🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: sci.math.stat
Article #144

Re: REQUEST: Information regarding CSS: Statistica program

#144
From: Sharachchandra Lele;100 BldgT-4;...
Date: Thu, 30 Jul 1992 05:25
73 lines
3493 bytes
To: simon@mmpe.mineral.ualberta.ca
Subject: Re: Statistica/MAC: looking for comments or reviews
Newsgroups: comp.sys.mac.apps,comp.apps.spreadsheets
In-Reply-To: <simon.698512261@mmpe.mineral.ualberta.ca>
Organization: University of California, Berkeley

In article <simon.698512261@mmpe.mineral.ualberta.ca> you write:
>I noticed a full-page
>ad for Statistica/MAC by StatSoft in the March issue of MacWorld, with some
>impressive sample figures.  Has anyone out there used this package, and could
>they pass on a few comments pertaining to my earlier notes?  Recommendations
>would be welcomed.
>
>Regards,
>  Simon.
>

I don't have experience with Statistica/MAC, but I have evaluated
CSS:STATISTICA/DOS, and the comments might be relevant.

I found CSS (for short) good and bad: 
Good: menu driven (you as a mac user do not know anything else, but let
me tell you that the other statistical/graphics programs for DOS are
just modified versions of old Mainframe command-line driven programs!

Good: speed, much better than anything else i have seen.

Good: manual has lot of discussion of statistics: what technique to
use, etc.

Good: huge range of statistical and graphics capabilities
**** In general, it has many of the features that you mention.

LIMITATIONS:
1) TREMENDOUSLY complex structure, makes data opaque:
* there are separate "spreadsheets" for data, output of statistical
analysis, and graphics.
* You cannot simply have data in a file and choose the variables
to graph at random. everytime you go thru graphics, it generates its
OWN dataset corresponding to the graph chosen
* two kinds of graphics programs, three kinds of data editors, three
macro languages, ....
2) SERIOUS: can only do 10 y-series in a plot !!!!
3) inability to do quick lookup of data from other data files : in a
typical MERGE command, you have to combine all the variables from the
second file.
4) character variables can only be 8 characters long
5) Graphics: given the limitation of 10 y-series above, overlaying of
plots becomes imperative, but this cannot be done in the way SYSTAT
will allow you to overlay.
6) Does not have the easly aggregation capability of SPSS (or even the
limited one of Lotus 3.1!)

GLITCHES:
* many glitches in memory management, leading to frequent crashes
* DANGEROUS : incorrect import of long decimal numbers from Lotus!
* inability of screen drivers to drive many monitors,

Technical support was also quite lackadaisical.

I finally decided against the package for the above reasons. They do
have a 30-day trial period with full money back, so you might want try
it out yourself. I myself have been juggling between SYSTAT, SPSS and
QuattroPro.

Good luck! (Looks like your Trapeze was quite something!)

--Sharad Lele

Energy & Resource Group
Univerity of California, Berkeley

Electronic-mail:	lele@violet.berkeley.edu	(Internet)
Technical Details
Message-ID: <157ug7INN33b@agate.berkeley.edu>
Path: arc1-ca.pugleaf.net!sparky!uunet!elroy.jpl.nasa.gov!ames!agate!violet.berkeley.edu!lele
References: <1992Jul3.163117.9940@jupiter.sun.csd.unb.ca>

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!elroy.jpl.nasa.gov!ames!agate!violet.berkeley.edu!lele
From: lele@violet.berkeley.edu (Sharachchandra Lele;100 BldgT-4;26886;5106440692;NT23)
Newsgroups: sci.math.stat
Subject: Re: REQUEST: Information regarding CSS: Statistica program
Date: 30 Jul 1992 05:25:27 GMT
Organization: University of California, Berkeley
Lines: 73
Sender: Sharad Lele
Message-ID: <157ug7INN33b@agate.berkeley.edu>
References: <1992Jul3.163117.9940@jupiter.sun.csd.unb.ca>
NNTP-Posting-Host: violet.berkeley.edu
Summary: Statistica is good in parts