[Project-ideas] Introduction

[Nicolas Delvaux]

Hi,

You may also want to leverage a spell checker and a grammar checker.

Something useful in my opinion would be to have an easy way to define 
rules.
As I said to you in my last mail, one important thing to take into 
account is the user experience.
Translators and linguists are not software developers and it should be 
easy for them to add and share a rule like "in French, all exclamation 
marks shall be preceded by a narrow no-break space".

Food for thoughts :-)

Cheers,
Nicolas

Le 2013-04-19 11:17, Shaumik Daityari a écrit :

> Hi,
> 
> I found an open source tool, hjerson [0] which searches for the 
> following
> errors in translations:
> 
> - inflectional (morphological) errors, i.e. incorrect word forms
> - reordering errors, i.e. incorrect word order
> - missing words
> - extra words
> - lexical errors, i.e. incorrect lexical choice
> 
> You can find a more detailed description of the tool at [1].
> 
> Would I be going in the right direction by incorporating hjerson in 
> the
> project?
> 
> [0] http://www.dfki.de/~mapo02/hjerson/ [4]
> [1] http://ufal.mff.cuni.cz/pbml/96/art-popovic.pdf [5]
> 
> On Wed, Apr 10, 2013 at 4:30 PM, Runa Bhattacharjee <runabh at gmail.com>
> wrote:
> 
>> On Wed, Apr 10, 2013 at 1:11 PM, Shaumik Daityari 
>> <sdaityari at gmail.com>
>> wrote:
>> 
>>> So basically, what I understand is that both the translated and the
>>> original text would be the inputs, and I need to check for possible
>>> errors in the translations as well as the GUI structure.
>> 
>> 'Checking for completeness' is more like it and that would include
>> checking for errors of omission, commission, validation et al.
>> 
>>> What kind of UI are we talking about here? Is it just the web
>>> interface? Or something else?
>> 
>> As has already been mentioned, deckard implements a smallish
>> representation of this project idea for desktop applications that are
>> built using gtk+. It is limited in terms of using manual intervention
>> for the final leg of validation.
>> 
>> What I'd suggest is to first navigate around a fully localized 
>> desktop
>> environment to get a general overview of the comparison factors. 
>> Based
>> on this the 'translation/localization completeness' can be checked 
>> and
>> then evaluate whether the possible implementation of the idea (that 
>> you
>> may have in mind) can be restricted to desktop apps or can be 
>> extended
>> to web based apps or can be done simultaneously for both.
>> 
>> Runa
>> --
>> http://about.me/runa.bhattacharjee [1]
>> http://fedoraproject.org/wiki/User:Runab [2]
>> _______________________________________________
>> Project-ideas mailing list
>> Project-ideas at lists.ankur.org.in
>> http://lists.ankur.org.in/listinfo.cgi/project-ideas-ankur.org.in [3]
> 
> --
> Shaumik Daityari,
> Integrated M. Tech. (3rd year),
> Geological Technology,
> Department of Earth Sciences,
> Indian Institute of Technology, Roorkee,
> Uttarakhand
> 
> _______________________________________________
> Project-ideas mailing list
> Project-ideas at lists.ankur.org.in
> http://lists.ankur.org.in/listinfo.cgi/project-ideas-ankur.org.in [3]



Links:
------
[1] http://about.me/runa.bhattacharjee
[2] http://fedoraproject.org/wiki/User:Runab
[3] http://lists.ankur.org.in/listinfo.cgi/project-ideas-ankur.org.in
[4] http://www.dfki.de/~mapo02/hjerson/
[5] http://ufal.mff.cuni.cz/pbml/96/art-popovic.pdf