facebook and Twitter. And so, there has to be a way to keep track of
all this diversity and present the translator with a form of choice.

> One possible problem with this solution is that using the application
> might be too cumbersome for the user (with the method specified
> above). First, the user has to copy the text, and only then can he/she
> use the application (in my opinion, monitoring what the user is typing
> raises too many security issues) by clicking on some icon present. I
> would be grateful if someone could suggest a simpler use-case.
>
> I would be grateful for the community's comments on this idea.

The underlying premise of the idea sounds good. An aspect that you'd
probably need to think in addition with the above is the 'offline
mode'. A lot of translation happens in a disconnected manner ie. the
tool/hardware is not connected to an active network. The ability to
use the input from the user when translating and do in a somewhat
non-invasive way the changes and suggestions is a great one to have.
The evolution/augmentation of the glossary can happen in both offline
and online modes.

At the risk of repeating myself, please have a formal proposal in the
system. Having more discussions on the page/system helps the
organization and, the proposal lifecycle as well.


-- 
sankarshan mukhopadhyay
<http://sankarshan.randomink.org/blog/>