Conference azur::mcc

>================================================================================
>Note 1829.0                   OSI AM Questionairre                    No replies
>TOOK::MATTHEWS                                       42 lines  19-NOV-1991 15:41
>--------------------------------------------------------------------------------
>
>I would like to request help from those in the field that have contacts
>with customers. My team is developing an OSI CMIP AM. We have a prototype
>and are working with selected vendors to test our AM against objects
>they are designing. I am also getting prepared to open a phase 0
>for the AM. I would like to test the waters on a few issues and
>hope that you can provide me with the input I need.
>
>1. Do you have a customer who does not have DECmcc today but would buy
>a license for DECmcc if the OSI AM were a product today, Please
>send me the name of that customer. We already know from a previous
>note that JPL is such a customer.  
>
>2. Do you have a customer today who has DECmcc or is already considering
>buying it, who would buy a license for the OSI AM and install it
>immediately. Send their name along with the following information.
>
>2a. How much would they pay for the basic OSI AM.
>
>2b. How much would they pay for a GDMO translator which is similar
>in concept with the SNMP MIB compiler tool.
>
>2c. What objects would they manage via the OSI AM? 
>
>2d. Are they strictly an end user or are they interested in developing
>OSI objects as targets of management.
>
>3. If the AM could be available 6 months sooner without formal conformance
>testing, would they defer buying it until formal testing were concluded
>or would they buy it for immediate use and accept an update after
>conformance testing.
>
>4. Would the customer require a complementary OSI AGENT PM with the AM or
>is the AM alone sufficient to meet their needs.
>
>Please send all responses to: Took::MATTHEWS
>
>Each individual answer to this survey is not sensitive, but the overall
>pattern of responses are so I would request that you please send mail.
>Anyone who has an interest in the result, I will share the results but in
>a less public forum than this conference.
>
>wally
>

1) customer?  No - but I've talked to many that think they will need it,
	and gut feel is that it would help close deals even among customers
	that are not using it.  It gives DECmcc an 'osi' flavor that it claims
	but cannot prove today.

2) customer now?  No - bets are we could find a handfull if we had a PID.

2a) how much - considering that we are giving away "advantage networks" now,
	it would be hard to figure out how to charge for 'real osi'.  I
	spose if we package it in terms of 'NMF compliant' or something
	like that it would have value.  We were getting 6K pretty easy for
	the SNMP-AM before deciding to give it away.  I'd suspect that we
	could get 5K for an OSI/AM.

2b) how much for a GDMO translator?  I think they would expect it to be
	in the price of the AM.  To price it separately starts complicating
	the order process - something we don't need to assist.  However, just
	like with the IP AM, I think services could make some serious money
	providing working GDMO/MIBs, and even going to the point of loading
	them for the customer.

2c) What would they manage? - Other management systems - NMF objects would
	probably be the first market.

3) Available - Many will buy it to know they have it - and won't use it.
	So, I'd vote for the six months earlier with no formal conformance.
	We can market around that, and it will help sell the rest!

4) OSI Agent PM - I think this would occur as part of a NMF package - and
	it would be nice - not needed first release - but that six months
	afterward release would be nice to have it.  It would help drive
	the market towards adopting NMF standards.  A NMF package (or
        "solution set" as the new corporate buzzword goes) could probably
       	sell for 5-15K above BMS, depending on what it actually did.  It 
    	would be a good addition to the EMS customers, to offer them
    	something for the extra money they have spent.
    
    bill

    to .1
    Yes and No. Both will become the same OSI AM in the future. The goal
    set of both were identical, but the current prototypes have significant
    differences. Each was trying to address different requirements and
    development viewpoints, so each developed different functionality
    in their prototypes. We are currently in the process of merging them.
    So, if you are asking about whether their conceptual goal is the
    same, the answer is yes. If you are trying to identify the
    functionality behind this request by using the current TNMP phase
    review documents, the answer is No.
    
    We have been working with vendors who want to define OSI objects
    for nodal processors, switches, directory services, etc. They need
    to support a variety of objects via a common AM and need it to
    support all CMIS functions following the OSI CMI. I am trying to
    identify the potential market size for additional vendors who
    want help during the prototyping stage of their objects or customers
    who want early support for in house development work. 
    
    We are using a customer willing to pay as a litmus test to seperate
    the religious rhetoric surrounding OSI from the near term business
    needs. In the past, we at DEC to often have made engineering trade
    offs based on religious rhetoric rather on business opportunities.
    
    No one is questioning whether we do an OSI AM or not. What is being
    questioned is how soon it is needed, how much money will it make,
    and how many resources should be allocated to it. 
    
    Lacking, solid information to answer these questions, I am trying
    to gain information via all the channels that are available to me.
    
    wally

to .1 (and .3)

Peretz,

As Wally said in the previous reply, our 2 prototypes are quite different, as 
the original starting points were far apart and the initial requirements also 
slightly different, but the long term goals are quite similar.

This similarity of goals made it desirable to have only 1 OSI AM product in the 
future and if there was some delay in answering you, it was because we were in 
Littleton last week to discuss these topics. 

Our 2 prototypes (PNMP and NME) been different, have some complementary and 
merging the 2 prototypes should bring more functionalities quicker. We are 
currently working on that with NME.

Best regards,

Marc.