This is reminiscent of the operating system wars and browser wars. In the end there will only be 2 models that can survive. 1: give it away for free or 2: locked in with top notch hardware.
Not for every situation. CLI is great for coding agents (and I'd agree, far better in most cases than MCP). But it requires some execution runtime somewhere to actually run. So for app use cases where you don't want to build out your own tools for every integration, MCP can be a solid option.
Also, in my opinion, it's much easier to build a good MCP interface than it is to build a good CLI interface - and afaik there's support for MCP tools to return things like images from an MCP tool call directly to the calling LLM that is a bit tricker to do via CLI.
There is definitely some debate going on on mcp/cli/api etc, but is quite obvious that mcp is being adopted as standard for integrating third party applications to the major clients => chatgpt apps, claude connectors, mistral, cursor etc.. they all connect to external apps using mcp. Of course it's possible for you to tell claude to use some cli directly but it's much easier to connect the mcp with one click
I've been sitting in the same camp recently. We maintain both an internal MCP and CLI for our app which our devs use locally. The CLI so far feels like a much smoother experience both in terms of setup, control and performance.
But i can see how MCP being able to plug into a remote agent that doesn't have terminal access is very useful. Seems like it's a best tool for the job conversation or am I missing some other advantage?
CLI certainly is better than local MCP. But nowadays, most MCPs are remote and the comparison fall short, at the notable exception of `gh` in a coding environment. But having CLI already authenticated is not guaranted either!
MCP makes a lot, lot more sense when you think of it as as a auth standard and not a comparison with CLIs. It obviously does more than just auth, but having standardised auth (which CLIs definitely do not) is the real 'killer' feature.
Yep, that was exactly my point with the ask to elaborate.
MCPs and CLIs feel like two wholly different things. Contrasting them feels even more confusing than comparing Skills vs MCPs, which would also be wrong in my personal view. Different runtime requirements, different access patterns, different distribution, different discoverability.
I agree with @martinanld that MCPs, especially with the advent of EMA, are a completely different beast thanks to a structured way for auth and access control.
I am so tired of people repeating this. Usually, this results from conflating two uses of MCP: local, which can indeed be replaced by CLI (and you can argue which one is better), and remote, which is entirely different, and there is no way to replace it with a CLI (note that you are making an implicit assumption that a CLI tool can be used at all, which is not always the case).
Please don't repeat this. It's like saying that apples are dead and oranges are the future.
Explain again why a cli can't access a remote service ?
The only difference that is not superficial is the token count. Everything else : format, doc, auth, discoverability, ... can be made the same (the cli could be an mcp client, and the mcp server could implement a local shell, after all)
> Tell me you don't understand what you're saying without telling me you don't understand what you're saying.
Please don't cross into personal attack, regardless of how wrong someone is or you feel they are.
Your comment would be fine without that last swipe, and even better if you had gone on to say what the two purposes are. Then we could learn something from it.
If you wouldn't mind reviewing https://qqrl.tk/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.
They do. The Chinese government gave them a special exemption, presumably because they wanted to build EV manufacturing expertise. Other foreign auto companies are not allowed to open their own factories in China; they have to do a joint venture with a local manufacturer.
BYD has been making batteries since '95, cars since '05, plug-in hybrids since '08 and EVs since '09. I don't doubt that China may have made use of Musk, but I severely doubt he's the one who "taught them how to build an EV".
If you think China can only make stuff by copying what other does, you're gonna under-estimate them.
The timing doesn't line up. BYD has already been selling EVs by the time Tesla opened a factory in China. Heck, they were selling EVs even before _Tesla_ existed.
And they clearly have their own expertise. There are videos of BYD and Tesla car teardowns, and you can see that they quite differ in design philosophies.
I think China was more interested in creating more competition internally, rather than just ripping off the technology.
It's not. China has literally _thousands_ of years of bureaucratic institutional memory. And it just keeps perpetuating itself.
Before the 20-th century, the Chinese officials had to study the classic Chinese literature and pass exams based on that knowledge. These works were completely abstract and literally useless in day-to-day work. And you had to follow all the rituals to demonstrate your allegiance and being-in-the-group.
Now they just swapped the Classical Chinese works with Marxist writings. Nobody cares about their content, but you have to know them and you have to follow the rituals.
it follows marxist principles and is building towards communism, which isn't overnight. It's currently in a socialist stage. Also, Xi is closer to the captain of a ship rather than an absolute monarch. He has a lot of power, yes, but that's because the party trusts him, not because he demands it
In the age of Mao, wasn't it closer to Marxism? There are more billionaires in China now, then there were back then. By that I mean, the wealth disparity in China is at an all-time high now, is it not? Xi removed the 2-term limit from his own position, and has been doing an excellent job at consolidating his power base, through all means necessary.
Disclaimer: I believe that pure "capitalism" and pure "communism" are marketing terms which both lead to authoritarianism, aka the "Horseshoe Theory of politics." To me, the natural end-state, if we survive the extremists is Social Democracy. However, it's boring and everyone appears to find the extremes far more exciting.
China was completely mucked up economically under Mao, especially around the cultural revolution. I went there in 1983 when GDP per capita was like $300 and it was a bit prison camp like. It's changed a lot.
I was not there, but I believe that history shows that you are correct. I am not trying to sell Mao at all. If anything, he is a yet another ideological-extremist cautionary tale. (yet again, killed millions of his own people through poorly thought out absolutism)
Until Xi, China appeared to be moving in a good direction.
Not really, marxism is a way of looking at the world, not an economic system in itself
>There are more billionaires in China now, then there were back then
They hadn't even built capitalism fully, so it makes sense that there was less capital
>By that I mean, the wealth disparity in China is at an all-time high now, is it not?
it is, and they're currently working on how to deal with that
>Xi keeps remove the 2 term limits from just position, and has been doing an excellent job at consolidating his power base, through all means necessary.
Sure, but that's just politics. Ultimately if the majority of the party had a problem with him he wouldn't be in power for long before a coup or a request for him to step down happened
reply