Archive of UserLand's first discussion group, started October 5, 1998.

Re: I don't need another graduate degree

Author:Arnold V. Lesikar
Posted:7/29/1999; 5:33:45 PM
Topic:The Doctor is In
Msg #:8902 (In response to 8899)
Prev/Next:8901 / 8903

With F5, it became somewhat more difficult to find my way around than it was in F4, but still F5 was enough like its parent F4 that getting around and doing things was still not too hard, particularly with guidance from the pros on the Frontier mailing lists.

F6 OTOH seems like a jungle. The Userland tutorials are like a path through the jungle. They're great. You can make great progress very easily - if you want to go exactly where the path leads. But if you want to go to some place just a little bit different, you rapidly find yourself back in the jungle. You can spend hours trying to cut your way through the thicket, and there are places where you can fall into a hidden hole.

The jungle is a good example. Adventure might be a better example. With the mR I have sometimes had the experience of finding myself in a network of twisty little tunnels, all the same.

What I'd want is not so much a tutorial as a map, showing how things connect. I'd want to know - of course - what calls what, but I'd also like to know what objects each routine uses and what objects it affects. I need to know what each routine needs in order to accomplish the job that it's supposed to do.

My most recent experience is with trying to index the Dome of the Sky at http://einstein.stcloudstate.edu/Dome/. (It's an academic site, but I have dreams of making money off of it one of these days. We're moving about 80 MB per day currently.)

Indexing the site was not quite a nightmare. The search engine tutorial pointed how to start, but it left out a good deal of necessary info. Nothing was in the tutorials about the role played by RSS files in indexing a weblog on the server side. As a matter of fact, there was nothing in the tutorials about indexing a weblog at all. Nor did the tutorials tell me anything about the getContent() script in the #search table. That script is essential if the indexing is to follow page addresses.

The tutorial assumed that my website is in a GDB. The Dome in fact isn't.

What with all the missing and mis-information, I spent at least two working days trying to index. This is time that could have been better spent on the content of the site or promotion. And several times I thought I had succeeded, only to discover in the mR log that the indexing had failed. The log offered few clues in how to proceed.

Tutorials won't help me that much. I need something that will help get around those twisty little tunnels better. What I want to know is how to customize the software to my own peculiar needs. Userland needs to make it easier for people to get where they want to go, not to where Frontier thinks they need to go.

The Dome for example is rather a mixture, serving static pages mostly. CGIs run in the Website Framework. The weblog and search routines run in mR. The Dome doesn't fit a standard model. Implementing the Dome entirely as a dynamic site in mR might make things easier, but I don't have a big enough honker that I'd dare try it.

Like it or not, that's my experience and my perceptions. I'm not a CIS pro, but those are my perceptions. Take them for what they're worth.

arn



This page was archived on 6/13/2001; 4:51:36 PM.

© Copyright 1998-2001 UserLand Software, Inc.