As former academic and current industrial developer, I suspect that the opportunities 
for academic/industrial collaboration are very
large, even for topics relevant to professional programmers and software engineers.  
The key or trick is recognize the impact on the
industrial organization of the selected research methodology and ensure, if possible, 
that the research has some direct, immediate benefit
to the industrial organization. 

For example, if what you want to do is take 50 experienced programmers, do a two day 
training, and then run a 4 hour study, your best bet is to 
ask a national research funding organization for the money and then hire the 
programmers yourself as contractors.   On the other hand, if what you say is that 
you'd like to send in three trained observers and have them observe 50 programmers for 
a week each to find out what actually consumes their time, tell me when you expect 
them to arrive and I'll see that my organization arranges office space and telephones 
for them. (You'll still have to find funding for their salaries and other expenses.)

Why the difference?  Methodology 1 translates into a 100% sure two and half day 
schedule slip; for most software managers who already dealing with
unreasonable schedules, agreeing to a two day scheduel slip is equivalent to agreeing 
to a salary cut.  Furthermore, at best, it will yield a handful of  superlative 
comparision, a is better than b, c is better than d, that may or may not be relevant 
to the organization in question, but, in the big
picture are going to be but single bricks in a very large wall.

 Methodolgy 2 might actually cost as much programmer time as methodology 1, but it's 
less likely to produce an overall schedule slip, and it might yield multiple ways in 
which that particular industrial organization can improve its practices.   In fact, 
for the manager in question, it might give
her/him the leaverage for their next promotion.

I recognize that using anything other than a classical experimental psychology with 
all the statistical trappings may create problems in
the home department of the reseacher, particularly if that department does not have 
enough social science expertise to recognize that
ethnomethodological and verbal protocol studies are well accepted approaches to 
behavioral research.  On the other hand, the ease of 
publishing this kind of research should convince even uninformed colleagues that this 
is a valid research approach.

Ruven Brooks

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Thomas Green
Sent: Monday, December 08, 2003 4:49 PM
To: [EMAIL PROTECTED]
Subject: Re: Penny for the guy: was RE: PPIG discuss: Loss of context experienced by 
programmers


Umm. The rift between academic and industrial reappears .... Here's something to bear 
in mind, maybe?

The rewards, intrinsic motivations, opportunities for action, and sources of esteem 
for academics and industrial people seem to be often very different. It doesn't make a 
lot of sense to criticise "them" for being different from "us". So don't let's let 
ourselves slip towards blame games. 

Opportunities are limited on both sides. Obviously, industrial people have to make 
some money, and may not have the opportunity to support research. Perhaps less 
obviously, most academics have training and experience in only one area of research. 
They don't have the opportunity to do any other kind, even if what they're trained in 
turns out not to be the most "relevant to real life" as they say.

The other thing to remember is that many different people are doing many different 
things in "real life" and therefore it's pretty well impossible to say that a piece of 
academic research is completely useless. "Real life" is not just professional 
programmers and software engineers dealing with horrendous amounts of code, it's also 
hobbyists and non-specialists and spreadsheet users and scripters and students and so 
on. We all need to be a little careful of making generalisations from our own everyday 
situation, and of judging outcomes only by one criterion.

We can't create collaboration by sheer will power. The phrase that goes around these 
days is evidence-based practice, in which the people who are actually doing something 
in the world make their decisions on the current state of the evidence. This can be 
encouraged by funding academics to create useful summaries of the current state of the 
evidence, and of course by creating research initiatives, funded at least partly from 
outside academia. This seems to work in some health-related areas, although I don't 
know in detail how well it works.
 
PPIG people have made attempts in these directions but there have been problems 
(though Derek Jones's mighty tome is maybe a success). If any of you folk out there 
see your way to furthering the cause of evidence-based practice, that would be good 
news and I'm sure it would be well received by both industrialist and academic types.
 
----------------------------------------------------------------------
PPIG Discuss List ([EMAIL PROTECTED])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

Reply via email to