Text Files Unfairly Compared To The Internal Combustion Engine

Alan Green compares Text Files to the Internal Combustion Engine. That is, a technology that is obsolete and despite all its short comings we all have to live with. Well I agree that the Internal Combustion Engine needs to go, that and "9 other technlogies that deserve to die". However, I disagree with his assessment that text files deserve the death penalty.

Alan Green in his piece cites a couple of options:

In a recent post, Chris Petrelli holds forth Smalltalk and Zope as examples of programming systems that don’t need text files. This paper by Roedy Green explores what could be achieved if we gave up text based source files. Others have been experimenting for years.

I used to have this belief, as do many, that the only way to manage information is to mold it into a strict formal structures. That's what conventional software engineering practices tell you to do. That is map your data into a relational data, create object oriented structures to encapsulate access logic and build UI's with rich UI components. However, somewhere along the way, the web has shown clearly off track we all are.

I wrote recently how the underated text is making a comeback. However, the plain and simple reason why text or text files will continue to remain ubiquituous is that it is the correct approach. It is the correct approach because it is the most loosely coupled approach among the alternatives.

The flexibility afforded by text files outweigh the need for structured data stores. The fallacy one needs to recognize is that perfect a-priori domain modeling is an effort in futility.

The success of Google is a good example of how deriving metadata (i.e. using titles, links, citations etc.) rather than relying on canned metadata (i.e. meta keywords, rdf etc.) can lead to unanticipated benefits.

Eclipse is another good example, with appropiate use of fast intelligent parsers, manipulation of text (i.e. code) has never been so easy. Now consider that Eclipse has its roots from the VisualAge system that has the same object database underpinnings as the Smalltalk system cited above.

One last reminder of the benefits of text files are Wikis. These simple constructs allow incredible flexibility in structured information. I'm coming to the realization that the perfect content management system or web framework may just be something derived from Wikis and a file based system. Come to think of it, isn't this precisely what Blojsom combined with Radeox can do?

Google, Eclipse and Wikis, shinning examples of why intelligent parsing is the more favorable approach to take as opposed to designing rigid structures to encapsulate information. The point is pretty clear, free your information and you'll discover benefits in unanticipated ways.