code in cvs is broken

[Linas Vepstas]

On Fri, Nov 16, 2001 at 03:29:53PM -0600, Rob Browning was heard to remark:
> linas@linas.org (Linas Vepstas) writes:
> 
> > There doesn't seem to be any strong technical reason for a call-back
> > design between c and scm in gnucash;  in most cases, it seems that it 
> > would be just as easy/quick/efficient to use scheme to push data into 
> > the c code.  Right?
> 
> In general I agree that simpler is better, and and so I might often
> side with you on this one, but I'm not certain enough about our needs
> ATM to be sure that calling from C to scheme won't ever be the best
> answer to a given problem.  

I didn't ask for a never-ever commitment. Just a 'usually' commitment.

> In fact, we already call from C to scheme
> quite a bit, 

yes, I noticed ...

> and I suspect there will continue to be many cases where
> doing so is the most appropriate solution.

Is it obvious that its the 'most appropriate'? That is what I was 
questioning. A lot of the cases seem to be 'gee it was the first 
thing I thought of', rather than a 'this is the best way'. 
Loading the currencies in the engine being my pet example.

> As a particular concrete example, consider the GTK code.  Until and

OK, well, yes, for gui code, yes;  I forgot about that.
So maybe I have to limit my comments to engine code ... 

--linas

-- 
pub  1024D/01045933 2001-02-01 Linas Vepstas (Labas!) <linas@linas.org>
PGP Key fingerprint = 8305 2521 6000 0B5E 8984  3F54 64A9 9A82 0104 5933