I'm with Larre. I myself have always worked under JES2. Closest I came to JES3 
was when the old Bank of America (San Frisco) was swallowing up the late great 
Security Pacific Bank (LA). BofA was JES3. In a meeting about obstacles to 
merging, some exec asked "How about first converting SecPac to JES3, then 
merging?" Bless his heart, he had no idea the barrel of worms he was proposing. 

I don't believe that either JES is inherently superior. They are two 
alternative churches sitting kitty-corner on a busy intersection. One proclaims 
"There Ain't No Hell", the other counters "The Hell There Ain't". They each 
have adherents who don't want to concede the struggle to the other. A third 
church that melds the best of both might be a promising compromise, but habits 
would have change. People love their habits. And what would it be called? JES5 
(2+3) or JES6 (2*3)? 

I sympathize with JES3 sites that will get bitten.  

.
.
J.O.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler 
SHARE MVS Program Co-Manager
323-715-0595 Mobile
626-543-6132 Office ⇐=== NEW
[email protected]


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf 
Of Larre Shiller
Sent: Thursday, February 28, 2019 3:27 AM
To: [email protected]
Subject: (External):Re: z/OS V2R5 Will be the Last Release to Include JES3

> Do you the think the z/OS overall ecosystem and the platform is made stronger 
> or weaker by getting to one JES?

Well... I guess there are a number of ways to look at that.  But I think that 
IBM runs the risk of losing some number of z/OS customers as a result of this, 
which could in the long term affect the overall IBM revenue stream and the life 
expectancy of the z/OS platform.  I have to assume that the number-crunchers at 
IBM have already factored this into their calculations and have come to the 
conclusion that the risk is worth the reward here.

In the near term, some larger and more complex JES3 installations, and 
especially those in the public sector that have constant external pressure put 
on them to "modernize" their systems (whatever that means... typically meaning 
"move to Linux" or "the AWS cloud"), may simply take this opportunity to move 
to a completely different platform.  Or simply speed up the already on-going 
effort to eliminate (what is perceived as) legacy software with a high 
"technical debt" cost by moving to a different/modern platform.  This is indeed 
a very real risk... and IBM has put many of us in a box.

For those JES2 shops out there... imagine how your shop has grown over the past 
30 years and what products/services that you support that have an API or in 
some way interface with JES.  Just imagine what would it take for you to be 
able to know for sure that this conversion will be 100% successful?  IBM can 
add as much JES3 functionality into JES2 as it wants to, but at the end of the 
day, you still have to do the conversion and ensure that your environment is 
100% functional the day after the conversion.  And how much fun do you think it 
will be to convert 16 SYSPLEXes and 300K MIPS-worth of workload (OK, so 30% of 
it is in a single SYSPLEX)...?

A JES3-to-JES2 conversion effort is a high risk change that requires 
essentially the same level of effort as a conversion from one platform to 
another--but it has the disadvantage of a 100% "must-go-right" overnight IPL 
switch.  And in the end, there's nothing to show for it... other than a 2 
instead of a 3 and a line item in the CxO's spreadsheet that shows the millions 
of dollars spent on R&D, additional products, test time, personnel costs and 
lost productivity.  Also, I certainly would not want to be the public official 
testifying in front of a Congressional panel the week after seeing the 
"Critical System Failure Affects Millions of Seniors" headline in the local 
paper.  Would you put your professional reputation on the line and tell your 
CxO that "everything is going to be all right" the day before a conversion like 
this?

On the other hand, moving to a different platform allows a staged application 
migration and a completely modernized environment--rich with new functionality 
and without the "legacy" problems that persist on z/OS while at the same time 
minimizing and compartmentalizing the risk to critical production applications. 
 So... why would a CxO choose to spend millions of dollars on a costly and 
risky conversion effort (even if it were logistically possible), instead of 
using those same funds to completely modernize that same environment...?

Similarly, very small shops that do not have the personnel or resources to 
perform a conversion will probably wait this out for a while and then simply 
move off the platform.  For medium-sized shops, it's probably a wash.  Overall, 
the number of customers affected here is probably rather small, but given the 
size of some JES3 shops, if this chases away enough JES3 shops, it could 
seriously negatively affect the overall installed MIP count and thus the long 
term stability of the platform.

So... perhaps the shops that remain on z/OS will be in a better position... at 
least for a while, until those remaining shops also start drifting away from 
the platform for one reason or another until eventually there's nobody left to 
turn out the lights.

Larre Shiller
US Social Security Administration
Office: 410.965.2209
 
“The opinions expressed in this post are mine personally and do not necessarily 
reflect the opinion of the US Social Security Administration and/or the US 
Government.”

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
[email protected] with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to