discussion:Goals Round II (fwd)

[Mike Prince]

---------- Forwarded message ----------
Date: Fri, 11 Nov 94 2:25:35 MET
From: Francois-Rene Rideau <rideau@clipper.ens.fr>
To: Mike Prince <mprince@crl.com>
Subject: Re: discussion:Goals Round II

>> End Users/Marketing Goals:
>> 
>> - must be a friendly intuitive UI
Interface, yes, but interface to what ? Losedoze & Matincock suck because they
have quite a nice interface to nothing. All the goodies in X-Win-gows are
seldom used because there is do not represent anything in the underlying
system.
So, I say *yes* to an interface, but let's build the interface over our
system, and not the system under the interface (which again does not mean
no work may be done about the interface before the rest is done, but that
certainly, this useful but ungrateful work should be adapted to any changes
in the underlying system, and not the system changed to cope with details of
the UI).
Do not put this first in the list.

>> - must have a lot of productivity apps (mail, wp, spreadsheet, ie necessary
>>   apps)
Yes, but we won't write definitive beautiful versions for the first releases
(if -- when -- there'll ever be).
Put this even later in the list.

>> - must fill a strategic niche (i.e. choose our customers well)
Anyone who wants a computer for a useful work (not just to please the boss)
is a good customer, which should be a better goal: be useful to everyone.

>> - must be maleable (no set ui, easy to personalize)
I'd say *fully programmable*; as using a computer *is* programming. Weenies
just program trivial things, which does not mean we should forbid them from
doing anything else (just like DOS/Windows/MacIntosh do). All the computer
users I know one day need to do more complicated things, and cannot do anything
when the system does not provide way to do it automaticall, whereas they could
(and thus should)

>> - use all the available resources (over net too)
Not only past and present, but also future.
Which means we should not provide code that is specialized for a large number
of existing resources, but is generalized for any resource, while still
conserving security. And *this* is where we can crush and the ridiculous
MacroShit software.
Do I'd say, support genericity at system level.

>> - all the above must remain constant over all platforms (who cares what
>>   hardware they are using).  My app here runs on all my platforms.  No
>>   multiple shrinkwrap copies.
Err, ultrafast routines could have hand-optimized versions for several
computers together with intermediate code as a portable version; but indeed,
no "FAT" binary should ever be. The common low-level language should not be
assembly but assembly-level abstraction should be supported -- ask JVS.


>> Programmers/Marketing goals
Again, programming *is* using the computer; So called "programmers" just
explore this art deeper than "mere" users.

>> - must have a good development environment
>> - must be able to use all available resources (look above too)
So an development environment is just a user environment.
Just natural extensions to the above.

>> - persistency:  This is not really addressed by the above, it could be
>>   considered to be part of the user friendliness thing.  Critics loved
>>   the way the Mac saved its state when you shut it down...  This takes 
>>   all that one step further, not only is the ui the same, but the apps
>>   and all objects in the system are exactly the way you left them.
A much better example for persistence (and limited genericity) is the
HP[24]8 pocket calculator series. It's always been my dream no more to have
to get a calculator or launch a multi MB application to compute or manipulate
simple algebraic (or numerical) expressions.