Conference heron::euro_swas_ai

195.0. "FWD: Lisp Paper-fyi" by HERON::ROACH (TANSTAAFL !) Thu Apr 26 1990 15:29

Printed by: Pat Roach                                    Document Number: 011504
--------------------------------------------------------------------------------

                  I N T E R O F F I C E   M E M O R A N D U M

                                        Date:     25-Apr-1990 10:32pm CET
                                        From:     PAPAGEORGE
                                                  PAPAGEORGE@AITG@HERON@MRGATE@HUGHI
                                        Dept:      
                                        Tel No:    

TO:  ROACH@A1NSTC


Subject: FWD: Lisp Paper-fyi

From:	AIADM::MACARTHUR      24-APR-1990 11:12:14.10
To:	@AITCSTAFF.DIS
CC:	MACARTHUR
Subj:	Lisp Paper-fyi

From:	AITG::LEARY        "Louise Leary DTN 291-8007" 24-APR-1990 09:48:57.93
To:	AIADM::MACARTHUR
CC:	LEARY
Subj:	June...Could you please forward this to Dennis' staff?  Thanks, Louise

From:	PIPE::KEATING      20-APR-1990 20:50:23.95
To:	AITG::LEARY
CC:	KEATING
Subj:	Louise, Don't know how to reach Dennis O'Connor.  Do you know who his 
secretary is?  Can you help me in getting this to him and his staff??   /Sue

From:	PIPE::NMAIL-DAEMON 20-APR-1990 20:37:59.04
To:	KEATING
CC:	
Subj:	Report on failed mail

Errors were detected when processing your mail message
 which was entered at 20-APR-1990 20:36

From: KEATING
To:   AITG::OCONNOR
CC:   KEATING
Subj: Dennis, did not have your Staff distribution list, could you please have 
your secretary forward to your staff?  Thanks, /Sue

----------------
The following error message was returned whilst sending to
AITG::OCONNOR

    %MAIL-E-NOSUCHUSR, no such user OCONNOR at node BIZET

This is a hard (irrecoverable) error.
No more attempts to send to this address will be made.

----------------
The text of your failed mail message follows:


From:	PIPE::KEATING      20-APR-1990 20:36:39.04
To:	@SDTDIS.DIS,@DSSG_STAFF.DIS,NM%CRA::FULLER,NM%SADMAN::JOHNSON
CC:	NM%AITG::OCONNOR,NM%LPEARL::PRATT,KEATING
Subj:	Lisp Paper

	TITLE:  LISP - THIRD PARTY PRODUCT CHOICE

	Written By:  Christine Pratt
	             SDT Product Manager
		     April 11, 1990

	

	PURPOSE:  
		Recently, a decision was made by SDT (Software 
	Development Technologies) Management to aquire a LISP product 
	through a third party rather than continue a development ef-
	fort which was underway in the AITG (Artificial Intelligence 
	Technology Group).  The purpose of this paper is to discuss 
	the reasons for this decision.  It is believed that third 
	party software will be used more often in the future as a 
	means of competing in the Software space.  Use of a third 
	party product offers a means of transferring vital Engineering 
	resources to the revenue stream vs. employing them to develop 
	a marginally profitable product offering.



	THE MARKET:

		Over the past three years there has been a steady 
	decline in the sales of the Digital developed VAXLISP product 
	for VMS and ULtrix platforms.  The average decline from year 
	to year was 18%.  The sales base on which this decline occurred 
	ranged from a high of 569 licenses to the current FY 90 fore-
	cast of 375.  The average MLP for the product is $7.5K. On the 
	average, 31% of these sales are to Universities and, therefore,
	are heavily discounted.  In short, this product has not been 
	a major revenue producer for the Company.

		In the fall of FY89 a contract was negotiated with 
	LUCID - a Lisp vendor from the West Coast.  This contract was 
	put in place to provide a LISP on the RISC platform in a shorter
	time frame than Digital estimated for internal development of a 
	similar product.  The Lucid product is a market recognized 
	product which is used as the Lisp offering by Sun, IBM, and 
	HP (Hewlett Packard) - to name a few of the major users.  In 
	addition, the Lucid Lisp product is an integral part of some 
	A.I. tools being offered by Digital such as Epitool from 
	Epetech in Sweden.

		Recent market research indicates the choice of a third
	party is a good one for the RISC Lisp product and should be 
	explored for the Lisp offerings on the other platforms as well.  
	This would result in a single-sourced, consistent Lisp offering 
	to our customers.  Input from the customer base indicates they 
	are not using the full functionality of the current VMS/Ultrix 
	VAXLISP product.  Therefore, we should look to build the most 
	often used features into future offerings.  In addition, 
	papers by outside market research firms such as the Relayer 
	Group indicate although LISP is a fine language for programming, 
	it is too complex for the general market.  Predictions are 
	Lisp use will decline as languages such as C++ become more 
	popular for Object Oriented programming. The Relayer group 
	believes  the LISP market will shrink over the next five years.  
	The residual market is expected to be a small niche of highly 
	technical users.





	THE PRODUCT COMPARISON:

		A point by point comparison would be difficult here
	and probably of not much benefit to the reader.  However, it
	is planned to have such a comparison compiled to support the
	migration of the current installed base to the Lucid Lisp pro-
	duct.  This information will also support the marketing effort
	for the Lucid Lisp product.  In general, the goal of the nego-
	tiations with Lucid are to obtain the key features of VAXLISP 
	in the Lucid product to support an easy migration of our cur-
	rent customer base.  The major points being discussed are:

		- call-out and call-back compatibility
		- alien structure compatibility/the interaction
	          with non-lisp data
		- condition system compatibility to support
		  error handling
		- stream definition compatibility which extend
		  I/O system device drivers
		- Motif interface functions and user interface
	
	Other points which are being requested are:

		- X3J13 Pretty Printing
		- International Character sets for Oriental languages
		- Installation and command line compatibility
		- various VMS support

		The results of the negotiations will tell us which of 
	these and many other lesser features used by the customers 
	will be included in the product from Lucid.



	THE DECISION:

		A look at the market for the Digital developed Lisp 
	product, coupled with preliminary financials which did not in-
	dicate a profitable situation, support the decision to look 
	to a third party for Digital's Lisp offerings.  This decision 
	is not a reflection on the internal skills or abilities to 
	deliver a product.  This is a decision which is based on good 
	business judgement.  We will fill a need of the customer base 
	with a product which both is recognized in the Lisp Market and
	which meets a large percent of the functionality. And, as the 
	market declines we will not own the retirement process and
	maintenance of a non-profitable product.  This allows our 
	scarce engineering resources to be used to support the devel-
	opment of new, revenue producing product without ignoring the 
	current customer needs.




	SUMMARY:

		The decision to supply our customers a Lisp offering 
	through the use of the Lucid product is a solid one.  The 
	customer need is filled.  The engineering resources are used
	where they benefit Digital the most.