SV: Rexx compression in TSO?
- From: thomas.berg@xxxxxxxxxxxx
- Date: 30 Aug 2005 10:12:21 -0700
Yes, it is a silly license. A classical example of economical suboptimization by IBM.
Thomas Berg
-----Ursprungligt meddelande-----
Från: IBM Mainframe Discussion List [mailto:IBM-MAIN@xxxxxxxxxxx] För Charles Mills
Skickat: den 30 augusti 2005 18:45
Till: IBM-MAIN@xxxxxxxxxxx
Ämne: Re: Rexx compression in TSO?
Replying to the points made by Don Imbriale, Walt Ferrell, and Thomas
Berg:
Why do I want to embed the source code in the load module? Answer: because you have to do that to exploit the Alternate Library.
Why do I want to exploit the Alternate Library? Answer: because otherwise customers who did not license the Rexx compiler could not run the load module at all.
Why do I complain that uniquely the Rexx compiler requires the target machine to license the compiler in order to exploit its object code? Isn't the Alternate Library free? Doesn't the Alternate Library solve the problem?
Answer: this is a widely misunderstood area. This one comes around on IBMMAIN about once a year. Yes, the Alternate Library is free, but all it lets a customer do is execute in normal Rexx interpreted mode the source code that the Rexx compiler optionally embeds in the object module and hence the load module. It does not allow the customer to execute the faster object code that the compiler generates. Without the Alternate Library, the developer would have to ship two distributions: load module and "classic" Rexx exec. The Alternate Library provides the sole benefit of allowing those two distributions to be combined into one
package: a load module that contains both compiled object code and a classic Rexx exec.
In other words, the Alternate Library does NOT provide access to any of the benefits of the Rexx compiler. A customer who has the Alternate Library is no better off using a compiled Rexx load module than using a Rexx exec. IBM specifically states in the documentation (I am paraphrasing from memory here) that a compiled Rexx program runs no faster with the Alternate Library than the original source code would with the normal Rexx interpreter. I suspect, in non-technical terms, what the Alternate Library does is simply pull the Rexx source code out of the load module and pass it to the normal interpreter in a form that the interpreter can handle.
I can't change IBM's licensing or technology but at least I would like to be able to pull the comments out of the source code that I am effectively forced to ship to customers.
Charles
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to listserv@xxxxxxxxxxx with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to listserv@xxxxxxxxxxx with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
.
- Prev by Date: COBOL callable services?
- Next by Date: Disaster Declarations due to Katrina
- Previous by thread: SV: SV: Rexx compression in TSO?
- Next by thread: SMF TYPE 50
- Index(es):
Relevant Pages
|
