Moving on to the next stage of virtualisation
I refer to your recent articles on virtualisation, and Jon Rolls' letter noting that desktop virtualisation was not covered (Computer Weekly, 27 February).
I would like to add that application virtualisation takes the process a step further - allowing otherwise conflicting applications to run side by side on the same PC.
Watching Altiris SVS running different versions of Microsoft Office simultaneously on one machine demonstrates the power and flexibility of this technology, and may one day lead to all applications being shipped as virtual layers, which we just load and run on our PCs. No more DLL hell and uninstallation problems.
Microsoft too has recognised the power of application virtualisation by buying Softgrid - so watch this space!
The real power of the grid is the ability to collaborate
Alex Hardisty, Grid Centre manager, Cardiff University
It's a shame Arif Mohamed's article on enterprise grids (Computer Weekly, 20 February) places so much emphasis on the notion of a grid as just a vehicle for delivering massive computational capacity to the enterprise.
Grid does that but Mohamed fails to highlight the real potential - that is, grid's ability to help people from different organisations to dynamically share IT-based resources whilst collaborating on projects of mutual interest.
In their landmark paper, "The Anatomy of the Grid", Ian Foster, Carl Kesselman and Steven Tuecke assert that "the real and specific problem that underlies the grid concept is co-ordinated resource sharing and problem solving in dynamic, multi-institutional virtual organisations".
There are very few virtual organisations (consortia) in existence today where the full potential of IT-enabled collaboration is routinely achieved.
The situation can be likened to that of the mid-1990s, when every organisation had an intranet but almost none had an extranet, and when they did it was bespoke.
The grid is like a teenager's brain, experiencing the change from child to adult. Not all connections are properly wired. Some have yet to be switched on. It is in transition: first generation from academia to industry and commerce second generation towards a service oriented architecture.
For the grid to succeed we need appropriate open standards to be widely adopted, and a shift in socio-cultural understanding of what the grid is all about. Thence, in the early years of the next decade, we will see the real power of the next generation grid as an enabler of collaboration. Understand and nurture your teenager!
Don't be short-sighted about iris recognition
David Alexander, Vistorm
I must take issue with Andrew Meredith, who cast doubt on the security of biometrics (Letters, 6 March). I am an information security consultant and have studied this subject extensively.
While I agree with his comments about fingerprints - the work of Matsumoto and others in Japan has clearly demonstrated the flaws in the sensor technology - I cannot agree with his comments about iris recognition. The "popular TV programmes" Meredith referred to are not always known for the rigour of their background research and this would appear to be one of those occasions. It is perfectly possible to build an iris scanner that is highly resistant to Type 1 and 2 errors and almost every attempt to deceive them.
I suggest that the iris scanners used in the programmes Meredith saw were old technology and not representative of the latest devices.
A final point is that iris recognition is the only practical way of differentiating between identical twins, which even DNA testing cannot do.
Anyone needing further data is advised to look at the web pages of John Daugman of Cambridge University, who has done much of the work on the algorithms used in this field.
Think laterally and don't follow the ITIL pack
Adrian Montagu, infrastructure manager
I have been reading your Next Move column and I am amazed at everyone's lemming-like following of ITIL.
Wasn't ITIL created/instigated by a government department? I cannot think of one government project in IT that has been on time, on budget, or has functioned properly. On that basis, ITIL should be avoided and some lateral thought applied.
Do you have a fresh take on someone's opinion on this page, or something to say about a Computer Weekly article? E-mail email@example.com. Please include a daytime phone number.