Comments on: Tim O'Reilly: Open-source purists trying to answer the wrong question

Open-source purists continue to fixate on the wrong argument, worrying about free riders when they should really be thinking of how to create data-driven businesses, Tim O'Reilly argues.

[Crosbie_Fitch]

I suggest that free software is about liberating software engineers and the public in general from the shackles of copyright and patent, and the epiphenomenal effects of such privileges (the closed source tendency of the proprietary software development process and copyright based business model).

It is not about making source code visible for the purposes of collaborative development, making software free of charge, nor is it about preventing free riding or compelling reciprocation from others. Those may be epiphenomena of free software, but they are not the objective.

The mission of our times is to liberate the people, to restore those freedoms suspended by copyright and patent, ultimately to abolish those unethical and anachronistic privileges unsuited to the information age.

[pentest]

All open source licenses depend heavily on copyright. The author of the code retains copyright in open source code.

[odubtaig]

As far as I'm concerned (I know. Who am I? Just some schlub on the internet) it's not going to become about the data because for me it's always been about the data. If I put my data into SQL Server or OracleDB or Google's BigTable then they are the gatekeeper, they hold the key and I'll always have to go through them to get my data, often at cost. Even with word processing if I use a closed source program or I'm paying a subscription for an online service and something happens that means I can't use that original program I'll probably have to pay the same people again to access my own data.

In those situations I don't actually own my own data, the vendor does.

With something that is at least built around an open standard, even if it's not F/OSS, if the same situation occurs I may have a choice of who I pay to be able to use my data again. With enough money (such as a company may have) a bespoke software may even be created as is the case with many companies needing to meet an uncommon neeed. I could even learn how to do it myself if I had a lot of time on my hands. Ultimately, I would own my own data. This doesn't automatically mean it costs nothing to access it but it does mean that company X doesn't hold the only route to accessing it.

The software, though, is still hugely important because it does something that is required to make any data useful: it turns it into information. It is the tool required to interpret and manipulate the data.

This is why groups like the FSF focus on the software because it is required to do both these things and if the software is F/OSS then the tools that make the data useful and, by virtue of being exposed through the access methods used by the software, the data are free.

[hozelda]

Software lock-in is powerful. FOSS is not just a developer phenomenon, it's a user phenomenon. While many companies look for ways to lock-in customer, customers don't want to be locked in. In the FOSS world, users are frequently developers to some extent so they can do something about it. The (A)GPL and similar license help even the playing field to prevent a very strong player from using lock-in to dominate. Many developers like that license because it gives the edge to those that contribute the most rather than to those that contribute the least.

[realmerlyn]

Hey, thanks for mentioning FLOSS Weekly. In return, I mentioned your article in the opening of the upcoming show on Saturday. Props!