The next file format

[Andreas Xavier]

---- On Tue, 26 Aug 2014 13:54:51 -0700 Bruno Coudoin<bruno.coudoin at gcompris.net> wrote ---- 

 > 
 > Le 26/08/2014 22:13, Andreas Cord-Landwehr a écrit : 
 > > So the question is, how important do we rate the use-case of (non-)programmers 
 > > opening the ZIP archives and editing the vocabulary files by hand in 
 > > comparison to the "cost" of another library. (And that together with the quite 
 > > heavy dependency of Boost.) 

I would like to keep the data at rest unzipped to encourage people to look at their own data.
I think open source and open data are co-existing goals.  

What I expect of 1000 users is:

990   - users download a course and use it
   9   - users open the editor and work on a course for themselves/others
   1   - user opens a text editor to adjust something that annoys them
   0   - users download source and fix something.  I think that these users are fictitious :)


 > > With an eye on making the live for porters (to Mac, Windows, Android,..) 
 > > easier and taking into consideration that we are talking about a library and 
 > > not an application, I am in favor of reducing external dependencies where 
 > > possible and reasonable. 
 > Hi, 
 >  
 > Just answering on this point, I agree with you. I am reluctant adding a  
 > Boost dependency just for Yaml. 
 >  

I hear everyone, no Boost.

I still think that YAML is a better solution than JSON.  

If there is way that doesn't add any additional dependencies can we still use YAML
, or do people have other objections?


 > Bruno. 
 > _______________________________________________ 
 > kde-edu mailing list 
 > kde-edu at mail.kde.org 
 > https://mail.kde.org/mailman/listinfo/kde-edu 
 > 


Andreas