I was about to try opencode after using claude code for quite a while.
I think understand the fundamental difference in how they work (acp against existing agentic loops with toad vs a single agentic loop for all models with opencode) but I’m curious why we might want toad over something like opencode, which lets me use any model under the sun.
I suppose toad gets to use the highly specialized agentic loops for each cli. And has a nicer (? opencode is pretty slick from my brief usage…).
Curious to hear about why you chose to built this way and what advantages you see.
Citron (who founded Discord) also was involved in OpenFeint, a mobile game social SDK that started out originally as a game for the iPhone called Aurora Feint. So it's kind of cyclical in a fun way - game, social tool, game, social tool.
I expect this is a pretty common story on a wider basis and people sometimes overlook it. How many bug trackers or build systems started out as internal tools for managing some other product? Etc.
We do! I'm in Japan right now, taking photos of as many markers as possible that I'll add into the app when I get back! We're also releasing a 'submission app' in early 2025 that will let you submit markers and get them added into the app so we can expand into more places!
I've always been pretty quick to see the hidden image, but I could see these faster than ever - almost instantaneously, even the first one. I wonder if it's something to do with me having stared at a screen for a decade since the last time I saw a stereogram...
The article mentions it, but you can sum up print debugging as selectively enabling verbose/trace logging. “We’re about to do X” or “We just did Y, here is Z”.
A debugger gives you insight into the context of a particular code entity - expression, function, whatever.
Seems silly to be dogmatic about this. Both techniques are useful!
Unfortunately, it’s not possible to express this sentiment via election participation. Abstention ends up supporting one candidate more than the other. What seems to be an affirmation of neutrality is not that in practice.
The national election is an exercise in partisanship. Your opportunity to feel represented is what the primary is for. And for once I'm not sneering at the sentiment because basically neither side ran a primary (the Ds managed to not run one twice!)
> Your opportunity to feel represented is what the primary is for
This is why it was a major issue for me that the Democrats did not hold a primary and just decided Kamala would be the candidate. If a major part of your campaign is "vote for us or democracy dies!" it's pretty hard to swallow if you increasingly feel that your voice doesn't matter in your own party.
Given that Biden dropped out of the race fairly late, what was the option? I agree with you in principal, but there doesn't seem like there was any way to actually implement it in the available time, assuming that the population take adequate time to make an informed choice.
There were three candidates and two of them were jokes. Everyone that would have been a real threat hung back, hoping to rise either under Trump or after him. He sure wasn't taking it seriously, neglecting to even register in some states, causing the Haley loss to 'none of these candidates' in Nevada
Do you think the real threats were colluding with Trump? Otherwise, I don't see how Trump being extremely popular within his party means the primary wasn't legitimate.
You misinterpreted the massive disagreement of the population (20M people) with an affirmation of neutrality. One can hope the dems will not misinterpret it (as they often do, unfortunately, and they already started on twitter and the mainstream media). Hopefully they can recognize and acknowledge that a large portion of the left disagrees with their policies and start listening to their base, otherwise they will keep losing more votes every 4 year.
I and another registered democrat neighbor had this exact discussion and conclusion outside of the voting center we live next to on election day. Assuming we're not the only ones.
Some jurisdictions have a "none of the above" option. I think Nevada has one. An early report there show 1+% voted "none of the above", or something similar to that.
It was just a glance on one of the shows during the returns last night, I maybe be completely wrong.
> Unfortunately, it’s not possible to express this sentiment via election participation.
Not just election participation, no.
You do have to use the generally-free-to-use, generally-globally-accessible publishing systems that are available to nearly anyone with a computer to explain why you refused to vote. (This is my big issue with the "Refusing to vote is meaningless, because noone will know why you didn't vote." counterargument.)
Whether your assertions that you didn't vote because -for instance- none of the available candidates were people you wished to see in the positions they were running for get deleted because they are "Election misinformation" or similar is an open question.
Yes exactly, at some point I asked to maintain it and kinda redid it. Now I kinda consider it "done", as in "maybe some more work would be put into it, but by end large I don't think it's going to change in the future".
I'm a managing consultant, so I don't have much time to do actual work (due to having to chat up clients and tell other people what to do). With cursor I can do what would have taken me an afternoon, previously, in an hour.
This is huge for me, because I can realise ideas and show people how to solve problems whereas bc (before cursor) I was reduced to saying "I'm sure that will work if we do it that way" and hoping that my folks would follow through on the thinking themselves.
Cursor supports me in two ways.
1) It gets things moving - I can put all the boilerplate and infrastructure to build something in place in a few minutes.
2) It removes road blocks. When I get stuck it gives me fixes.
Now I recognise it's all from Stack Exchange, basically, but it's a terrific interface for that knowledge.
What it doesn't do is provide new knowledge or ideas for fixing things. That has to come from upstairs still, but that's never been an issue for me.
I was about to try opencode after using claude code for quite a while.
I think understand the fundamental difference in how they work (acp against existing agentic loops with toad vs a single agentic loop for all models with opencode) but I’m curious why we might want toad over something like opencode, which lets me use any model under the sun.
I suppose toad gets to use the highly specialized agentic loops for each cli. And has a nicer (? opencode is pretty slick from my brief usage…).
Curious to hear about why you chose to built this way and what advantages you see.
reply