The ?¬?rst testing
was performed after having de?¬?ned over half of the language, so its basic structure was
clear. The second testing was done when the language was already de?¬?ned, something
you may consider a beta version. Most of the feedback and comments were related to
code generation: ?¬?nding a consensus among the four programmers involved on a good
structure for the Python code to be generated. Finding good structures for the code was
not just for the sake of DSM as that was needed anyway. Most testing work was done
by the language de?¬?ner by modeling about 10 small to midsize mobile applications
using various constructs of the language. The language for Python for S60 was
relatively easy to test as the domain could be examined by running the applications
within the mobile phone, either in the real target device or in a PC-based phone
emulator.
8.3 LANGUAGE FOR APPLICATION MODELING
The objective of the language was to make application development as easy and
natural as possible. The approach to achieve this was by using modeling concepts
directly based on the phone??™s services and UI widgets and by following the already
used ???cartoon-style??? of UI application speci?¬?cations. Thus, domain concepts such as
???Sending text message,??? ???Note,??? ???Form,??? ???Pop-up,??? ???Browsing web??? became
candidate concepts for the language.
Pages:
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311