DaisyDay
I Did Nothing Wrong!! ~~Team Deep State
- Jan 7, 2003
- 40,821
- 19,236
- Country
- United States
- Gender
- Female
- Faith
- Unitarian
- Marital Status
- Married
- Politics
- US-Others
None would be my guess. Why would anyone put COBOL code in a database? The way it usually works is that the code accesses the database using some protocol such as SQL in a black box function.I just have access to the beginning of the linked article. But MILLIONS of lines of COBOL code is an incredible number, I wonder how many of those lines of code are part of the old Social Security version of a database.
What they might not realize is that COBOL works with modern relational databases.What the DOGE people realize is that a modern relational database is needed.
They wouldn't have to anyway; just the interface would need adjusting when the database is replaced - and maybe some reformatting of the data.They don't have to rewrite the database code, there are open source relational databases available.
Process checks? I doubt they actually process "checks" much these days as most payments are by direct deposit these days, but they still need to process some as the old legacy systems that use paper checks get phased out. COBOL processes gigantic batch runs very well. You actually have no clue what fraction of the old number will be needed to process millions of transactions nightly, daily, weekly, biweekly, monthly, quarterly and annually. The system has to process a huge variety of inputs from he IRS for every dollar (over $600) reported, withheld, remitted and categorized. Each of those inputs generates transactions to various other systems - treasury, banking, IRS, states and other municipalities, employers and employees (workers who pay into SS). Then there are the business rules that have to be known and understood - even from decades ago - that still affect present-day payers or recipients.The code needed to process checks, etc. would be a tiny fraction of that old number.
Upvote
0