How Program History Can Improve Code Completion
Romain Robbes ~ Michele Lanza
ASE 2008 (12% acceptance) ~ to appear ~ IEEE CS, 2008
his pageweb is not up to date
I can send you his Phd
On Dec 1, 2008, at 12:50 PM, Damien Cassou wrote:
On Sun, Nov 30, 2008 at 7:16 PM, Stéphane Ducasse
<[EMAIL PROTECTED]> wrote:
I wanted to share that with you. I'm reading the PhD of Romain
Robbes and
he has an excellent chapter on measuring different strategies for
code
completion.
One of the best algorithm is to keep a list with the recently
modified
methods
and the messages they send.
His benchmark shows a drastic improvement even compared with
ecompletion
with RoelTyper.
So david do you store the recently modified methods globally for
all the
browsers?
Is it easily accessible?
He also mentioned that we get even better results by combining
RoelTyper and then this approach... any taker? this is important to
have
much better
editing tools.
I guess Romain already did something to be able to compare the
results. Where is his work?
--
Damien Cassou
http://damiencassou.seasidehosting.st
_______________________________________________
Pharo-project mailing list
[email protected]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
_______________________________________________
Pharo-project mailing list
[email protected]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
