Mathematics that our universe does not support is still mathematics. Also you're viewing this through a human lens. How do you know it doesn't support what you think it doesn't support? Mathematics after all is a very human pursuit. Some alien species may have started off with ternary or even continuous logic. We have no counterexamples to show that that is beyond our universe.
We have just convinced ourselves it is statistically safe, much like most of medicine. Most drugs are not made from molecular simulation of an entire human body.
Of course we have also convinced ourselves before that cocaine in drinks, lead in petrol, asbestos in walls… were all safe…
This reminds me of the Feynman interview where the interviewer asks "how do magnets work" and he goes on this rant of how that's unanswerable and you have to decide on what is it you really want to ask. You can't expect to know the full chain of knowledge because at some point you will be asking about quarks and gluons and then hit a wall where "nobody knows". Similarly you cannot simply just give up any investigation at all because then you'll end up recommending people to put lead in their cars.
My favorite part of Feynman's magnetism rant about how the question is terrible and can't be answered is that he seemingly accidentally gives a very good explanation: that magnetic repulsion is a highly concentrated version of what repels his hand from the arm of his chair.
We also actively cull them, for the rats. Recently watched an episode of survivorman where he was on this tropical island infested with rats. There's absolutely nothing on that island yet there's rats all over the place. Having no humans they may explode in population.
Umm… there are a lot of people out there, more than you can imagine… I'm sure there's quite a considerable community who want to upload their minds to skynet.
Every time I see something like the Panama or Epstein files I just get reminded how many people there are in the world doing weird shit.
For me it's not coolness it's just context, 16:9 is too short to fit enough lines and scrolling and jumping is as annoying as tabbing in and out.
This is also one problem with ereaders for me, especially for anything technical. With paper when the textbook says refer to fig 13.2 or we covered this in section 4.3 I can just flip to that page with my fingers as a bookmark, read it, and go right back. Just not as tactile on computers. Even if you did like lsp assisted jumping all over the place you quickly forget what the heck was going on and where you came in from.
Of course maybe someone will come and correct me but this is just what I experience. It's still better flipping pages for technical stuff. It's like scribbling a DJ disc, imagine doing that with a mouse or keyboard. It's just… different.
Still don't get what's the advantage over just adding a special API key or a wrapper cli or whatever mechanism that achieves all of that without being a "protocol" and with all the context bloat. Like the github cli is a good example. You give it proper auth keys etc. and for sure there's some telemetry in there about usage as well. If there isn't then it's easy to do from the api side too.
Even if you don't own the code or infra, like say a frontend team who wants models to test out the backend apis and do something. Well in that case how do regular devs do it? Do they also get unfettered access in the past? Surely there's still some mechanism you can repurpose for agents to use?
I'm not trying to argue I'm just saying I didn't catch on the first time mcp was a thing and I still don't know what it's doing now.
If a full coding agent can access a CLI tool. that agent can almost certainly access the API keys being used by that tool. They can go as far as decompiling binaries, or rewriting them to log the key before it is used.
If you are worried about a prompt injected agent stealing your keys, that's a problem.
(There is a way around that: you can use an HTTP proxy that inserts those credentials but otherwise lives outside of the agent's realm of influence. MCP is a whole lot easier though.)
> If a full coding agent can access a CLI tool. that agent can almost certainly access the API keys being used by that tool.
So, don't do that then?
Why do you need to use a full coding agent as the interface between the model and the CLI tool?
A 10-line program can do the wrapping of any existing CLI program so that environment is not leaked to the model, while providing the CLI program with the environment as well as restricting what programs can be called to a whitelist.
If you CLI program is echoing its keys in the response, or the endpoint is echoing keys back, that's not a problem that can be solved with MCP anyway.
So you're building a custom harness here that provides tools, and you're wiring up your custom harness to effectively do a subprocess execution of a CLI script for every tool call the model request?
One reason to switch to MCP here would be to avoid the overhead of forking a new process for every tool call, and to enable maintaining state between tool calls.
(That performance overhead is so trivial as to not be worth caring about, but the state thing may be useful - keeping a stateful browser session running between tool calls is harder with a CLI, for example.)
> So you're building a custom harness here that provides tools, and you're wiring up your custom harness to effectively do a subprocess execution of a CLI script for every tool call the model request?
Well, yeah. Subprocess execution is on the order of double-digit milliseconds. The "wiring up" is maintaining a whitelist of what tool commands map to which executable. It's a lookup table with very little maintenance required.
> One reason to switch to MCP here would be to avoid the overhead of forking a new process for every tool call, and to enable maintaining state between tool calls.
I feel like I am taking crazy pills :-/
The overhead of forking, on the ancient machine I call my desktop, is at most double-digit milliseconds. The state isn't being tracked by the MCP server anyway, it'll be tracked by the harness and/or the model, no?
My main reason for adding MCP support is so that existing callers that want to use my WebApp(s) can just use it without needing any changes on their side.
IOW, I am going to add it at some point, but not for the reasons you give. I'll add it to be compatible.
reply