zbentley

8 ms·

121.
by zbentley 2mo ago
AI can benefit significantly from type checking . This provides a minimal-setup-required feedback loop that it can use to address errors. User defined types are also a greppable “what do I need to look at in this codebase to implement feat
122.
by zbentley 2mo ago
Franchise failure rates don’t bear out that being a real risk. However, McDonald’s is publicly traded, and investor pressure to improve at the margins may make a business that is fundamentally well served by being change-averse do stupid th
123.
by zbentley 2mo ago
> What in my comments gave you the impression that I'm trying to argue for this at all? Of course they don't care, otherwise they'd be maintainers themselves. You said: > When something isn't packaged for these use
124.
Marketers are Addicted to Bad Data (2020)(jacquescorbytuech.com)
42 points by zbentley 2mo ago|48 comments
125.
by zbentley 2mo ago
2004 was a weird time.
126.
by zbentley 2mo ago
The existence of a build step at all, even a nice and simple one, is enough to deter lots of users who would find a lot of software useful. Remember that nontechnical users exist (e.g. folks who only know how to install software via a GUI).
127.
by zbentley 2mo ago
> It has worked for decades and many thousands of packages just fine And yet insane quantities of programs used by millions are only available via raw .deb file downloads, or random alternative PPAs, or via programming language package m
128.
by zbentley 2mo ago
That’s naive. Most users want to use a given piece of software . If they can’t, they usually consider that a defect of the software in question. The users that know and care enough to complain to their distro maintainers are not representa
129.
by zbentley 2mo ago
I don’t think that’s an architectural solution to this class of bug. I replied to a similar comment here: https://news.ycombinator.com/item?id=49281533
130.
by zbentley 2mo ago
Maybe, but there’s nothing about this particular bug that’s due to SQLite not being a networked database. Postgres is just as likely to have TOCTOU races in its checkpointer, which works roughly the same way.
131.
by zbentley 2mo ago
I mean, ProxySQL and SQLite are at very different tiers of reliability. I encountered multiple unreported data-corrupting (and some resource exhausting/connection mis-pinning) bugs in ProxySQL within a few months of using it for the
132.
by zbentley 2mo ago
Eeeh, this particular bug was a race when applying WAL to the data files in a checkpoint. Postgres’s checkpointing is theoretically just as vulnerable to this class of bugs as SQLite, though it doesn’t seem to have equivalent issues today.
133.
by zbentley 2mo ago
This particular bug doesn’t seem to arise from SQLite’s “lite” nature. It’s a TOCTOU inside the DB when applying WAL segments in a checkpoint, which is a pattern used in extremely similar ways by Postgres and MySQL. They don’t seem to have
134.
by zbentley 2mo ago
The subset of the list that supports in-process calls and which allows for zero-copy (and, realistically, probably mutually mutable) data sharing of its types is shorter still. I think the useful definition of "ABI" includes more
135.
by zbentley 2mo ago
This has been somewhat argued to death, but even if you put aside operational concerns with static linking (security/size/independent upgradability), many attempts to do away with shared library ABIs end up reinventing them--at le
136.
by zbentley 2mo ago
I wonder how many of them only work because of things like https://github.com/WebKit/WebKit/blob/main/Source/WebCore/pa... Which overlaps with, but isn't the same as, JS quirks mode.
137.
Weekend at Bernie's(nesbitt.io)
2 points by zbentley 2mo ago|0 comments
138.
by zbentley 2mo ago
Even if it's better than Argentina's, it's still not good: https://www.eatingpolicy.com/p/grasping-at-paper-straws Contracting policy/regulation is an area that accumulates the worst tendencies of g
139.
by zbentley 2mo ago
I don't think that follows. Wikipedia's sourcing rules overwhelmingly favor publications released for non-pageview-based purposes (academic writing, books), or journalistic productions (whose pageview-based revenue is almost entir
140.
by zbentley 2mo ago
Eh, it's hard to prove absence, and time is a beneficial quantity here. The longer something sits on nightly, the greater the chance that bugs are identified before it reaches stable and its usage significantly increases. A lot of bugs
141.
by zbentley 2mo ago
> some people are just not very into building step by step like "okay first let's check the JSON shape the endpoint returns", "okay now let's make a simple form", etc That's accurate. There are a lot of
142.
by zbentley 2mo ago
I'm choosing to interpret this as very dry sarcasm.
143.
by zbentley 2mo ago
The comparison to banks is actually pretty interesting. Unlike app developers, banks are by default directly liable for fraud that happens on their system. In many cases, even when someone gets their bank to send money to a scammer, they
144.
by zbentley 2mo ago
Fortunately, it's not a linear scale. All sorts of technical and political options exist which don't fall cleanly into the anarchy or baby-proofing spectrum. Random incomplete examples, in no particular order and off the top of my
145.
by zbentley 2mo ago
Dude, write your own blog posts. LLM writing obscures whatever points you were trying to make. It’s obtuse, repetitive, verbose, ugly prose. This is not compassionate, educational, or likely-to-garner-engagement communication. If you care a
146.
by zbentley 2mo ago
I don’t think writers have automatic oracle status or anything, but they tend to issue commentary that has a better track record of being borne out than what is communicated by business leaders with a very strong vested interest in advancin
147.
by zbentley 2mo ago
Or sales/FDE deployment
148.
by zbentley 2mo ago
ERPs are often used as a lot more than a system of record: they’re directly engaged with as the LOB software point-of-contact by many expert employees, and function as a sort of encoding of organizational behavior. They’re a lot more than a
149.
by zbentley 2mo ago
> they aren't lying, they might be deceptive I think that’s well within the bounds of what most people would consider to be a “lie”. The legal system has more specific definitions, though.
150.
by zbentley 2mo ago
> like 99% sure Claude would've mentioned where the majority of the project came from. I doubt that. I’ve seen it produce thousands of lines of code or text with like a 2% string distance from something available online and not attr

More ›