Open Source Debate
Open Core – What Is Acceptable?
Henrik Ingo is the latest in a string of bloggers debating the merits and evils of Open Core, but he takes this debate one step further and asks 'if Open Core is unacceptable, then what is acceptable?'
Firstly, Ingo states that if a vendor sells primarily closed software, but markets themselves as an open source vendor, it is unacceptable. 'True' open source companies have put time, effort and money into promoting open source, turning it into a well-known and respected term. In Ingo's opinion, it is unfair to capitalise on all this hard work, or to exploit well-meaning community members who may inadvertently contribute to closed-source software, which they believe to be open source.
Open Core further muddies the waters, because the mixture of open source and closed software the label describes isn't particularly uncommon. Brian Prentice drew attention to this at his blog, stating that “the idea that a functionally complete, proprietary solution is somehow unique because it was built atop an open source base fails to recognize the fact that many proprietary solutions are being built using open source components.” In fact, Google, IBM and Microsoft all use elements of open source software in their products. Logically then, they could claim to follow an Open Core business model, or to be an 'Open Source Company,' to use another term for dual licensing. Again, the end result is that customers who are not already familiar with open source, receive a confused message that 'open source' can sometimes equal proprietary software. “If the vendor markets his open core product as open source, it adds up to deceit and it waters down the open source brand in general,” Ingo concludes.
Ian Skerrett also recently called for an end to the term 'Open Source Company,' using the same argument: that this label was often used by primarily closed source companies leveraging the open source reputation for financial gain.
So, what is the solution? Ingo proposes that if a product contains both open source and proprietary components, it should be called closed source. Ideally, he would like to see open source and closed components kept strictly separate, with clearly-distinguished differences in branding and product name. He gives an example: Apache is open source, but not everything labelled 'MySQL' is, in fact, open source. Apache code may wind up in closed software (IBM's Websphere contains Apache code,) but it is never called an Apache product.
Furthermore, Ingo advises that the three main Linux distributions - Red Hat, Ubuntu and Debian – should block open core brands. He already sees some evidence of this; Red Hat has been careful to keep its distribution free from closed source kernel modules, and Debian renamed 'Mozilla Firefox' to 'Firefox' as a workaround for Mozilla's trademark policies.
Finally, he sees educating companies about open source and Open Core, as an important area. He believes that some companies may not even be aware that their business doesn't qualify as 'true' open source, they might simply be following the example set by MySQL and other prominent mixed open and closed software providers. “Once people know about this issue, it will already have some effect as open core vendors need to reconsider if the closed source software is so valuable to them that it offsets the loss of community goodwill,” he concludes.