gOS(h)

As you may have heard, Google have decided to step into the Operating System arena by releasing one aimed at revolutionising the desktop OS market.  At launch, it will be targeted at the lightweight ‘netbook’ range of devices specifically for people who require regular, rapid access to the Internet.   Information coming from the Mountain View chocolate factory is that it uses the Linux kernel using a ‘new windowing system’, presumably their own GUI rather than relying on X Windows and KDE or Gnome.  Presumably it is a rework of the Android platform to better support larger screens, keyboards and mice, although they do say that the two projects are not intertwined.  The OS also has Chrome Browser built in and is designed to be an interface into ‘cloud’ computing resources that have been talked about over the last couple of years.  Because the big G now provides ‘office’ resources such as E-Mail, Documents, Spreadsheets and Presentations online, they don’t see the point in having massive computing resources on the desk to write a simple document, when their server infrastructure can provide that for free (on the basis that they’ll sell advertising to you) provided you can get online.

Its an interesting approach to take, and it will be interesting to see how it pans out.  However, I do have a few concerns.  Firstly,  the Internet is NOT yet all pervasive.  Sure, you might have it wall to wall at home, and you may have access in the office.  You may even have 3G on your phone or via a dongle to get online whilst you’re mobile,  but you don’t have to travel very far to enter an internet black hole.  For example in our Office,  I can get HSDPA access with T-Mobile and 3 Internet,  but BT/Vodafone barely even has a 2G signal.  Likewise driving home,  even though I travel the M6 corridor, one of the busiest routes on the motorway network,  there are at least 3 spots where the signal drops out completely.  At the moment, this isn’t a problem, because you don’t need access to the net 100% of the time to write a document or update your e-mail.  But if you’re sat there with a blank screen because your network has dropped out, frustration levels soon start to set in.   

Of course, you can get round all this with offline synchronisation, which is the one thing currently missing from another cloud computing environment you may be exposed to – Citrix.  However,  you’re then NOT working in the cloud when you’re working offline, which means more storage, processing and memory requirements, defeating the point of a lightweight computing model.

Maybe my fears are unfounded,  or will be addressed when 4G (WiMax or LTE) appears, but I still can’t see full coverage if you’re stuck up a mountain somewhere miles from the nearest transmitter.  Still, I await this project with interest and see how it pans out in the battle with Microsoft and the soon to be released Windows 7.   Its good to see Linux continuing to make inroads into the end user environment,  I just hope it doesn’t become the pervasive kernel standard to the detriment of other Linux projects.

BootNote:

You may have heard the tale attributed to various IT Thinkers of the early 20th century but generally now attributed to be Thomas Watson of the IBM Corporation.  The quote believed to be from 1943 goes “There is maybe a world market for maybe five computers”,  ironic considering how ubiquitous computing now is.  But the point is that Google is now building a single massive global computer (a mainframe if you will) that we all have access to via what could be pretty ‘dumb’ devices.  This computer currently runs the worlds most popular search engine,  it hosts videos,  allows you to go shopping and send Emails.  In 50 years time,  will computing be done not at the desk, but in ‘the world’?  Will we access the internet in the same manor that we access electricity today?  We currently plug an electrical appliance into a socket and it works using electricity.  In the future, will we plug a terminal into a socket and the internet ‘just works’.   Don’t be surprised if we move to a “pay for what you use” ‘net where you don’t buy access to the grid, but you do buy access to the resources you use just like the utilities of today.