Shane Ginnane wrote: <snip>
The (HZR) logger dependency starts to look a bit silly on a (non-sysplex) ring of 5 or 6 MONOPLEX systems.
<snip>
As many seem to ask for things like this, it raises an interesting general question. In how many ways should we implement things? Intuitively speaking, the greater flexibility offered by more choices always seems good, but...
Someone I work with uses the metaphor "surface area" to describe much of the system's componentry. Surface area can allude to both the number and nature of externals and to internals of various kinds. The more surface area you design in, the more you need to do, both to implement a function now or change it later. (Meanwhile, someone else here has embarked on a rather systematic course of hair removal using the time-tested grab-and-pull method, in large part because the thing he's working on changing has "a lot of surface area.")
I think adding surface area needs some justification beyond, "It would be nice if...."
I do get that converting to OPERLOG on systems that run along quite happily with entrenched SYSLOG-based systems is harder than it might appear at first glance from our end of it. On the balance, though, it would have cost more development time to support both SYSLOG and OPERLOG, added to code complexity and support cost, added to PFA's documentation, and added more complexity and cost to future changes in the same area. Would it really have been worth it? If I asked the same question 10 years from now would the answer change? What function would you have given up in return?
Just some food for thought. -- John Eells z/OS Technical Marketing IBM Poughkeepsie ee...@us.ibm.com ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
