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

Richard Eng richard.eng at rogers.com
Wed Jul 18 20:21:02 UTC 2007


On 7/18/07 1:35 PM, "Ramon Leon" <ramon.leon at allresnet.com> wrote:

>> Seaside may not be for beginners, but that does not mean you
>> shouldn't try to lower the entry barrier or minimize the
>> learning curve. Even experienced developers may have a hard
>> time if they come from a Java or C background.
>> Not everyone in the world is an object-oriented guru.
> 
> Maybe not, but if you aren't pretty good with OO, Smalltalk really isn't the
> language you want to be in.

With all due respect (and I *do* respect you enormously!), Smalltalk is a
great language with which to learn OOP. I was briefly introduced to C# and I
didn't like it. Smalltalk is so much more elegant. You shouldn't have to be
an OOP expert before using Smalltalk.


> 
>> Seaside should not take an "elitist" position (an undesirable
>> word I pointed out in the "new Seaside homepage" thread). It
>> should be accessible to the broadest web development
>> community.  
> 
> There's nothing elitist about recognizing the simple truth that beginners
> and experienced programmers want different things from a framework and
> targeting both is very difficult.

You are not understanding me correctly. You should not dumb down a framework
for beginners. Seaside *ought* to be a power tool. All I'm saying is, lower
the entry barrier--minimize the learning curve--with accurate, up-to-date
tutorials based on non-trivial, full-fledged web applications. Provide great
documentation. *Streamline* the learning curve for beginners (even beginners
who have years of Java experience).

>From what I've seen of Seaside so far, there is no reason whatsoever its
power can't be made more accessible to the "masses."


> 
>> It should share the same aspirations as Ruby on
>> Rails, which is trying vigorously to conquer the world (and
>> apparently making huge progress!).
>> 
>> Richard
> 
> I have to disagree, if I shared those aspirations, I'd use Rails.  Seaside
> is targeting a different market than Rails, and it's doing pretty well as
> is.  The community only needs to be large enough to ensure its survival, it
> hardly needs to conquer the world.

I certainly do not share this view. There's nothing wrong with being a niche
player, but I have greater hopes for Seaside.


> Bringing a framework to the *masses*
> would involve dumbing it down...

Not true at all.


> 
> A good example from the Rails community is scaffolding, a feature even DHH
> regrets including.  It's the wrong approach but newbies flock to it, and
> abuse it, and then complain when *scaffolding* isn't fit for production.
> Well duh, that's why it's called scaffolding.  Yet, it's probably the number
> one thing that makes people try rails.  Without scaffolding, Rails wouldn't
> be nearly as popular as it is, but it'd probably be a better framework, with
> a community less likely to shoot themselves in the foot with bad code.
> 
> Templates and scaffolding are features that makes things easier for the
> beginner, but they're the wrong approach because they encourage copy and
> paste programming.  Reusable components are a much better approach but
> customizing them usually requires subclassing and overriding and makes for a
> much less effective 15 minute screencast demo.  Making a framework easier to
> use doesn't automatically make it better.

Absolutely right! Don't offer scaffolding and templates!

Don't dumb down the framework. The framework itself doesn't have to be
easier to use--just the infrastructure support such as documentation,
tutorials, etc. (For example, it's a pain searching the Seaside archives for
specific topics.) Lower the barrier to entry.

But Seaside can *still* aspire to greatness.

Regards,
Richard




More information about the Seaside mailing list