In message <199509020524.WAA28298@MindBender.HeadCandy.com>,
	"Michael L. VanLoon -- HeadCandy.com" writes:
>On the other hand, it suffers badly when scaled to large sites.  So, I
>could also make the argument that more small sites use YP, but the
>largest sites use Hesiod.  Now, how do we count which is used more?
>:-)
[...]
>Iowa State has roughly 850 Unix workstations in a homogonous,
>distributed network, with roughly 25,000 user accounts, all
>coordinated comfortably by Hesiod databases, and authenticated with
>two Kerberos servers.  And that doesn't count the Eudorized PC's and
>Macs all over the place, or Linux/FreeBSD/NetBSD boxes connected to
>the ethernet in peoples' dorm rooms.  Last I heard, MIT had around
>1500 (probably more now) Unix workstations, with a similar number
>accounts.  I have no idea how many machines DEC (an Athena proponent)
>has on their internal net, but I'm sure it's a large number.  I just
>don't see any sites this size using YP (or at least doing so without
>cursing everyone in site).

Just to add a case study, for what it's worth:

I am currently sitting at a Major Corporation (one who's products
most of you are probably using right now), who has extensive networks,
and uses YP for everything.

I've spent most of my time fixing problems with their YP.  Would something
else work better?  Most certainly.  Do I still prefer YP?  Most
certainly.  Would they ever switch?  Absolutely not.

--
Jason Downs           | GCS/CM -d+ H(+++) s+:++ !g p? !au w+ v-(*)
downsj@xenotropic.com | C++(+++) UBAVHS++++$ P--- E--- N+++ W--- M-- V--
---------------------/ \-------------------------------------------------------
     Only the Feds could save children from abuse by burning them to death.