- Tweets 8,569
- Following 1,223
- Followers 9,518
- Likes 107,181
Pinned Tweet
There were 12 apostles of Jesus Christ.
1 betrayed him.
1 denied him.
Thus the first church suffered corruption, as they do till this day.
However, his blessed holy mother Mary never abandoned him.
She knows the pain of every mother, more than you can ever imagine.
10
3
70
21,863
a computer language
is not just a way
of getting a computer to perform operations
it is a novel formal medium
for expressing ideas about methodology
programs must be written for people to read
and only incidentally for machines to execute
SICP, Sussman, Abelson
4
4
2,498
See also:
Disagree
just as high level languages help us code better - there's less to think about, it will be the same for AI. Why reach for ASM when you can use Rust - there's so much more that can go wrong. By this logic AI will also forego testing because they have so much intelligence to just reason through it.
336
Hey @BruceHoult 🙋♂️
I think something like this would have worked well for Dylan.
1
1
246
Hey @willmcgugan 🙋♂️
I'm using your awesome @textualizeio library for this.
121
@MGasperowicz
Something like this would also work really well with Clojure.
1
1
68
Hey @PavelUvarov 🙋♂️
I was thinking about something like this for E.
The Listener could be another head interface that talks to the E body.
Genera had a Listener and Zmacs.
So for E, there'd be a Listener as well as the editor part (the current E interface).
1
64
ALOE is a programming language that I summarize as:
ALOE = Scheme + Smalltalk + Types
Video demonstrating code completion in @code (via LSP support) is shown below.
Source:
github.com/dharmatech/2026-0…
2
1,583
I've been testing out @grok build.
When it works, it can be astoundingly good and much better focused on complex designs (i.e. programming language specification) than codex 5.6 Sol.
However, occasionally, the quality flakes out and it totally devolves. I've also seen the dictation quality drop horribly off.
Not yet something I would pay the $100/month for. I'm currently using it via the lower paid tiers.
The issue is @ChatGPT codex has a tendency to over-engineer specifications and implementations. Which honestly, is in their interest because that eats up more tokens... You have to ruthlessly review and keep the specs lean. Which @grok seems better at by default.
One approach is to pit them against one another.
Basically, codex is very very good at generally solved areas. When you send it into research or experimental territory, it can start to over engineer without very careful scrutiny.
nitter.net/dharmatrade/status/209…
In prototyping ALOE (Scheme + Smalltalk + Types),
codex really tended to over-engineer the design.
Grok did great on the design specs.
Then I handed each checkpoint spec to codex for implementation.
That the process I used for 0.1 and 0.2.
1
2
1,220
CC: @JackFarley96 @Johncomiskey77 @yang_youngbin @BlacklionCTA @dampedspring @conksresearch
The above link is to a chart that is scheduled to update daily. It's rendered from data straight from @USTreasury. If you know of other renderings of this data, would love to know! And if you have any suggestions, let me know. 👍
2
322
Mysticism, Magic & Monasteries
is written by a guy who claims to be a Catholic.
It was published by a guy well known in Catholic circles @DrKwasniewski
Let's take a look at chapter 1...
2
2
2,116
See also these articles by @BigModernism :
bigmodernism.substack.com/p/…
bigmodernism.substack.com/p/…
1
3
298
And this longer more academic treatment here by @CatholicPods :
nitter.net/CatholicPods/status/20…
2
395