UK Politics
General Discussion for politics in the UK.
Please don't post to both !uk_politics@feddit.uk and !unitedkingdom@feddit.uk .
Pick the most appropriate, and put it there.
Posts should be related to UK-centric politics, and should be either a link to a reputable news source for news, or a text post on this community.
Opinion pieces are also allowed, provided they are not misleading/misrepresented/drivel, and have proper sources.
If you think "reputable news source" needs some definition, by all means start a meta thread. (These things should be publicly discussed)
Posts should be manually submitted, not by bot. Link titles should not be editorialised.
Disappointing comments will generally be left to fester in ratio, outright horrible comments will be removed.
Message the mods if you feel something really should be removed, or if a user seems to have a pattern of awful comments.
!ukpolitics@lemm.ee appears to have vanished! We can still see cached content from this link, but goodbye I guess! :'(
view the rest of the comments
Yep honestly in genral this is true.
But gender historically has not been a field expected to change. This is an easy fix when such medical information has no relevant on the use of the data. IE pretty much every other database used.
But when you make a change where historical data needs to be kept. The age and design of such systems makes the changes much more complex and expensive.
As a retired legacy systems software engineer. Where updating old database structures and all the systems related to them was my job for 20+ years.
This would involve hunting down old and badly documented systems all over the nation.
It is the need to address what would have been implemented as a fixed variable never expected to change. But also used to decide on many medical pathways. Even more so over the last 30 years as the medical industory has been correcting male dominated research bias on treatment pathways.
And having to find every instance connected to it and ensure it can handle something that is now a data type with a historical status.
Unfortunatly developers have historically implemented code with the expected prejudices society embodies. We were after all only human so did stupid human things with out knowing the potential harm.
Oh so that's what they need AI for...
LOL. AIs help will would be significant, but not adequate on its own. It's just not at that level yet. I did this for Y2k in the financial industry and worked for the US medical industry in the early 2000s doing similar legacy migration.
The issue is distribution of solutions. The NHS has 50 years of random IT roll-outs, handling in very different ways with little unification. The system would need a huge change to unify the data formats and sharing policies. This on its own would take years to manage over such a large organisation that basically cant do the shut a system down and see what breaks diagnosis used by many to investigate old servers long since forgotten.
It's doable. AI will help staff do it faster. But it will still miss a lot, and live testing is way more dangerous than would ever be expected.
The best approach in my opinion would not only be to re-write everything and roll out a whole new system within every NHS and linked system. With a collection of APIs allowing all the old data types and interfaces used to link to that system. Then spend a year or 3 running them in parallel with staff using the old system. Using AI and skilled developers to hunt for and fix areas where data fails to get added/updated to the system as expected. Then slowly start moving staff over to the new system.
But its also the most expensive option and does little to address the human problem in such situations. IE basically risking repeating the London ambulance service IT roll out like issues.