> I think a better long-term story is to have a documentation repository
 > where we can place -- and update -- system design information for
 > future people who have to maintain this stuff.  Scattering it on
 > obliquely named project pages (how far can I see on a "Clearview"? and
 > are avian carriers involved in "Volo"?) doesn't help organize or
 > explain anything.

I think the naming issue is a separate matter.  Further, I think even with
the oblique names today, there is sufficient tie-in to allow developers to
find what they need -- e.g., both Clearview and Volo purposely used their
project names in the changeset log comments so that future developers can
connect the dots and pull up the relevant background behind a particular
code change.  That is quite valuable.

 > Projects make sense to me while they're going enterprises and still
 > doing useful work.  I'm much less convinced that they're of lasting
 > value historically.

I disagree -- I have often found decade-old documentation (when I can find
it) useful for understanding the rationale behind non-obvious changes.

 > But even where they do have useful information, we still need to avoid
 > having it just pile up in the middle of the road where it's at least
 > slowing traffic, if not yet causing accidents.

How is the project page "slowing traffic"?  Are these really the long pole
in the website migration project?

I'm all for disposing of pages for abandoned projects, and for updating
the pages for delivered projects to make it clear it is only of historical
note.  However, purging them and moving their content to somewhere else
seems of limited utility, and makes it harder for people to find relevant
content.

-- 
meem
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to