cocoon-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Joerg Heinicke <jheini...@virbus.de>
Subject Re: JXForms vs. Woody vs. KISS
Date Tue, 28 Oct 2003 21:09:02 GMT
On 28.10.2003 21:23, Ugo Cei wrote:

>> The point is not whether the user can write a flow script in 
>> JAvaScript or XML. I don't want to use XML as a scripting language 
>> either. All that is needed is ability to define basic flows  that 
>> don't involve significant logic beyond what can be done using JXPath 
>> using XML. Is that possible ?
> 
> 
> I don't know. If we're talkin web "applications" here, sooner or later 
> the  logic driving page flow is going to be so complex as to require a 
> real programming language to express it clearly. If what we're talkin 
> about is web "sites", then maybe Cocoon's sitemap and core components 
> can be all you're going to need.

I'm not convinced, Ugo. I'm sure it's possible to have a declarative 
flow approach. Only the interpreter or generator must be intelligent 
enough. Of course a declarative approach is always somewhat restricting, 
but isn't this desired? It works for the sitemap itself. And there are 
also workflow markup languages on the run. But such an XML based flow 
interpreter engine might be a big thing and independent of Cocoon.

>> There are currently XForms books on the market (correct me if I am 
>> wrong, I believe an O'Reilly one just came out). Woody is not ready to 
>> be documented by a publisher yet.
>>
>> XForms is more likely to be accepted by an XYZ MegaCorp in the U.S. 
>> than Woody/Cocoon Forms.
> 
> 
> The problem with XForms, at present, is that we have to transform XForms 
> markup to HTML, if we want to support the current browsers. And since 
> XForms wasn't designed for this, transforming XForms to HTML is a PITA, 
> at least IMHO. Better use something that is designed from the start to 
> support server-side processing of HTML forms.
> 
> When (and if) XYZ Megacorp intends to support an XForms-capable user 
> agent, all the infrastructure you put in place to support XForms 
> server-side isn't going to be worth a penny. What is going to be 
> worthwhile is your knowledge of XForms, but applied in a totally 
> different context. Well, when (and if) that happens, I think I can reach 
> your level of proficiency in a few days.

+1

>> Antonio pointed out earlier that "From mid last year to today I needed 
>> to learn: Java, HTML, XHTML, CSS, XML, XSLT, POI, FOP, SVG, SITEMAP, 
>> XSP, XPATH, Javascript, OJB, JDO, Original DB Actions, Modular DB 
>> Actions, Forrest." There is a limit to how much a single person can 
>> learn, and the amount of time one is allowed to spend learning on the 
>> job is finite. The less to learn the better  and more productive 
>> developers can be.

The need for learning much is not a point in this time. Time goes by and 
much is developed. The question is not, if it is something new or not, 
if you have something to learn or not. It is about worthiness of the 
invention/development, it's about (not only) technical progress. This 
means you have not always to invent the wheel as we didn't for Woody, we 
made it only a bit rounder.
(Did I wrote all these pathetic sentences?? :-) )

Joerg


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org


Mime
View raw message