Hacker Newsnew | past | comments | ask | show | jobs | submit | fanf2's commentslogin

(2019)

The IETF’s review has an amazing title “The Helminthiasis of the Internet”

https://www.rfc-editor.org/rfc/rfc1135


Before apt, the main user interface for dpkg was dselect, which was written in C++

But doesn't this actually strengthen my point? Debian transitioned from a tool written in a more demanding language to a tool written in a less-demanding one.

At a quick skim this looks like they reinvented something very similar to phkmalloc, but they didn’t cite phkmalloc nor include it in their benchmarks.

https://phk.freebsd.dk/sagas/phkmalloc/

https://cgit.freebsd.org/src/tree/lib/libc/stdlib/malloc.c?h...


Instead of representing atoms as string literals, you can represent them as global variables, eg

    const char conti[] = "conti";
Then you can use pointer comparison instead of strcmp().


You'll still probably need the `strcmp` because the pointers won't be the same unless you check for them and make them the same.

You may be thinking about how `eq?` (reference equality) works in scheme. That's usually done by hashing the identifier string. Which is the more general solution to this equality problem.


The atoms strcmp()ed by the interpreter are all created by the compiler so you can ensure the pointers are equal by construction.


You're right `virtmach` only works on things that are output from `compile` and maintaining the invariant that virtmach lisp uses those pointers isn't difficult to do in with how the evaluator is presented.

It gives virtmach lisp and scheme different ontology, but I can't think of any practical reason why that would matter other than it makes things a little bit more complicated. But, then again if I'm thinking practically scheme should be using hashed identifiers, and then there's no reason for them to have different ontology and conceptually we're right back where we started with virtmach lisp and scheme using identifiers as objects.


A newish solar farm near here has fixed panels https://maps.app.goo.gl/DCw7DfNb5bDTRu1E9


There are lots of news stories in recent years about Oracle sending lawyers round to demand money from orgs both commercial and non-commercial who are using Oracle Java SE, eg https://www.theregister.com/2025/05/09/users_advised_to_revi...


Auditing is only for paying customers of commercial products who've agreed to it in their contract. There's obviously nothing to pay for because the JDK is free. Oracle doesn't even collect contact information when downloading the JDK.

What could be happening there is that an Oracle customer uses their customer account to get access to a non-free JDK, such as an update that isn't offered on the public website because it's past the free update period.


Yeah, the longstanding hpn-ssh fork started off by adjusting ssh’s window sizes for long fat pipes.

https://github.com/rapier1/hpn-ssh


Yeah, there’s a replacement for scp that uses ssh for setup and QUIC for bulk data transfer, which is much faster over high-latency paths.

https://github.com/crazyscot/qcp


Reminds me of Teletext, the BBC Micro’s mode 7, and Bedstead https://bjh21.me.uk/bedstead/


Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: