Analysts fail on open source

perspective Industry analysts can play a valuable role. But their shortcomings are particularly evident in their coverage of open source software. And, apart from analysts, what viable alternative information sources exist?

Despite their critics, analysts at their best can bridge a significant gap between vendors and software users. Vendor marketers who create information for prospective customers are understandably sales-driven and rarely grasp how their products are deployed.

The result tends to be material strong on superlatives and weak on substance. Certain features are emphasized while the processes involved in implementation are largely ignored, even though these may be more significant and costly than the software itself.

Users, on the other hand, have a deep understanding of their own problems but cannot always put it into a wider perspective. They have difficulty seeing the direction technology is taking, a direction influenced by a complex mixture of technical and commercial factors.

This gap leaves an opening for analysts, who should be relatively neutral about the success or failure of products or vendors, to stand back and form a view of the trends.

Provided analysts can also stay close to the practicalities of implementation, they can translate product features and trends into strategies capable of successful deployment. All this helps IT as a whole.

Of course, the limitations of the analysts have become well known. Researching, writing and speaking on IT involve a good deal of hard work, for which analysts want payment. Revenues have always come from vendors as well as users and ideally were more or less balanced to avoid undue influence from either side.

But analysts have not always kept the balance, and services such as free conferences have been provided on the back of payments from vendors.

Over time, it has become increasingly difficult to generate revenue from users, throwing the analysts into the arms of the vendors. Some analysts have gained a reputation for being little more than vendor mouthpieces.

A small number of analysts have adopted what they describe as an open source model for much of their writing, releasing information freely. This approach inevitably involves a reliance on generating some kind of fee-earning work and only a limited number of analysts are able to achieve this combination.

Within the smaller analyst houses there is often a feeling that efforts are inadequately rewarded. And the notion of open source analysis is only loosely connected with open source software.

The general reliance on vendor cash clearly leaves most open source software out in the cold. That lack of coverage is a pity because there is much high-quality software in the sector and it is sometimes developed rapidly as skilled software writers join successful projects.

But open source projects rarely have significant funding to afford commissioning analysts. That lack of coverage compounds a problem with the presentation of open source software.

Unlike commercial software, where claims are grandiose but vague, nearly all open source products are described in detailed technical terms.

In both cases, the putative user is left unable to discern how the product might solve actual problems. The honorable exception to this tendency perhaps includes the successful dual licensed products, such as MySQL, that straddle the open source and commercial worlds.

In principle, this is a clear opportunity for industry analysts, who have the skills to translate the technical representations of open source developers into the practical conceptions that are understood by potential users.

In fact, sophisticated users are painfully aware of the need for analyst coverage of open source. Despite the excellence of much open source software, there is often management prejudice against non-commercial software, and independent commentary is a useful balancing factor.

A few analysts have indeed made an effort in this direction but are finding it a struggle. Large vendors can make life a lot easier for analysts. They have staff dedicated to analyst relations and are able to provide a range of helpful meetings and backup material.

Open source projects cannot usually afford these luxuries. Another problem for analysts treading the open source path is that smaller analyst firms find it very difficult to secure substantial advisory business from large user companies.

What alternatives are there for making sense of the already vast and rapidly proliferating world of open source software? A variety of information exchanges are growing up, although many more are started than ever thrive.

An example that is showing more signs of life than most is Ohloh. But because of the sheer number of projects, if exchanges present any comparative information at all, it is normally based on crude automated metrics.

As a result, even someone committed to working largely in the open source sector finds it hard to identify relevant projects to meet a particular need. They are often stumbled upon largely by chance, and evaluation is usually hit and miss, limited by the time available to achieve practical results.

What is ideally needed is a shift in attitude by buyers and especially larger buyers. Every open source project is by no means a success but the open source model is a powerful way to deliver very high quality software.

It is also very efficient, partly because it eliminates many of the most costly overheads of software production and partly because of its ruthless "evolutionary" character that stems from the impact of the main open source licenses.

So long as people see open source simply as free software, it has little chance of reaching its full potential. As in many other areas of IT, buyers would serve their own best interests by putting more resources into encouraging the most valuable sources of innovation and development.

And that might include spending a lot more money on good analyst coverage of open source software, funding analysis that steers between the rocks of empty rhetoric and arcane technology to reveal true value.

Martin Brampton is founder of Black Sheep Research, an independent consultancy providing research, writing and speaking services on a wide range of business and technology issues. Martin was previously a director at Bloor Research, and has worked with IT as a user and analyst for over 20 years. He is a longtime contributor to ZDNet Asia's sister site Silicon.com and his blog can be found on his Web site.