[ Home | Contents | Search | Post | Reply | Next | Previous | Up ]
From: Why pick on SSADM?
Remote Name: 188.8.131.52
Not saying SSADM doesn't suck, but it seems to me that is only an extreme example of all that it wrong with development methodologies in general.
They all seem to have at their root some variation on the 'infinite monkey theorum'. That is to say, they assume that one need only find an arbitrary number of random dolts (the exact number may or may not be specified by the mehodology), ensure their slavish adherence to some set of methods and procedures, and a superrb system will result:-
HOW TO BUILD A SPACE SHUTTLE: Step 1 - Perform feasability study Step 2 - Analyse requirements Step 3.....
This assumption does not consider important a) the business, political and economic context in which development of a given system occurs b) the talent, creativity and motivation of the individuals doing the developing Without taking these factors into account, any discussion of which particular diagramming technique to use or what review process to employ is irrelevant. These are highly unlikely to be decisive factors regarding the success or failure of a project.
Most meths (the ones I have seen) gloss over the stuff that's actually HARD i.e. the actual analysis and coding. Dogmatic adherence to any given methodology seems to me to be no substitute for genuine analytical ability and ingenuity. Miss these, and you'll get a bag of nails, SSADM or otherwise.