@theorytoe
Is that D'n'B? 🤔
@r000t
No developer — even godlike, is knowledgeable enough how it would work in real-life conditions, but neither is LLM.
True — I might be wrong and a lot of DB-devs never go through such thorough trials. On the other hand, I doubt said devs ever deal with such critical infrastructure. Again, I might be wrong in general and it's all amateur bullshit at this point 🤷
@r000t
Which were supposed to introduce further performance enhancements, but not under all conditions — we did all that shit and at times we still had to resort to buying more expensive hardware — WAY more expensive! It was as far as possible from "This just won't work!" coming from some seemingly smart kid, knowledgeable about databases — we did real testing!
@r000t
I agree with you on the general principle, what I doubt is — how it works on that scale. I have to admit: I did have to deal with MS SQL — no company names of course, but it was critical. And what we did over the weekend on introducing performance enhancements is — we poured n-last-days of data from production into it and we were reproducing the most likely scenarios from production database with three most recent releases of MS SQL with different versions of our .DLL-shims…
@r000t
I don't mean you're wrong though — the company could hire less qualified developers to do the optimisations, and to a degree — they would still be successful (I doubt it would be in the 60%-90% ballpark on that scale — hundred of thousands of queries), but still — it might be not be as linear, and the final result could be less obvious.
@r000t
True, but — saving 60%-90% time on queries running hundreds of times per day? They should and would have done it even without chatGPT long time ago, that would have saved their company generous sums on hardware, especially considering lots of business processes still aren't properly scalable and can't be parallelised and distributed over as many nodes as possible — they have probably had to resort to buying more and more expensive hardware every time the load increased noticeably.
Void Linux users... buckle up ⚡
📦 **xbps-tui**: A TUI for the xbps package manager.
💯 Supports search, filter, package info & more!
🦀 Written in Rust & built with @ratatui_rs
⭐ Source: https://codeberg.org/lukeflo/xbps-tui
#rustlang #ratatui #tui #voidlinux #linux #package #manager #terminal
@romin
…do as the Romans do 🤪
@kravietz
Oh, I know exactly what you are talking about!🤣
> migrated to PL/SQL
Or TSQL! A more straightforward migration path from MS Access. Then high-throughput parts have been rewritten with pluggable .DLL-libraries here and there — because otherwise it's slow… And "solutions" like that have been accumulating over decades — and no one remembers how any of this works except for that one man who currently works in a different department and is nearing the retirement age himself.
Regarding the US scandal about Social Security database[^1], a completely unrelated anecdote from an IT practitioner.
I once worked for one firm developing an IT system for managing corporate pensions. I managed to erase the most traumatising memories, but just imagine this: a system developed since 1980’s that started as a set of complex electronic spreadsheets (think early MS Excel), then migrated to early local databases (think MS Access), then migrated to PL/SQL where textual representation of the user interface was implemented in SQL.
Please re-read the last part above, so that you can fully imagine this little coding wonder - text and then HTML being printed from inside SQL. I know most people likely won’t believe this, but please believe me that was true - and then think for a minute about maintainability of such code. And say a short prayer for its developers.
It was not done by choice, of course - that’s the end result of 40+ years of continued development of a system that constantly remained in production, with production data loaded, on which real people’s pensions depended.
On top of it add dozens of changes in pension law, which must have been accumulated in the code because you must differently calculate pension before and after say 1994. On top of it add dozens of code variants created for different clients with different needs and jurisdictions.
Now about the data. This is the part most misunderstood by people in EU - in UK and US there’s no central database of citizens, like for example we have PESEL in Poland. At birth you’re assigned a PESEL identifier and it accompanies you the whole life and uniquely identifies you for your salary, pension, healthcare and everything. Britons say it’s “for privacy”, which is why when we apply for a bank account or rent a car we need to bring a pile of bank statements and utility bills for the last three years, but let’s not dwell on this subject.
The second part to this is the legal culture of declaring your identity rather than having it certified by government agencies, so if you once write “Annie” and then 10 years later you write “Ann” or “Anne”, this ultimately creates three database objects that are really one physical person. On top of this add typos and reading or OCR errors. The consequence of this is terrible inconsistency and duplication of data because there’s no single reference id.
When read article on US Social Security database issues such as the one below, this rings a bell. All the above factors is precisely how you end up with ~400m entries in a pension database in a country where ~340m people are really living. And that’s how you end up with people with incorrect birth dates and no death dates[^2], for example because nobody legally confirmed their death.
But this doesn’t necessarily mean that US SSA is a crooked system that illegally pays 400m pensions to non-existent people. The total amount of payments that were not legitimate was estimated at just ~0.84% and that’s less than private pension funds[^3].
Of course, Musk could easily fix it if US had implemented an EU-style single national id database to use as a reference record, but good luck with that…
[^1]: https://www.factcheck.org/2025/02/online-posts-misconstrue-data-on-social-security-numbers/
[^2]: https://www.ssab.gov/research/social-security-and-the-death-master-file/
[^3]: https://www.politifact.com/article/2025/feb/17/are-150-year-old-americans-receiving-social-securi/
@newt
Time to made a child with the girl. You are getting used to it 😅
@yayfee @neural_meduza
Нет, но мы можем сообщить об ударах молний ранее 😅
@Rezard
Но если нет разницы, зачем платить больше? 😏
Happy #303day to all who celebrate.
It took me a whole year to finally upload last year's jam...
None
Just in case: DMs/PMs simply don't exist on this instance as concept — don't use them, use the other instance if you absolutely have to, or send an email to any address at m0xEE.Net or .Com or .Org, but I prefer keep most communication public.