UK Banks Still Run Software Code Written More Than 60 Years Ago
Posted3 months agoActive3 months ago
computerweekly.comTechstory
calmmixed
Debate
40/100
Legacy CodeBanking TechnologySoftware Maintenance
Key topics
Legacy Code
Banking Technology
Software Maintenance
UK banks are still running software code written over 60 years ago, sparking discussions about the reliability and maintainability of legacy code, as well as the potential role of new technologies like LLMs.
Snapshot generated from the HN discussion
Discussion Activity
Moderate engagementFirst comment
50m
Peak period
8
0-3h
Avg / period
4
Comment distribution24 data points
Loading chart...
Based on 24 loaded comments
Key moments
- 01Story posted
Sep 22, 2025 at 5:57 AM EDT
3 months ago
Step 01 - 02First comment
Sep 22, 2025 at 6:47 AM EDT
50m after posting
Step 02 - 03Peak activity
8 comments in 0-3h
Hottest window of the conversation
Step 03 - 04Latest activity
Sep 24, 2025 at 3:35 AM EDT
3 months ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
ID: 45331180Type: storyLast synced: 11/20/2025, 2:33:22 PM
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
In the past I've heard that some banks put a decimalisation layer on top of their existing business logic, that would translate between the old Pounds, Shillings and Pence currency, and the new decimal currency. I wonder if there are any banks out there which still have Pounds, Shillings and Pence at the heart of the computer systems.
I'd expect there was still ugliness as the conversion ratio was 2.4 to 1, not making for an elegant quick shift.
Also yielding:
I wonder how they handled those values with their mapping layer? Did the add a layer for handling fifths of an old penny?(The 2.5p coins, then eventually also the 0.5p "new pennies" went out of circulation.)
FTA: “the reason it lasted is because it was very simple and it worked properly and it was high volume, simple transactions,” he added. “The banks are moving away from these systems because the people who understand them are leaving, and no young professionals want to learn languages like Cobol.””
I guess young professionals do not want to write software that is very simple and works properly (/s, but only partially)
Maybe it makes sense to look at this through the developers' career-management lense.
Specializing in a potentially niche technology such as mainframe COBOL means limited job options as time goes on.
With few employers and few employees in that market, small changes could drive salaries much higher or much lower. (I'm speculating.)
It would make sense for developers to demand higher salaries to justify that risk, but AFAIK banks have a reputation for low developer pay.
This is an organizational problem or just plain neglect. I learned cobol back in the seventies in one day. I used it for 40 years and I never had any problems understanding programs written by others.
Sometimes at different banks which were subsequently merged with, sometimes by a consultancy which has long since gone bankrupt, sometimes with a compiler that is no longer available. It's a miracle the system works at all.
That's quite exceptional. Have you never worked on legacy code or with below average devs?
Granted, perhaps banking isn’t the ideal place to experiment with this type of technology, but it does seem like a promising use case.