The Soap-box - I Hate OOP

by Robert Sorrells

Published 1998-06-01    Printer-friendly version

(Editor's note: "The Soapbox" is an arena in which a reader can express his or her opinion on a Clarion-related subject. The viewpoints contained in this column are solely those of the author and do not reflect the opinions of Clarion Online or its staff. Submissions for "The Soapbox" should be non-inflammatory. Clarion Online reserves the right to refuse any submissions)

I HATE OOP! Okay... I said it! I'm not politically correct by saying it, but this is my present opinion. I HATE OOP, I HATE IT, I HATE IT!!!. I hate it, and I think TopSpeed is making a mistake by embracing it. I went to the East Tennessee Clarion conference in May (which incidentally was very well organized) and I met a lot of good people. ETC did an excellent job organizing the conference, and I learned a LOT (plus great food, margaritas, and BBQ). But after listening to days of OOP this and OOP that, and OOP being crammed down my throat every where I turned, the conference really reinforced my resistance to OOP and ABC.

To get a feel for the background of the audience, several of the speakers did an informal survey prior to their speech, and it usually went something like this: "Who here uses CW2003?", and about 10 hands go up. "Who here uses CW4a with Clarion templates?" and 98% of the hands go up? "Who here uses CW4a with ABC templates?" and I saw only about 20 hands each time, out of an audience of 120 or so.

Lesson learned: I'm in the mainstream. I use both CW2003 and CW4a with Clarion templates. Good so far.

Day 1: Tom Moseley was the first guy to speak on templates. He did an excellent job in the morning, and got us over the fear and the learning curve of templates. What an excellent concept -- programs that write other programs. Very worthwhile. In the afternoon, however, he was showing us an example by using a tiny little simple "hello world" example with ABC. Well, we spent ALL afternoon trying to get OOP/ABC to work and, with glazed over eyes, I phased in and out, hearing phrases from the loud speaker "...and then with ABC you have to type this, and then you have to go over here and add that, and of course, it's intuitively obvious that you must add this to point to that, and ...". Well, I don't think the example ever worked, and I forgot what the question was.

Lesson learned: Templates are cool, but ABC/OOP sucks! It's difficult to use, and development is very slow. Tom Moseley is one of the smartest people that I listened to at the conference, and if HE has trouble with ABC, there ain't no way in hell I'm going to master it.

Day 2: We heard from Roy Rafalco, the president and COO of TopSpeed. Message: Synchronizer is coming. "Borg" is coming our way at Warp 9, and I can't wait for these cool sounding super wizards. CWIC sounds like a great technology. I bought it and hope to try it on my next project.

Roy also said that Clarion will not make any changes to the traditional Clarion templates; rather, they will focus all of their energies on ABC.

Lesson learned: I can't wait for these super-wizards, but I better learn this ABC/OOP stuff because that's the future.

Next, Andy "Cowboy" Stapleton. What an awesome presentation. This guy walks the talk. "SQL is more OOP than OOP!" is a quote. Also, of all the people from around the world at the conference, he was one of the few folks there that didn't have an accent. <g> (I'm from the south, also).

Day 3: Ross Santos talked about OLE/COM, and that was interesting. Complex, but interesting. I really liked his philosophy of reusing what's out there, with a focus on controlling it within Clarion rather than re-writing. Very useful. [Editor's note: Ross Santos will be writing an article on that very subject for Clarion Online in the near future]

Next, Dave Harms' "Thinking in Objects". There are certain people that you can just tell, who really know their stuff, and Dave Harms is one of those folks. This guy has forgotten more than I'll ever know about programming. He did a good job of explaining objects and OOP. However, the more I listened, the more strongly I concluded that this OOP junk ain't for me. First of all, he uses a $2000 diagram program (Select Enterprises?) just to keep up with the documentation of these OOP UML diagram things. Hey! I've never done documentation on my code in the last 10 years, so why should I start now? (Lets see a show of hands of people who did a flowchart AFTER writing the program in college. Another show of hands: Who out there deletes the "Help" button the very first thing after a Wizard generates your form?) Who's got time for this stuff? I believe Dave also mentioned that he found a lot of information from the Microsoft Developer CD ROM subscription, at $2500 per year. Who's got time to read that? Lets see... what else. Oh yeah. It's very easy to cause "memory leaks" if you forget to add Free() or something, somewhere. No errors or warnings from the compiler, yet my customer must reboot daily because their 128 Meg workstation runs out of memory. Boy, I can't wait. Sounds more and more like "C" every minute. Give me a break. Oh! Here's another benefit of OOP. You can't troubleshoot it because you don't know exactly how the program will execute, so you better get it right the first time. And you need to get C++, so you can get some kinda class viewer software from it. Again, like Moseley on day #1, Dave never got the example app to work.

Lesson learned: ABC/OOP may be great for geniuses, and may be more efficient for weather-prediction algorithms, but as far as I'm concerned, for "get it out the door" rapid business database development, it sucks! I don't have time to document code. I don't have time to read an encyclopedia CD from Microsoft every quarter. I'm not going to spend a nickel, much less $2000, on a diagramming program. Memory leaks with no compiler warnings? Give me a break.

I couldn't stand OOP anymore after this. I checked out of the hotel, even though I had paid through Saturday's conference, drove home that night through hail and tornadoes, and re-installed VB5 and ordered the latest Sheridan datawidgits CD's. Of course, like the last 9 times that I've broken up with Clarion and reinstalled VB, I concluded in about a day that VB5 sucks, and Clarion (with the traditional templates) is still lookin' good.

From my observations over the years, there are two camps of Clarion folks (and programmers in general). There's the majority of us, who are focused on getting programs out the door, who don't have time to write code; and the minority who wouldn't think of doing anything but hand code, savoring the elegance and beauty and poetic form of every line of code. Another analogy: The majority of us would like to build a pool with a giant backhoe, and would actually prefer a computer-controlled phaser linked to a brain scanner. The minority would prefer the tools of an archeological dig -- tweezers and brushes, savoring each and every grain of sand, and the form and elegance and placement thereof. Hmm... let's see... what category does this OOP stuff fall into?

The reason that 95% of us chose Clarion in the beginning was because you could develop a stand-alone, kick-ass, database business application with minimal programming. I'll never forget hearing from a "purist" programmer who, after I told him that I used Clarion, told me that Clarion was for people who didn't know how to program. Well, I thought to myself, Clarion is for people who get paid by the job, and not by the hour. Well - I don't know anymore.

Unfortunately, the minority seems to have the ear of TopSpeed, and seeing that VB5 has all of this OOP/Class junk, they convinced someone at TopSpeed that OOP/ABC is great and wonderful and the future. I think it's a giant step backwards. Who cares if OOP produces smaller code or executes faster? Hard drives are a dime a dozen. Time to market is key, and from what I saw from the experts, developing with OOP/ABC doesn't exactly set any speed records.

To leave on a positive note, there are some heros at TopSpeed working on this super-wizard technology. I can't wait to buy it. The goal of TopSpeed should be a "Wizard"-like system that will prompt you through 98% of your entire program effort and time - without even having to populate a "To-Do", much less having to write a single line of code. The goal should be to have a product that can produce a "Quicken-like" application in a day, including reports. And if OOP is in the background, who cares. I don't want to see it, or even know it's there.

Sincerely,

Robert (Bob) Sorrells

Forest Enterprises

 

Printer-friendly version

Copyright © 1999-2008 by CoveComm Inc. All Rights Reserved. Reproduction in any form without the express written consent of CoveComm Inc., except as described in the subscription agreement, is prohibited.

Clarion Magazine ISSN 1718-9942