lordmauve

18 ms·

1.
by lordmauve 4mo ago
We need to see DeepSWE scores. SWE Bench Pro is junk.
2.
by lordmauve 4mo ago
Where did you get that idea? It uses mini-swe-agent, same as SWE-Bench. https://github.com/datacurve-ai/deep-swe
3.
by lordmauve 4mo ago
I don't know if DeepSWE is genuinely a good benchmark. It's more important that their analysis demolished the validity of SWE-Bench Pro, objectively: it is being mismarked. I think that buys enough credibility to propose an alter
4.
by lordmauve 4mo ago
Given DeepSWE just blew apart the SWE-Bench Pro benchmark and handed a 14-point lead to GPT-5.5, it looks pretty bad that they've listed SWE-Bench first in the model release and no DeepSWE. Like, this isn't obviously an answer. Or
5.
by lordmauve 5mo ago
From a lot of game jam games with custom engines, I have a pretty standard approach: I set the game clock to zero timestep, and push a pause gamestate onto the gamestate stack, where top gamestate gets all input. The trick is that you need
6.
by lordmauve 9mo ago
Are you using Stop hooks to keep Claude running on a task until it completes, or is it doing that by itself?
7.
by lordmauve 1y ago
It's written by AI. I have made LLMs adopt the same tone by prompting to be engaging. Short sentences. Every point intended to land with impact. Artificial gravitas. I consider that a failed experiment and rewrote it, rather than posti
8.
by lordmauve 1y ago
Finally, the "AI" turns out to be 700 Indians. We now have the full loop of humans asking machines asking humans pretending to be machines. Civilisation collapses
9.
by lordmauve 2y ago
I'd argue this is the wrong design goal: correctness is more important. I'm in the UK but my work PC's Internet exit node is in New York due to enforced use of corporate proxies, so the time shown to me is 5 hours out. Javasc
10.
by lordmauve 2y ago
I don't think that is incompatible at all. It's a restatement of the same thing: If someone is right a lot they are likely not making assertions about things that they know nothing about. If they are making assertions about a doma
11.
by lordmauve 2y ago
> > Nobody in their right mind would want to learn programming in BCPL > I agree, but that's not what the author thinks: The author of this document is Martin Richards, the creator of BCPL. Of course he thinks you would want t
12.
by lordmauve 2y ago
What we call cornflour in British English, and cornstarch in US English, is flour heavily processed to remove everything except starch, giving a flavourless thickener. If you use wheat flour you will change the flavour and also add a slight
13.
by lordmauve 2y ago
Also pipx isn't reproducible - it re-resolves dependencies so you may end up with different versions over time or in different places, eventually causing something to break. If you have a shiv working it stays working, assuming you hav
14.
by lordmauve 2y ago
The finance industry mostly uses floats for currency, up until settlement etc. "What would I get for this share?" can be answered with a float. "What did I get for selling this share?" should probably be a fixed point va
15.
by lordmauve 2y ago
> However the following is fine: > > case $a in > > No field splitting occurs here This kind of bullshit is how I made a career rewriting people's buggy shell scripts in Python
16.
by lordmauve 2y ago
In reality it will just be a petrol car but they'll cheat on the emissions test.
17.
by lordmauve 2y ago
Good. As someone who is a paid up OpenAI user I absolutely don't agree that there should be a role for a team screaming to put the brakes on because of some nebulous, imagined "existential risk" of hypothetical future AGI. Th
18.
by lordmauve 2y ago
Or just keep options open for the time being.
19.
by lordmauve 2y ago
> Most popular solutions like Zip, gzip, tar, RAR, or 7-Zip are near or more than three decades old. If I can't extract .pack archives 3 decades from now, the use of SQLite 3 will be the reason.
20.
by lordmauve 3y ago
> I would be interested in examples of code that users believe require cancellation-specific async code, though. This happens all the time. For example, cancellation in the middle of sending a HTTP request. The connection is now unusable
21.
by lordmauve 3y ago
Python imports are much more principled than sourcing bash though. They are executed in a new namespace, and subsequent imports reference that namespace directly instead of re-evaluating the code. C extensions don't significantly chang
22.
by lordmauve 7y ago
If you cannot do that you have Aphantasia: https://en.wikipedia.org/wiki/Aphantasia
23.
by lordmauve 8y ago
Please don't use "word" to mean 16 bits. In an era when machine words are generally 64 bits, we're not talking about an anachronism from the previous generation, but the one before that - in a language that is completely
24.
by lordmauve 8y ago
Mu uses the QScintilla editor from https://pypi.org/project/QScintilla/ , which is GPLv3.
25.
by lordmauve 11y ago
I've heard that they will be available commercially next year through partners, but won't be distributed free to all students again.
26.
by lordmauve 12y ago
Don't make it a tuple. A tuple indicates heterogenous data (a record), while a list indicates a sequence of data. If nothing else, use lists because the syntax means it stays as a list even if you someday remove all but one option and
27.
by lordmauve 12y ago
Code review: default_reasons doesn't need to be a dict. Use a list. You don't need random.randrange(). Use random.randint() to select a random integer and random.choice() to select a reason from default_reasons.