@lauriewired@karpathy a lot of engineers forget that a lot of people don't think like engineers. many people don't realize they even have a problem until someone points it out
oh come on. You think your grandma wants to make her own app? Much less maintain it.
Everyone neglects the mental energy it takes to even *think* of what it is exactly you want.
The entire principle of apps relies on some faith that designers; and the collective feedback of their users, can come up with a workflow or design paradigm that is *better* than what you; an individual could come up with.
This is true for 99% of cases.
I’d be floored if even 1% of global users want bespoke applications for uber-specific needs. Not saying it’s useless; but it’s so far away from what average users want.
@pierceboggan@VSCodeChangelog@code Why do we need this review interface at all? Why not integrate agent changes review with the (already great) Git review? (incl. stage, commit, stash, etc.)
@Love2Code I'm not an expert, but @deno_land and Deno Deploy make it really easy. Minimal configuration, simple commands to deploy, or just push the code
@udolele@iamdavidhill@iamdavidhill in that sense, I'm more interested in being able to trace the conversation history (walk through thinking chunks, skip tool calls), and zoom into subagents, rather than just seeing the running ones
@udolele@iamdavidhill Sometimes when the main agent does exploration/research (esp. using Perplexity MCP), it can be interesting and useful to look at how it formulates the query (is it on track?) and review the results (can be useful for learning)
@dkundel DeepWiki is great for working with libraries that don't have much documentation. It answers tricky questions about their internals based on the code itself.
Perplexity MCP is almost always better than a normal web search. Very useful for "is this a known issue?"
@diegohaz I like to do the same in @deno_land, just one config with the standard import map. No tsconfig at all, and you can do the same @/ without workarounds.
@logseq What is this? An attempt to hype it up? It's been coming together for years. It's been almost in public beta for months... "In a month" was 5 months ago 🤦♂️ eternal alpha with a second-system syndrome 😞
@zeddotdev I love how fast Zed is overall, but sometimes it hangs completely because of the Git integration. For example when you switch from one branch to another and have too many files that were ignored on the first one, it can't handle the number of changes trying to lead them eagerly
@iamdavidhill It would be nice to see the (average?) token/s and correlate the animation with it. It's just a way to see at a glance how fast different models are while working with them. Maybe tokens are not the right metric (considering tool calls), but it's something
@iamdavidhill@opencode It's good, but only when you're absolutely sure that the user doesn't need the context of what's behind the modal, because this UI doesn't offer an easy way of switching back and forth and completely streaks the focus. It has to be a "mode" to warrant a modal.
I consider myself a "just in time" developer.
Like I have no idea how to work with sqlite databases in Python. At least not without an ORM. By the end of the day I will.
Let me tell you the secret of how I do that...
I read docs and play with code until it works. 🤫