@ltratt@mastodon.social
@ltratt@mastodon.social avatar

ltratt

@ltratt@mastodon.social

Shopify / Royal Academy of Engineering Research Chair in Language Engineering.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

ltratt, to random
@ltratt@mastodon.social avatar

One thing that surprises many folk is that pauseless GCs pause, but they don't pause where GCs traditionally pause: in other words, pauseless GCs are only pauseless if you consider one particular part, pausing in other parts. Pause to consider that. https://x.com/richardstartin/status/1796162611886653726

dotstdy, to random
@dotstdy@mastodon.social avatar

There's a bit of stuff in this article which phrases it in terms of changes over time, e.g. compute capability has grown and we no longer need big data. But it seems closer to reality that it was never required, and continues to not be required. (looking forward to the same style of post happening in a few years vis-a-vis microservices)

https://mastodon.social/@ltratt/112518285832831004

ltratt,
@ltratt@mastodon.social avatar

@pkhuong @dotstdy There are definitely people who have big data --- just not very many. Some years back I saw a well known big data (I think before that term became commonplace) company give a demo. After they completed someone in the audience sheepishly asked "maybe I didn't see correctly, but does the amount of data you showed us that fit on your laptop?" "um.... yes" "do you have any bigger data to show us?" "this is the biggest we've tested on".

ltratt, to random
@ltratt@mastodon.social avatar

One thing I think about a lot is "how much time should this organisation spend making tools to help with its main software tasks?" Experience has taught me that most invest far too little in this, but I've struggled to find a good way of defining what "too much" might look like.

ltratt, to random
@ltratt@mastodon.social avatar
ltratt, to random
@ltratt@mastodon.social avatar

Clausewitz's concept of friction applies to every area of life, including software. I could not agree more with Hillel https://www.hillelwayne.com/post/software-friction/

ltratt, to random
@ltratt@mastodon.social avatar

Probably the most surprising thing I've seen this week -- and I'm a few years late on this -- is that printf is Turing complete https://github.com/carlini/printf-tac-toe

ltratt, to random
@ltratt@mastodon.social avatar

There's nothing open-source maintainers like to hear more than a company saying "Fix our problems now because our paying customers are suffering". https://twitter.com/FFmpeg/status/1775178803129602500

pervognsen, to random
@pervognsen@mastodon.social avatar

I've been trying to watch Tour de France: Unchained on Netflix and it's very hard to suspend disbelief when the interviews are all filmed like reality TV where the interviewees are speaking in present tense about events as if they just happened hours before when the interviews are clearly filmed long after in bulk.

ltratt,
@ltratt@mastodon.social avatar

@pervognsen I haven't watched "Unchained", but I was highly sceptical of the "let's listen to the team talking to the rider thing" last year, and as the days went by the stilted "chatter" confirmed my scepticism. Then came Pogacar's "I'm dead. I'm gone" and my jaw hit the floor. Truly one of the most astonishing moments I've seen in sport.

ltratt, to random
@ltratt@mastodon.social avatar

An excellent debugging story from inside a garbage collector (always tough bugs!) from https://mastodon.social/@cfbolz https://www.pypy.org/posts/2024/03/fixing-bug-incremental-gc.html

ltratt,
@ltratt@mastodon.social avatar

I was of course amused by the use of multitime for "run a command for at most N times, stopping earlier it fails". This is such a frequent want of mine that I eventually packaged it up into try_repeat https://tratt.net/laurie/src/try_repeat/

ltratt, to random
@ltratt@mastodon.social avatar

An interesting economics-focussed take on open-source https://www.bloomberg.com/opinion/articles/2024-02-26/open-source-software-is-worth-a-lot-more-than-you-pay-for-it. One factor that I think is often over-looked is the extent to which open-source spreads learning. If you want to learn how to program complex systems, you now have many to study!

ltratt, to random
@ltratt@mastodon.social avatar

What books do you recommend that talk deeply about what software is? What makes it different or similar to other things; how we can think about it. Ideally something that starts from basic principles and builds upwards.

ltratt, to random
@ltratt@mastodon.social avatar

An interesting (and, in the best sense of the word, opinionated) article on trade-offs in ISA design, a topic about which I know nothing https://queue.acm.org/detail.cfm?id=3639445

ltratt, to random
@ltratt@mastodon.social avatar
ltratt, to random
@ltratt@mastodon.social avatar

The UK is having one of those annoying combinations of cold weather + renewables in the doldrums. Not everyone is in a position to shift their electricity usage but for those who are I can reoffer my "Scheduling my Electricity Usage" post https://tratt.net/laurie/blog/2023/scheduling_my_electricity_usage.html

ltratt, to random
@ltratt@mastodon.social avatar
patina, to random

In early versions of Rust, types with destructors had an extra bool added to their definition: the drop flag.

Its value dynamically tracked whether the destructor should be run for "maybe/partially initialized" values. Any time a value was moved, Rust would zero out the old location to clear all drop flags.

This was very inefficient, so the design was replaced with implicit local variables, removing support for more complex dynamic initialization of things like arrays.

https://rust-lang.github.io/rfcs/0320-nonzeroing-dynamic-drop.html

ltratt,
@ltratt@mastodon.social avatar

@patina IIUC Rust still has drop flags for some use cases but they're stored on the stack not in the type (https://doc.rust-lang.org/nomicon/drop-flags.html). I must admit that I was rather surprised recently when I mused on some minor Rust semantic issue and was pointed at this page!

ltratt, to random
@ltratt@mastodon.social avatar

I didn't remember this paper from a decade ago that shows that x86 is Turing complete just via (carefully constructed) page fault sequences -- well worth a read if, like me, you don't know it. [Thanks to Stephen Dolan for pointing me at it!] https://www.usenix.org/system/files/conference/woot13/woot13-bangert.pdf

ltratt, to random
@ltratt@mastodon.social avatar

One day the world will believe that programming languages are not intrinsically "compiled" or "interpreted". Until then, we must fight the good fight -- this time Eddie is doing his bit by showing that Python isn't an "interpreted language" https://eddieantonio.ca/blog/2023/10/25/python-is-a-compiled-language/

ltratt, to random
@ltratt@mastodon.social avatar

It seems remarkably hard to find a good USB hub -- I've now seen 4 in a row (different manufacturers!) that have occasional, mostly minor, issues, typically around disconnections. Is there actually a good USB hub (not worrying too much about speed etc)?

ltratt, to random
@ltratt@mastodon.social avatar

10 days left for you to express interest in attending PLISS (Programming Language Implementation Summer School)! https://pliss.org/2023/

ltratt, to random
@ltratt@mastodon.social avatar

Statically compiling Python is harder than it looks, even with static types https://bernsteinbear.com/blog/typed-python/

ltratt, to random
@ltratt@mastodon.social avatar

In my "Why Split Lexing and Parsing In Two" post, I mentioned in a footnote a 2008 interview I'd done with the late Eelco Visser on parsing. Listening back to it highlighted to me his energy and knowledge, but audio-only isn't for everyone -- so I've produced a transcript at https://tratt.net/laurie/blog/2023/my_interview_with_eelco_visser_with_on_parsing.html.

I also included a little bit of background on the interview -- neither Eelco or I really knew what we were doing, but by the end of the interview I think we'd hit our stride.

ltratt, to random
@ltratt@mastodon.social avatar
  • All
  • Subscribed
  • Moderated
  • Favorites
  • JUstTest
  • ngwrru68w68
  • everett
  • InstantRegret
  • magazineikmin
  • thenastyranch
  • rosin
  • GTA5RPClips
  • Durango
  • Youngstown
  • slotface
  • khanakhh
  • kavyap
  • DreamBathrooms
  • provamag3
  • tacticalgear
  • osvaldo12
  • tester
  • cubers
  • cisconetworking
  • mdbf
  • ethstaker
  • modclub
  • Leos
  • anitta
  • normalnudes
  • megavids
  • lostlight
  • All magazines