 |
|
|
 |
||
 |
||
 |
||
 |
||
 |
||
 |
 |
|
 |
||
 |
||
 |
||
 |
||
 |
||
|
||
 |
Your independent resource on business integration and network computing through middleware and message brokering
Open source middleware: a review of the implications
Tom Welsh
Consultant
Management introduction
A recent IDC report reveals that Free and Open Source Software (FOSS) has spread far beyond Linux, and is gaining "enormous momentum". Based on survey replies from over 5,000 developers in 116 countries, the IDC's analysts declared that FOSS is the "most significant all-encompassing and long-term trend that the software industry has seen since the early 1980s".
IDC thinks that FOSS will eventually play a role in the life-cycle of every major software category, and that it will fundamentally change the value proposition of packaged software for customers. Based on the survey sample, FOSS is already being used by 71% of the world's developers and is in production at 54% of their organzsations.
In this analysis, Tom Welsh examines the growing impact of FOSS on middleware. He considers:
the major FOSS players, including the Apache Software Foundation, the Eclipse Foundation, the Free Software Foundation, Mozilla, Red Hat/JBOSS, ObjectWeb and others
the reactions of some of the leading vendors of proprietary middleware - specifically BEA, TIBCO, IBM, Microsoft, Sun, Oracle and Iona.
Figure 4.1: JBOSS JEMS architecture
Figure 4.2: TIBCO General Interface
Management conclusion
Some years ago there was a common belief that FOSS would never catch on in commerce, industry, government and other 'serious' IT markets. Today, that belief has been exploded. Successive waves of FOSS are eroding the corresponding 'closed-source' markets with astonishing rapidity, so that virtually all commercial software suppliers are now anxiously planning to minimize the impact on their own businesses.
First GNU, then GNU/Linux and BSD, then Apache, Netscape/Mozilla/Firefox, Eclipse, JBoss... the list goes on, almost without end. So far, FOSS has already dominated some distinct segments - for example, mass server farms. Companies like Amazon, eBay, Google and Yahoo would have to spend much more for IT if they did not use FOSS on a large scale.
No fewer than 74% of the top 500 supercomputers run Linux. Amazon powers more than two-thirds of the HTTP servers on the Internet. Eclipse is the most popular Java IDE while JBoss is the most popular Java application server. Meanwhile, MySQL's installed base is getting to the same sort of size as Oracle's, even if it is not nearly so lucrative.
Middleware - "the software that nobody wants to pay for" in Chris Stone's famous words - looks to be next in line. Indeed, those forms of middleware that have been adequately standardized (such as Java EE and CORBA) are already beginning to be dominated by FOSS. With the advent of Web Service standards and AMQP, SOA and MOM may be next. There are already several high quality FOSS ESBs available.
Some vendors will be able to hang on, charging high prices for what are still perceived to be 'top quality' products with added value over FOSS solutions. As time goes by, however, this strategy will probably become less and less feasible. For those that depend entirely on proprietary closed-source software, the writing is on the wall.