[Gllug] Universities and degrees in computing.

[Sean Tohill]

On Sat, 28 Dec 2002, Xander D Harkness wrote:

> My brother who has a US degree is seeking to do a PhD in the UK in 
> computing.
> 
> What universities would people recommend?
> 
> It needs to be Internationally recognised and not just the latest fad.
> 

It's not as easy as picking somewhere to do a BSc, there are many more
problems if he ends up doing something he doesn't want to do.

Some questions he needs to ask himself:
why? - there will be times when the research or other tasks becomes
difficult, a drag or generally onerous. a good reason to be there and
suffering may help. motivations will be different for different people.

does he have the qualifications to do a PhD? usually a first, maybe an
upper second. some departments want the applicant to do their MSc first,
which adds an extra year. this is due to the variable skills graduates
have. hopefully they may know what a computer is after the MSc :) 

does he have the money or does he need a bursary or grant? having the
money opens the doors of most universities, especially if he is classed as
overseas - they pay more. a good candidate who can support themselves
should be able to negotiate reduced or even no fees. times are tough for
students and less want to do PhD's now. 
a bursary or grant will be in the control of an academic and it will be
for research in a particular area, usually of interest to that academic
and maybe even tied to a larger research project.  will need to get on
with that academic for 3 years. there is only one winner if they fall out.

what area does he want to research in? this is tricky because the research
will need to be very specific and he may only have a rough idea. for
instance wanting to do research into networks will not be good enough. it
needs to be more tightly focused than that, even at an early stage. 

Which university? this is quite difficult unless he doesn't care what he
researches into. if he doesn't care then he just needs to surf around a
few web sites and see what takes his fancy. if he was my brother and
didn't care i would point him at the following: edinburgh, manchester,
imperial, southampton, cambridge, maybe oxford. they are all strong in
many areas. there may be others but those would be my preference. if he
has some idea what area he wants to research then it becomes harder. for
instance UCL are very good at many aspects of research into networking,
Queen Mary's in distributed and parallel computing. other universities may
have built a reputation in other areas. the web pages and in particular
the number of published papers a research group has produced (and if
recent) will help here.

What area now? given he has now done some research into what different
groups are looking at he may have more of an idea of the research area he
wants to look at. taking the networks as an example he may find various
themes such as: applying formal methods to network protocols, simulation
of WAN's, performance aspects of fiber, ......... he should read some of
the papers in a sub area that interests him and decide if it the sort of
thing he wants to do. if yes contact the research admissions tutor.

Potential problems. 
being very clever doesn't mean a Professor has good supervisor skills or
even any people skills at all. the biggest gripe research students have is
that their supervisor is crap, or words to that effect. the only way to find 
out is by talking to some of his/her current students. 
also the supervisor on paper may not be the actual day to day supervisor.

supervisor leaves to a better job elsewhere. this happens but a good guide
is to find out about the turnover of the staff. if they are happy they
won't be looking to leave. it is not unknown for several staff to leave
for another organisation leaving no one to supervise certain stuents.

you are just the dogsbody of a Professor, reader etc. this happens too. if
they want a particular aspect to be researched then you will be made to do
it. they won't be interested in you going off at a tangent, especially if you
have a grant or bursary associated with a specific project. all funded
projects will have deliverables and not delivering will penalise the
department when they apply for funding in the future. it is not unknown
for some academics to expect research students to act as unpaid tutors or
even to take on some work such as research group sys admin, latex expert,
network fixer ....... - again usually unpaid but expected.

other thoughts. try to get a copy of a book by phillips and pugh (i think)
called something like how to get a PhD, or maybe how not to get a PhD.
try to find out the success rates for departments and individual
supervisors. this may be difficult to find out, but very valuable. current
research students or better still some who have just completed might be
able to tell you this. ask them about what sort of culture the group or
department has too. some places are nicer to work in, have happier people
etc.

hope this wasn't too boring. if you get along to any of the gllug events
feel free to ask me questions.

sean



> Kind regards
> Xander
> 
> -- 
> Unix soit qui mal y pense
> 	[Unix to him who evil thinks?]
> 
> 
> 
> -- 
> Gllug mailing list  -  Gllug at linux.co.uk
> http://list.ftech.net/mailman/listinfo/gllug
> 


-- 
Gllug mailing list  -  Gllug at linux.co.uk
http://list.ftech.net/mailman/listinfo/gllug