Was he entirely wrong? Have you tried to dump the stored proc into a frontier model and ask it to refactor? You'd probably have neat 20 stored procs with well laid out logic in minutes.
I wouldn't keep a ball of mud just because LLMs can usually make sense of them but to refactor such code debt is becoming increasingly trivial.
Yes. I mean... of course he was?. Firstly, I had already gone through this process with multiple LLMs, from various perspectives, including using Deep Research models to find out if any other businesses faced similar issues, and/or if products existed that could help with this. That lead me down a rabbit hole of data science products related to regulatory reporting of a completely different nature which was effectively useless. tl;dr: Virtually all LLMs - after understanding the context - recommended us doing thing we had already been urging the business to do - hire a Technical BA with experience in this field. And yes, that's what we ended up doing.
Now, give you some ideas about why his idea was obviously absurd:
- He had never seen the SP
- He didn't understand anything about regulatory reporting
- He didn't understand anything about financial derivatives
- He didn't understand the difference between Transact SQL and ANSI SQL
- No consideration given to IP
- etc etc
Those are the basics. Let's jump a little bit into the detail. Here's a rough snippet of what the SP looks like:
SELECT
CASE
WHEN t.FLD4_TXT IN ('CCS', 'CAC', 'DEBT', ..... 'ZBBR') THEN '37772BCA2221'
WHEN t.FLD4_TXT IN ('STCB') AND ISNULL(s.FLD5_TXT, s.FLD1_TXT) = 'X' THEN 'EUMKRT090011'
END as [Id When CounterParty Has No Valid LEI in Region]
-- remember, this is around 5000 lines long ....
Yes, that's a typical column name that has rotted over time, so noone even knows if it's still correct. Yes, those are typical CASE statements (170+ of them at last count, and no, they are not all equal or symmetric).
So... you're not just dealing with incredibly unwieldy and non-standard SQL (omitted), noone really understands the business rules either.
So again... yes he was entirely wrong. There is nothing "trivial" about refactoring things that noone understands.
The issue is that AI will be creating software at whatever abstraction layer it is asked to produce. Right down to ASM maybe even machine code if someone actually wanted or needed that. Perhaps not the AI of today but given a few years I'll be quite surprised if it still can't.
If we can take a computer as powerful as today’s laptops and make it crawl because of the amount of inefficiencies in software like Teams, I’m not holding breath for embedded. If you apply the same kind of engineering principle as Anthropic, you’ll be laughed out of the room.
> For example wood carvers, blacksmiths, butchers, bakers, candlestickmakers etc etc.
Very, very few of those professions are thriving. Especially if we are talking true craftsmanship and not stuffing the oven with frozen pastries to create the smell and the corresponding illusion of artisinal work.
They are thriving where I live. There is a huge artisinal market for hand crafted things. There are many markets, craft centers, art fairs, regular classes from professionals teaching amateurs etc. In most rural communities I have visited it is similar.
They're existing, not really thriving. Artisanal things have become more popular as a hobby, but even people who get into them commercially rarely make real money off of it. The demand exists, but purely as a novelty for people who appreciate those types of things, or perhaps in really niche sub-markets that aren't adequately covered by big businesses. But the artisans aren't directly competing with companies that provide similar goods to them at scale, because it's simply impossible. They've just carved out a niche and sell the experience or the tailoring of what they're making to the small slice of the population who's willing to pay for that.
You can't do this with software. Non-devs don't understand nor appreciate any qualities of software beyond the simplest comprehension of UX. There's no such thing as "hand-made" software. 99% of people don't care about what runs on their computer at all, they only care about the ends, not the means. As long as it appears to do what you want, it's good enough, and good enough is all that's needed by everyone.
The problem for software artisans is that unlike other handmade craftwork, nobody else ever sees your code. There's no way to differentiate your work from that which is factory-made or LLM-generated.
Therefore I think artisan coders will need to rely on a combination of customisation and customer service. Their specialty will need to be very specific features which are not catered for by the usual mass code creation market, and provide swift and helpful support along with it.
I'm the exact age as the author and this post could have been written by me (if I could write). It echoes my story and sentiment exactly right down to cutting my literal baby teeth on a rubber key ZX Spectrum.
The anxiety I have that the author might not be explicitly stating is that as we look for places we add genuine value in the crevices of frontier models' shortcomings those crevices are getting more narrow by the day and a bit harder to find.
Just last night I worked with Claude and at the end of the evening I had it explain to me what we actually did. It was a "Her" (as in the movie) moment for me where the AI was now handholding me and not the other way around.
> The anxiety I have that the author might not be explicitly stating is that as we look for places we add genuine value in the crevices of frontier models' shortcomings those crevices are getting more narrow by the day and a bit harder to find.
That's exactly it. And then people say "pivot to planning / overall logic / high-level design," but how long do we have before upper management decides that AI is good enough at that stuff, too, and shows us all the door?
If they believe they can get a product that's 95% of what an experienced engineer would give them for 5% of the cost, why bother keeping the engineer around?
English is my second language so I'm not well tuned to picking up on the phrases that expose writing as AI generated. Even so it doesn't really change the sentiment being conveyed nor the fact that it's better writing than I could muster.
I really don't understand the demise of usenet as a way to have a public message board. It worked perfectly well for decades and then died off all at once when the bigtechs did everything in their power to squelch it and instead replace with their walled gardens.
This guy has completely lost the plot. Too bad he holds the entire regime in Washington by the short hairs so they're all too happy to comply with his delusions.
I wouldn't keep a ball of mud just because LLMs can usually make sense of them but to refactor such code debt is becoming increasingly trivial.
reply