[comp.sys.next] Improv API and related complaints

dwatola@NEXTASY2.EECS.WSU.EDU (David Watola) (03/31/91)

R. Antony comments:
>>[stuff about Improv deleted]
>>What the developers can and should do is to provide hooks
>>to which 3-rd party develpers and users can attach
>>a custom plotting module.
>>Why nobody has such a thing?
>
>If you would care to read the release notes to Improv, you would find
>that Lotus did exactly that. Just go ahead and order your API disk.

i really don't think the API will be quite that flexible.  if you use
appinspector or gdb to look at the remote messaging functions supported by
presentation builder, you get just about the same functionality as you get
from the inspector panel.  in any case, this API disk is vaporware...
the release note makes it sound like all you have to do is pick up the phone
and you've got your API disk in a week.  i called lotus and they said it
was not available-they would send me a postscard describing it (wonderful).
john corey from next told me they don't have it either.

in general, i think improv is pretty slick.  but i also believe lotus is so
smug about their revolutionary spreadsheet interface that they don't see how
they have crippled the program with some of their limitations.  for all its
ease of use, it is certainly not as functional as 123.  easy-to-use formulas
are not as good as macros.  hell, you can't write a decent LIFE simulation
using improv :(  ok, sure, spreadsheets were originally intended for financial
crapola.  but i seriously doubt that financial planning constitutes the 
majority of 123's use across the country.  Face it, 123 gets used for
EVERYTHING under the sun.

more arbitrary limitations:  how come i can't plot 2000 data points?  why can't
i get decent x-y plots (rather than just scatter plots--i want connected lines).
how come when i do scatter plots i have to have those HUGE symbols at my data
points?  etc. etc. etc.
i suppose i can't complain too loudly though.  it was free.  i certainly
wouldn't pay any money for it, though.

 
 
originally intended for financial analysis, but it has grown beyond that.
123 is a general-purpose tool with a lot of power--i know of one MAJOR
engineeering/manufacturing company that uses it to handle vast amounts of
production data and perform statistical analyses...  there are just some silly
arbitrary limitations imposed by lotus on improv's capabilities.   

rca@cs.brown.edu (Ronald C.F. Antony) (04/02/91)

In article <9103310744.AA09707@nextasy2.eecs.wsu.edu> dwatola@NEXTASY2.EECS.WSU.EDU (David Watola) writes:
>from the inspector panel.  in any case, this API disk is vaporware...
>the release note makes it sound like all you have to do is pick up the phone
>and you've got your API disk in a week.  i called lotus and they said it
>was not available-they would send me a postscard describing it (wonderful).
>john corey from next told me they don't have it either.

I was told it would be available within 4-6 weeks. Also there is not
only messaging but also the possibility to dynamically link code into
Improv. This should solve most problems. If it doesn't then you are
probably better of writing a new application anyway.

>in general, i think improv is pretty slick.  but i also believe lotus is so
>smug about their revolutionary spreadsheet interface that they don't see how
>they have crippled the program with some of their limitations.  for all its
>ease of use, it is certainly not as functional as 123.  easy-to-use formulas
>are not as good as macros.

I know. There are also a couple of things wrong with how they handle
summary functions. But considering it is a 1.0 version and something
completely new, it's ok. Lotus is working on macros though.

>majority of 123's use across the country.  Face it, 123 gets used for
>EVERYTHING under the sun.

I know a person that wrote a graphical user interface in 123 by using
xy-plots and connecting lines. It's sick... Sure, 123 can simulate a
Turing machine. Maybe someone writes even a C-compiler for 123 (I mean
c2-123 not 123-2c)?  :-)

>more arbitrary limitations:  how come i can't plot 2000 data points?

This has probably to do with how they used postScript. There are
limits on the size of arrays. But then again look at the PostScript
code produced by WingZ. Well you better don't. It looks awful.


Ronald
------------------------------------------------------------------------------
"The reasonable man adapts himself to the world; the unreasonable one persists
in trying to adapt the world to himself. Therefore all progress depends on the
unreasonable man."   G.B. Shaw   |  rca@cs.brown.edu or antony@browncog.bitnet