Information Week for its December 6th ,2004 issue turns to the topic of The Future of Software. eWeek and Infoworld and your local newspaper are sure to follow. These reviews are tremendously backward looking and often miss major trends just coming down the pike. Give Information Weeks Charles Bacock credit he caught half of the action on the client and desktop. Yes rich clients are a growing trend – but the Presentation Server model is getting re-inforcement too with JSF, better standards compliant browsers and new innovations in JavaScriopt and DHTML.
In fact this is typical of whats happening across the board in IT – everything is up for grabs. New OS are vying for for a place in the Sun – with Linux, Sun Solaris for free and the perpetual leader and runner-up Microsoft Windows. Leader by dint of its 90% market share on the desktop. Runner-up on the server side because Windows server are tragically flawed because they were developed with “Just Good Enough” software development tools and attitudes – where the availability, reliability, security always took second place to new functionality and GUI usability.
But there is also an open contest between SOA, ESB, and good old fashioned event/transaction and messaging processing loosely coupled with Corba, Web Services and EDI hacks as the performance and security of proposed standards like .NET or Web Services primarily get confounded by fast moving BI, SCM, BAM, CRM vendors and service providers. The simple fact of the matter is that SOA and ESB are being worked out and testderiven – and the verdict appears to be not Web Services predominately but more like Web Services compositely – as controlling glue but having to defer to other technologies for performance, reliability, security and other reasons. In short, Web Services is not going to be the pure play taht certain parties have been selling.
Ditto for fast evolution in the database and storage world where open source and commercial are about to collide and duke it out for leading market and mind share. The duel is whether centralized and on demand will be able to contend with and displace open, distributed, agile/trim and replicating database technologies from SleepyCat, IBM/Cloudscape, Sybase, MySQ and a number of others. ETL, replication, and long transaction savvy – will be as important as load balanced, parallel, and an acid transaction ripper.
In short, no major part of software is being left untouched to move along docile-like towards some service plateau. Yet from the trade press we just do not get this perspective that literally everything is up for grabs in the IT world. Everything. And thats because the trade press has been gutted by the cutbacks in IT spending. So the coverage and dekbate is lost to the weblog ether. And as a result some major issues and drivers are getting short schrift attention.
For example, the two major IT drivers –> the Cheap Revolution which makes solution makers anywhere in the world potential bidders for software and application maintenance and development and Moores Law Renascent –> the doubling rate of computing power will continue indefinitely because computing technology reboots itself. Just like the Internet quickly builds both connection and technology workarounds to obstacles/impediments – so does computing – it provides the tools for its own progressive rebooting. Unfortunately there is not a Byte or Computerworld or Information Week with enough editorial insight and clout to “see ” these trends and things. The editorial smarts has been dissipated in the IT downturn and in the flight to weblogging ether. There are many contenders but no IT editorial must read.