miniblog.

Playing with perf today! It's really interesting to see low-level details of where compute time is going. Branch prediction works well most of the time! (At least for this workload.) Based on https://jvns.ca/blog/2014/05/13/profiling-with-perf/ and https://users.rust-lang.org/t/profiling-in-rust-application/18195/2

Playing with perf today! It's really interesting to see low-level details of where compute time is going. Branch prediction works well most of the time! (At least for this workload.) Based on https://jvns.ca/blog/2014/05/13/profiling-with-perf/ and

The dominance of statistical models in AI, our bias towards embedding human knowledge, and the effectiveness of large, generic compute: https://www.incompleteideas.net/IncIdeas/BitterLesson.html

@cwebber@octodon.social Yep, ever since I read 'Intel x86 considered harmful' I've been amazed by how many CPUs exist inside a modern computer. Any could have bugs. There have been some verification efforts for RISC-V, and the ISA seems to have a bright future, so I'm hopeful there. I hope it's economically viable though. I loved the OpenMoko project but it wasn't commercially successful. I understand that companies wanting strong guarantees do custom, verified, single-purpose hardware (e.g. with Galois).

@kensanata@octodon.social Wow! Thanks, I will bear that in mind. (This is a problem that's been solved in newer IRC competitors I suppose.)

@cstanhope In theory it's not giving them much more data: I already have to give the company meter readings. https://www.telegraph.co.uk/money/consumer-affairs/six-reasons-say-no-smart-meter/ and https://www.the-ambient.com/guides/smart-meters-uk-guide-418 persuaded me against getting a smart meter though. The former link suggests I won't save money or escape confusing bills, and the latter suggests the early standards haven't won. It'll probably win eventually, for better or worse.

@srol Ooh, excellent example! I think that the reuse is a little friendlier than "copy your friend's CSS", so there are some good themes out there.

Other programmers using your code is often a high compliment. Making something from scratch is fun, but learning someone else's API is work. Rather than the common "I built one, you should use it", I try to ask "what would persuade you to use this?".

Scarcity versus abundance mindsets in software design, and shipping rough prototypes early: https://breakingsmart.com/en/season-1/rough-consensus-and-maximal-interestingness/

What's the threshold for politely pasting text in IRC channels? I pasted two lines and my client sternly warned me and mentioned pastebins.

An Experimental Evaluation of the Assumption of Independence in Multi-Version Programming, by John Knight and Nancy Leveson https://sunnyday.mit.edu/papers/nver-tse.pdf If you take bunch of programs written independently, are you more likely to reduce bugs by taking the most common output?

Excellent discussion of a Unison meetup, discussing their design (globally consistent content-addressed codebases!), type system, and tooling: https://unisonweb.org/2019-04-04/first-meetup.html#post-start It's really impressive how much they've achieved.

Friday's xkcd is blunt but fair: it's very hard to secure any part of a modern computer stack, and we depend on all of it: https://m.xkcd.com/2166/

Apparently you can buy smart power sockets now, e.g. https://www.amazon.co.uk/Lightwave-L42-Smart-Socket-Stainess/dp/B075XT62RC Whilst a consumer might enjoy the remote control, these devices also monitor power consumption. I imagine this would be particularly useful in a commercial property context.

Different computing platforms have different notions of 'rich text pasting'. Pasting styled text from a browser to a document editor is fraught with all sorts of compatibility challenges. Is it worth it? Text formatting is very complex and representations depend on application.

The US Navy is exploring ships that are more highly automated. The crew is smaller, and their roles are generalist problem solvers. https://www.theatlantic.com/magazine/archive/2019/07/future-of-work-expertise-navy/590647/

Emacs trunk is now using HarfBuzz for text rendering! https://lists.gnu.org/archive/html/emacs-devel/2019-06/msg00123.html

How good are your error messages? Wouldn't it be nice if there was a tool that told you if your error messages were unhelpful? Proactive Detection of Inadequate Diagnostic Messages for Software Configuration Errors https://homes.cs.washington.edu/~mernst/pubs/inadequate-diagnostics-issta2015.pdf

Today I learnt that Pharo has integrated history! This isn't a VCS, this is just a convenient way of reverting changes at the granularity of a method ('message' in ST terminology). It's a good level of granularity (cf file-level undo in a text editor).

I used to know people who sold thin client devices rather rather traditional PCs. The thin client model didn't gain much traction. Looking at it today, were they ahead the time?