Roger Peppe's Avatar

Roger Peppe

@rog

CUE, Go, fiddler, climber, gardener, curry maker.

1,358
Followers
159
Following
387
Posts
25.04.2023
Joined
Posts Following

Latest posts by Roger Peppe @rog

sf hard enough to move the markets. is that a first?

27.02.2026 17:44 ๐Ÿ‘ 0 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

tech debt never gets cleared. we'll have to declare bankruptcy.

24.02.2026 09:14 ๐Ÿ‘ 0 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

5 authors I've read at least 5 books by:

E J Swift
Adam Roberts
Alasdair Reynolds
Christopher Priest
Adrian Tchaikovsky

22.02.2026 01:47 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 1

I just read that text and the rhythm of the prose instantly says "AI slop" to me and I ignore it. Sorry.

16.02.2026 21:54 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

I found it one of his most accessible - I liked it!

16.02.2026 10:28 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

If only the wi-fi had worked and the bluesky app had notifications, I'd have seen your recommendations :( oh well... I'll check the slides out later!

02.02.2026 07:09 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

Any talks you'd recommend going along to?

01.02.2026 08:51 ๐Ÿ‘ 0 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

see you there!

30.01.2026 15:53 ๐Ÿ‘ 0 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

Anyone I know gonna be at #FOSDEM or #cfgmgmtcamp ?

30.01.2026 11:20 ๐Ÿ‘ 8 ๐Ÿ” 0 ๐Ÿ’ฌ 2 ๐Ÿ“Œ 0
Preview
proposal: net/http/httptest: synctest support ยท Issue #76608 ยท golang/go Proposal Details I use the httptest package a lot, and in particular NewServer rather than ResponseRecorder, because it picks up subtle real-world issues that don't appear when not using the entire...

Nice! Maybe I stimulated that by creating github.com/golang/go/is...

29.01.2026 11:15 ๐Ÿ‘ 8 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

Lots of my friends there... :) Enjoy!

28.01.2026 10:28 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

Went to order this and found I've had it on pre-order since July 2024! Only another couple of months now... can't wait!

28.01.2026 10:27 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

good luck!

26.01.2026 22:34 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0
Preview
Proposal: Generic Methods for Go ยท Issue #77273 ยท golang/go Proposal: Generic Methods for Go A change of view. Background For clarity, in the following we use the term concrete method (or just method when the context is clear) to describe a non-interface me...

Interesting expansion of generics in #golang ๐Ÿ‘€

23.01.2026 00:35 ๐Ÿ‘ 36 ๐Ÿ” 8 ๐Ÿ’ฌ 3 ๐Ÿ“Œ 0

Perfect! Exactly the design choices I was after. Thanks for mentioning it. ๐Ÿ‘Œ

23.01.2026 18:18 ๐Ÿ‘ 3 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

I remember enjoying Air, but can't remember the first thing about it!

23.01.2026 11:05 ๐Ÿ‘ 0 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

I'd like to know more about this risk, as I'd assumed the risk was similar to visiting an arbitrary URL or clicking on a link on a website you don't necessarily trust (something that's pretty common to do and hard to avoid tbh). Are there security risks that are specific to QR codes themselves?

21.01.2026 16:15 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 3 ๐Ÿ“Œ 0

Still @nickharkaway.com's best IMHO. Although I never re-read it, so it's possible I'll be disappointed when (not if) I do ...

19.01.2026 20:30 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

I actually genuinely went back to the beginning immediately after finishing and started again (didn't finish again mind!). Another book this brought to mind: Gone Away World.

19.01.2026 19:44 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

Worth saying! I have seen so many utterly worthless, nay negative-worth, unit tests in the past. And testscript itself is still so nice to use - we use it a lot.

16.01.2026 08:50 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

If the local indexes are all within ~/.claude, the total size of that directory (as reported by `du -s`) is just 55MB.

14.01.2026 15:03 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

I have no idea! Are they all loaded into memory? BTW that memory size is _resident_ memory. Virtual memory use is 77GB!

14.01.2026 15:02 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

I don't understand why Claude is such a resource hog when it's not actually doing that much locally. For example, it's currently idle on my machine and it's coming up as using just under 500MB of memory. WTF?!

14.01.2026 14:59 ๐Ÿ‘ 3 ๐Ÿ” 0 ๐Ÿ’ฌ 2 ๐Ÿ“Œ 0

Nice! I've implemented your "naive approach" in the past but this is much nicer.

13.01.2026 13:19 ๐Ÿ‘ 2 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

great idea. joined!

09.01.2026 15:41 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

"Is it too much to expect that a hotel should provide an easily accessible mains power socket within a relatively short distance of the bed?", I think as I delve under my bedside table for the 5th night in a row. #firstworldproblems

09.01.2026 00:53 ๐Ÿ‘ 2 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

Good info, thanks! It would definitely be nice to have a cheat-sheet for this kind of optimisation. Maybe restricted to optimisations that reduce allocations and/or big-O behaviour.

08.01.2026 17:43 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0

... because naively that's what I'd expect to happen. I wonder if there's a comprehensive list of optimisations like this that we can refer to.

08.01.2026 17:25 ๐Ÿ‘ 0 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

This idiom is interesting. I presume there's some compiler optimisation that says that `append([]byte(nil), make([]byte, n)...)` does not _actually_ copy n bytes from a newly allocated empty slice?

08.01.2026 17:24 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 1 ๐Ÿ“Œ 0

my main reason is unfortunately a "by design" issue that will never be fixed AIUI: no cursor warping. Without that, the editor I use is substantially broken.

04.01.2026 16:42 ๐Ÿ‘ 1 ๐Ÿ” 0 ๐Ÿ’ฌ 0 ๐Ÿ“Œ 0