I agree with both your statements, I fully believe in the importance of
running code - from early on. Question just is where to get the necessary
resources from...
I agree with Alberto. These are the same priorities. One goes faster in
having models when one discusses theory, and running code when developping
standards. This is the basics of cybernetics (improving efficiency through
analogical thinking based upon models obtained from feed-back).
Let me also second Alex comment encouraging you to post to the list - I
think the entire WG can benefit from your experience with ICAP and your
thoughts on the process.
To address this and the list problem, I would simply suggest that we turn
the default respondto to the list. From personnal experience most of the "a
parte" are just due to the cumbersome "reply to all, remove all execpt
list" one keeps forgetting. We might still have delays but at least in
sequence.
My 1.78 eurocents
jfc
Thanks
Markus
I agree with both
Alex Rousskov wrote:
On Wed, 14 May 2003, Alberto Cerpa wrote:
Since you guys are following this path, I would like to provide my 2
cents. Try to write a reference implementation at the same time that
you are writing the OCP/OCPTRAN draft, even if it takes more time.
Having running code is of invaluable help when trying to debug some
protocol issues in the draft. It will be worth it in the long run.
I agree. I do not know whether there will be enough
demand/interest/resources to write a reference OCP library, but we
should definitely keep that in mind. As the first step, we need to
agree on basic message structure and encoding principles, before we
can start writing any code. I will try to post a rough draft ASAP to
keep us moving.
Alex.
P.S. Please post to the list if possible.
---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.478 / Virus Database: 275 - Release Date: 06/05/03