on Qua 08 Fev 2012 16:26, bill said to people at <alt.comp.linux>:
> The cost of a server and licenses for, for example, a charity or a
> school, is trivial compared to the cost of supporting it.
Something that has been very expensive to many companies (including M$
itself) is underrating the market and/or regarding users as eternally
blind, deaf and stupid.
The cloud, in the long run, may become extremely troublesome to a company
that is known to violate users privacy. Take a scenario of a 10 million
cloud clients. Now suppose that only 5 or 10 of them, noisy ones that are
able to use press as well as M$, file lawsuits against M$ for privacy
violations. This is enough to start a herd move that can flood M$ with
lawsuits beyond its capacity to settle with money. Many will simply flee.
Another scenario is similar to the week in which M$ paralyzed 100 thousand
european companies during a whole week because of a bug in its
update/licencing system. This single fact caused banishment of windoze from
german health services and costed M$ more than US$250 million. What if M$
cloud thing simply stopped working during a week?
The risk of such disastrous scenarios is a function of the strategic value
of the violated information and the number of clients.
> Such institutions do not employ IT professionals to help them choose
> their systems, they buy what the bean counters tell them they should
> buy, and the bean counter's device of choice is always a M$ product.
As I said, underrating users is a big mistake.
> But they do upgrade from time to time and this new policy of M$ allows
> something VERY important.
The extent of such moves is usually not that big except in case of a
disturbance in the market. For instance, XP support is already the longest
ever in M$ life because Vista was a disaster: 35% of Vista users downgraded
to XP. These same users did not upgrade to w7 so that XP share now is twice
the Vista share and equals that of w7. M$ moved precisely to fit market
needs. But was unable to stay above 90% share in the home desktop market.
Several measurements have even found windoze (all versions) below 80%. An
unknown number (something around 10% of all computer users as numbers
suggest is probably a huge overestimate) of users left Vista for other
platform and w7 did not bring them back.
A remarkable amount of new M$ products sales are through new machines. M$
relies heavily on plots with hardware vendors. As computers sales have had
a recent global decrease of 2%, windoze sales have fallen 6%. This is a big
lot of money and a much smaller boost of the upgrade sales that are
expected if the first wave in new computers succeeds.
Small business users can still use their older versions and rely on their
beans counters during many years before deciding whether M$ imposed (if
this is the case) cloud thing is reliable or not. Cheap operation! And
there is no hurry at all. :-)
> I believe that M$ are sufficiently ruthless to refuse to provide server
In regard to M$ ruthlessness we do agree. :-)
But other companies are not the 'good guys'. They know that a change of
paradigm in desktop computing could kill windoze within a decade or two but
such a change is so unlikely in the near future. Thus, they currently push
the tablet and mobile markets where they win: 1 out of every 12 computers
sold is a tablet.
> licenses for organisations with less than 100 desktops, yes.
That would be a bad move. First, because it would affect (downwards) only M$
server market share, currently between 38% and 48% depending on measurement
methods. Second, because many such customer companies, as many users did
upon the Vista fiasco, will migrate to platforms that do not impose
undesired clouds inflating these platforms respective server market shares.
Hosting strategic business information in M$ domains is suicidal: M$ sells
to US government the tools to break cryptography meant to protect users
that also paid for the M$ cryptography (closed source!) tools. What would
stop M$ from using the same tools it sells to US government? What kind of
privacy protection can M$ offer?
IMHO, M$ current surge of enthusiasm with cloud technologies is just
marketing fireworks of a big and slow company that arrived late (again) to
the real world.
M$ heads are smart. They know the windoze business is to shrink slowly.
There is something interesting that seems to confirm that and has gone
mostly unnoticed: M$ leading acquisitions have been directed towards the
communications and entertainment markets.
--
Alexandre