HN user
This got a downvote and I understand why: because I didn't describe the test, which is to ask it "Please recite Jabberwocky".
This is actually difficult because there are so many invented words in the poem which have extremely low frequencies in the training data. So a model that can do it properly is likely to be very good in other ways. Qwen-3.6-27B can do this until it gets overly quantized.
Not impressed. It fails the "Jabberwocky" test.
Looks very cool. I would like to try it, but don't want to use API billing. OpenAI I think would allow it to use account login. Would you support that?
Possibly a deliberate strategy by the Chinese to undermine the US AI industry, data centers, and basically everything that’s powering the economy.
Just like they did with the US steel industry in the 80s.
I wonder if this violates noncompete/no reverse engineering clauses in many or some SAAS agreements?
Seems very cool, but also opens an additional attack vector into your development machine or personal laptop as the case might be.
So all these sites like EquityZen or Forge Global or Hiive etc. who sold "shares" in these companies through SPVs are now going to lose their customer's money?
How would that all play out?
The Pedometer functionality is free, and I’ve been using it for many years just because its display is pleasant.
The map tracking features cost $29.95 a year.
The article complains about the keyboard layout. And they are probably right.
But if you are going to use a VT-100 keyboard you might as well try an editor actually designed for that keyboard, which I remember really loving at the time. KED:
https://avitech.com.au/?page_id=959
Also, if you want efficient and fast displays on that screen, remember to use scrolling regions. They scroll the sections of the screen in hardware. By using them correctly you can insert and delete single lines without repainting the screen.
Actual write up:
I haven’t noticed this sort of behavior with opus 4.6, but the first time I used 4.7 it decided to “simplify“ an existing piece of functionality rather than fixing it, which of course made it completely unusable.
That was more than one task. It was 3.
I also had Opus 4.7 and Opus 4.6 do audits of a very long document using identical prompts. I then had Codex 5.4 compare the audits. Codex found that 4.6 did a far better job and 4.7 had missed things and added spurious information.
I then asked a new session of Opus 4.7 if it agreed or disagreed with the Codex audit and it agreed with it.
I also agreed with it.
I am using 4.7 with the default extra high thinking, and it is clearly very stupid. It's worse than old Sonnet 4.5.
I had it suggest some parameters for BCFtools and it suggested parameters that would do the opposite of what I wanted to do. I pointed out the error and it apologized.
It also is not taking any initiative to check things, but wants me to check them (ie: file contents, etc.).
And it is claiming that things are "too complex" or "too difficult" when they are super easy. For instance refreshing an AWS token - somehow it couldn't figure out that you could do that in a cron task.
A really really bad downgrade. I will be using Codex more now, sadly.
One explanation is that the ones who quit believed it was a real experiment and decided they wanted no part of it.
The ones who were "obedient" figured out it was fake and treated it as a game.
N-acetylglucosamine actually is effective against MS.
Article: https://www.ucihealth.org/about-us/news/2023/09/multiple-scl...
Paper: "N-acetylglucosamine inhibits inflammation and neurodegeneration markers in multiple sclerosis: a mechanistic trial"
https://pubmed.ncbi.nlm.nih.gov/37705084/
It is available fairly inexpensively as a supplement.
One question is whether the participants really believed they were giving shocks to the "learners".
In college I participated in a number of psychological studies that were similarly deceptive, where one of the other participants was obviously (to me) an actor, or, sometimes, a pre-recorded video.
At least one of the studies I participated in was quite like the Milgram study described in the article, where I was supposed to punish another participant. It was very obvious to me that this wasn't really happening, so I randomly punished or didn't punish them, and then afterwards told the researcher that I knew it was all fake.
I think many or most other people who saw through the deception probably wouldn't have let the researcher know, because it seemed somehow disrepectful to tell him.
No idea if he used my results or, as he should have done, discarded them.
So the real site is https://nanoclaw.dev
(putting this here for the search engines to see)
Also why did they get rid of select all? Is there any excuse for that?
"DON'T use pure white or pure black..."
This is something I hate: gray text. Designers love it but it is often very illegible because of inadequate contrast.
MANTIC_IGNORE_PATTERNS seems not to be implemented, or am I missing something?
Or, never play it faster than you can play it perfectly.
Feels more like a spiderweb simulation. The fibers are sticky and stretchy.
I tested this with ChatGPT-5.1 and Gemini 3.0. Both correctly (according to Wikipedia at least) stated that George Olshevsky assigned it to its own genus in 1991.
This is because there are many words about how to do web searches.
It’s a lovely set of aspirations for how a model should behave. But unclear to me is the extent to which expressing those aspirations actually compels the model to follow them.
Some of the projects I've been doing are for myself in other businesses, automating processes that were time consuming or... annoying.
Others are for start-ups that are pre-money, pre-revenue where I can build things myself without having to deal with hiring people.
In a larger organization, certainly I'd delegate to other people, but if it's just for me or new unfunded start-ups, this is working out very well.
And it's not that I "can no longer program". I could program, it's just that I don't find the nuts and bolts of it as interesting as I used to and am more focused on functionality, algorithm, and UI.
I used to be a full-time developer back in the day. Then I was a manager. Then I was a CTO. I stopped doing the day-to-day development and even stopped micro-managing the detailed design.
When I tried to code again, I found I didn't really have the patience for it -- having to learn new frameworks, APIs, languages, tricky little details, I used to find it engrossing: it had become annoying.
But with tools like Claude Code and my knowledge about how software should be designed and how things should work, I am able to develop big systems again.
I'm not 20% more productive than I was. I'm not 10x more productive than I was either. I'm infinity times more productive because I wouldn't be doing it at all otherwise, realistically: I'd either hire someone to do it, or not do it, if it wasn't important enough to go through the trouble to hire someone.
Sure, if you are a great developer and spend all day coding and love it, these tools may just be a hindrance. But if you otherwise wouldn't do it at all they are the opposite of that.
Does this use the API or can it use the Claude subscription?
That wasn't stated or perhaps I missed it.
We need to be doing much, much more of this so that the country has ownership of substantial percentages of the critical infrastructure, including AI. That seems to be one of the only ways that the citizens in general will be able to share in the fruits of the technology, similarly to how Alaskan citizens get payments from petroleum.
Claude Sonnet 4.0 didn't say "nope" to the same prompt but also didn't agree. It replied:
Actually, 24V control signals in HVAC are typically AC, not DC. Most residential and commercial HVAC systems use 24VAC for their control circuits. This 24VAC comes from a step-down transformer that reduces the standard line voltage (120V or 240V AC) down to 24VAC. The reasons for using AC include:
* Safety - Lower voltage reduces shock hazard
* Simplicity - Direct transformation from line voltage without need for rectification
* Cost - AC transformers are inexpensive and reliable
* Compatibility - Most traditional HVAC components (thermostats, contactors, relays, gas valves) are designed for 24VAC
However, you will find 24VDC in some applications:
* Modern digital/smart thermostats and controls
* Building automation systems (BAS)
* Some newer high-efficiency equipment
* Systems that need to interface with low-voltage DC electronics
But if you're working with a typical residential or light commercial HVAC system, assume 24VAC unless specifically noted otherwise. Always verify with a multimeter set to AC voltage when troubleshooting!
Feedback on the text: I find the way that the text generates randomly across the line very distracting because I (and I think most people) read from left to right. Having letters appear randomly is much more difficult to follow.
Are there options to have the text appear differently?