Thursday 16 February 2012

Give penguins a chance

Would switching to open source software save public money? I don’t know, but we should at least try to find out.

The Windows logo versus the Linux mascot. A little-known but very bloody feud.

I know software testing is a very absorbing activity, but in between bouts of testing you might have noticed there’s a bit of a financial crisis going on. As tax rises, benefit cuts and axing public services don’t go down that well with the public, the government is keen to find less painful ways of saving money. This, in part, was the idea behind the Spending Challenge letters that went out to all public sector workers shortly after the 2010 election asking for ideas to save money. The ideas ranged from the pragmatic to the ridiculous, but one suggestion that caught my eye was to switch proprietary software for free open-source alternatives. This is not an unthinkable as you might expect; the Lib Dem manifesto said they’d look into this, and George Osborne himself is said to be interested.

I’ll be open and upfront here: I use Linux, LibreOffice (effectively the successor to OpenOffice) and other free open-source products wherever possible. It’s partly I don’t want to pay for software when free stuff does the job, and partly because I have problems with the way Microsoft uses its dominant position to make life difficult for people who use competitors’ products. But I don’t believe in imposing my views on other people, and I’ll help out with any IT problems whatever software they’re using. (Indeed, a software tester who doesn’t is a short-lived one.) I wouldn't push savings too much with a charity (Microsoft usually heavily discounts software for them). I’d also be hesitant to encourage a small business to switch to open-source when everyone they work with expects them to do all things Microsoft. The public sector does not have that problem – they mostly communicate with each other, and they’re big and ugly enough to insist anyone else works with their software if they wish – but any move away from Microsoft or any other proprietary software must save the public money, and not just be done to prove a point.

Open source is a far better option that it used to be. As little as ten years ago, getting Linux to work was a nightmare for even the most tech-savvy users. Nowadays, however, it’s as easy to learn how to use a Linux-based computer as it is to learn a Windows-based one, and for the most basic layman’s tasks, the differences between the two are trivial. Microsoft Office still offers a lot more features than LibreOffice, but most people don’t use the advanced features anyway. Those people who claim that you’d have to pay if you’re a business, or that they might start charging you later, are mistaken – the licence used makes this impossible. There’s a lot of talk over the costs of acceptance testing or retraining, but it’s broadly similar to the hyperbole against moving from IE6 which I’ve already been over.  And if you still think Linux is only for geeks in their bedroom, the continuing success of Linux-based Android tells a different story.

But Microsoft does make one valid point: there’s more to the cost of IT in business than the licence. The term Microsoft keeps banging on about the Total Cost of Ownership, and much as I hate buzzwords, it has to be taken seriously. There’s labour costs associated with installation, maintenance, fixing problems before they disrupt your business, and the hardware needed to support your system. Microsoft also claims that if software’s free, there’s no-one on the end of a phone if things go wrong. That’s not really true any more; the major Linux distributors sell Enterprise packages that include this support, but the fact remains it costs money. The bottom line is that Microsoft claims their software works out cheaper when you factor in everything. I find some of their anti-Linux claims to be dubious, but that’s just their marketing department doing their job, and I wouldn’t be surprised if Microsoft and Canonical do the same.

Anyway, here is my idea. It’s a suggestion which the Government is welcome to take up without any need for acknowledgements or royalties. It’s a tried and tested method which works in every other area of government business when different companies claim to provide the same goods or services for less money.

Without further ado, the solution is …

[Drum roll]

… put it out to tender.

At the moment, public sector IT contracts generally are a choice between company A providing Windows and MS Office, company B providing Windows and MS Office, and company C providing Windows and MS Office. That’s not good enough. I can’t think of a single example other than this where it’s considered acceptable to choose one company without considering any competitors. It doesn’t have to be a choice of all Microsoft or no Microsoft; it’s perfectly possible to run LibreOffice on Windows, Microsoft Office on Linux, or mix and match pretty much any combination of open source and proprietary components. Claiming Microsoft is the only option doesn’t wash any more – government bodies elsewhere in the world have made the switch and managed. Claiming it’s what everyone uses is a poor excuse for any government that believes in free and fair competition. If 90% of motorists drove Skodas, would anyone argue the Government should help make it 100%?

What should we consider when awarding the contract? Anything we think is important, just as long as all sides get to make their case. Does Microsoft believe their software is cheaper to maintain in the workplace? Are their servers easier to maintain? No problem – let Microsoft make their case, let the open source vendors reply. Is there a problem with a Microsoft lock-in? Their licensing arrangements? Let the open source vendors say why there is, let Microsoft say why there isn’t. Does Microsoft or Linux offer better security? Which is faster? Which is more reliable? For all of these questions, we should be asking the vendors to make their case themselves, rather than picking one and dismissing the others out of hand.

And what if the winner is Microsoft, Microsoft and more Microsoft? It will still be worth the paperwork. Experience shows competition is good for Microsoft products. Microsoft moved on from the horribly outdated IE6 because of competition from Firefox. When the XBox’s standing was threatened by the revolutionary Nintendo Wii controller, they responded with the equally innovative Kinect. There have been advances in Windows and Office in the last two decades, but two things in particular have never really been addressed: why it’s necessary to pay hundreds of pounds for software when you only use 10% of the features, and why the processing power needed to run them balloons as quickly as processing power of computers. With real competition to the office market and something might be done about this.

Will this happen? On the one hand, if the Cabinet Office consider upgrading from IE6 to be too difficult/complicated/expensive, there isn't much hope. On the other hand a consultation was launched last year on this area, and although it seems to be confusing open source with open standards a bit, there are signs that the Government is starting to recognise the need for proprietary and open source software to compete on fair terms. The Government is in a far better position to bring competition back to IT than any other company, and if they stick this course, it could be rewarding for everyone.

No comments:

Post a Comment