HNHacker News
TopNewBestAskShowJobs

Jtsummers

29,848 karma · joined October 6, 2009

[ my public key: https://keybase.io/jtsummers; my proof: https://keybase.io/jtsummers/sigs/2v_NBwmgy9lrspxf8k7DN4UwIi-73h22Wxr_7kF01RM ]
submissionscomments
Jtsummers··on A technical vision for the Haskell ecosystem
> For the beginner, there isn't enough syntax on offer to discern who's doing what to whom here.

If you throw a beginner at any non-trivial code in any language without teaching them anything about the language, then, yeah, it'll be hard for them to figure it out. That says nothing about the language itself, though. That's pretty obvious to everybody.

> Is there a "function signatures for dummys" somewhere?

Yeah, of course. There's actual beginner material, here's one example:

https://learnyouahaskell.github.io/types-and-typeclasses.htm...

Not hard to find, and not the only introduction to the language that covers this material. I guess you never bothered to look, or even to ask Gemini.

Jtsummers··on Writing correct software is hard and math alone won’t help us (2016)
Engineering mostly isn't building either. AEs don't build the planes, they design them. EEs may prototype some boards, but as a profession it's also mostly design (intended for manufacture). Unless you're an EE working primarily on one-offs, and small ones at that, you don't build most of the products of your mind. MEs don't build the cars. CEs don't build the dams or bridges.

Engineering is principally a design discipline, with a small fraction actually doing the hands-on building of their designs at least once they hit a certain scale of manufacture or get past prototyping. Technicians, skilled and unskilled labor, and machines do the vast majority of the building of those designs.

Jtsummers··on Writing correct software is hard and math alone won’t help us (2016)
CS has always had two (probably more) schools of thought. It's usually clear when you visit or work with people at different universities. Schools where CS came out of their math department are generally more theory focused, ones where it came out of an engineering (EE, specifically) department will be less theory focused and more practicum focused. Of course, this can also shift over time as new faculty come and go, their research areas and backgrounds will skew the focus of their department.
Jtsummers··on Writing correct software is hard and math alone won’t help us (2016)
To draw it, select an initial arrangement (put all the disks on one tower, this will become one of the three triangle corners, with the other two equivalent states at the other corners).

Now draw a dot to represent this state, then consider the two consequent states that can come out of it (move the smallest disk to one of the two empty towers). Draw a line representing the transition between the initial and those two consequent states (and the reverse, you don't need arrows here as all movements are reversible). Now examine the two new states, you'll see you can draw a line between them and also introduce two more states (move second biggest disk to the still empty tower).

Repeat this process, and select the right arrangement of the nodes, and you'll get something approximating the sierpinski triangle. You may need to play with the layout a bit to get it right.

Start with just one disk, then two, then three to see how it gets more and more like the sierpinski triangle as the disks increase.

EDIT: https://www.cut-the-knot.org/triangle/Hanoi.shtml - found this which illustrates it and already has the right layout so you don't have to retry it. You'll notice that it's not quite the same as the Sierpinski triangle, the middle sections are hexagons instead of triangles.

Jtsummers··on Why I wrote the BEAM book (2025)
> Erlang runtime semantics (cooperative "async threads")

Preemptive, not cooperative. Cooperative scheduling is based on the programmer releasing control manually (much like manual memory management) or having yield points baked into the language, often at function calls or IO events. Processes on BEAM can be preempted, they're allotted a number of "reductions" and when those are exhausted, the process is preempted (the process may also yield earlier). This guarantees every process is able to get a turn. Cooperative is fine, but it requires much more care. Go prior to 1.14 suffered from cooperative scheduling, a tight loop (not that you would write this one specifically, but as an example) could hog an OS thread:

  i := 0
  for {
    i++
  }
The conceptual equivalent in Erlang would only execute the loop so many times before it would be preempted.
Jtsummers··on My AI agents found an unmapped Maya site?
> I can’t move past

Come on, George, you're clearly a ramblin' wreck from Georgia Tech. You've taken nearly every course GT has offered over the past century, you can get through a captcha. Live up to your name.

Jtsummers··on Clegg: Getting Too Close to Trump Was a Mistake
Archive link: https://archive.is/G7Dgd

Gift link: https://www.theatlantic.com/ideas/2026/11/silicon-valley-tru...

Jtsummers··on Let over Lambda–50 Years of Lisp (2008)
> I think it's better to learn lambda calculus directly, without the Lisp "gimmick".

The book isn't about teaching lambda calculus, though. It's covering things you can do in Common Lisp, specifically around macros.

Jtsummers··on Iran was behind cyber-attack on French supermarket freezers, Paris says
> Israel and USA have been intent on invading Iran for decades. But they didn't. Then suddenly they did.

There is a war between the US/Israel and Iran, but there has not been an invasion. Attacks from a distance are not invasions. Contrast with the Russia-Ukraine war, where there actually has been an invasion.

Jtsummers··on Iran was behind cyber-attack on French supermarket freezers, Paris says
Russia has demonstrated that they can conduct so-called "hybrid warfare" as long as it doesn't lead to serious injury or death and the response will be meager. But it still creates pain for those targeted nations.

Iran could be seeing that this approach, rather than more blatant attacks, is a useful way to create problems for the US's allies with a lower risk of elevating those US allies to direct enemies. The intention would be to get the US allies to put pressure on the US (in particular) to back out of the US/Israel war on Iran.

Jtsummers··on When a language's creator and its owners strongly disagree [pdf]
> Python is over 30 years old and it's going fine. You don't see people ragging on Guido everywhere.

Guido stepped down from that role 8 years ago. It now has a 5-member steering council along with the core developers as the primary maintainers and decision makers.

Jtsummers··on Drones built to look like birds evaded detection at Air Force tech exercise
This is the size and sort of drones a unit would use for scouting the local area. The range isn't great, but it is suitable for that and less conspicuous than the gas-powered drones used 15 or so years ago and the buzzing quadcopters of today.
Jtsummers··on Making any integer with four 2s (2025)

  7 = ⌈ 2.2 + 2 + 2⌉
Would still work.
Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
Depends on the schools for networks. I attended a college that finished adding wired networks to their dorms in 2000 or so. It just wasn't a campus priority in the 90s, fewer students had computers back then. If it was a school with more money or an engineering school they got to it earlier, the rest added it when they added it.
Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
no longer available
Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
You're going to run that on a computer from the late 1980s or early 1990s? And how do you get the data onto those computers when they're not connected to a network, or the same network, as your other systems?
Jtsummers··on Democrats launching congressional investigations into SpaceXAI, META and Apple
Cowards like them are why I quote comments. They claimed that Democrats (not in power at the time) blocked DOGE and that DOGE was actually going after corruption, both lies. Then they deleted it when called out.
Jtsummers··on Democrats launching congressional investigations into SpaceXAI, META and Apple
> I already have zero trust in government due to the kind of corruption that DOGE was going after. But the Democrats shut that down.

DOGE never actually went after corruption, though, and only idiots ever believed that was their intention. They said they were pursuing savings by finding fraud, waste, and abuse and then ignored one of the biggest sources of all three: The DOD.

And then the Republicans decided that the remarkably efficient and perennially audit failing Pentagon should get an extra half trillion dollars. Largely to pursue contracts with companies owned by Musk, Thiel, Lucky, and other Trump inner circle types.

And how, by the way, did Democrats shutdown DOGE? Republican president, Republican Senate, Republican House. They controlled everything they needed to to continue operating DOGE. DOGE is squarely a Republican failure to do anything useful.

Jtsummers··on Democrats launching congressional investigations into SpaceXAI, META and Apple
Vetoes can be overridden by Congress, though Republicans will have to vote with Democrats since the Democrats don't have a significant enough majority. And many of the surviving Republicans are going to have to evaluate how safe they feel given the recent election. Do they continue to fail their constituents and risk losing in 2028 to kowtow to a lame duck president, or do they actually do something worthwhile for their salaries and the trust that's been given to them.
Jtsummers··on Democrats launching congressional investigations into SpaceXAI, META and Apple
Go to archive.is (or one of the other archive sites), paste in the URL you want to get an archive copy of, then copy the resulting URL and paste it here.
Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
No. DOD still has a lot of air-gapped systems, some for legitimate reasons, and many out of momentum or lack of understanding of how networks work.
Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
Useful aphorism in general, not terribly applicable in this case. Even if it had been tested regularly, it was cut around the same time the primary failed. Regular testing doesn't stop construction crews from cutting a line.

A more useful comment might have been to say something about the depth of the backup, having only one backup connection not being enough or something.

Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
> Do they seriously not have a backup connection?

That was the backup connection. The primary failed, and concurrently the backup had been severed by a construction crew.

Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
Fortunately the choice isn't between 1980s hardware with floppy disk requirements and rented servers in someone else's building. There are a lot of options in between those two.
Jtsummers··on US FAA says it will phase out floppy disks from air traffic control by 2028
I've worked on systems this decade that still relied on floppy disks. It's getting harder and harder for them to operate because the disks are getting rarer, as are the drives. And often the essential component is not the disk itself. There are some other computers in the system that cannot be upgraded to any more recent device, systems from the 1980s and 1990s that force them to continue using a floppy disk.

If it was just two modern computers and transferring data between them, the floppy disk would have been dropped in a heartbeat. But it's the old computers still in the critical path that are holding everything back. And those are really hard to maintain. Last year we phased out a system that was dependent on computers from the mid-1990s, getting replacement parts was practically impossible, they mostly just don't exist. You'd buy old systems off ebay and hope that they had enough working parts to scavenge, and the right working parts.

EDIT: Light editing because I had some extra words.

Jtsummers··on Infant Bleeding Surged as Parents Rejected Vitamin K Shots, Studies Find
"Advanced persistent threat": https://en.wikipedia.org/wiki/Advanced_persistent_threat
Jtsummers··on Infant Bleeding Surged as Parents Rejected Vitamin K Shots, Studies Find
They're a troll, read their comment history. There's nothing of value in any of their comments, and they only post so they can stir up controversy. Flag and move on, stop feeding them.
Jtsummers··on Newsom Commutes the Sentences of California's 548 Death Row Inmates
What's the confusion? The US is a federated system, each state is responsible for determining and carrying out the punishments for crimes committed within that state. Some offenses will result in federal charges, in addition to or instead of state-level charges, so those sentences and punishments would be handled by the federal government instead. Prisoners charged with crimes at the state level will typically be held in prisons in that state. Prisoners convicted of a capital offense (one that can result in the death penalty) in some state will be held in a prison in that state on "death row" (colloquial term meaning a section of a prison dedicated to inmates sentenced to death).

The quoted bit is just literally saying that CA has more death row inmates than other states. That doesn't mean they sentence more people to death, a large part of the reason is that they haven't carried out an execution in 20 years. So the numbers will continue to grow instead of shrinking until the inmates die from some other cause (age, illness, suicide, etc.). This is in contrast to states like FL and TX that continue to carry out death sentences, so the numbers will grow as people get sentenced but shrink faster than in CA as they carry out the executions.

Jtsummers··on California Gov Newsom commutes death sentences for all 548 people on death row
As a political move it's not a bad one. It saves the state money (eliminating the mandatory appeals), and it forces the pro-life fiscal conservatives who want to oppose it to explain that they're actually pro-death and in favor of spending more money, not less. Which puts them in a conundrum. Espouse their actual views and let their opponents use it against them, or stay quiet and irritate their constituents who want them to be in favor of executions (irrespective of the cost).

And all the while, it really changes nothing about the status quo of executions in the state, they weren't going to happen before this commutation, and they won't happen after.

Jtsummers··on YC's Garry Tan uses AI agents so much, he spends up to $200K a year on tokens
I'm sure one of his agents could answer that question if you posed it to them.
← PreviousPage 2 of 34Next →