[AFRINIC-rpd] Academic IPV4 Allocation Policy - Draft 1

[Guy Antony Halse]

Hi

On Tue 2013-01-15 (21:28), Seun Ojedeji wrote:
>    On the FTE, I don't agree with that, unless we don't want to face the
>    reality on ground.

The concept of full-time equivelency is used as a measure of resource
allocation.  It is not a concept that is unique to higher education, and is
generally well understood.  It goes by many names; FTE is just one.  We must
be care not to confuse the concept with its specific application to
students.

> Part time students are entitled to everything a full time is entitled to
> on campus as it relates to resources.  I don't think

The problem is really the non-contact (distance) students; the students who
are only actually present on a given campus for a very short period (usually
measured in days or weeks per year).

In the case of our specific institution, we have a large number part-time,
non-contact students, mostly in adult continuing education programmes.  They
represent a significant proportion of our "registered students" (13%). 
However, they only represent ~ 5% of our FTE count and they currently only
represent about 1% of the devices registered on campus (and those devices
only appear on our campus network for a few days a year).  They are also, by
definition, not residential students.

This means that there's significant scope for oversell.  The same IP address
could be dynamically assigned to a large number of non-contact students,
provided their visits to the campus do not overlap.

In our case the concept of FTE students is a very simplistic measure of the
number of students one can expect to be active on a campus at any one time. 
In practice it doesn't quite work like that.  However, in practice not all
students have multiple devices (many of ours have none).  Because this
policy is all about averages, that shouldn't matter.

>    On a lighter note,I wonder why we don't really count staff as part of
>    the user..for debate. Those numbers really count, especially if we are

This concerns me too.  The staffing ratio differs widely between HE
institutions.  In places where the ratio is comparitively high (usually
small institutions, like ours), they can represent a significant proprotion
of the total number of people.  Like students you have a part-time vs
full-time problem.  In South Africa we report on SCUs (staff complement
units), being the staff equivelent of a student FTE.

By way of a concrete example, in 2012 we had 7476 "registered students".  We
also have 880 academic and support staff, or about 11% of the number of
users who might use our network.  (I've excluded service staff, who tend not
to use devices requiring IP addresses.)


What we really need is a metric that provides a realistic estimate for the
number of active users who might be on a campus at any time; a realistic
measure of full-time equivelency as it applies to a network.  Unfortunately
"registered students" is not that metric...

If there's an easy way of determining (FTE + SCU) that applies generically
across Africa, that would be the right(tm) way IMHO.

These concepts aren't difficult to codify into policy, nor are they
difficult numbers to compute.  I'd be surprised if other HE institutions
throughtout the region are not already using some sort of full-time
equivelency statistic for their internal resource-planning purposes.

However, the resulting numbers can be very difficult to *prove* unless
you're already working to an audited method.  (Conversely they're very easy
for me to prove, thanks to the Department of Higher Education and Training's
reporting requirements in South Africa.)

- Guy
-- 
Manager: Systems, IT Division, Rhodes University, Grahamstown, South Africa
Email: G.Halse at ru.ac.za   Web: http://mombe.org/   IRC: rm-rf at irc.atrum.org
*** ANSI Standard Disclaimer ***                                    J.A.P.H