From: bmanning@vacation.karoshi.com [mailto:bmanning@vacation.karoshi.com] Sent: Tuesday, February 22, 2000 6:47 AM
bmanning@vacation.karoshi.com Sent: Monday, February 21, 2000 5:11 PM
While I think that Bill Manning's DNS TXT suggestion is clever, and nicely distributed, it requires a lot of effort.
actually, I think both efforts have about the same level of effort. In one case, its concentrated in a single place, in the other, its all over the place.
The nice thing about central is that things will get done, or not done, consistantly. You tend to strike a closure point at some time. In the distributed "all over the place" model, things may never be
consistent, nor
will they ever be complete. It is also difficult to enforce quality standards.
Hum, is this an argument for reconsituting "Ma Bell"?
No it isn't, you mistake my intentions. I was just weighing out each end of the spectrum.
One of the strengths of the Internet is its distributed nature.
Tell that to DOC/NTIA/ICANN. I'm with Stef and the ORSC in this, aka; the opposition (see: www.dnso.net).
And the trend is in this direction with the addition of new IP aware products (cell phones, networked "gadgets", et.al.) Trying to retain a centralized structure is (IMHO) doomed to failure. Its better to have broadly available standards that can be enforced at the provider/subscriber boundary and then let the market "bloom" rather than have a single forcing function that everything must run through before proper operations can occur.
Finding that sweet-spot has been elusive.
We might as well argue for the reconstitution of the InterNIC and the abolishment of RIPE, APNIC, ARIN. You'll get "things ... done, or not done consistantly." and will "strike a closure point at some time".
Actually, RIPE/APINIC/ARIN is too centralized for my taste. <grin>
Striking the balance is hard but I expect that the trend is away from centralized services.
Not according to DOC/NTIA/ICANN <sigh>.