It’s been an aim of mine for a while to find time to use what I know and can do to help others. But I’ve never had much time for local LUGs since uni (and they don’t really need me), and the charitable tech organisations (like Geek Corps) that I found weren’t something I could spare the time to get involved in, since I was busy trying to make ends meet and get my career to a place I felt comfortable with. I knew I could spend some time, but that I’d prefer to do it from home and make use of my broadband link. Also it meant not being parted from the wife, which work has a habit of doing. It struck me that what was missing was a sufficiently large organisation that provided support, primarily remotely, for people that needed it in developing countries. I wanted – I want – to do my bit, but in a way that makes best use of the resources at my disposal. I could donate a fiver a month, but wouldn’t a few hours of my time (worth tens or hundreds of pounds depending on how one measures it) be more useful?
Being in the Linux/Open Source business for a few years now it’s struck me that uptake of Linux in particular in schools (and other orgs with scant cash resources that could be better spent elsewhere: like on teachers) is hampered by the lack of low cost, dependable professional support. This situation is identical to that faced by businesses who did not have the coinfidence that Linux can return benefits over the long term, supportable and supported by both internal staff and an industry that can provide professional Linux/Open Source services. Like Red Hat, like IBM, like SuSE, Canonical, Novell, Spikesource and a growing raft of other companies.
Now consider a set of schools in a developing country and the resources available are exponentially smaller. If it is dependable, professional support that will help, then low cost is not good enough – it must be free (as in beer). This realisation got me thinking that if a sufficiently large group of charitably minded FLOSS geeks (desktop, server and network admins, etc) got together then we could provide a reliable, professional set of services for vertain locations globally. With enough people then this thought of open sourcing services could, with great organisation, energy and will of it’s members, provide industry strength support at zero cost for the end user: an org/school/charity/medical centre etc that could vastly benefit from the enabling power and freedom of Open Source.
So what would it take? Well I know from experience that it is possible to provide 24×5 enterprise Linux Help Desk support with less than 10 people (I did it with 7) globally for over 1000 end users. However I think hundreds, but probably thousands are needed. Why? Momentum, growth, scale: more volunteers means more help can be given. The vision is wide ranging but it is possible to start small. Perhaps with as few as 10 people. What would that give us? Well if 10 people could pledge and give 2 hours per week, then that is 2×5 support (2 hours, 5 days a week) with 2 people on at any one time, to cover for illness etc. More personnel would be needed as the services grow in depth, coverage and type. Apart from people it would also take resources, personal and organisational. Personal resources are easiest: basically a computer connected to the internet is all that is needed to get started. Organisational would be stuff like web hosting, bandwidth, mirroring services, Legal/business help, marketing etc – but that’s for later.
I’ll just braindump the rest of my ideas for now which have been building up for months. classroom-in-a-box, internet-cafe-in-a-box ideas where the only install required (http/ftp/cd/dvd) is to the server with a distro selected from a range available, wrapped in kickstart (or equivalent), the freshly installed server then itself fully configured to be the kickstart (or other) server for the other clients be they thin, fat or something in between. I forsee a matrix of distros on the top, with install types (school, cafe, medical centre, office) on the left, that build distro-independent identical servers via a unified build environment, that can then install any client computer of choice. More tomorrow….
