alfs at linux2themax.com
Tue Nov 22 12:17:44 PST 2005
I suggest 2 things.
1) daemon be done in C for right now as a proof of concept and clients
be done in C & perl/python?
2) make 2 proof of concept's one daemon in c one in perl/python.... and
one of each for the client.
I agree with Gerard's prevous comment about xml im not overly fond of it
either. So far the syntax is a bit odd in how it works(or not),
documentation may fix this i dont know(still working on that part).
In about a weeks time I hope to start learning C my self and will see if
I can start contributing here more.
as for the authentication. is it posible to use say 2 certs from ssl to
be a verificataion for server/client. not nessasarly send the commands
through ssl but if a ssl connection is formed and a couple
instructions/data is properly passed between the two its authenticated.
or evin ssh keys or somthing
Gerard Beekmans wrote:
> Your Python code is probably as cryptic to me as C code is to you. In
> the end that can't be helped. Everybody prefers their language of choice.
> I can see benefits from a scripting language in that it saves having
> to recompile every time you make a change. C is kind of annoying for
> that. I'm not concerned with speed penalties of scripted languages
> (perl and python seem to be our two most supported choices) vs. C or C++.
> I don't think we can really make a choice based on which is more
> powerful. All languages get the job done. In the end it's a matter of
> preference I think.
> I don't mind holding a vote, but I can see it happening already:
> everybody prefers their own languages and we don't get anywhere still.
> Thomas, what are your thoughts on this? You said you were comfortable
> with both C and Python. Any preference as to what you think the
> official ALFS programs should be coded in?
More information about the alfs-discuss