Open source in the Enterprises: a compelling move?

Evans Data Corp., a market research company focused on software development, just completed a survey it started in 1999 about the Open-source penetration within large enterprises (400 development managers were interviewed at corporations with more than 2.000 employees).

Results are mixed: in 1999, 17.5 percent of the managers claimed that Linux was running on at least one server in their company and this percentage was boosted to 36.1 percent in December 2000.

But we've got to keep things in perspective because most of this huge surge took place before June 2000 (35,5% at that time), which means that Linux would not have gained any new customer within large corporations for the last six months.

The question we should ask ourselves is whether the Linux penetration rate is decreasing because the Linux has already found its market and as it only applies to some specific applications (among which, of course, web servers accommodation), it does no longer manage to deprive proprietary software of some of their marketplaces, or whether this present situation is only a short break in the market, meaning that enterprises need to digest these new applications before they get to the next stage.

It is still too soon to decide which is the right answer. But we already know for sure that, given the utilization ratio today, the hour of truth has come for OSS.

It might happen that Linux, PHP and Mysql are confined to some given fields (web servers, web sites development) and let proprietary software deal with the other aspects of the information system (CRM, back-office,) or else OSS keeps on expanding within the enterprises and in such case, we might then talk of a technologic revolution since it would mean that it is the very heart of the information systems that switches to the Open- source.

As long as OSS was only used for new applications (typically for web sites), somehow disconnected from the basic process of the information system, most development managers let it happen.

What's more, managers often only heard about those initiatives after technical choices had been set up by web developers keen on using OSS for their every day work. As those developments did not endanger the existing applications, they let it happen.

But enterprises can no longer ignore the fact that web applications need to be interconnected with their own computing devices or those of their partners.

Any enterprise needs to choose a mode of development at once.

Switching to OSS is a challenge for the years to come and development managers will have to decide whether or not they want to take such a big step and commit themselves to OSS.

As of today, very few have made that choice since only 4.2% of the enterprises that were interviewed in the Evans Data study declared running Linux on the majority of the servers.

It goes without saying that going from a proprietary software to a free one is not only a matter of replacing one type of software by another one, it involves a whole philosophical move.

First of all, it proves necessary to adopt a pro-active approach as far as developments and computer updates are concerned since OSS requires to surf the Internet and take part in chats to find an answers to your development problems and it means that you do not have to wait for the editor's answer. The last option was easier but the enterprise had to wait till the editor was willing to give him an answer.

And yet, the community of developers who has just spent five years working on OSS has managed to set up a whole set of functions and tools that answer most needs in corporate enterprises (online catalogue management, auction service, ads management, mail server, FTP server, etc).

To the exception of Microsoft, that often competes with the Open Source functional richness and might even surpass it, I think that most proprietary software in the eCommerce (Intershop…) will find it increasingly difficult to manage to impose themselves within the corporate enterprises.

What's more, the main argument in favour of proprietary solution, namely a stable structure for heavy charges, is no longer what it used to be since web sites that based themselves on free software prove that open source databases are just as good as the ones that were based on a proprietary software (even though they still have some progress to make in the management of simultaneous access).

As is often the case, the answer will come from the degree of commitment shown by OSS developers.

And let's admit that proprietary editors have managed to establish very strong links ("sponsored" training programs, commissioning policy…)

Sources : Evans Data - Techweb