'Free' is the new 'cheap' for software tools

(continued from previous page)

"The whole notion is to develop the community and to grow the adoption. (Free databases) is one way to stop the open-source database adoption--it's not really going to generate revenue per se," Yuhanna said.

"The real battle here isn't revenue, it's developer mindshare."

--RedMonk analyst Stephen O'Grady

Forrester estimates that the open-source database market, which includes support, service and license revenue, was about $300 million last year and will grow to $1 billion by 2008. Demand is being driven in large part by lower costs and maturing products, said Yuhanna, who predicts that 20 percent of "mission-critical," or essential, corporate applications will run on open-source databases by the end of the year.

Freely available software has been around for some time. But free products, which encourage developers to try out software, combined with open-source communities around those products, can be very compelling for a software company, noted RedMonk's O'Grady.

Open-source communities tend to spawn the creation of add-on products, such as plug-ins to a browser. They also foster the usage of open-source components in a certain combination, such as LAMP (Linux, Apache, MySQL and Perl, Python or PHP), he said.

"The real battle here isn't revenue, it's developer mindshare," O'Grady said. "By leveraging open source, (open-source database provider) MySQL and others have been phenomenally successful at capturing developer mindshare."

The prices for development tools, which are often used in tandem with databases, have shrunk down toward zero, as well. The popularity of Eclipse, an open-source framework, has made it difficult to charge for a basic integrated development environment (IDE), analysts have said.

Last November, Sun Microsystems made all of its development tools free to programmers who sign up for a yearly subscription to the company's developer network. And Borland Software, which traditionally focused on selling IDEs, has revamped its strategy over the past three years to selling suites of lifecycle tools that address testing, modeling and coding.

In another example, Adobe Systems on Wednesday is expected to revamp the pricing for its Flex Flash development tools.

The first version of its Flex tool set was about $15,000, but the price was throttling its adoption, said Jeff Whatcott, senior director of product marketing at Adobe's enterprise and developer business unit.

"The goal is to get to a million developers building rich Internet applications," said Whatcott. "To do that, you need a good product line. But you also need a licensing model that supports viral, development-to-development marketing. That doesn't happen with a $15,000 product."

[rariane]

mp3

I wanna downroding and It's free?

[tech_junky]

Of course, they get people back to their...

currently free product and then snag em when they suddenly start charging for it again.

[fafafooey]

Who has done that?

The "free" versions are usually resource limited or license limited (developer only) and remain that way.

They give away the free versions for small installs and developers, and charge for the "enterprise" or "commercial" versions.

How dare they want to make money...

[mlcntrprs]

Growing or slowing???

"Forrester estimates that the open-source database market, which includes support, service and license revenue, was about $300,000 million last year and will grow to $1 billion by 2008."
Question...
Isn't $300,000 million = $300 billion and isn't that larger than $1 billion?

[mlamonica]

Fix made on numbers

That was an error and it has been fixed. Forrester's numbers size the open source database market at $300 million in 2005.

[Cargill_Biff]

The problem is...

These version are not just resource limited, but there are also lots of 'higher end' features that are not enabled in the developer versions (eg. high availability), which I think this is ok.

However, the problem I see is that the vendors have the right to arbitrarily change what exactly is free and what isn't. If I build small solution that relies on a certain feature set, and want to upgrade to a newer version of the dbms (to get some bug fixes for example), there is no guarantee that the features I have used will still be supported in the free version.
For example, lets say that IBM notices that non-core feature x is used by almost all developers in the free version, and they are not seeing revenue developing out of the 'free' developer base. Nothing prevents them from making an 'update w' or 'version y' which disables this feature, forcing users to either rewrite or upgrade. So much for free.
MS has already demonstrated signs of this when the changed the limitations in their move from MSDE to SQL Server Express. Granted the changes they made may impact many people, but there is no reason why they could not have.

I know you get what you pay for, but the 'free' closed-source database adoption will only work if you can trust the vendors...yeah right.

[fireball74]

Vendor trust

I have more trust in IBM than M$ or Oracle these days. If anything, I think IBM will add features to the free version of DB2. M$ can shoot themselves in the foot over "lost" revenue all they want. All that will do is lose them potential buyers, and support in applications that could get them into some new and niche markets. Of course, M$ is all about the bottom line and vendor lock-in. There are now alternatives to SQL Server, so if you don't like what they're doing, switch and show them how much you appreciate it. :)

[hberenson]

Microsoft had a "free" database long before open source

Microsoft has had a couple of "free" database offerings from long before Open Source databases became a significant part of the picture. Access' JET database engine has been essentially free since the early 90s and SQL Server's MSDE has been available for free since 1998. SQL Server 2005 Express is just the latest incarnation of MSDE, with its characteristics tuned to better target the market. Certainly Open Source offerings influenced that tuning. Microsoft's success with MSDE, and the increased focus on this area with SQL Server 2005 Express, as much as the Open Source offerings, are what forced Oracle and IBM to come up with their own free databases.

[fireball74]

I don't know about that....

What makes M$ JET "database" so unappealing is the lack of multi-platform support. To top that, Access is slow, and definatly not a SQL class database. Basically, you're comparing apples to kiwis. There's no competition to SQL from JET at all.

The SQL Server 2005 Express is too limited, but so is Oracle's "free" offerings too. IBM isn't limiting theirs near as much, which tells me they want the niche markets as well as better support for their platform in OSS projects. It's clear that neither M$, nor Oracle, really want to compete with the Opensource offerings. They just want to throw out a line to see if anyone bites so they can lock them in. Not gonna happen with this developer.

[cagerattler]

Lite Software isn't that effective...

It seems logical that by giving away "Lite" or "feature discounted" versions of the biggest database product brands will lure developers away from open source solutions (i.e. competition) and potentially expand the customer base.

However, in the real world, "lite" versions don't work that way. What happens in many cases is that the customer service group gets inundated with calls from frustrated developers trying to figure out how a certain feature would work IF they had the full product. Customer service calls cost money. (It would might be better if the first department to field "Lite" version software calls was the Sales dept. At least they'd be able to convert a percentage to the full version.)

In addition, the developer who elects to go Open Source does so because they are operating under a budget, or they haven't quite decided that the solution they are building will end up being the ideal solution over the long haul, or they just don't like dealing with The Man. They are usually evaluating more than one potential solution at a time, so working with fully-equipped solutions during evaluation results in a better final evaluation/recommendation.

I think that the big brands need to embrace the open source market and educate developers on all products available (not just their own). They should also provide developers who use open source a migration path to their products (and provide the tools to facilitate this).

Basically, the big brands should tell the developers - "Hey, we understand. Whether you try us now or open source now is fine. At the point where you determine that you really need the long term support infrastructure and guaranteed performance, call us. We'll still be here."