> Most models of software development, as backed up by my own personal
experience, indicte that requirements/specification work, not coding,
is the dominant programming actvity. Coding only accounts for at most
25% of development time.
I am not sure what area you are working in but for me coders do not exist. such posts died out in the 1970s when coding sheets and punch cards fell out of use. The people I employ are programmers and this for me is part analysis, part design, part coding and part unit testing. We are not working directly on customer applications so I guess we might be a small select bunch. Saying that coding takes only 25% is for me a useless statistic since my people use source code as a medium in which to experiment with design and to a certain but significantly less analysis. In effect we experiment with ideas using code as the medium of expression iterating to a solution to a problem and then we refactor (not necessarily using the overworked, over-hyped?, XP semantics to this word) to produce production quality systems. The source code is therefore an automatic output from the analysis and design process.
> People who are really good at writing clear, well structured code are
about the least valuable members of a programming team. They're the
ones we hire as contractors and let go as soon as the coding phase is
over. A good testor is worth three or four of a good coder.
I fully agree about testing but I cannot see the rest of this since a good programmer is a person who writes good unit tests for their code. The two go hand in hand and cannot be separated.
> The people we hang on to even when times are tough are the ones who can design useful applications that are feasible to develop. Such people do know a great deal about coding; otherwise, they couldn't design things that were feasible to implement. More important, though, they understand application requirements and can translate these into software architectures.
I can certainly connect with these sentiments but my problem is I cannot see why you have "coders" at all.
> If I could determine the direction of software psychology research, it would be to study this software design process. Unfortunately, it's not easy to study. It's pretty easy to write several programs that solve the same problem; it's much harder to come up with several problems that are different in content yet equivalent in complexity. Still, as long as the requirements gathering and design process dominates software development, any knowledge that can be gathered in this area is immensely valuable.
I am no psychologist but I can say that the best software developers I know all have an excellent grounding, albeit informal and often indirect, in what is in effect epistemology. Scientist learn about scientific method in order to codify what it is they know; how to structure an experiment in order that the facts can be turned into information and knowledge. Far too many computer science courses ignore this and turn out people who can write code but cannot debug it. These people are generally also the ones who cannot design a software system. The core problem is that they cannot reason about what it is they know and can know from the code and the test cases written. This is sad. If more universities focussed on this in the computer sciences courses we would have more good programmers.
For me this is why I think you value good testers, they understand what is known and what can be known about the program they are dealing with.
-- Russel. ===================================================================== Dr Russel Winder +44 20 7585 2200 41 Buckmaster Road +44 7770 465 077 London SW11 1EN, UK [EMAIL PROTECTED] |
-- Russel. ==================================================================== Dr Russel Winder Chief Technology Officer OneEighty Software Ltd Tel: +44 20 8680 8712 Cygnet House Fax: +44 20 8680 8453 12-14 Sydenham Road [EMAIL PROTECTED] Croydon, Surrey CR9 2ET, UK http://www.180sw.com ==================================================================== Under the Regulation of Investigatory Powers (RIP) Act 2000 together with any and all Regulations in force pursuant to the Act One Eighty Software Ltd reserves the right to monitor any or all incoming or outgoing communications as provided for under the Act |
PGP signature