On Thu, Feb 6, 2014 at 5:07 PM, Ward, Mike S <mw...@ssfcu.org> wrote:
> John McKown " It also gives our end users the idea that z/OS is incapable > of easy to use data access." John, what do you mean by this statement? > > The "problem" with data access here is that _all_ our data is stored in VSAM KSDS data sets. As opposed to, say, Windows where it is on MS-SQL Server and Oracle databases. So every time the users want a data extract, they must request that a program be written and then run. The resulting extract file must then be downloaded so that the data can be put in the proper place. Either in an SQL database, or maybe the user can use it directly. This makes "ad hoc" requests very difficult compared to just developing an SQL SELECT which can not only retrieve data, but also do some aggregation. Or can be integrated into an .NET program which the person, or a developer, can write faster than we can get a new COBOL program written and run. This thanks to the fact that z/OS has _good_ change control which is _enforced_. I don't know how they do Windows development. But, in any case, writing COBOL using ISPF in a compile/run/debug loop is much "slower" than using a Windows IDE. No, we won't get RD/z. Too expensive. We do have a product, PowerExchange, which can do SQL type queries against a VSAM KSDS, or even a sequential data set. The first problem is that it needs somebody to supply a "schema" of the data set. The second problem is that everybody who knew how to use the product has been RIF'd. But this is regarded as a mainframe problem because data access should be "brain dead simple". I agree that __end-user__ access should be simple. But that costs money (as in get DB2 or some other "package", for which we have no budget). Now, due to the above, an end user request for data access takes much longer on z/OS than on, say, Windows. And so the end user perceives that z/OS is inherently more difficult to get information out of. Add to that the fact that IT management here basically regards z/OS as obsolete technology, and you probably see the problem. -- Wasn't there something about a PASCAL programmer knowing the value of everything and the Wirth of nothing? Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
