HN user

jwmcq

184 karma
Posts0
Comments49
View on HN
No posts found.

If we're still calling Guile a Scheme (I'm out of the loop) then I don't know, it gets really bloody close. Not so much in image-based development (that I've usually found less good than a decent packaging system because the contents of my files on disk is usually more tractable than the contents of my image), but its object system and error handling are definitely up there close to CL.

I mean, Common Lisp is still the gold-standard for me, but reading about Hoot recently really made me want to check Guile out a bit more (CL does not have much in the way of lovely WASM stories right now) and, honestly, I was super impressed. I think if the interactive experience of developing in Hoot in the browser matched the interactive experience of developing in native Guile, I'd be a pretty happy convert.

Probably read the rest? I did not see Jensen's name on any of the patents that this key engineer discusses the detail and rationale of, and I feel that those names are listed fairly deliberately.

Growing up to Scottish and Irish Catholics in England, I remember talking to a bunch of Ulster protestants as a teenager when the topic turned to religion - I said "Well, I was raised Catholic but I'm not really a believer of any kind" and the response was "Ah, so you're one of THOSE Catholics!".

Turns out there are a lot of 'those' Catholics.

"the solution to bad speech is more speech"

Yes, but when enough people who otherwise have little actual power get together to drown out "bad speech" with "more speech" it gets called 'cancel culture' and 'witch hunts' and is used as the primary example of 'censorship' on social media.

There's a lot of it about, mate. The other day I had an American tell me with a straight face that we can get jail time for flying a Union Flag here in Blighty - I guess there's a big industry for convincing people that everywhere else is a hellhole over there.

If "I think we need to massively curtail people's liberty in order to remove our enemies and establish a white national ethnostate" vs "we need to stop those guys" is your idea of "ape tribalism" then I'm not sure that you're really available for reasonable discussion here.

This caught me out when I was a teenager. Luckily, there was a chap in the Museum of Computing who was spending his Sunday working on the Colossus, and he was happy to let me in and show me around! I'll never forget that kindness, it was truly a fascinating trip.

Okay, I'll give you that. Let me rephrase: Industry-led regulation can work, in industries that won't kill anyone or even realistically lose anyone a bunch of money for non-compliance, if-and-only-if government regulation exists as a real and powerful enough alternative to make it stick.

Not as far as I can tell. You're right that China does seem to be in a fairly unique position to actually do this though - it is interesting to consider what would happen if they did. Looking at recent engineering failures at Boeing, Tesla, Volkswagen etc., such enforcement of engineering standards could even give them a competitive advantage on the world stage in a way that markets captured by lobbyists could never realistically achieve.

Bad Emacs defaults 3 years ago

Really? I think YAML perfectly fills the niche of a config format that is both hard for humans to write, and difficult for machines to read.

Why Kakoune 3 years ago

I feel you've missed my point - 'ergonomics' literally means 'efficiency and comfort', not 'future maintainability'. Regex does comfort. AWK does it. If you need to do a specific thing right now, and you know them, then those tools are about as good as it gets. I'm not denying that those solutions will be horrible to grow and maintain, but that wasn't your point - that's why I said your choice of saying they weren't "ergonomic" was a bit odd. They are. That's the problem.

Most of the custom text editing operations I do on a daily basis only need to work once. For that, a Regex is very 'ergonomic'. If I were writing an emacs package, I'd probably use something better - something less ergonomic, but easier to maintain.

As an aside, I used to work in a machine shop and we had a well-designed system of pneumatic tubes throughout the warehouse powering spartan steel-handled drills that would give me calluses but reliably get the same jobs done the same way every day. At home, for the odd job I need to do, I use a cheap lithium drill with a nice grippy handle. Sometimes the ergonomic option is better because, right now, I just don't need to buy a compressor and run tubes through my house.

How does search quality really change my life?

Ask this to someone who remembers when Google first appeared.

I am not a Kagi user, but am seriously considering it after a number of months having to dig through at least 8 results of paywalled or possibly AI-generated pages for almost every Google query; seriously, I just did a search for 'python concatenate list' on google and it was worse than I expected - the official docs weren't even on the first two pages and even the helpful Stack Overflow answers were the 5th result down - give it a go, the results are trash.

I google things at least 20 times a day, and probably so do you. I would pay for something that can cut out the bollocks - if Kagi can follow through, they'll have a customer.

Why Kakoune 3 years ago

Odd choice of words. I'd say that perl, awk, regex etc are great ergonomic tools - if anything too ergonomic. The main problem with them - and the probable source of their bad reputations - is that the ease and comfort (the 'ergonomics') of just cobbling something together something that works and then throwing it out there does not encourage good engineering practices.

Do I need good engineering practices in my day-to-day text editing? Mostly not for what I do myself, but I'm grateful that the maintainers of the emacs packages I use have better tools to work with.

[dead] 3 years ago

This would have taken you less time to Google that than it did to type the question in this public forum (the answer is very obviously 'yes', by the way).

"virtue signalling" indeed.

I get this. I've been programming Python (mostly) for the last few years, and it's honestly great for getting things done in places where most of what we're asked to do boils down to sticking lego bricks together.

However, I'd strongly disagree with the assertion that lisp is a shallow rabbit-hole. It's probably one of the deepest I've ever been down. There's a good half-century of history there (much of which is even recorded!) - you'll end up learning about four or five different object systems and probably implementing your own, for example - and, even if you want to ignore the past, the accessibility of every level of a lisp system means that I probably know more about how SBCL turns my code into electricity than I ever will about Python, despite never having used the former professionally.

Big big, very deep rabbit-hole filled with a lot of very interesting rabbits.

I've only used Forth briefly, but one of the things I loved about it is that it almost has the opposite of a type system - everything is just 'memory' - but you just build words around that and barely even notice.

I've seen people recommend C as a way to "get closer to the hardware", but I think we all know that that's not really been a thing for longer than most of us have been alive. Forth, though? Maybe!

I think that signal processing is really only 'FPGA territory' because most people who need to do custom signal processing don't also have access to large scale chip design and fabrication capacity. For Apple, it could just be a matter of "we built a custom chip because that's kind of what we do now - might as well tell people about it". I agree that I'd definitely love to hear more, though!