Lack of these qualities will be most visible as use
of the environment grows in time and size. On the time dimension, the evolution of
metamodels??”and the parallel evolution of the environment itself??”will be slowed
and soon stopped by the weight of unknown and unintended consequences of changes.
On the size dimension, the tool will face dif?¬?culties in scaling to support multiple
models, modeling languages, and users.
MostDSMenvironment developers have consistently shown a disregard for earlier
work. Of the tools whose ?¬?rst version was released in 2006, not one matches even the
power and simplicity of GraphTalk, released in 1988, let alone that of mature tools
such as MetaEdit+ and GME. Paradoxically, it was the success of UML that
rehabilitated modeling after the disillusionment caused by the failure of CASE, and
yet that same success was instrumental in the decline and fall of many promising
metamodeling tools of the 1990s. Dying before Google and the Wayback Machine,
their experience is effectively lost to current developers, even if they published widely
in journals and conferences.
If there is one area where the new DSM environments must stop and rethink, it is
their meta-metamodels. All are strongly in?¬‚uenced by UML Class Diagrams, a
language for a domain completely different from that of graphical modeling
languages.
Pages:
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764