Quote:
Originally Posted by GirlySports
Anyone here know COBOL?
|
I worked in the mainframe programming ecosystem at the beginning of my career, PL/1 as the programming language and IMS and DB2 as the databases. IMS was a hierarchical database primarily of the 1970s era.
I didn't work with COBOL, but it's a bit of a misnomer to call something a "COBOL database." COBOL is a programming language and most COBOL applications in existence would have used IMS, CICS, or DB2 for data storage. I have read that the Social Security system in question is so old that it uses a proprietary database called "MADAM", so the non-standard "1875 epoch" explanation is entirely plausible.
The following article has some information about the SSA's systems.
https://ccianet.org/wp-content/uploa...205%202009.pdf
There is some entertaining reading buried in this article, and it seems surprising to me now that Musk's team of supposed IT geniuses have completely missed the boat on what the real problem is.
Quote:
|
We cannot afford to store our vital national records in the electronic equivalent of ancient hieroglyphics whose meaning can only be gleaned by an ever diminishing handful of experts. As described in the documents currently available to the public, SSA’s strategy for migrating from MADAM to DB2 without changing its obsolete data formats is the equivalent of transferring a set of ancient hieroglyphics from a crumbling stone wall to a modern plate of stainless steel – such a strategy modernizes the substrate, but does nothing to modernize the archaic form of the text itself. With each passing year, the cost and technical challenges of creating a truly modern database for SSA’s citizen information become greater, as does the risk of a catastrophic failure of access.
|