Network access tying was(RE: Python Shareware?)

Alex Martelli aleaxit at yahoo.com
Fri Jun 8 10:24:37 EDT 2001


"Mike C. Fletcher" <mcfletch at home.com> wrote in message
news:mailman.991996993.4902.python-list at python.org...
> But, but, isn't that the basic idea behind Windows and Office XP :o) .
That
> seems to require a constant connection to (the Microsoft version of) the
> Internet to do even the most basic of things?  Please tell me that this
    ...
Let's see, wasn't it Microsoft that coined the motto
"The Network *IS* the Computer"...?-)


> Microsoft might be able to do network tying because their market share
> approaches unity and they can spend hundreds of millions brainwashing
people
> into ignoring the problems, giving up their privacy etc. But most
shareware

Maybe.  It does seem that they had to backpedal *VERY* fast from
their original idea of Passport's license conditions (the ones
which enabled them to freely use any intellectual property that
you happened to store or transmit through their servers:-), so
their brainwashing may not be quite as effective as all that.

I think it's more likely that network tying can work for those
products where being connected to the network is a clear added-
value proposition -- which can be a fair amount, if one can
structure things right (games can use a net connection to let
you play with others, financial apps can use a net connection
to get updated stock info, program-development tools can query
an error/fixes/workarounds online DB for latest info, etc, etc).

Presumably one would have to let the program have SOME use even
without a connection if feasible, but a much higher utility
when the connection is available -- and the access to that
"higher utility" would require some form of payment to work.

A subscription-model might work best -- a sales-model appears
to offer better initial cash-flow, but, as your expenses to
maintain and update the online service are mostly ongoing rather
than up-front, it creates a cost/revenue models mismatch which
spells eventual problems IMHO -- subscription is a better way
to go, and it's *HARD* to *TRANSITION* to it if you start out
with sales instead... though it's doable [we did it!-)], you're
better off starting with the right business model at once...


Alex






More information about the Python-list mailing list