Thomas, Ruven,

>> I think programming performance is still dominated by ability, not training.
>
>I don't think it's training, as such, but experience, especially experience
>that has been thought about and brooded over. And there have been many
>studies in many branches of skills showing that experts (i.e. people with
>many, many hours of experience) regularly have large and well-practised
>cognitive structurings available that novices don't have.

Ericsson's deliberate practice paper makes this case very well.

> This has been
>demonstrated for programming too.

As I understand the research it has shown that one group of people
do things differently than another.  Which is not quiet what Ericsson
was getting at.

My point about the difference between experts in chess and programming
is about the levels of expertise.

In chess we have the traditional pyramid.  World champion at the top,
then International Grand Masters, Grand Masters, and so on down until
at the very bottom there are the likes of me.  There is a qualitative difference
between the people at these various levels.

In software development there tend to be two levels.  Those who can do
it and those who are good at it.  My argument is that there is not the
incentive, economic structure, or method of measuring performance
needed to create the pyramid of excellence structure seen in chess,
violin playing, etc.

For instance.  Say there are half a dozen companies making widgets,
employing 10 developers each.  There will be a mixture of abilities and
a small number of these 60 people will considered the experts.  Where
is the incentive to become a better widget software developer?  Personal
goals, yes.  How is a more expert widget developer measured, how
much opportunity is there to learn from what others know, how long will
the company continue to employ a widget developer wants the product
is complete?  Where is the large pool of talent spending investing the
10 years of deliberate practice?

 From Ruven Brooks email
>programmer on our team almost invariably begins a discussion of how to implement a 
>new requirement with "well, on the xxx project we did it this way because ... and we 
>encountered this kind of problem."  He also frequently reuses old programs to produce 
>new ones - "I took
>the program I wrote to do N and modified the input so that it now does M."

This is not the sign of a real expert.  He only has his own experience to
make use of.  Listening to chess commentaries on the TV (I don't know any
grand masters) they always seem to be talking about what other people have
done, more than what they have done.  It is certainly my experience of
recognised experts that they know a lot about what others have done.

> I'm not too happy about any comparison between the relative effects of
>ability and experience. I frankly don't know how that comparison would make
>sense (takes us straight back to the nature-nurture controversy). And I
>don't think it has great importance in real life - or does it? As a
>lab-based psych, I stand to be corrected there.

I think that the current state software development is such that the framework
necessary to create the 'true experts' in the sense used by the expertise
researchers does not exist.  What we have is a two tier system of plodders
and plodders with ability.  A very truncated pyramid.

Will such an environment ever exist?  Ten years ago there were 10+ word
processing programs.  Now there is Word and a few niche products.  Where
is the opportunity to become a true expert in writing word processors?  The
same can be said of other applications.  I think writing software is a temporary
blip.  Once we get over this blip there will be time to become experts in
maintenance.  But this is another issue...


derek

--
Derek M Jones                                           tel: +44 (0) 1252 520 667
Knowledge Software Ltd                            mailto:[EMAIL PROTECTED]
Applications Standards Conformance Testing   http://www.knosof.co.uk



- Automatic footer for [EMAIL PROTECTED] ----------------------------------
To unsubscribe from this list, mail [EMAIL PROTECTED]  unsubscribe discuss
To join the announcements list, mail [EMAIL PROTECTED] subscribe announce
To receive a help file, mail [EMAIL PROTECTED]         help
This list is archived at http://www.mail-archive.com/discuss%40ppig.org/
If you have any problems or questions, please mail [EMAIL PROTECTED]

Reply via email to