Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
When I did mainframe bean counting in the last century, we basically kept a database with the 'end of year' and each 'end of month' amounts for each account, and a MTD (month to date) summery. We kept the in the month journal entries with each transaction so they could be updated during the month (and that would be used to keep the MTD amounts up to date. When we closed out each month, we archived the previous months journal entries. On occasion we had to go back and back post journal entries, and re-run end of month or end of year posts, and then re-process all entries from that point forward. That kind of processing was the exception rather than the rule. I did write 'audit' programs, to allow checking the amounts. This kind of program I found as pretty much 'required' because we were consistently in a 'production maintenance' environment. Yes, I understand abhoring anything but fully updated information, and having to re-process old transactions. But after supporting both tax departments and general accounting (I stayed away from HR and payroll like the plague -- different issues there), being ready to explain your system and how you can validate it to auditors is a "good thing"(tm). The process wound up being basically the same as time has come forward. But instead of using VSAM or ISAM, we use SQL language databases. Instead of keeping tapes with journal entries, we kept disks with 'flat files' or sequenced entries in SQL databases. IMHO, the tapes made more sense (especially with the automatic processing of generation data groups ... but that is dating me even further back in the dinosaur age). I hopes this helps.
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |