- From: Urs Eppenberger <Urs.Eppenberger@switch.ch>
- Date: Tue, 08 Dec 1998 19:44:49 +0100
- To: discuss@apps.ietf.org
I observed that we have often recurring discussions in a BOF on what a working group should tackle. The participants have interest in various areas, generally there is not time enough to do a thorough job in all areas and in the right sequence. In the current, complex Internet world a working group needs to go through a number of steps: 1 Document what is around already. Challenge: make it complete and finish at a certain time. 2 Identify the problems which need to be solved. Challenge: agree on the problems 3 Collect the requirements: Challenge: keep this to a minimum to allow a solution 5 Define the architecture: Challenge: agree on a common view for the future 6 Specify the protocols: Challenge: find dedicated authors and keep community interested until the very end In earlier times it was OK to roughly touch 2 and then directly jump to 6. All project management theories (or software development strategies) show something among the lines of the 6 steps above and definitely warn to ignora step. This takes time and needs a lot of work. But since the world got more complex, we need better thought out solutions, and following the easiest rules of software development would help. Urs Eppenberger, SWITCH
Received on Tuesday, 8 December 1998 13:46:10 UTC