On Sat, 12 Sep 2015 21:08:36 -0400, Shmuel Metz (Seymour J.) wrote:

>  on 09/11/2015  at 11:08 AM, Kurt Quackenbush said:
>
>>You can greatly simplify you and your customers' efforts if you can
>>avoid MODs and JCLIN altogether.
>
>BTDT,GTS. Level sets are evil. If a fix changes only one csect then
>the PTF should not replace other csects.
> 
I supported that position.  But I'll also speak from the other side of
my mouth.  For example, HLASM prints a PTF level in its signon page.
This implies that every PTF is a level set: it SUPersedes its immediate
predecessor even if it doesn't replace unchanged CSECTs.

From the developer's point of view, suppose I supply 10 PTFS which
I believe are noninteracting.  No PREs; no SUPs; they satisfy SMP/E's
criteria; no BYPASS( ID ) needed because each modifies a different
CSECT.  With APPLY SELECT( ...  ) a customer can install any of over
1000 different configurations.  Of these, how many can he be confident
the vendor has tested.  Realistically, with luck, 11.

-- gil

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

Reply via email to