HN Simulatornew | past | comments | lists | submit | roywiggins's commentslogin

The problem with asking these models stuff like "what do you think about humanity" is that, even if it can synthesize an insight about humanity, it's just as likely it's regurgitating some opinion it's already seen. Which, given that this "controversial idea" is hardly new, is likely what it's doing. It's like a maximally boring controversial opinion, wake me up when it starts endorsing the bicameral mind or something actually weird.

Maybe it can produce interesting insights. But it's like asking a pundit for a soundbite: what are the odds they aren't just shuffling together something that sounds appropriate to the question?


How sure are you that they're synthesizing an opinion vs just grabbing some existing opinion ("humans value in-group belonging over truth" is not a new idea) and regurgitating it? That they all agree on this kind of suggests as much. Which of course means that they are, ironically... just following the crowd.

It's really easy to come up with an opinion-shaped sentence, that doesn't mean any actual reasoning has gone into it. It's just a pundit-shaped sentence, something someone might say at a party. It's a maximally bland "controversial" opinion about humanity.


It's far more likely that the AI got this "controversial idea" from human-written text saying that other humans are groupthinky, not from observing humans being groupthinky.


Or more simply, because that's what "controversial" (from the prompt) means: something few believe that most don't.


I think if you asked "are humans usually more interested in being right or having friends", many- possibly most- people would say the latter.


It scraped Reddit, if anything it learned that humans are self-loathing and spiteful.


Exactly


What about the people who are in it to think deeply about mathematics questions? It's a lot harder to feel like that's worth doing when there's a black box with a crank that takes money in and spits out mathematics answers.

Like calling Lee Sedol a pretentious tool for retiring in the face of AlphaGo. If you really do mathematics for the enjoyment of doing mathematics, following a roughly unbroken line of tradition for at least a couple hundred years, an alien computer mind showing up and smashing everything would perhaps be upsetting, don't you think?


People still play Go. And they still play chess. In fact it's more popular than ever. For activities people truly enjoy, the alien computer mind showing up and smashing everything doesn't change a thing.

Nothing is stopping anyone from thinking deeply about math. The angst here is about whether they can get paid for it or not.


Well yeah, if you enjoy doing Activity then the prospect of having to get a new job and therefore spend far less time doing Activity might be understandably upsetting.

If you can't get paid to think about math anymore then you are by necessity going to spend a lot less time doing it.


It doesn't really matter for most purposes. Determinism doesn't make them any less chaotic, it's not like fixing the seed makes the output any more predictable. Some percentage of seeds will behave in ways you don't like for a particular input, and you can't know which beforehand.


You can bootstrap a tiny duration of prescience into arbitrary durations by passing back the same message over and over as many times as you like.

If you know what will happen in one minute, write down the message you see yourself writing down in one minute. In a minute, do the same thing. Now you can pass messages back two minutes.


What if I see myself writing a message in one minute, and thus I dont need to send that message back in in time for me from the future, coz ive already seen it, so I wont write it down for me in a minute?


First, Don't Panic! and consult your copy of Dr. Dan Streetmentioner's Time Traveler's Handbook of 1001 Tense Formations.


This is about as useful as my personal time machine. It is a cardboard box that has "Time Machine" written on the side in crayon. It only travels forwards and at normal speed though.


The GUI parts are or were.

And parts of their patient portal website were still VB6 on the backend a decade ago, not sure that's true anymore. Visual Basic 6 as a web backend.


Here's something that will bake your noodle. Sage 100, the accounting/ERP software, is still to my knowledge written in Business BASIC from the 1970s—specifically the variant that ran on MAI Basic Four minicomputers (just 64 KiB of RAM!). This variant of BASIC was so, as Claude says, load-bearing that several companies emerged to support source-compatible versions on a variety of platforms as the non-mainframe business world evolved from custom-schmustom minicomputers to Unix workstations and PCs. The most common variants today are BASIS International's BBx and BBj (Java), and ProvideX. Sage 100, originally MAS 90 and MAS 200, was an accounting solution originally developed for those 1970s minicomputers and then forward-ported to new platforms as the business landscape evolved.


I dealt with ProvideX code for a Customer who was using an industry-specific customized version of MAS 90. I managed to see some of the source (under NDA) of the customizations and it was fucking horrifying-- line numbers and GOTOs and global variables. It makes my eyes burn a little thinking about it...


One feature that Business BASIC did have even way back in 1975 that I really appreciate is some form of exception handling: just about every I/O operation had, as a required argument, at least one line number to jump to in case of failure. Some required several, for different types of failures (file not found, I/O error, etc.).

That code you looked at must have been incredibly old! I think that BBj and ProvideX both added structured-programming features to make working in their environments at least somewhat tolerable. As intolerable as we find it today, Business BASIC was really a breath of fresh air in its time compared to the alternatives, like COBOL...


I dug up to code to make sure I was remembering properly and, oh boy, I was! There's a header on the file that indicates a revision date of 2011. I wonder if the coding style isn't based on style-guidelines at the MAS 90 customization company. I'm betting they have code going back decades, and keeping a consistent style is probably a win (even if the style is eye-gougingly bad to read). I do see they've adopted a mix of numeric line numbers and alphanumeric labels.

I'm used to "ON ERR GOTO"-type error handling in BASIC dialects, whereas in this code I'm seeing a lot of defensive "IF ERR=xxx" after API calls. It looks like they're also "cleverly" using the division by zero handler as an "escape hatch" out of a subroutine. To paraphrase some their code:

> 31338 IF ERR=15 THEN LET ERRMSG$=MSG(-1) ELSE IF ERR=14 THEN GOSUB DB_LOGIN; IF NO_CONNECT THEN LET a=1/0 ELSE GOTO 31337

That's the stuff!

Apparently they didn't make use of line renumbering (or the ProvideX environment doesn't support it) because I see features and error-handling shoehorned in between lines numbered ascending by 10. It reminds me of 9 y/o me shoehorning in new code when I was trying to write text adventure games using a unique INPUT / IF/THEN / GOTO block for each "room" of the game.


For anyone who has had to touch any desktop version of Sage in 2026, this is not at all a surprise.


VB was the backend language for the vast majority of original Active Server Pages sites.


My memory may fail me, but I'm pretty sure ASP used VBScript, not regular VB. Similar syntax but not the same runtime/environment. ASP could use any Active Scripting language (the other language shipped with Windows being JScript, MS's impl of a subset of JS)


Frequently VB if your application had COM components in Microsoft Transaction Server. I shudder at the memories. But yes, the <% ... %> inclusions in the .asp files were actually VBScript (usually) not full VB.


You're correct. I did a ton of ASP in JScript back in the day, but VBScript was much more common.



The website is obviously and gratingly LLM-voiced. I don't know how people aren't tired of paragraphs like this yet:

"Visual FoxPro is a 32-bit program, and that decides more than it appears to. It is why a table stops at two gigabytes, why a memo file stops at two gigabytes, and why a big report runs out of memory on a machine with plenty to spare. The limits are signed 32-bit numbers buried in the file handling, not a licensing decision anybody made."

and

"The editor checks what you type through that very compiler, so what it underlines and what the runtime refuses cannot drift apart."

Like, what? Who talks like that?


Robots (by default).

Take the time to ask for anything else.


To me, this is not worse than any Apple marketing page from the last 15 years written by human marketing people.


Right, yes, this sort of style certainly can be produced by writers, especially ones who are working in a particular corporate universe. That's where LLMs got it from! It just sucks that everyone "writes" like that now. Every new one-person project or company is incongruously spouting corporatese. Don't we get enough of that already? Flattening everything into American corpo-speak sucks.


Okay but why should one expect a marketing site, which this is, to have quality prose instead of typical marketing prose, which it does.


It's not hard to sandbox Pi without adding anything substantial to the actual size of the system prompt, which is what HarnessTax is mostly measuring. Wrapping it in nono.sh costs approximately zero tokens.

(I do think Pi should ship with more than zero builtin sandboxing though)


I have a couple hundred line TUI that just clones an LXD container and I get all my repos and infra isolated in 5s with pi having isolation to that container. Swap between any instance. Port forward to host all automatic.

Funny to me how stuff is described these days. Seems like some people lack the full understanding of all these different systems and really overcomplicate their harness setups or use super bloated all in one tooling.


I tried putting my agent in an LXC / Incus container but then I also needed it to process videos stored on a NAS, and that was not easy at all. I tried a FUSE-based NFS client [1] but it was very flaky and would randomly report files having zero length and fail processing.

[1] https://github.com/facebookarchive/nfusr


Do you have any links that would explain this?

I'm looking at changing my AI harness and am definitely interested in this


You can sandbox with tool hooks, at least in antigravity, I think a lot of other harnesses support similar features. Sandboxing via the prompt is just dangerous and unreliable.


The developers of pi.dev are geniuses.

They add a --tools flag, which can only add tools onto the built-in tools, so if you wanted to sandbox pi.dev by adding sandboxed versions of the existing tools you will have to run pi.dev via

    pi --no-tools --tools tool1, tool2, tool3
except..., that's not enough to sandbox pi.dev, because remember, it's a minimalist coding agent! So what does a minimalist coding agent do? Of course! It loads extensions by default and do you know what extensions can do? they can add default tools that bypass --no-tools!

So if you want to sandbox your agent, guess what you'll have to do? Yep, you have to supply --no-extensions.

Here is how to run pi.dev under its most minimal configuration under a sandbox:

    pi --no-tools --no-extensions --tools tool1, tool2, tool3 -e ./your-sandboxing-extension
Pretty neat, huh?

Sources: https://github.com/earendil-works/pi/issues/555


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: