Letter to the editor: RAD tools need short learning curve

By Roland Giesler, Special to searchWebServices 01 Mar 2002 | SearchWebServices

Digg This!     StumbleUpon     Del.icio.us

Editor's Note: The following piece was send to searchWebServices in response to a recent article on Magic Software's updated eDeveloper application development tool. (See the link below to read the original article.) SearchWebServices welcomes reader feedback, but does not guarantee the accuracy and validity of the material submitted.

Dear searchWebServices,

I have used the older DOS version of Magic and was suitably impressed.

To me the matter is quite simple though. If you can save a substantial amount of time developing an application by using a CASE (computer-aided software engineering), RAD (rapid application development) or whatever tool, then great! Of course maintenance is just as important, but typically these tools, like Magic, make changes a breeze.

If a tool is too difficult to learn and will not incorporate changes back into the model (older Clarion e.g.), then its counter productive. On the Internet, products like Microsoft FrontPage make it very quick to create a rudimentary site, but lack easy access to documentation of advanced features like creating your own styles and customizations, so the more serious developers typically ignore it.

The fact that FrontPage tends to drop (or worse still, attempts to fix) anything it doesn't understand, unless you specifically tell it that this is "outside" code, doesn't help either. Once you have a set of toolbars, styles and buttons defined, created a layout, etc. it's actually to quick and simple to create a new page, it's a no-brainer. The problem is many people have tried FrontPage and it's created a barrier in their minds: Tools that do it for you are limited.

On the other hand, one irony is that many developers shy away from using tools like Magic (or other fully integrated dev tools) because the marketing departments of these companies don't market the product well enough. Although Microsoft created a problem with FrontPage marketing, it has gained (at least here in South Africa) a huge mind share through the training seminars it provides.

It gave away t-shirts, caps, PDA's, stereos, and tons of its own software to entice people to come to these sessions, and in doing so it showed developers and managers how to do funky stuff, be it sometimes nothing new to those who were really knowledgeable. But still the feedback, co-operation, participation and mind share they got helped win them the local market.

Companies like WordPerfect and Novell tried too late to do similar things, and their efforts were quite pathetic -- in my opinion -- and didn't get them anywhere.

Make no mistake; I'm not a Microsoft lover who just swallows anything they feed me. I'm a firm believer and advocator of (open operating system) GNU, GPL (General Public License) and free software in general and, to that end, I predict that Microsoft will give away most of its software for free "offline" use in four to five years. It will make money out of services that function seamlessly within their software. It doesn't make the software "free" yet in the same way GNU is, but it does mean people will be able to get it for free. Microsoft knows how to look after their developers, better than any other company I know.

But back to the RAD issue. I use Microsoft Visual Studio for as much as I can (migrating to .NET), since it makes things easier and faster for me without sacrificing flexibility. Intellisense probably saves more time when I do stuff I don't do everyday (the more I consult, the less I have time to try out stuff!), than any other feature.

I've tried other products including the CoffeeCup HTML Editor, the XMLSpy IDE (I still do, since VS doesn't "know" XML), SoftQuad HoTMetaL Pro and many others. Although each has its strengths, the total environment counts more than just a single great feature that saves some time. CoffeeCup has great potential, but if it doesn't understand what my server can do, it takes more time to get things right.

So I'd give Magic every chance it deserves, but still don't think it will make a significant impact. If they were to give it away for a few months, it would change the picture dramatically. Magic should start charging for upgrades and support after the giveaway period, but it would gain a market presence and penetration that it lacks currently. (Upgrades to a few million clients could be more profitable than a new product to only a few hundred thousand, if their client base is that large!)

Maybe Magic can do everything (including real magic) but unless I can see it work for me and for a substantial number of others, I won't believe it and neither would most others. So why pay money for something that may or may not solve a problem that is surmountable? The risk of failure is too great. I've been down that road many times.

I would say the points that are crucial for a RAD tool to succeed in the Web development environment are:

1. Intelligent error checking in the server context, i.e. the environment must understand the server and client(s). If a product that understood more servers than IIS alone came along, offering automatic exposure of used object properties and methods, hurray! The casual developer typically sticks to his tools, so help them when they need it.

2. Open Architecture allowing third-party plug-ins.

3. Low or no cost of entry. Microsoft has cornered a large part of the market for beginners, but also a significant part of the serious and professional market. A high cost product entry into that market is too risky, especially since Apache doesn't cost a cent to buy either, but runs on more platforms. Integration into MS is essential.

4. Excellent online documentation and resources, preferably community maintained, but initially product developer compiled and delivered with the product.

5. Readable source to allow manual fixes should it be necessary. It also makes a product easier to learn, by reading the output of a process to learn what it does (only for advanced developers).

6. Maximum marketing effort through creative means.

I believe that all these are needed for any product to have significant impact in the current RAD market.

Roland Giesler
Green Tree Consulting
Hermanus, South Africa

FOR MORE INFORMATION:

CLICK for the original article: Magic Software recasts developer tool for Web services

CLICK to send us your opinion on this article or any other content you see on searchWebServices

CLICK to interact with your peers in searchWebServices' Best Practices Forum

Digg This!     StumbleUpon     Del.icio.us

RELATED RESOURCES 2020software.com, trial software downloads for accounting software, ERP software, CRM software and business software systems Search Bitpipe.com for the latest white papers and business webcasts Whatis.com, the online computer dictionary