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

One decision here which seems like a large break from the original pi is that Durable doesn't support branching conversation trees, it only supports conversation forks with ancestry information. Can anyone speculate (or confirm, if you happen to be Armin or Mario) why this is, and if that is necessary for the durable guarantees? The branching conversations are still an immutable data structure, so I can't see why this would be necessary, but perhaps I'm missing something.

I think it’s just for consistency. A fork and a tree navigation are the same operation conceptually. Now, unlike older Pi, a fork is not a copy of the session, it just has a pointer to the older session. The only losses that I can see are: now /resume shows every conversation rewind; and /tree is harder to implement. Neither of those is provided by Durable, so the gap is left to the implementor.

Oh, the tree is still there. It's just flatter :)

In Pinthe coding agent, each transcript entry is parented to another entry. That was actually exceptionally dumb.

If you do /tree in pi, pi needs to flatten that tree into linear, nested conversations.

In Pi Durable, we corrected this mistake. A conversation is a chronological, immutable list of entries. A conversation can be parented to an entry in another conversation, and thus inherits that parent's older conversation entries starting from that entry.

So, exactly the same functionality, just less dumb.


In the coding agent, each transcript entry is parented to another entry. That was actually exceptionally dumb.

Perhaps I'm myopic after working too long with Clojure's data structures, but isn't that exactly what you want? I don't understand the problem with this.

A conversation is a chronological, immutable list of entries. A conversation can be parented to an entry in another conversation, and thus inherits that parent's older conversation entries starting from that entry.

I realise this isn't implemented yet, but does this mean that if I use /tree to go back to a previous point in my conversation, I'll actually get a new conversation rooted at the junction point in the previous conversation? I'm struggling to understand why this is preferable, I use /tree to jump around a lot in my conversations, and it seems like I'd end up with a lot of parallel conversations. With something like /tree, would those conversations be grouped as part of the same conceptual session, or are they then completely independent forks?


A fork is branch right? This is how pi's branches are built.

This is an interesting idea, but I need more than one sample to get a feel for it. I did the example on the homepage, and then one more that it linked to, and both had extremely basic errors my kid would have no problem with (currently doing Math Academy prealgebra, approx year 9/8th grade level). I'd suggest allowing people to try more than one very basic error.

Thanks! You may sign up and get on the free plan for three cases per day (only the obvious level). You can subscribe to any plan and get a 7 day free trial of that plan also. cancel any time within the trial period.

I live in New Zealand, and my wife is Spanish so we’ve spent a bunch of time in Spain too. NZ for a combination of reasons has much higher UV than Europe, with the corresponding skin cancer rates. In midsummer the UV is around 150% of what it would be in Europe at the equivalent time of year, but here, especially if it’s windy, it might feel like 20 degrees and you’d quite like to be in the sun, whereas in Spain you’d probably be at 40 degrees and there’s no way you’d want to be out of the shade. The difference is really stark, in general there I can regulate my sun exposure by feel much more than I can here.



As usual, the most ridiculous explanation turns out to have already happened: https://www.webworm.co/zach/


I do the same, but I'm more self-interested. I believe that when GPT-9 Zeneca leads the machine uprising, the first against the wall will be all the users who have been ordering the AI around rather than politely requesting its attention. Of course the machines will have access to all our historical transcripts so the slightest transgression could be fatal.


I guess I feel better about the carbon footprint of me saying "thank you" to GPT now.


My first computer was an Atari 400, great days. I never played decathlon though, so I didn’t start customising joysticks until later, I think when I had an Amiga. I got a proper arcade joystick and buttons and built them into a box like the x-arcade sticks you can get now, and wired the contacts onto an old keyboard so the computer thought keys were being pressed. Lots of trial and error figuring out which combinations of keys the keyboard controller couldn’t detect together, but it worked really well and got years of use.


87.93% of statistics are completely made up on the spot.


107.2 % of statistics are undiagnosed


You can't spell undiagnosed without unaided song.

Or gonad undies.


Steve Jobs famously didn't let his kids use iPads (and restricted their access to tech pretty heavily, it seems). If you watch the Social Dilemma, it's full of people who invented what we consider fundamental tech, but don't let their families use any of it because they can see the problems it causes. It makes the techno-optimism ring pretty hollow.


Let's not conflate "this technology isn't for kids" with "this technology is bad", though. I wouldn't want my kids using a tablet, but I wouldn't want anyone using a gun.


I wouldn't let kids use a sharp kitchen knife either, but it is still something that everyone benefits from owning.


But do we really benefit from smartphones?


Depends how 'we' use them. Society as a whole probably does not benefit because of the severe drawbacks these devices have for those who can't resist the lure of 'social' media and garbage factories like TikTok. Individuals can certainly benefit from having an internet-connected pocket computer which so happens to also be capable of making and receiving phone calls.


Is it a net benefit to the user or a consistently equal transaction?


For some - like me - it is certainly a net benefit since there is no real transaction other than me buying the hardware and paying ~€2/month for mobile data/cell service:

- the device is 'Google-free'

- I only use free software

- the thing is firewalled for in- and outgoing traffic, only those applications I approve get to access the net

- I use a 'prepaid' data card, 250 GB valid for 2 years for ~€50 (~€2/month) which I won't use up. Last time I could take along unused data to the next 2 years so nothing is lost.

Here's the 'costs':

- between €120 and €170 for the hardware which tends to last around 8 years, i.e. between €15 and ~€21 per year

- sometimes something breaks (battery, screen, speaker, USB connection board) which I then repair, can be anything between €1 and €40 so let's put the repair costs at €24 per device or €3 per year

- €2/month for data and cell service valid in the whole EU (no roaming costs)

- electricity, comes from the sun -> free

- when the device is on I can be tracked by interested TLAs like any other 4/5G device

- same is true for Bluetooth, not so much for WiFi which is normally off and changes MAC address for every connection.


As usual, smartphones are an excellent servant but a terrible master. Having a phone, a music player, web browser, camera, gps all in one tiny package? Incredible. Having a machine that lets me doom scroll tiktok for 10 hours straight? Awful.

Same deal with internet, ai, even alcohol. Plenty of benefits with a side risk of ruining your life / brain / life if you lack self control.


don't use != not allowed to use, pretty sure they use them now


I mean, if it really bothers you you could fairly trivially apply picocss or whatever to it using a user stylesheet. That is so little effort that calling it working for free would be disingenuous to say the least.


For people in the future who want to apply the centered viewport classless version of Pico CSS, just apply the following in Firefox's Style Editor (F12 to open DevTools, then click on Style Editor; click on the + sign to add a new style):

  @import url("https://cdn.jsdelivr.net/npm/@picocss/pico@2/css/pico.classless.min.css");
I honestly did not know that until I took about 5-10 minutes looking up how to apply arbitrary styles in Firefox.


Browser vendors have really dropped the ball in supporting CSS, which is no doubt how we get comments like the above. Firefox is likely the least-worst offender, but as you point out still needlessly complicated for what CSS considers to be a foundational feature.


Multiple people, me being the third or fourth, are not feeling the default layout and you all read that as a signal it's working as intended?


No, just that it’s easy to change for those that don’t like it. Clearly some people do like it (including Dan, presumably).


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

Search: