Software Engineering is dead, says Tom deMarco

...in a fascinating paper over here (hat tip Coding Horror for the link). He notes that:

I'm gradually coming to the conclusion that software engineering is an idea whose time has come and gone.

Software development is and always will be somewhat experimental. The actual software construction isn't necessarily experimental, but its conception is. And this is where our focus ought to be. It's where our focus always ought to have been.

Two things that are interesting are:

(i) We told ya so - see here points 4 and 5;

(ii) But, for someone as revered as Tom deMarco to say it is tectonic plate shifting stuff in geekland

In our thought piece linked to above we were extrapolating the trend from Web 2.0, Open Source libraries and self-designing software agents (eg Genetic Algorithms) and we predicted that :

(i) The Software Release cycle would be virtualised - we are already putting together systems from existing libraries of proven code. Software engineering is now about assembly.

(ii) We can imagine code automation increasing, just as 3GLs and then 4GLs and so on automated coding of earlier systems, so we can imagine smarter and smarter algorithms being used to assemble code blocks, and eventually automation of engineering.

After all, its happened in every other type of engineering application (for example a mechanical engineer no longer does real calcs on pressure vessels but uses trusted standards), and as the current crop of technologies become understood, standardised and classified the need for engineering per se goes away. As de Marco notes, proper engineering is always in the new stuff, the experimental end.