Don't mention "open core" to Larry Augustin. Trust me, not a good idea.
He gets a small twitch in his face and enough clouds build that, even though I've got a few inches, a lot more pounds, and some martial arts training on him, I felt it prudent to take half a step back when I stupidly lumped SugarCRM in with open core products in a barroom conversation with him last month.
The clouds dissipated and I, and my drink, emerged unscathed. But I did get a thorough explanation from the CEO of SugarCRM of what open core is, and why it is not the same as dual-licensing.
And that, it seems, is a common error: when I first noticed the term "open core" trundled out on the blogosphere, I thought it was a catchy new way of describing the dual-licensing approach that a lot of commercial open source vendors have taken to sell their products. That is not correct.
The classic example for dual-licensing is MySQL, which has two licenses: the free (as in beer) version under the GPL and the not-so-free non-GPL commercial licensed version. With the commercial license, customers can sell a MySQL-based product without making the product open source. SugarCRM, it should be noted, has a similar dual-license approach, with the Community Edition under GPLv3, and the commercial versions (Sugar Professional and Sugar Enterprise) under a proprietary license.
Open core is sorta-kinda similar (hence the confusion) but only on the surface. Open core doesn't rely on the licensing of software, so much as the features of the software itself.
Take MindTouch, a web-based wiki collaboration platform. It's an open core product (indeed, MindTouch CEO Aaron Fulkerson is credited with coming up with the open core business model), which means that the freely downloadable MindTouch Core is licensed under the GPLv2. But additional add-ons to the MindTouch Core product are sold as proprietary products. So, as the name implies, the basic element of an open core project is always free or open source, but the extra peripheral goodies come with the proprietary price tag.
Subtle, but definitely different.
It goes without saying that open core has a few critics. To be fair, I am unsure if Augustin disagrees with open core on principle or that I'd erroneously labeled SugarCRM as open core. Take it up with him--carefully.
One known open core critic is Simon Phipps, who this week fired a direct shot at the business model in a Computerworld UK article titled "Open Core Is Bad For You." And, in case you think Phipps' headline was too vague, here's the intro to that article:
"The open core model is being feted as the new default open source business model. But I assert it does not deliver and sustain the principle that delivers cost savings and flexibility to the customer--software freedom. As a consequence, businesses who live-or-die by open core risk the fate of Compiere ERP unless they can manage the incredibly delicate balance their customers will discover they demand."
That Phipps is not known for beating around the bush.
Of course, there were rejoinders to Phipps' arguments, the most notable coming the next day in the same publication from former MySQL AB and current Eucalyptus CEO Mårtin Mickos:
"Simon has some great points in his posting yesterday, reminding us all that the non-open features or services a company provides to its customers may lead to lock-in and reduction of freedoms for the customer. He also comments that open core businesses ;stand to benefit massively; from this. It seems that he is arguing that this is a bad thing. My main point is the opposite: by having vendors in the open source space that benefit massively, we will have a stronger world of free and open source software..."
Before the open source world gears up for another round of epic but ultimately useless flame-warfare (GNOME vs. KDE, vi vs. emacs, Linux vs. GNU/Linux), it is helpful to put things into perspective, which was done most eloquently by Redmonk analyst Stephen O'Grady.
O'Grady argues that for all practical purposes, open core is the same as dual-licensing, in that it's just another way of marketing open source in a predominately commercial software world. O'Grady isn't defending open core--he emphatically states he doesn't like it at all. But he acknowledges that there is little open core opponents can do about it, since there is nothing in the open source model that can put the brakes on the open core model.
Critics of open core can either keep complaining about it, or they can compete with "pure" open source products. That may seem tricky, until you read the examples O'Grady puts forth:
"Historically, remember, the largest open source projects in the world have been neither open core nor dual licensed. As popular as the dual licensed MySQL database is, Linux is yet more popular still. And as commercially prevalent as open core is today, none of the products sold in that fashion are remotely as popular as the Apache web server," he wrote.
In other words, let the market decide which model works the best.