troupo

17 ms·

391.
by troupo 5mo ago
A "tool call" is literally a JSON-RPC call with a predefined schema. You'd know that if you actually did any of the following: - read the specification - implemented an MCP server - observed communication between client and s
392.
by troupo 5mo ago
This is a great example of the AI-hype-induced reply. > to an LLM that can call tools (since MCP essentially turns an API into a LLM tool). "Tools" is literally an API call > With MCP, the agent sees a bunch of tools it can
393.
by troupo 5mo ago
> One advantage is the MCP advertises itself to the agent with its schema and api shape. So, OpenAPI/Swagger for REST? GraphQL? SOAP schemas? All of these (and more) exist. What does MCP add that these don't have?
394.
by troupo 5mo ago
> I have some hope that this'll all lead to a revival of semantic web / microformats / etc Why would it? Do you see any agents or models use that? No, instead vibe coders at Anthropic vibe-designed a bespoke protocol that
395.
by troupo 5mo ago
> All of those are just implementation details to the much more important point: our AI agents are getting access to services they otherwise would never have had access to. As in: if your models and agents were as good as you claim them
396.
by troupo 5mo ago
What a nice ad hominem you decided to employ in lieu of an actual argument. Try again.
397.
by troupo 5mo ago
There's also Armstrong's Corollary: "All sufficiently complicated distributed systems contain ad hoc, informally-specified, bug-ridden, slow implementation of half of Erlang". It remains to be seen how Coalton fares ther
398.
by troupo 5mo ago
> I know hating CSS is a fun pastime for folks around here, but maybe it’s just that building good, rich user interfaces that people can use is an inherently hard problem. That CSS and web never really addressed did they? There's al
399.
by troupo 5mo ago
And the irony is that the author of that talk spent that same decade busy shoving as much Javascript into browsers as possible. After all he's the originator and the main promoter of web components where every single thing including bu
400.
by troupo 5mo ago
Google themselves are the primary driving reason why most websites are unreadable. And since now they explicitly aim to never drive any human traffic to any websites, it will only become worse.
401.
by troupo 5mo ago
Who cares? It's on Google not to degrade their search with bullshit AI. I mean, it would be if Google gave a damn about search anymore. Now we are all just reverse centaurs
402.
by troupo 5mo ago
> This isn't a unilateral Google project. I can list dozens of unilateral Google projects that are pushed under the guise of "standardization" and then released despite no consensus or against explicit objections from othe
403.
by troupo 5mo ago
The other devs only care about Google and Google's own priorities: https://news.ycombinator.com/item?id=48204981 It also helps that Google is a promotion-package-driven company.
404.
by troupo 5mo ago
> The keyboard issue was prolonged only by the nature of Apple's hardware release cycle. You know that "meticulous planning" is supposed to include mechanical tests, catching edge cases etc. right? Instead, it persisted fo
405.
by troupo 5mo ago
That's such a non-argument that I struggle to even comment on it. By this measure Microsoft is the most meticulously planned company in history with a spotless track of achievements.
406.
by troupo 5mo ago
> But wasn’t that just a setting changing it to default to fn They relented and added this settings two years in, IIRC. It wasn't there from the start
407.
by troupo 5mo ago
You're conflating what I wrote with "you're talking about commercial success". Broken keyboards shipped for years, "we didn't think about thermal throttling", "we have no working replacement for a wid
408.
by troupo 5mo ago
> I think you confuse long-term planning with long-term success. No. Long-term planning presupposes actual planning and execution, including failure modes, usage patterns etc. There's markedly less and less of that under Cook's
409.
by troupo 5mo ago
From the discussion linked in the Webkit repo: --- start quote --- Philip: First, google slides is written in svg, so that won't change with this. But google docs is using canvas, so they might be a candidate. … they might want to inte
410.
by troupo 5mo ago
> his is a web standard and expected to get buy in from Safari and Firefox before shipping to users. 1. It's not a standard. It's a scribble on a napkin in a working group's repo: https://github.com/WICG&#x
411.
by troupo 5mo ago
In a significantly reduced, dumbed down and non-functional way
412.
by troupo 5mo ago
> You're underestimating Apple's meticulous planning, which has only become more intense in the Cook era "Meticulous planning" and then: butterfly keyboard, wireless charging, Mac Pro, Liquid Glass. And those are just
413.
by troupo 5mo ago
> Before opus 4.5, I was doing a lot of hand holding and was coding a lot myself, but I have not written code since that day more or less. I still must hand hold it every day, as it always does things wrong. Especially after it got serio
414.
by troupo 5mo ago
He has the most measured (and often quite detailed) posts on LLM and LLM progress, and is the opposite of hype.
415.
by troupo 5mo ago
> Why aren’t they dogfooding their own products to replace such roles? Oh, they do. That's why they buy actual engineers to fix the vibe-coded slop they produce en masse: https://x.com/jarredsumner/status/2
416.
by troupo 5mo ago
Yes, you do hide behind buzzwords. Because your actual front page is this: https://www.stainless.com/ not the sdk subpage --- start quote --- Best-in-class interfaces for developers and agents Great agent experience is buil
417.
by troupo 5mo ago
My own initial reaction is... why not? :) Especially if the author knows C# already.
418.
by troupo 5mo ago
Likely AI-generated. LLMs love putting emojis on lists. Edit. From Authorship section: This ksharp interpreter implementation was coded originally by SWE-1.5 and 1.6 with significant contributions from Kimi K-2.5 and 2.6 and Claude Opus&#
419.
by troupo 5mo ago
> There are an incredible amount of elements That is, a dearth of them. Most of them are still there defining basically just text content. > Including for changing just a piece of text (b, i, strong, em, and many more) Not as many as
420.
by troupo 5mo ago
> I feel like this is a bad example because “card” is a presentation thing, not a content thing. It is both, and herein lies the problem with HTML and the quest for purity. The content you display in a card differs from the content y

More ›