|Think about Loose Coupling|
But it doesn't change that it is a very prevalent attitude. Especially in large corporations and government departments where the building of new platforms for internal and external distributions takes takes 9 to 18 months of development, cost many millions and target 3 year stability periods. And that's when things are good. As many companies found over the last 2 to 3 years, when economies are bad, old platform lives get extended, new platforms get deferred and life-cycle targets get increased.
doesn't do .0 releases anymore
All that means is that x.1 becomes the new x.0. People aren't so easily fooled.
From first hand experience, when large government departments are picking and choosing which releases of third party software are integrated into their platform builds; they aren't fooled by such transparent trickery. They don't pick on the basis of version numbers. The release and bug histories of each package are considered -- going back several releases, to the beginning in the case of some newer packages and if -- as Perl historically has -- a product is seen to require 2 or 3 point releases in quick succession after a new major version in order to work out the kinks; then a new version will not be considered until it reaches a similar point of stability. And if that means sticking with a mature point version of a back-level major release, so be it.
And when a government department targets a 3 year life, they seek 5 years of active support. Look at the support period histories of XP. SAS, Oracle, RH, ...
With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
"Science is about questioning the status quo. Questioning authority".
In the absence of evidence, opinion is indistinguishable from prejudice.
In reply to Re^7: What is a really old version of Perl? (Instability == death knell)