Slashdot Mirror

http://www.cetus-links.org/

database.store(object);

What is small? If you rather would deliver products without bugs, and there is a huge payoff here, you should have a look at Eiffel. It really supports a better style of programming. There are two commercial Eiffel compilers and one Open Source compiler. There must be hundreds of companies using it and thousands of programmers. You ain't gonna get a competitive edge if you use the same language as everyone else.

If you are a startup, you usually don;'t have a second chance. You don't have the luxury of Microsoft or Linux (SSH/sendmail/wu-ftp/...) programmers, who can use C/C++ and get away with buggy or insecure code and patch later. Will the programmers in your company produce better code with C++ than Microsoft does? That's hardly likely. Will they with a pure, clean language that supports Design by Contract?

Take for example the interview with Bjarne Stroustrup:

My rule of thumb is that you should have a real class with an interface and a hidden representation if and only if you can consider an invariant for the class.

That is exactly Eiffel's position. But not only that, it allows you to write that invariant and it will make sure that upon violating that invariant you get an exception. That really helps, instead of having some abstract invariant or precondition that is burried in the documentation and which you only detect when your 500 million dollar rocket has blown up.

And Eiffel is infectious. People who use it and who got its ideas, don't even want to go back to inferior languages.

But for more information see the Eiffel Cetus pages.

Eiffel is a very powerful OO language. There are tons of links about Eiffel at: http://www.cetus-links.org/oo_eiffel.html. Check them out!

While this is a good and useful book, it does not do much teaching of OO design. If you don't already have a good understanding of OO analysis and design then the text becomes more of a superficial cookbook than a source of theoretical underpinnings.

You would do better with Object-Oriented Software Engineering: A Use Case Driven Approach, by Ivar Jacobson, or Object-Oriented Analysis and Design With Applications, 2nd Edition, by Grady Booch

There's more good info here.

Although that is the biggest problem I have noticed, I don't think its a problem of OOAD at all. You run into this problem using any programming method. Its not a problem of OOAD at all.

OOAD can be applied to problems in any field. However, OOAD does not solve any problems itself. Its a method modeling and developing a solution. OOAD is not specific to any programing language in particular, think of it more like a methodology. You still need to create an efficetve solution. Its a tool for doing that, not a solution in and of itself.

If you can map out some goals of what you want, I don't see why you should not be able to produce an effective design to do it. There are many sources of information about this (you can find links to quite a few on http://www.cetus-links.org)

If your not comfortable with object oriented design, then I would suggest that the problem may be in choosing a problem solving approach you are not comfortable with; and not a problem with OOAD itself. If the people who will be ultimately responible for maintaining this software are not OO people, then it doesn't seem like the right tool for the job. You can't attribute that problem to OOAD, though.

ISE Eiffel (CTO is Bertrand Meyer) are the inventors of Eiffel, and they've got the best-known, commercial IDE+compiler.

Pebbles are a feature of their IDE, and there's support for in their GUI library, but that's it.

Here are a lot of useful Eiffel links, including links to the free OS SmallEiffel compiler.

I just heard someone say that they found an old procedural program of theirs that used too many global variables and too many parameters. Rather than blame his bad programming or lack of knowledge about procedural/relational organization techniques, he blamed the paradigm and used it as a sorry excuse to proceed with OOP.

Ironic given his very long rant essentially pointed at bad OOP implementations rather than OOP as a paradigm!

Of course, I run Cetus Links so I may be biased...

--

Self: a little outdated now but there are still some interesting ideas in it (many of them were used to develop Java). You can find it somewhere on sun's website.

SELF wasn't really a big new paradigm, though. It worked on a few interesting ideas: dynamic compilation (which became Java's JIT), prototype-based OO, and a new widget system (Morphic, which lives on in Squeak).

I think prototypes are an interesting alternative to classes for OO programming, and it's something that hasn't been very seriously pursued since then. It's a purification of OO in some ways, as too is SELF's insistence that all access to instance variables be through method calls.

Alan Kay, who coined "object oriented", said before that he really named it incorrectly, because he felt message passing was a more interesting and powerful concept. I think SELF moved more in that direction.
--

Given that the ArsDigita Community System is so heavily database-driven, I was wondering what tools you use for data modeling and schema management.

What is your opinion of modeling tools like Sybase's PowerDesigner and Platinum's ERwin? What kinds of tools do you think are necessary to facilitate the development of highly portable, vender independent database designs? Finally, what is your opinion of UML and to what extent does ArsDigita use it?

"The axiom 'An honest man has nothing to fear from the police'