The next morning the whole group gets to ooh and aah over the pretty symbols, but
the mood is generally tense. The meeting is scheduled for shortly after lunch, and so
far the generators seem far from producing full working code. Looking at the code
produced and imagining the time it would take to write the rest by hand and get it
working, the team is feeling understandably nervous. Quite often, on the second day
some team members have to go off to do other urgent tasks, but the scope for parallel
or group work at this point is more limited anyway.
One developer generally works on the generators, running them on the model and
comparing the output with the known good application code.Asecond developer may
work on a thin domain framework layer, writing functions to reduce duplication in the
generated code. If there are other developers, they may be set to work normalizing the
known good application code into the more consistent format the generator will
produce, debugging problems in the current code produced by the generators, or
writing batch or make ?¬?les that will build the code into an executable.
If all goes according to plan, lunch will be a well-earned respite, safe in the
knowledge that the code compiles and runs perfectly, and all that remains is to polish
up the demo for the presentation.
Pages:
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634