Neither free as in speech, nor free as in beer

Maybe it was a coincidence but two recent blog posts have highlighted questions over the value of source code in terms of enterprise adoption of open source software and the wisdom of open source vendors in focusing on low cost.

Savio Rodrigues noted that his experience backs up recent data from Forrester that suggests that “most developers don’t really care that OSS is libre (free as in freedom); they care that it’s gratis (free as in beer)” while Bruno von Rotz asked whether enterprises really care whether the code is open source.

While there are undoubtedly some enterprises for which access to source code is important (either practically or theoretically) for many enterprise users, access to source code is irrelevant.

Savio suggests that the apparent change in attitude is actually the result of expanded adoption of open source software moving from “religious adopters” to “regular developers”. I would suspect that this has something to do with it.

In any survey, who you ask is important. Our report into the impact of open source in the database sector sector included a survey of people responsible for database purchasing, rather than database developers or administrators, and we found that just 6% of them cited access to source code as the most important reason for deploying open source (compared to 41% citing cost cutting).

The results were somewhat different from those seen in other surveys focused on database administrators or developers.

Our survey indicated that database purchasers were more likely to increase their adoption of “Express edition’ (free as in beer) versions of proprietary database products than they were to increase their adoption of open source database software and, intriguingly, that the planned adoption of Express Edition products was markedly higher among existing open source database users.

Meanwhile, as Bruno von Rotz points out, much of the open source software consumed by enterprises is presented to them in a way that access to the source code is irrelevant. We have seen increased use of commercial licensing by vendors generating revenue from open source software, and the increased inclusion of open source software within larger commercial packages.

Arguably, the end user does benefit from the code being open in these cases thanks to the ease of adoption and the benefits of lower development costs and higher code quality being passed on to them (although that in itself is debatable) but the user gets no direct benefit from the source code if they have no intention of or inclination towards accessing and altering it.

Marketing open source on the basis of free or freedom in these cases is a waste of time as the software consumed is neither free as in beer, nor free as in speech.

Meanwhile, one of the criticisms of Express database software is that it is limited in terms of the number of users, or processors, or the amount of data that it can handle and that adopters eventually have to pay considerable sums to migrate to expensive proprietary licensed versions.

The criticism is valid, but so is the criticism that some enterprise open source users will find the functionally of the open source edition limited at some stage, only to find that the answer to their problems is to adopt a proprietary-licensed “enterprise edition”.

I have previously noted that the assimilation of the some benefits of open source software (low cost, collaborative development, ease of adoption) might prompt something of an identity crisis for free and open source software.

Savio has repeatedly warned against open source vendors focusing on cost, and I would suggest that any vendor that is selling their software on the promise that it is low cost, while also hawking traditionally-licensed enterprise version is perpetuating confusion while also playing into the hands of its proprietary rivals.

Our earlier report, Cost Conscious, highlighted that while lower cost was the biggest driver for the adoption of open source (cited by 43% of respondents), it was considered less important after adoption (31%).

It would be easy to assume that the adoption of open source software had failed to deliver cost savings. In fact 84% of respondents said their adoption of open source had met or exceeded their cost savings expectations.

What had actually happened was that other benefits such as flexibility, performance and reliability were considered to be more significant benefits after adoption than they were drivers before adoption.

As open source matures factors such as these should arguably become the primary focus of open source vendors. Open source software is not free and it shouldn’t be claimed to be.

Collaborative development and lower development costs are benefits that vendors gain from the open source model that can be passed on to users in terms of lower adoption costs, but low cost is a means to an end not an end in itself.

If open source is to become more relevant to mainstream enterprise users then the focus has to shift from decreasing cost to issues such as flexibility, performance, reliability and increasing revenue generation opportunities.

Tags: , , , , , , , , , , , , , ,

9 comments ↓

#1 Jay Lyman on 04.15.09 at 9:51 am

It’s interesting to watch and hear what the biggest drivers of open source software are from the perspective of enterprise users, and it seems to be changing over time and across different sectors. One thing I would point out is that the availability of source code, regardless of whether it is accessed or used, goes hand-in-hand with other benefits of open source, primarily avoiding vendor lock-in.

JL

#2 Matt Asay on 04.15.09 at 10:01 am

Speaking of “who you ask is important,” have you considered that Savio has a vested interest in prospective buyers not fixating on cost? He works for a company that sells incredibly expensive software!

What your (otherwise very good) analysis also misses is that while cost *is* a primary driver of open source, the primary reason I see enterprises *staying with* open source is because it offers them more flexible licensing (something that “express editions” can’t match), more innovation (ditto), and better performance (depending on the product, of course). My customers come to Alfresco for a lower-cost ECM solution, but also because they’re looking for a content-services platform that none of our proprietary competition can match, and a big part of that comes down to open source.

Your warnings that proprietary extensions may obviate *some* of the benefits of a free entry cost are valid, but to the extent that this “open core” model proves to be a bait-and-switch, it won’t work for vendors *or* buyers. In other words, the market will ensure that it’s open enough to remain a great model for both customers and vendors.

#3 Matthew Aslett on 04.15.09 at 10:23 am

I was using Savio’s blog as an example. Our research backs up his views.

Also, as I wrote above “other benefits such as flexibility, performance and reliability were considered to be more significant benefits after adoption than they were drivers before adoption”

That’s what I was suggesting vendors should focus on. Maybe I didn’t express that very well. I’ll edit.

#4 Steve on 04.15.09 at 4:17 pm

In my research on open source in identity and access management, the issue of transparency has come up as a major advantage of open source over proprietary. For these customers, both cost and source code access are important, but it’s how the software is packaged that turns out to be the most appealing.

#5 451 CAOS Theory » Open source tunes into the channel on 04.15.09 at 5:12 pm

[…] as discussed in Matt’s recent entry on the drivers of enterprise open source, cost may be what leads customers and users to open source […]

#6 Roy Schestowitz on 04.16.09 at 7:09 am

Savio Rodrigues noted that his experience backs up recent data from Forrester that suggests that “most developers don’t really care that OSS is libre (free as in freedom); they care that it’s gratis (free as in beer)” while Bruno von Rotz asked whether enterprises really care whether the code is open source.

I very much doubt it. it’s the same Forrester that attacked Linux at Microsoft’s behest (and budget). Some leaked documents show IDC doing the same thing by choosing a population survey that is rigged (tilted towards Microsoft) by design.

#7 The cost value proposition of open source redux « rand($thoughts); on 04.16.09 at 2:33 pm

[…] Posted by Savio Rodrigues under Open Source No Comments  Matt Aslett’s post yesterday is one that will surely attract attention and possibly eggs from commercial open source […]

#8 The cost-value proposition of open source redux « BREAKING IT NEWS FOR BUSINESS on 04.18.09 at 5:43 am

[…] Aslett’s post yesterday is one that will surely attract attention and possibly eggs from commercial open source […]

#9 Military Open Source Software (Mil-OSS) WG3 | Transcending Technology - Michael J. Russell on 08.30.11 at 12:55 pm

[…] to any team that attacks a problem aggressivelyWikipedia4. From the 451 CAOS Theory Blog, ‘Neither free as in speech, nor free as in beer‘ by Matthew Aslett, April 15, 2009, downloaded 4/17/095. ‘Turning the Tables?‘ by […]