[comp.unix.large] ph

aglew@crhc.uiuc.edu (Andy Glew) (09/11/90)

Oh, well, I cannot resist...  As people who have read my .signature
over the past few months will know, I am a fan of the "ph" nameserver.
"Ph/mqi" is a nameserver database and mail query system developed at
UIUC CSO.  I am unconnected with ph except as a user.




Ph is a tool for finding out about things.  Mainly other users,
although things like the worth, course calendars, pizza joints, etc.
can also be found under ph.

Ph can be used interactively, eg.
    > aglew ksh dwarfs.crhc.uiuc.edu /usr/dwarfs/aglew 
    > ph glew
    ----------------------------------------
		name: glew andrew forsyth
	       phone: XXXXXXX
	     address: 1608 hunter
		    : urbana il 61801
     office_location: No office; often found at CRHC lab
		    : Coordinated Sciences Lab 6-110
		    : 1101 W Springfield
		    : Urbana IL 61801
	  curriculum: grad
	       hours: CSG lab (CSL 6-110) typically 10am-6pm M-F
	       other: Advisor: W-M Hwu
		    : CSG/CRHC: my group is changing its name.
		    :     Try CSG if CRHC doesn't work, or vice versa.
	    email to: a-glew@uiuc.edu (aglew@dwarfs.csg.uiuc.edu)
    ----------------------------------------
UIUC has ph tied in to the INTERnet "whois" command;
other sites have it tied to finger.


Ph is also coupled to email at UIUC, so, for example, you can email
andy-glew@uiuc.edu.  In fact, given somebody's name and the fact that
they're at UIUC, you've got a reasonable chance of pulling an email
address out of thin air, and getting to them first time, using ph.




Is this a large systems issue?

    Yes - particularly for large distributed systems.  Maintaining
alias files (for email) on hundreds of machines is an impossibility.



Why have I posted this?

    (1) To invite discussion about nameservers and whitepages
utilities like ph.

    (2) To encourage other sites to start using ph.  Ph is not
research; it is a tool used by several sites, at least one of which
has several thousand users (I'm sure I'll be corrected if I'm wrong).
    One aspect of tools like ph is that a variety of fully featured
tools is worse than a wide distribution of a single tool, that might
not have so many features.  Ph may not be ideal (although it is better
than most other whitepages systems that I have been able to use) but
it is working, and used at enough of the systems I connect to outside
of UIUC that I can almost count on it being there.

--
Andy Glew, a-glew@uiuc.edu [get ph nameserver from uxc.cso.uiuc.edu:net/qi]