@jmoiron https://t.co/ZFSHy9dOrc
jmoiron plays the blues
@jmoiron Adeus, Astrud 😢
@jmoiron It's wild how much of a cultural touchstone things like Chrono Trigger or Cowboy Bebop have become; things that (in the US) felt incredibly niche in their time.
(also, to the question, Primer or 12 Monkeys) https://t.co/8ckWtamLEw
@jmoiron @JackMurphyRGR Shows how little people incorporate evidence into the way they build their world view when even corroborating evidence is dismissed.
@jmoiron @enpeta7police Yes :)
@jmoiron The problem with ORMs is mostly a social one: it's not that they aren't a useful tool for SQL experts, it's that they are foot canons for SQL novices, and the way they're sold to novices as "a way to avoid writing SQL" prevents them from building this judgement. https://t.co/LLBI...
@jmoiron.bsky.social Is there a consensus as to why Mimas is spherical but Proteus is not? Is it Proteus' slightly higher density enough of an explanation? Does temperature play a role?
@jmoiron @jordanorelli I still see this occasionally, and people have told me that it is "more readable" which is both false and not the inspiration behind doing it in the first place.
@jmoiron Times Square got this much better and it's still absolutely terrible. https://t.co/DDtd8XKy1y
@jmoiron Threw on Guitar Romantic this weekend, just sat and listened. Hard to believe it's been 20 years. There's a re-master due to come out this week. Wonder if that period is due a revival soon.
@jmoiron @inowland DRAM isn't persistent, but even there repeated access to the same DRAM page is faster than accesses to different pages. Everywhere you look, sequential access is going to be advantaged, whether it's in cpu cache or prefetching.
@jmoiron Lesson: linear access has dominated the physical storage layer for ~50 years. Even when vendors improve random workloads, physics (c) will continue to advantage linear over random, and hardware will continue to bias for linear workloads because you design for what people run. ht...
@jmoiron The only reliable 10x multiplier in engineering is having solved the same problem before
@jmoiron RT @BlackwolfOates: https://t.co/90jSpfG2N0
@jmoiron.bsky.social replied: Mine:
ei.org (reed elsevier, now RELX), winsec, attila technologies, advance internet (conde nast), shopwiki, flavors.me / HiiDef, turntable.fm, datadog
@jmoiron No, it's not in an environment. It's been towed beyond the environment. https://t.co/w9R0KJl3va
@jmoiron I remember seeing a demo of this a long time ago on some thread in the selectbutton forums. This was a really great game and more people should play it. https://t.co/3BaNfAHdAN
@jmoiron As people run more temporal workloads, this will only need to be learned (at tremendous cost) over and over. LIFO isn't the only mechanism that can reduce temporal locality and kill goodput. https://t.co/KlCeO4td6S
@jmoiron.bsky.social Using my domain as a handle does seem nice and portable, but I've never been totally convinced on de-cen social protocols. Scale without centralization just can't reach the efficiencies we see, and 99% of people will just use the New Thing.
@jmoiron @RayS Thanks Ray! 🙇
@jmoiron To this day I don't have a lobsters account because I have always felt weird asking for one. That said, I wouldn't mind a bluesky invite, if anyone has one they are willing to send my way.
@jmoiron @PreetamJinka "You're not uncool, it's just hard to make friends as an adult" :( https://t.co/68o6lDP9C9
@jmoiron @goinggodotnet Readability has a similar subjective vagueness to it. Usually these words mean "my preferred style". I'm not claiming all styles are equal, but if you can't articulate the improvement in some other way, it's just a style change, and these kinds of changes are low...
@jmoiron A task to make code "Cleaner" is a flag: it's almost always a no-impact medium-risk refactor with no measurable positive impact. Those cases where it's not should always describe the change and its benefits directly; "Clean" has very little value on its own.