WYFP is our community's Saturday evening gathering to talk about our problems, empathize with one another, and share advice, pootie pictures, favorite adult beverages, and anything else that we think might help. Everyone and all sorts of troubles are welcome. May we find peace and healing here. Won't you please share the joy of WYFP by recommending?
It's post 7.45pm CDT and a WYFP diary hasn't been posted yet...... and I've got a doozer of an issue.
Monday, I (we 4 contract employees) were accused of shopping for an answer to a question that we've been asking for over a month now. Do you want us to consolidate DB2 sql disasters and DB2 'not available' return codes where we can or not?? The answer we've been given is analyze it..... well.... that's a shit, non committal answer if I ever heard one. Then we go through one design review where I had consolidated codes and got lambasted for it. So, I put it all back and then make sure that all error handling is taken care of and get lambasted for doing that and adding return codes...... Ok, fine screw it. I'm not touching anything that doesn't absolutely doesn't NEED to be changed in this conversion.
We were instructed to convert around 100 pgms, 15 tables, untold copylibs, jcl, packages, and binds from IMS to DB2 and while we're in them clean up the code. I think that there's an average of 4000 lines of code per pgm.
The next problem is the time frame to complete this project. The lead pgmr wants us to 'buy in' to no more than an average of 2 weeks per pgm (to design, write test cases, code and unit test). We're telling her that that idea is insane. Her response is basically to threaten to recommend that our jobs get shipped to India if we don't agree. She claims that the conversion will go faster as we move on in the pgms. Maybe we will, maybe we won't. We won't know until we get further along.
We've pretty much decided that when push comes to shove, our reply will be - Based on J's assessment and estimation that the pgms will become easier as we progress, ok.... Then when we're getting closer to the dead line and things aren't complete, we can refer back to J's assessment and that she was incorrect and what we had been trying to tell them was more accurate. She KNOWS we're more correct in our estimation than 'they' are.
Needless to say, I'm pissed.
We're designing on the fly now. We're finding issue after issue. There may be a problem with the new databases not containing all the information from the original. There are 2 tables that had a few fields with the exact same name. The system architect and user have decided that they contain the same data. J, the senior pgmr and lead project pgmr isn't so sure and hasn't been able to prove it yet.
This is what the design specs are going to look like, based on J's comment that they don't need to be so detailed:
1000-Mainline Section.
No changes
1100-Primary-edit Section.
Change IMS names to DB2
1200-Edit-Repair-Date Section.
Change IMS date time ddmmmyy format to DB2 ccyy-mm-dd format
1300-.......
I had been doing it much more detailed. So detailed that the 'coding' and compile phase took me only 12 hours on a 10,000 line pgm to get a clean compile.
I will probably continue doing the same thing for myself. But what I give to management is what I wrote above. Cause I have a feeling that we're going to be specing, writing the test cases and they'll have someone else to the coding and testing...... cause those outsourcing places work faster..... they'll get crap back. But, that's not my problem.
They've decided that this project isn't really worth investing the proper amount of time and resources despite the fact that this conversion is the foundation for all of their future plans. So why should I give them exactly what they need to accomplish their end goal??? They decided that we're the 'hired help' and basically mercenary coders.... That's what they're going to get.