🚀 go-pugleaf

anderson@macc.wisc.edu (Jess Anderson) writes:


>In article <1992Apr10.052251.6145@athena.mit.edu>
>jgshir@athena.mit.edu (John G Shirlaw) writes:

>>Steve jobs and NeXT in general use the phrase "Mission Critical App" a lot as
>>they see it as there main advantage

>>I always have problems with deciding quite what this phrase actually means...
>>any one got a good example eplanation and/or examples...preferably ones that
>>don't include the financial sector and or DTP.

>No answer to your question, but I noticed on television the
>other night that Unisys is also using the phrase
>"mission-critical" in their ads.

>-- 
>Jess Anderson <> Madison Academic Computing Center <> University of Wisconsin
>Internet: anderson@macc.wisc.edu <-best, UUCP:{}!uwvax!macc.wisc.edu!anderson
>NeXTmail w/attachments: anderson@yak.macc.wisc.edu  Bitnet: anderson@wiscmacc
>Room 3130 <> 1210 West Dayton Street / Madison WI 53706 <> Phone 608/262-5888

Well, I'd say "mission-critical" has nothing particular to do with the
financial sector, DTP, or any other business sector or computing activity.
It has to do with anything that a company would deem critical to what
it does (i.e. its "mission").  Mission-critical apps are software that
assist in those critical activities.  Fairly reasonable?  Well, you
said an example.

I work for a furniture sales company.  With strokes from a very large
brush, you could paint a picture that showed two major mission-critical
aspects of our Information Systems: managing the flow of inventory,
and processing the results of sales activities.

Within these two IS segments is room for lots of mission-critical custom
apps, for example in the sales area.  Executive will want to know all
sorts of weird and wonderful things about who sold what where and to whom.
We spend a lot of time coming up with custom ad-hoc reporting to satisfy
all sorts of information needs in the sales analysis area, and more time
turning requests down because we don't have the personpower to deal with
them.  So for us, a mission-critical custom app might be something that
looked like this:

Attach to our Oracle database and grab raw information about sales
activity.  Display in slices of a basic 3D cube with sales location along
one axis, time along another, and product along a third.  For any slice
of the cube displayed, show quantity of units sold, dollar amount sold,
and margin; summarize along each of the two axes shown on the slice.
Allow dynamic drill up & down in the detail along each axis.  E.g. sales
location could vary from region within corporation, to province within
region, to store within province, to salesperson within store; time could
vary from month with the last 12 months, to day within the last month;
product could vary from major group within product group, to minor group
within major group, to category within minor group, to item within category.

Now, we could spend incredible time developing an X Window application
that did this.  Or we could spend a lot less time doing a NeXTstep app
that would do it (if we had any NeXTs, which we don't and probably
won't).  Or we could do something we like even better, and get a good
shrinkwrapped tool that can handle the guts of the work, and integrate
it with our system (for any Improv fans, the description of my vaporapp
will sound familiar).

We've done the latter in a way.  We have a PC (yes, I mean IBM clone,
386) package called Powerplay which can do the multi-dimensional stuff.
But getting the data into it involves running flat reports, stripping
data from them, transferring them to our mainframe, downloading them to
a PC, then importing them into Powerplay and building the view (which
can take time).  When you combine this process with the fact that the
initial sales activity reports that drive it are run off sales history
summary data that takes a minimum of three days for our retail system to
generate, you may be able to perceive why we would prefer to provide a
more real-time solution that could summarize and view sales activity
data "online".

Well, where NeXT is trying to focus is on the areas where you don't have
the convenience of going with an existing tool to do your work for you.
Or if you do, you require substantial support & integration work to build
stuff around it (e.g. the Improv API could be used to make Improv an
engine for a custom app).  Because the app is "critical", you can't
turn it down due to shortage of personpower, and you can't take a year
to build, test and deploy it.  It would also be nice if fast development
also meant fast enhancement later.  I dunno about anybody else, but our
executive sometimes seem to change their requirements daily.  :~)
-- 
Royce Howland   --   Everything is IMHO   |  "I can eat enormous quantities
royce@splunge.uucp        (NeXTMail OK)   |   of ice-cream without being sick,
or royce%splunge@atlantis.uucp            |   Mrs. S-C-U-M."  --Mr. Neutron
or alberta!atlantis!splunge!royce or ...  |   C86Net: Mr._Neutron@RT.Alta