[Seaside] another reason Rails gets market share andSeasidedoesn't

Conrad Taylor conradwt at gmail.com
Sat Jul 21 20:41:54 UTC 2007


On 7/20/07, Cédrick Béler <cbeler at enit.fr> wrote:
>
>
> > I disagree that you wouldn't want to do Smalltalk if you don't know
> > OO. Smalltalk is a great first OO language (taught lots on the job).
> > If anything, Smalltalk is distasteful to others who've used other
> > languages and maybe never really got objects (seen it over and over
> > teaching OO to people who've learned it on their own).
> I'm (was ;) ) a beginner/newcommer/journeyer... Before using smalltalk,
> I had only basic notions of some languages as html/http... Smalltalk is
> great to learn object and OO... I've learned a lot since but mastering
> is another problem...You quickly realize that object modelization is not
> simple even if often said to be a natural, simple way of thinking...
> "Write once and only once" has a implicit wrong meaning of simplicity...
> It's more a question of beauty, easy maintenance but not coding
> easiness...
>
> >
> > If you like PHP or Rails or servlets, that's cool too. It's just
> > different.
> Sometimes, I do some php, it's fun because rewarding even if the final
> code smells... you can quickly copy/paste/adapt hundreds of lines ;) ...
> Problem of smalltalk is that adding an extra functionnalities, or
> wrinting a fix is often one of two lines of code and that's not
> encouraging/rewarding for newcomers especially if you spend 2 hours
> "writing" it...



You're missing the point entirely of Smalltalk with your example of
producing a crappy solution in PHP. If I can write couple lines of code in
Smalltalk to implement a feature or a bug fix, this allows me to spend less
time doing maintenance activities and more time designing new features for
the system.  I have done the PHP and it's a maintenance nightmare sometimes
to locate bugs and/or add new features to an existing system due to poorly
implemented code.  I would rather spend 2 hours of writing 2 well crafted
lines of code than to write many more lines of code that isn't.  I have been
on projects where the code base got to a point where it wasn't maintainable
after adding both bug fixes and features.  Thus, the companies solution to
the problem was to redesign the entire system from scratch.

Finding the reward is hard in smalltalk, but I think I've reach the
> point I've too much *fun* doing some smalltalk...I'm not being objective
> anymore ;).


The fun in Smalltalk is being able to implement well designed solutions.
 BTW, I also do Ruby development for some of my clients and that community
is agreement with Smalltalk (i.e. implementing small methods that's focused
on a single objective).  Also, there are a few blogs dedicated to design and
code re-factoring.

What's fun in other languages are the results you get, and in smalltalk
> it's the environment and the result... but first you need to understand
> and be at ease with the environment.


I agree with you here 100% here with the addition that one needs to obtain
understanding of both the Smalltalk language and OOD.

Cédrick
>
> _______________________________________________
> Seaside mailing list
> Seaside at lists.squeakfoundation.org
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20070721/57d37dd9/attachment-0001.htm


More information about the Seaside mailing list