why open source?

This newsletter, as I said in the introduction, open sauce is all about digital liberation: “liberat[ing] your digital life from its technological prison”. So why have I only suggested open source alternatives to proprietary software applications like Microsoft Office and various video conferencing platforms? Why not suggest proprietary software that isn’t sold by a monopolistic tech giant like Microsoft or Adobe? Why not suggest freeware? Why is a particular form of software licensing so intrinsic to my model of digital liberation?

In this post I’ll outline some of the reasons why using open source software is so fundamental to digital liberation. This will include practical and ethical considerations with the proviso that the ethical considerations are infinitely more important to me than the practical considerations so just skip to the end if you want the real reasons for using open source software.

cost

Cost can be a factor in using open source but is never the driving factor. This comes down to the distinction between the two meanings of the English adjective ‘free’: gratis and libre. All the tools I’ve suggested in this newsletter so far are gratis (free as in beer ^[1]) but that is incidental to being libre (free as in freedom). There is plenty of open source software which is sold at a monetary cost. Conversely there is plenty of proprietary software which is available for free: this includes freeware like Adobe Reader or Skype.

Freeware could be seen as a means of digital liberation: sometimes it’s from smaller developers and allows you to get out from under the thumb of big corporate developers like Microsoft. I won’t advocate freeware because it’s not actually liberatory: it’s only gratis and not libre. For example, in librarianship there is a widely-used metadata editing application for Marc bibliographic records called MarcEdit. MarcEdit is freeware developed by Terry Reese who makes MarcEdit available for free but has not open source licensed the code. Thousands of libraries rely on MarcEdit for their cataloguing work and it is a single-point-of-failure for those thousands of libraries. At any point, Terry Reese could stop updating MarcEdit, remove the downloads page, and set thousands of libraries on a slow slide to the software’s obsolescence. Every library that relies on the software is ceding that small portion of their technical freedom to be controlled by a single person (even if that person is entirely benevolent with their control). Because it’s not open source, it’s not liberatory and it’s dangerous to rely on it.

Even open source software that is gratis is often only nominally so. Even if the software itself is provided at no cost, there can be often-huge hidden costs for labour and hosting. The technical skill required to run some software is a cost that you either take on yourself if you’re someone like me who installs and runs their own shit or that you take on by paying for the labour of another person.

security

Open source software is generally better software than proprietary software. Nowhere is this clearer than in regards to security. Security expert Bruce Schneier articulated this well in his newsletter when he said that “[i]n the cryptography world, we consider open source necessary for good security; we have for decades. Public security is always more secure than proprietary security. It’s true for cryptographic algorithms, security protocols, and security source code. For us, open source isn’t just a business model; it’s smart engineering practice.”

The basic reasoning here is that open source software is more secure than proprietary software because it has to be. Since the source code for proprietary software cannot be publicly seen, their security is often security through obscurity: it’s only actually secure because its concealed and if someone gains access to the source code, they gain access to all the underlying security mechanisms. The user is forced to accept whatever level of security the vendor is willing to develop. By contrast, open source software is open—open to customisation; open to modification; open to review—and so security mechanisms have to be actually cryptographically secure to everything except the key. This is known as Kerckhoff’s principle after the cryptographer Auguste Kerckhoffs: a cryptographically-secure system should be secure even if everything except the key is public knowledge. In other words, an adversary can see exactly how the lock works and yet still can’t open it without the key.

There are other security benefits to open source software: more eyes on the code means more opportunities for reviewers to spot security holes and plug them; more ability to customise the code means that, if you’re really paranoid, you can beef up the security of code yourself. Open source software is generally more secure than proprietary software by its very nature.

customisability

I’ve worked in various UK universities for more than a decade now and I’ve seen how every university is different: different subjects; different student bodies; different ways of doing things. Despite this diversity—and indeed the strength of this diversity—UK universities love to purchase expansive, generic proprietary software systems that reduce the university down to being the same as every other university. This is particularly egregious for library and publishing systems which reduce the richness and uniqueness of a library collection or a centre’s research outputs into the cookie-cutter model which will fit into a big commercial proprietary system specifically designed to appeal to the broadest possible commercial market.

Open source code allows for the extensive customisation that something culturally rich like a university deserves. This modification can be as simple as renaming a metadata field to reflect your institutional context or as extensive as forking a codebase to form an entirely new piece of software. When I worked at Soas Library, the open source nature of the VuFind library catalogue I helped to implement meant that we were able to customise it to represent the characters of the diverse languages of the library collection without errors.

Most open source software is used without customising the code. I run almost entirely open source software for personal and professional use and for 95% of that software I run it entirely without modification. But it’s the capability for customisation that confers digital liberation: you have the freedom to customise it even if you never actually change or even look at a single line of code. It’s that potential that gives you freedom and that gives you control over the software even if you never choose to exercise that control. As Cory Doctorow has succinctly put it, “If you can’t open it, you don’t own it.”

community

For me, the most important reason for using open source software is community. At first sight, there’s perhaps something of a contradiction between the practical considerations above which tend towards an individualistic (left-)libertarian ideal of individual freedom from the corporate constraints of proprietary software and this ethical consideration about using open source software because it’s good for the community. But I think this is where the praxis of open source software as digital liberation dovetails with my anti-capitalism to say that there’s something morally beautiful in sharing and sharing what you work on openly with other people in society.

Developing open source software and using open source software lets you give your work back to others so that they can benefit from it without the base restrictions of the market economy. When I write a software script for my own specific purpose and open source license it so anyone else could use it, I’m reaching out to share knowledge for the collective good. In the sense of Lewis Hyde, I am offering my work as a gift or, when using someone else’s software, accepting a gift from others. This knowledge sharing and gift economy disrupts the capitalist market paradigm.

As with gratis freedom, anti-capitalism is not a necessary part of open source software and open source licensing is not necessarily anti-capitalist liberation. Open source software is often provided by corporate software companies like Red Hat, Inc. or Microsoft: Microsoft owns GitHub which is a major hub for the sharing of open source code. Rather, open source software enables anti-capitalist liberation: it allows for divesting from corporate software and corporate control.

I used to work in librarianship and was occasionally frustrated that the same ethical principles behind embracing open access publishing—divesting from big corporations exploiting academic labour; the drive to share knowledge openly and communally^[2]—were not equally applied to embracing open source software in librarianship. I’m very fortunate to now work on the Open Book Futures project as part of Copim where principles of openness are baked into everything we do particularly through the key value, “No open access without open infrastructure.” As my Thoth colleagues put it, “open access publishing will not prove to be viable in the long term unless the entire scholarly publishing network connecting authors with reviewers, publishers, funders, repositories, and institutions is managed via interconnected, collectively managed open protocols.”

So why use open source software? Because the struggles for open are all linked: we can’t embrace open access without also embracing open source. The fight for liberation includes digital liberation and taking some measure of control over the software that we choose to use. Open source software, in its best form, can be a foundation for divesting from corporate control over our lives in the spirit of anti-capitalist liberation.