[comp.sys.next] Spreadsheets

zimmer@calvin.stanford.edu (Andrew Zimmerman) (04/13/91)

    While some of you are worrying about whether NeXT will switch to the 88110
or stay with the 68040, and what that would mean to your software investment,
I worry about whether NeXT will be around in a few years.  No matter what the
technical merit of a machine, its success depends on the software that is
available for it, and not just any software, but the big 3.
    Spreadsheets
    Wordprocessing/Desktop Publishing
    Databases

    In terms of wordprocessing, NeXT has TeX, FrameMaker, WordPerfect, and a
few other tools (ie Drawing programs).  Not only are these nice packages,
but they are compatible with the rest of the world.

    In terms of Databases, I understand that the NeXT has Sybase.  I don't
use databases, so I will leave further comment to those that do.

    Now, spreadsheets.  Right now, there is only one that you can get, Improv.
(I know about Wingz, but last I talked to Informix, it wasn't shipping for
the 040).  I have a lot of problems with Improv.  
   1.  It lacks some graph types that many people want.
   2.  I tried to use it just as a spreadsheet.  It did not seem obvious to
me how to use it.  Something as simple as swapping two rows took me 20 
minutes to figure out.  And even then, I had to do a copy and then a delete.
Copying a row caused all of the formulas to break.  Changing the name of the
row to the name of the other deleted row didn't help, the formula was still
in error.  All I had to do to fix the formula was to delete any character,
and then type in the exact same character.  I finally gave up when I had
problems with the IRR formula.  It seems that all of the cash flows have to
be in the same category which for my application was not what I wanted.
(I don't like the IRR functions that come with any of the major spreadsheets,
but thats a different issue)
But, the worst thing about Improv is that it is not compatible with other
spreadsheets.
(Yes, it can almost import 1-2-3 spreadsheets, but it is not compatible)

    I feel that lack of a compatible spreadsheet will be the downfall of NeXT.
It will prevent businesses from moving from IBM/PC's and Mac's to NeXT's.  
Instead, they will go to high end PC's, or high end Macs, or Sun machines.
All of these run industry standard software.  

    I just hope that NeXT realizes the need for a standard spreadsheet, and
gets either Lotus to port 1-2-3, or Microsoft to port Excel to the NeXT.
WingZ is not a bad product, but it lacks features that both of the above 
support.

Andrew
zimmer@calvin.stanford.edu

root@NEXTSERVER.CS.STTHOMAS.EDU (Max Tardiveau) (04/13/91)

In article <1991Apr13.115926.4528@neon.Stanford.EDU> zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:

> While some of you are worrying about whether NeXT will switch to the 88110
> or stay with the 68040, and what that would mean to your software investment,
> I worry about whether NeXT will be around in a few years.  No matter what the
> technical merit of a machine, its success depends on the software that is
> available for it, and not just any software, but the big 3.
>       Spreadsheets
>       Wordprocessing/Desktop Publishing
>       Databases
> [...]
>    In terms of Databases, I understand that the NeXT has Sybase.  I don't
>  use databases, so I will leave further comment to those that do.

You forget Oracle and Ingres. They're not exactly lightweight products.

---------------------------------------------------------------
Max Tardiveau
Department of Computer Science
University of St.Thomas
St.Paul, MN  55105
Internet : m9tardiv@cs.stthomas.edu
---------------------------------------------------------------
"Ban the bomb.  Save the world for conventional warfare."

sksircar@shade.Princeton.EDU (Subrata Sircar) (04/14/91)

zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:
> [...] the big three:
>    Spreadsheets
>    Wordprocessing/Desktop Publishing
>    Databases

This is an outdated catagorization of software, IMHO.  Where do Mathematica,
Matlab, compilers, and graphics programs fit here?  Any business that is buying
computers just for these three uses will buy cheap IBM clones, which can do
this and nothing else.  If all you want is text/data entry, you don't want a
personal computer - you want lots of terminals hooked to a minicomputer or
mainframe-type machine.

>    Now, spreadsheets.  Right now, there is only one that you can get, Improv.
>(I know about Wingz, but last I talked to Informix, it wasn't shipping for
>the 040).  I have a lot of problems with Improv.  
>   1.  It lacks some graph types that many people want.

Actually, no spreadsheet will suit everyone's graphing needs.  If I want
to graph something on a Mac I'd use Kaleidagraph, which
can do damn near anything you want.  If I'm on a Unix box, I use gnuplot.

>   2.  I tried to use it just as a spreadsheet.  It did not seem obvious to
>me how to use it.  Something as simple as swapping two rows took me 20 
>minutes to figure out.  And even then, I had to do a copy and then a delete.

My experience differs from yours.  When Improv was demonstrated at Princeton,
I was able to goof around with it for a little while.  As someone who had
used a spreadsheet once before (to keep track of rotisserie baseball stats) I
was quite pleased.  Being able to flip rows and columns with a keystroke 
impressed me a lot, as I always wanted to be able to do that.  It seems to me
to be ideal for introducing people to spreadsheets, and hence perfect for an
organization that is starting out.

>But, the worst thing about Improv is that it is not compatible with other
>spreadsheets.
>(Yes, it can almost import 1-2-3 spreadsheets, but it is not compatible)

I'd like to respond to this in two ways:  first in my normal tone of voice,
then in a nasal whine :<)

If it can import 1-2-3 spreadsheets, that's all you need.  If you're just
starting a business, you don't have old documents to worry about and you can
certainly get other documents in 1-2-3 format.  If you're upgrading, and your
old spreadsheets can't read/write 1-2-3 format, you are a very rare organism
indeed.  Improv can also write 1-2-3 format, I believe, so you're covered 
there.

Secondly, compatability between platforms is highly overrated.  Name me any
Mac software that was compatible with IBM when they started out.  Even now,
with the advent of Windows, only applications released by the same company are
even remotely alike.  Microsoft products look the same across different
platforms, and no doubt if they released WordNeXT it would look similar, but it
looks nothing like any Claris product.  Why do you expect the NeXT to be any
different?

Thirdly, compatibility is on the other end of a balance from innovation.  If
everything is backwards compatible, you can't ever make new strides.

>    I feel that lack of a compatible spreadsheet will be the downfall of NeXT.
>It will prevent businesses from moving from IBM/PC's and Mac's to NeXT's.  
>Instead, they will go to high end PC's, or high end Macs, or Sun machines.
>All of these run industry standard software.  

Industry standard software?  If you mean "each of these runs the standard 
software FOR THAT MACHINE" you'd be right but irrelevant.  If you're saying
that businesspeople will be scared away because "it's not 1-2-3" ... well,
you may be right, but I'd like to think that people are willing to try new
and possibly better products...

Lastly, have you considered the possibility that Improv, and things like
Interface Builder, might become the NEW industry standard?

I'm not sure if NeXT will succeed.  But if they don't, I don't think software
incompatibility with other machines will be the reason.  A reasonable
platform will persuade people to port/write software for it.  After all,
needs must, when the devil drives.




Subrata Sircar | sksircar@phoenix.princeton.edu |Prophet& SPAMIT Charter Member
	I don't speak for Princeton, and they don't speak for me.
"May their souls rot in easy-listening hell!" - Johnny Melnibone, GRIMJACK #76
"I seem to suffer from irrelevant flashbacks." - Paul, PAUL THE SAMURAI #1

erik@zeus.opt-sci.arizona.edu (Erik Schumacher) (04/14/91)

OK - somebody who really did some work with IMPROV has to defend that gem - 
stone among spreadsheets.  I used it extensively for my case studies in 
marketing, and to tell you the truth: not only am I still getting more and
more enthusiastic about it (I can only compare it with Excel and 1-2-3 - and
it shines!), but it also took the breath of the people I recently gave a 
presentation to with grafics from IMPROV.  There is simply no comparison to
that program, when one has to analyse complex and multidimensional business
data.  I agree that as a scientific user I also have a small wishlist, but to 
get nice plots there is dissplay, gnuplot etc.! But for my business applica-
tions - which is what IMPROV is primarily made for, I am sorry - IMPROV is a
dream!!

Erik Schumacher
University of Arizona
zeus.opt-sci.arizona.edu

dcarpent@sjuphil.uucp (D. Carpenter) (04/14/91)

In article <1991Apr13.115926.4528@neon.Stanford.EDU> zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:
>
>    Now, spreadsheets.  Right now, there is only one that you can get, Improv.
>(I know about Wingz, but last I talked to Informix, it wasn't shipping for
>the 040).  

I've been using Wingz on my 040 Cube since I upgraded.  There may be
some bugs, but I haven't noticed them yet.  So I'd say there are
definitely TWO spreadsheets available, not one.

-- 
===============================================================
David Carpenter            dcarpent@sjuphil.UUCP                    
St. Joseph's University    dcarpent@sjuphil.sju.edu
Philadelphia, PA  19131   

zimmer@calvin.stanford.edu (Andrew Zimmerman) (04/14/91)

In article <1991Apr13.115926.4528@neon.Stanford.EDU> zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:
>   2.  I tried to use it just as a spreadsheet.  It did not seem obvious to
>me how to use it.  Something as simple as swapping two rows took me 20 
>minutes to figure out.  And even then, I had to do a copy and then a delete.
>Copying a row caused all of the formulas to break.  Changing the name of the
>row to the name of the other deleted row didn't help, the formula was still
>in error.  All I had to do to fix the formula was to delete any character,
>and then type in the exact same character.  I finally gave up when I had
>problems with the IRR formula.  It seems that all of the cash flows have to
>be in the same category which for my application was not what I wanted.
>(I don't like the IRR functions that come with any of the major spreadsheets,
>but thats a different issue)
>But, the worst thing about Improv is that it is not compatible with other
>spreadsheets.

    Since I posted this note, I have played with Improv some more and (with 
the help of some other people) figured out how to move rows.  Here's how to
move a row:
1. select the row to be moved
2. in the edit menu, choose cut
3. move to the row right above where the new row is to go
4. in the edit menu, choose paste
   How to move a row to be the top row of the spreadsheet is an exercise left
   to the reader. 

If you delete a row that is used in a formula, and later give the exact same
name to a different row, there is an item called "check formula" in one of the
menus that will fix all of the formulas.

So, to be fair, it does seem that Improv does have the same functionality as
the other spreadsheets.  In the process, I also found a few other features
that are nice.  It just doesn't seem to be too obvious how to do some things.
(Note:  this doesn't mean that it isn't better, its just not obvious.)

Andrew
zimmer@calvin.stanford.edu

zimmer@calvin.stanford.edu (Andrew Zimmerman) (04/14/91)

In article <8242@idunno.Princeton.EDU> sksircar@shade.Princeton.EDU (Subrata Sircar) writes:
>zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:
>> [...] the big three:
>>    Spreadsheets
>>    Wordprocessing/Desktop Publishing
>>    Databases
>
>This is an outdated catagorization of software, IMHO.  Where do Mathematica,
>Matlab, compilers, and graphics programs fit here?  Any business that is buying
>computers just for these three uses will buy cheap IBM clones, which can do

The catagorization is not outdated, it is a fact of life for computers.  The
prodcuts you mention, Mathematica, Matlab, and compilers while important 
pieces of software, they do not make or break a machine like the big 3 will.
I consider graphics to be part of both spreadsheets (graphing) and Desktop
Publishing (Drawing).  BTW, before you knock IBM clones, all of the software
you mention is available for the IBM clones and the Macs.  (In fact, one of
the best symbolic math packages is only available for the IBM PC. (Derive)

>
>>    Now, spreadsheets.  Right now, there is only one that you can get, Improv.
>>(I know about Wingz, but last I talked to Informix, it wasn't shipping for
>>the 040).  I have a lot of problems with Improv.  
>>   1.  It lacks some graph types that many people want.
>
>Actually, no spreadsheet will suit everyone's graphing needs.  If I want
>to graph something on a Mac I'd use Kaleidagraph, which
>can do damn near anything you want.  If I'm on a Unix box, I use gnuplot.
>

A number of people have suggested gnuplot.  Gnuplot on the NeXT is deficient 
in two areas.  It is not interactive. (Or at least to my knowledge it is not,
or you use one of the X servers) and second, it doesn't provide a nice way 
to manipulate the data you wish to graph.  For that, you need a spreadsheet
to determine how you want to look at raw data.  Gnuplot under X solves the 
interactive problem.

>>   2.  I tried to use it just as a spreadsheet.  It did not seem obvious to
>>me how to use it.  Something as simple as swapping two rows took me 20 
>>minutes to figure out.  And even then, I had to do a copy and then a delete.
>
>My experience differs from yours.  When Improv was demonstrated at Princeton,
>I was able to goof around with it for a little while.  As someone who had
>used a spreadsheet once before (to keep track of rotisserie baseball stats) I
>was quite pleased.  Being able to flip rows and columns with a keystroke 
>impressed me a lot, as I always wanted to be able to do that.  It seems to me
>to be ideal for introducing people to spreadsheets, and hence perfect for an
>organization that is starting out.

I agree, there are some nice features to Improv.  

>
>>But, the worst thing about Improv is that it is not compatible with other
>>spreadsheets.
>>(Yes, it can almost import 1-2-3 spreadsheets, but it is not compatible)
>
>I'd like to respond to this in two ways:  first in my normal tone of voice,
>then in a nasal whine :<)
>
>If it can import 1-2-3 spreadsheets, that's all you need.  If you're just

Wrong!  Pages 9-7 through 9-15 of the Improv manual tells how to import a 
1-2-3 file.  It can be done, but I would not like to have to do it on a 
regular basis.  Not only do things have to be converted, but you lose all of
your macros!  Later in the chapter, it tells about exporting files.  It is 
not clear how well this might work.  

Contrast this to my view of the world, a world where you have 1-2-3 on the
NeXT.  At work, you use the large screen and speed of the NeXT to work on
your spreadsheet.  When you have to go on a trip, or home, you copy your
file to a DOS disk and can look at you file using you IBM portable laptop.
When you come back from your trip, or home, you take you IBM Disk and put
it in your NeXT.  
    SoftPC will almost allow you to do this, but you lose speed and your
nice big screen (which is a god send for spreadsheets).  (Actually, my 
experience with SoftPC was on a DEC 3100.  Maybe on the NeXT, it will do a
larger then normal PC screen.)

>starting a business, you don't have old documents to worry about and you can
>certainly get other documents in 1-2-3 format.  If you're upgrading, and your
>old spreadsheets can't read/write 1-2-3 format, you are a very rare organism
>indeed.  Improv can also write 1-2-3 format, I believe, so you're covered 
>there.
>
>Secondly, compatability between platforms is highly overrated.  Name me any
>Mac software that was compatible with IBM when they started out.  Even now,
>with the advent of Windows, only applications released by the same company are
>even remotely alike.  Microsoft products look the same across different
>platforms, and no doubt if they released WordNeXT it would look similar, but it
>looks nothing like any Claris product.  Why do you expect the NeXT to be any
>different?
>

Compatability between platforms in not rated nearly high enough.  NeXT
actually realizes this.  Why do you think you can write an IBM DOS disk on the
NeXT.   
  You are correct that not much of the Mac Software was compatible with IBM
when it came out.  There are two points to this arguement.  First, much of
the software was not available for the IBM PC at that time (the desktop
publishing aspects of the Mac).  The second point is that lack of
compatibility hurt the Mac as it did to other machines.  All can now read
IBM PC formatted disks (after the PC went to the 3.5 in disk) and all either
have products to run IBM software, or the company had boards that could
be added to get the functionality.  I might also point out that Mac's have
at best 10% of the market, while PC's have 80+%.  (I don't have hard
numbers for this, maybe the net can help)

>Thirdly, compatibility is on the other end of a balance from innovation.  If
>everything is backwards compatible, you can't ever make new strides.
>
>>    I feel that lack of a compatible spreadsheet will be the downfall of NeXT.
>>It will prevent businesses from moving from IBM/PC's and Mac's to NeXT's.  
>>Instead, they will go to high end PC's, or high end Macs, or Sun machines.
>>All of these run industry standard software.  
>
>Industry standard software?  If you mean "each of these runs the standard 
>software FOR THAT MACHINE" you'd be right but irrelevant.  If you're saying
>that businesspeople will be scared away because "it's not 1-2-3" ... well,
>you may be right, but I'd like to think that people are willing to try new
>and possibly better products...

Yes, industry standard.  Excel runs on three of the major players (IBM, Mac, and Sun (I have seen the first, the third I was told).  1-2-3 runs on IBM,
Sparc.  WingZ runs on IBM, Mac, Sun, and almost on NeXT.

>
>Lastly, have you considered the possibility that Improv, and things like
>Interface Builder, might become the NEW industry standard?

No, because they won't.  The industry standard for spreadsheets will probably
be the Gnu spreadsheet currently in the works.  As for IB, it also will not
be the standard.  The standard will use Motif and C++.

>
>I'm not sure if NeXT will succeed.  But if they don't, I don't think software
>incompatibility with other machines will be the reason.  A reasonable
>platform will persuade people to port/write software for it.  After all,
>needs must, when the devil drives.

No, a reasonable platform will not persuade people to port/write software,
Market share and incentives from the hardware manufacturer will.  
Dec has a reasonable platform in the Dec 3100 and Dec 5000 platform.  Yet for
some reason they have to go out to companies and encourage them to
port to that platform.
    An even better argument can be made about the NeXT vs IBM PC's.  Which
is a more reasonable platform for music?  Which one is Coda trying to port
it software to run on?

>
>Subrata Sircar | sksircar@phoenix.princeton.edu |Prophet& SPAMIT Charter Member
>	I don't speak for Princeton, and they don't speak for me.
>"May their souls rot in easy-listening hell!" - Johnny Melnibone, GRIMJACK #76
>"I seem to suffer from irrelevant flashbacks." - Paul, PAUL THE SAMURAI #1

Andrew Zimmerman
zimmer@calvin.stanford.edu

louie@sayshell.umd.edu (Louis A. Mamakos) (04/14/91)

In article <1991Apr14.015900.17956@sjuphil.uucp> dcarpent@sjuphil.UUCP (D. Carpenter) writes:
>I've been using Wingz on my 040 Cube since I upgraded.  There may be
>some bugs, but I haven't noticed them yet.  So I'd say there are
>definitely TWO spreadsheets available, not one.

While it probably works just fine, try to actually buy a copy of WingZ
to use.  I had a copy on order from July 1990 through Novemeber 1990
when I finally canceled my order.  I never got a straight answer WHY
they couldn't ship to the product to me.

I finally said to myself: "Screw this.  I can get Improv for free.
Why endure the hassle?"  So I canceled the order, saved the University
of Maryland a few hundered dollars and am now using Improv.  If the
vendor won't sell and deliver a product, or even provide some
explanation, then what am I to think?

I can't really blame the free Lotus Improv offer this for this, as I
had an order in for it long before Improv was available.  I also had
an *immediate* need for a spreadsheet product.  They missed out on a
window of opportunity to build a user base in a vacuum of available
products.  They blew it.

louie

hardy@golem.ps.uci.edu (Meinhard E. Mayer (Hardy)) (04/15/91)

Since the discussion that precedes is too long to quote -- just a
remark about 1-2-3 and SoftPC:
It works fine and reasonably fast -- like a 10-12 MHz 286/287  
(I have not received my IMPROV yet -- anyone at NeXT reading this?), 
except for printing, which is erratic; I managed to
print graphs, but not spreadsheets; Insignia blames this on the
serial port (which I use for printing) and claims it will work fine 
when I upgrade to 2.1.
So if anyone is really missing 1-2-3 and has $ 300 (and some
frustration) to spare, SoftPC may be the solution.

Greetings,
Hardy 
			  -------****-------
Meinhard E. Mayer (Hardy);  Department of Physics, University of California
Irvine CA 92717; (714) 856 5543; hardy@golem.ps.uci.edu or MMAYER@UCI.BITNET

sksircar@shade.Princeton.EDU (Subrata Sircar) (04/15/91)

zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:
>sksircar@shade.Princeton.EDU (Subrata Sircar) writes:
>>zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:
>>> [...] the big three:
>>>    Spreadsheets
>>>    Wordprocessing/Desktop Publishing
>>>    Databases
>>
>>This is an outdated catagorization of software, IMHO.  Where do Mathematica,
>>Matlab, compilers, and graphics programs fit here?  Any business that is
>>buying computers just for these three uses will buy cheap IBM clone ...


>The catagorization is not outdated, it is a fact of life for computers.  The
>prodcuts you mention, Mathematica, Matlab, and compilers while important 
>pieces of software, they do not make or break a machine like the big 3 will.
>I consider graphics to be part of both spreadsheets (graphing) and Desktop
>Publishing (Drawing).  BTW, before you knock IBM clones, all of the software
>you mention is available for the IBM clones and the Macs.  (In fact, one of
>the best symbolic math packages is only available for the IBM PC. (Derive)

First, I agree that IBM clones can do a lot of things - the big thing they lack
is that graphical user interface both as a standard, and with the functionality
that the Macinotsh and NeXT have (Windows is buggy, doesn't run some software
(like MatLab) and doesn't do all the nice things the Mac GUI does, but this
is a side issue and need not detain us).  My point here is just that if that's
what you want to do, you don't buy a NeXT (as price/perfornmance issues will
kill you) - you have to want other things.

As far as catagorization is concerned - those big three (while forming a large
chunk of my impression of a computer) are no longer make it/break it points for
a computer.  The scientific/engineering community doesn't want those, they want
powerful CAD packages, fast processing power, a killer development environment,
etc. and they are significant players these days.  You don't sell Lisp Machines
to business people wanting spreadsheets

>Wrong!  Pages 9-7 through 9-15 of the Improv manual tells how to import a 
>1-2-3 file.  It can be done, but I would not like to have to do it on a 
>regular basis.  Not only do things have to be converted, but you lose all of
>your macros!  Later in the chapter, it tells about exporting files.  It is 
>not clear how well this might work.  

I will comment that once you have Improv, you'll only have to import 
spreadsheets once or twice, and then everything will be in Improv's format.
Secondly, you are right on the mark about macros.  Hopefully Lotus will build
the wonderful macro capability of 1-2-3 into Improv - enough other users have
been vocal about that on the net.  Third, SoftPC should allow you to do what
you're asking - that was part of the point of the floppy drive.

>Compatability between platforms in not rated nearly high enough.  NeXT
>actually realizes this.  Why do you think you can write an IBM DOS disk on the
>NeXT.   

Well, my experience has generally been that if I want to move things between
different platforms I either a) make it plain text or b) hope that I'm using
the same program on both platforms (such as MatLab on the Sun and the Mac).
I generally use the (admittedly large and varied) computing power at my 
disposal in different ways - using the NeXT to develop my thesis applications,
using the Mac, the NeXT and the SparcStations for text entry, and using the
Mac and the Sun for computing power through MatLab.  Given the ubiquitous
nature of networking (something the NeXT wants to excel at), especially around
here, I don't need anything else.  Porting code, on the other hand, is a very
different story.  There I'd kill for everybody using whatever it is I'm moving
code to :<)

>>Lastly, have you considered the possibility that Improv, and things like
>>Interface Builder, might become the NEW industry standard?
>
>No, because they won't.  The industry standard for spreadsheets will probably
>be the Gnu spreadsheet currently in the works.  As for IB, it also will not
>be the standard.  The standard will use Motif and C++.

Abandoning the Interface Builder for Motif would be a tragedy.  I sincerely
doubt that Motif will be a standard either, but that's a side issue.  C++, 
maybe.  But who cares?  Using the IB you don't really care what language you're
writing in until you've finished the interface.  A tool like the IB will be
part of the future development standard, IMHO.  If Motif is so good, why did
IBM choose to port NeXTStep to the RISC 6000 workstation?  Clearly they see
some benefits there, or they wouldn't have done it.

>No, a reasonable platform will not persuade people to port/write software,
>Market share and incentives from the hardware manufacturer will.  

NeXT is approaching that.

>Dec has a reasonable platform in the Dec 3100 and Dec 5000 platform.  Yet for
>some reason they have to go out to companies and encourage them to
>port to that platform.

The Amiga is another example of your argument.  Wonderful hardware, lousy
software.

My point is that commercial viability is not solely determined by being able
to read/write other people's products; if you can, you do it providing better
products at at possibly cheaper price.  The jury is still out on whether NeXT
does that.



Subrata Sircar | sksircar@phoenix.princeton.edu |Prophet& SPAMIT Charter Member
	I don't speak for Princeton, and they don't speak for me.
"May their souls rot in easy-listening hell!" - Johnny Melnibone, GRIMJACK #76
"I seem to suffer from irrelevant flashbacks." - Paul, PAUL THE SAMURAI #1

waltrip@capd.jhuapl.edu (04/15/91)

In article <8242@idunno.Princeton.EDU>, sksircar@shade.Princeton.EDU (Subrata 
Sircar) writes:
> zimmer@calvin.stanford.edu (Andrew Zimmerman) writes:
	[...]
>> I have a lot of problems with Improv.  
>>   1.  It lacks some graph types that many people want.
> 
> Actually, no spreadsheet will suit everyone's graphing needs.  [...]
	Isn't that what the NeXTstep "services" environment is intended for?
	Shouldn't Improv be done in such a way that anyone can provide custom
	graphics services for it?  or so that Improv can offer spreadsheet
	"services" to other applications?
>	[...]
> Secondly, compatability between platforms is highly overrated.  Name me any
> Mac software that was compatible with IBM when they started out.
	That may have been a matter of luck.  It took Microsoft forever to
	come out with a usable Windows.  Now you have Windows for Macs and
	Motif (with Windows look-and-feel) for competing workstations.  And
	the Microsoft/DEC/Compaq/etc consortium in the wings.  Getting by on
	your GUI is a lot tougher these days.
> [...]
> Thirdly, compatibility is on the other end of a balance from innovation.  If
> everything is backwards compatible, you can't ever make new strides.
	Well, I think we can say that backwards compatibility generally doesn't
	hurt...especially when done cleverly.  NeXT has been rather clever.
	By using Mach/UNIX, they make it possible to port many UNIX apps
	easily and provide a variety of ways to provide a NeXTstep front end.
	The key is to carry innovation into the way you provide backwards
	compatibility.

	In addition, let's not forget that transitions are eased by SoftPC and
	compatibility is enhanced by tcp/ip and coXist (and mouse-X).  



c.f.waltrip

Internet:  <waltrip@capsrv.jhuapl.edu>

Opinions expressed are my own.