Re: Alloy paper



Date view Thread view Subject view Author view

Frank V. Castellucci (frankc@colconsulting.com)
Mon, 31 Jul 2000 12:20:58 -0400


Alan Cameron Wills wrote: > > Politics is the art of the possible. "Cry Haddock and let slip the Frogs of War!" :) > "Frank V. Castellucci" wrote: > [posit agreement snip] > I feel desperately misunderstood; doubtless my fault. > > I don't at all say "give over to the current winner". I say "work with the current 'winner', bending it gradually until it encompasses all our good ideas, and leaves behind its old bad ones". By the time we're finished, it will be just like what we originally thought > of, except it will be better. Yes, but in what, ten years time? If it is better now, then why not do both: Pursue adoption of valuable aspects into UML, while enhancing the base. Extend the base exposure through usage that extols it's benefits in a readily obvious way. > Yes, better. Because no matter what language any of us invents from scratch, it will always have shortcomings from the point of view of one user or another. Therefore any language is bound to be bad to begin with (from some points of view) and gradually improve with > time, provided that all the other other folk work together around it. Which is why the solicitation of feedback. Assuming that the domain has been analyzed, it is the realization into a form we are comfortable with that the opportunity exists. I am and will continue to use UML for the recognition factor, and adoption into the customer base I choose to service. BUT: UML is not a panacea, that is clear, and no matter how much time and effort is put in for the standard to become more universally correct, it will have been redefined. > By contrast, if people just say "this is hopeless, I'm going to invent my own", then we'll never have any common languages. We are in full agreement, but I ask you to consider the other intelligent option: investigate what else is already being done that does capture the ideals from the start and work with that. > > Now with the open source/free software tools arriving for UML, many of > > the project authors would really have no problem with adding a module to > > work with other methodologies and notation. > > Yes, but then whenever anyone changes jobs, they're going to have to learn different methods and languages: most of which are pretty certain to be someone's own local invention and not very good. That's the situation we're in at the moment, and it's a waste of > everyone's time and money. There are better things to spend our time on! I can't agree with that. My job is to provide the customer with support in their domain. If it is UML, I am UML, if it is Booch, I am Booch. If we can't arrive at a standard physics model, why do we persist in believing there is some universal salve for modeling? > > In the end, it isn't a question of one language being the 'winner'. The best thing about UML is not that it has any technical merit, but that there is a large number of people willing to agree to work with and on it. > Without getting into the social or economic disertation on this, I will posit that a uniform language in context is transient. > Alan > Frank


Date view Thread view Subject view Author view