Gallery

State of the nation

Posted on 28 Jan 2009 at 15:59

Simon Brock looks at the future of open-source software in light of the economic downturn, but warns that not all free solutions are best.

We've had this system in place for a couple of years and it has worked really well so far, but by the time you read this column we'll have replaced it and moved over to Google Apps for Enterprise. The inevitable question I hear you asking is "Why?" After all, I'm obviously an advocate of open-source software, and I do sincerely believe that all the products I've just mentioned are very good at what they do. So why change now?

To put it simply, email is very important to us, but the system above has become just too complex to maintain for a company of our size. For £25 per seat per year I can get Google to look after all our email and we won't have to do very much at all to make it work - and in case you didn't know, we will be able to use our own domain. All the features I mentioned above, including the mail filters, are included already, along with any new features that Google brings out. Assuming that I buy 20 mailboxes, which is probably more than I need, that will cost me £500 per annum, and for that I get all the management facilities I get from my current system plus a pretty good web interface. Each user gets 25GB of mailbox space (500GB in total) that would cost me £100 for two disk drives to implement with mirroring. Obviously, I then have to wrap up those drives inside a server or two, and then put the software onto them. By the time I'd done that and put the machines in racks, I'd probably have spent more than £500 already.

However, the real cost comes when it doesn't work or, more likely, doesn't work properly. While we've done pretty well in keeping our spam and virus filters automatically updated, every so often they've needed some manual intervention, and each one of these interventions costs time and therefore money. Most likely that means our email doesn't get delivered so quickly, and while everyone knows that email can be unreliable, nobody will accept that as an excuse whenever it happens.

With Google Apps I will also get a collection of other toys that we may or may not use. The "office apps" are nowhere near as good as real Office apps (neither the Microsoft or OpenOffice varieties), but they can be useful in an emergency or for collaborative working. We already use Chat, and I'm not really sure I want video chat!

The upshot is that while I don't actually know the precise TCO for my open-source mail system, it doesn't take much working out that the TCO for Google Apps is going to be lower. Even if there are problems I haven't foreseen, at least I'm following a well-trodden path. While I doubt I'll get the same level of service as The Daily Telegraph, which has recently moved to Google, I do know that there are at least another million enterprise users who are heading in the same direction...

End of open source?

So do I think that Google and other firms are going to destroy open source completely? The answer to that is a most emphatic "No!" Open source is here to stay and will continue to make a difference. Even though we've given up our open-source email system, there will be others who'll continue to develop and use open-source email. Other people may have different TCO models from ours that allow them to continue using open-source software, or for that matter Microsoft Exchange. Many of the concerns that people cite as reasons for not outsourcing their email (such as resilience and security) tend not to be such big issues as they think. After all, the people who have changed have managed to get over those problems.