Re: HTML and PDF Reports

[Joe Pluta <joepluta@xxxxxxxxxxxxxxxxx>]

Bradley V. Stone wrote:
> It's the understanding of how to use the markup language and it's tools that
> make a web application successful.  I don't need a huge toolset to do that.
>
> The lifecycle of most of my customers usually goes like this:
>
> 1.  They try a screenscraper/webfacing tool.  Hate it.
> 2.  They try Websphere or a similar tool.  Hate it.
> 3.  They contact me and say "show use what RPG can do".
> 4.  15 minutes later they're sold.  Their programmers love it and start
> hammering on the HTML, JavaScript and CSS tutorials all over the net excited
> about what they can do.
>
> RPG is dead.  Yes, we've been hearing that for years.  *yawn*
>   
Yup, if you have RPG programmers who have the time to learn HTML, JS and 
CSS, then using a tagged HTML technique is a great way to go.  At the 
same time, if you have someone who can learn Java, then JSP or JSF is a 
great way to go, as it opens up a wide range of pre-written Java 
libraries.  Even PHP is an alternative if you have PHP people in your shop.

No single technique is the best for every shop.  The question is time to 
market.  That's why I presented the challenge for the scheduling 
application - it's my opinion that EGL and RPG together are the fastest 
solution for web applications.  It's a little bit of a toss-up for older 
web applications: the traditional work with type panels where you list a 
set of records and click on one and drill down.  Those are pretty basic 
features and if you can't build one of those applications in five or ten 
minutes, you have a real problem.

But the learning curve for creating Ajax calls and updating a page or 
writing the JavaScript to access Google APIs is a significant level of 
complexity above just learning how to put tags in HTML and fill them in 
at runtime.  That's why I'd like to see the CGI equivalent of the 
scheduling application.

Joe