Sorry about the belated response. On 30 Jun 2005 14:58:54 -0700, in bit.listserv.ibm-main you wrote:
>In <[EMAIL PROTECTED]>, on 06/29/2005 > at 08:40 PM, Clark Morris <[EMAIL PROTECTED]> said: > >>Shmeul > >That's Shmuel! I claim bad typing because I looked up one of your posts to get it right. My apologies. > >>For the rest of the points raised I am making the following >>assumptions and need to understand if they are correct. > >No. PL/I supports BSAM. > >>4 Few if any applications programs in most shops (as opposed to >>system or home grown utilities) depend on BSAM and BPAM in >>installation written code. > >That's a hard one to judge; I've seen far too many installations >depending on programs for which they have lost the source code. I >doubt that it is possible to collect the statistics. > I suspect the need to handle things like long filenames and other new world stuff will doom these programs in the long term. Since z/OS support for CKD will be needed for at least 5 years and more likely 10 years after announcement of FBA anyway, I suspect the cheapest thing to do is keep the CKD code for existing programs and provide migration paths. The old programs won't be able to handle records expanded beyond 32K nor the blocks, longer names, Unicode, etc. anyway. One reason why I am not enthusiastic about maintaining the ability of old programs to access FBA files without a recompile is the shortcomings I see in PDSE. Note that many shortcomings I perceive may not be there but for example I think that we are still limited to 8 character member names. In looking toward what is needed in a platform for the 21st century, I have the sinking feeling that it would be far easier to build the z/OS reliability, scalability, work load management etc. into a Unix base than to bring long name, 64 bit for all compilers and all the other function that Unix and Windows people take for granted on servers into the z/OS environment. Given that we have shops still running 24 bit code, the maintaining of compatibility has its costs in terms of moving forward. If the z (now z9) series is to have a long term future, we have to determine what things can be broken as we move into a vastly different world. I can remember being unhappy when programs took over 100K to run under MVT. Now I can see the need for 1 megabyte records. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
