Open source is inferior to commercial software

I feel compelled to reply to Helen Barrett’s recent weblog post where she talks a little about Elgg and open source in general. Specifically, she says:

Of course that is one problem with open source software… without a business model to support the development, it can take longer to implement changes unless there is a regular funding stream. My experience with commercial tools shows that the companies are very responsive to their customer base, and have the resources to support ongoing support and development.

In my post Why does Elgg run PHP and MySQL? I pointed out a number of important reasons why going for the open source route might be a benefit to an educational institution. The big benefits are to do with the freedom this approach affords, which I’ll restate:

  • Freedom to copy
  • Freedom to modify
  • Freedom to reproduce
  • Freedom to redistribute modified versions

One of the big issues with commercial software is that it doesn’t always exactly fit an institution’s requirements, and many have found themselves actually altering the way they do things to fit the software! This is surely a backwards approach: software should be a tool designed to solve a problem. If you’re changing your procedures to fit the software, you’re actually creating a problem to fit the solution. It’s much better to be able to take a piece of software and knock it into shape to meet your needs; given that you can download and run Elgg for free, it’s much less expensive to be able to do this than with a commercial tool. There are a number of companies – Curverider and Aperto among them – that offer Elgg-related consultancy services, so you don’t even have to hire your own resource.

Generally, this is true of successful open source projects. Moodle, the Modular Object Orientated Dynamic Learning Environment, has a service network that provides a range of services depending on your requirements. Or, you can always take it out of the box and use it for free. The point is, you have more flexibility than you would installing (say) Blackboard, for a great deal less money. That’s money that can hopefully go towards teachers and other materials.

In all cases these companies are as responsive as any commercial company. And you don’t have to hire them – if you have your own technical resources, great! You can alter the software to fit your needs yourself, for no extra cost. Either way, this is both the business model and the funding stream, although it doesn’t have to be the only one – Firefox is said to get $1 per year per user from use of its Google search box, and Curverider has a non-programming paid service in the works.

This, however, leads me to my next point: the suggestion that commercial software providers are more responsive than their open source equivalents is plain wrong. The big example here is Microsoft and their Internet Explorer web browser. This corners something like 80% of the browser market, but is insecure enough for organisations like Penn State and the government of the United States of America to ask people to stop using it. Yet with all this negative press, security fixes still take three times as long as for Firefox. Microsoft’s advice for circumventing one security vulnerability? Don’t click on links; type in the URL instead.

Firefox is, of course, one of the big open source success stories. Like all open source projects, you can download it for free and modify it to suit your needs – hence interesting developments like Flock and Songbird. Its market share is increasing, it’s faster, and because it’s open source its guts are out for everyone to see. Internet Explorer is a black box; Firefox’s code is transparent. If you have the ability and the time, you can knock through it and see how it works. Do you really know that Internet Explorer isn’t sending back your browsing history to Microsoft?

This has obvious benefits in some very important areas. If the code for medical machines is open source, for example, problems can more easily be discovered and fixed. Similarly, take the voting irregularities in the US: the owner of Diebold, which manufactured most of the (closed-source) electronic voting machines, stated that he was “committed to helping Ohio deliver its electoral votes to the president.” With a closed-source system running often uncertified software, who’s to say those machines weren’t biased towards that result? Similarly, in education, you can prove that open source software used is fair and unbiased.

With all these benefits, why does open source still have such an unclean reputation? Josie Fraser recently highlighted the benefits of OSS Watch, but how much good are they and similar organisations actually doing in promoting the virtues of this free and open approach? Their own description states:

Perhaps you have heard that open source is a movement espoused by men (and some women!) sporting beards and sandals. Perhaps you have simply no idea what it is and you want to find out more. Whatever your motivations, it’s always a good time to start learning about free and open source software.

It’s great that they’re inviting people to come in and find out more. It’s not so great that they’re framing it as something espoused by beardy computer scientists. There’s a certain amount of marketing of ideas that needs to happen here, and we’re well past the point where software is just something for sandaled hobbyists in darkened rooms playing with motherboards and chipsets. It touches every aspect of our lives, and as a result it’s vitally important that we know what it does; and more importantly than that, any message we give about it has to be inclusive. The beardy image is something that is just going to send people away. While people like Richard Stallman – instigator of the free software movement – are doing some good work, the fact that they can sometimes come across as certifiable geeks isn’t helping anybody.

Similarly, organisations like the Open Source Initiative and OSS Watch would do well to learn from Mozilla and, to a lesser extent, Ubuntu. Make things friendly, emphasise that this is for everyone’s benefit (instead of just the computer scientists), and hopefully we can knock over this ideological barrier.






Leave a Reply

Your email address will not be published. Required fields are marked *