That they use some of their trillion dollar marketshare to solve it, why are you acting like this is a hard problem? It's not. They're just too cheap and greedy to do anything about it.
I have never had a problem using cli tools intead of mcp. If you add a little list of the available tools to the context it's nearly the same thing, though with added benefits of e.g. being able to chain multiple together in one tool call
Not doubting you just sharing my experience - was able to get dramatically better experience for multi step workflows that involve feedback from SQL compilers with MCP. Probably the right harness to get the same performance with the right tools around the API calls but was easier to stop fighting it for me
Did you test actually having command line tools that give you the same interface as the MCP's? Because that is what generally what people are recommending as the alternative. Not letting the agent grapple with <random tool> that is returning poorly structured data.
If you option is to have a "compileSQL" MCP tool, and a "compileSQL" CLI tool, that that both return the same data as JSON, the agent will know how to e.g. chain jq, head, grep to extract a subset from the latter in one step, but will need multiple steps with the MCP tool.
The effect compounds. E.g. let's say you have a "generateQuery" tool vs CLI. In the CLI case, you might get it piping the output from one through assorted operations and then straight into the other. I'm sure the agents will eventually support creating pipelines of MCP tools as well, but you can get those benefits today if you have the agents write CLI's instead of bothering with MCP servers.
I've for that matter had to replace MCP servers with scripts that Claude one-shot because the MCP servers lacked functionality... It's much more flexible.
I’m not sure it’s possible to conclude what hey actually believes from public statements. I do not trust him to tell the truth about anything related to AI.
Washington Post tried cheap "day pass" subscriptions and they didn't really work.
Publishers already relying on subscription revenue need to be careful: some portion of the people already paying $20/mo could save a lot by switching to $1/article.
Newsrooms also hate that approach because of the incentive structure. A lot of the most important stories aren't the ones people want to spend $1 to read.
I'd encourage you to ask a recording artist how they like the arrangement they have with Spotify.
But also, yeah, I do think the streaming financial incentives affect what music gets written and produced. Just not necessarily anything to do with cuss words.
The artists are there because of their own free will. They voluntarily signed their contracts with the record labels. How are journalists and independent journalism doing in comparison?
I think this is even harder to make work than straightforward micropayments with crypto or paypal or similar.
Is it the same subscription fee no matter what publications I read or how many articles? (If it varies directly based on what I'm reading then I think it is just micropayments.)
Publications with healthy subscription revenue like WSJ or the Economist are not going to be interested in participating unless they get paid a lot of money and/or can be assured it somehow will not cannibalize their direct sales.
Who owns the customer relationship? Publishers have been burned pretty much 100% of the time they cede that direct relationship to someone else.
Also, it's been tried: see Scroll, Apple News, Flattr, Coil, Brave BAT...
I agree Scroll seemed very promising but I'm not sure how successful they were. Did they provide more income than ads from subscription fees? Meeting that goal while burning investors money is less impressive.
This does not seem to reckon with the many, MANY times this has been tried and failed. (The LinkedIn post at least acknowledges attempts that "did fail in the 90s and 2000s" and quickly waves them away.) There have been at least a dozen serious attempts in the last 20 years that I'm aware of.
What about Blendle? They had NYT, WaPo and WSJ as launch partners in 2014 but give up on micropayments in 2023 citing "very low demand"
Or Flattr. Or Invisibly. Or Pico. Or Brave's goofy crypto token. Or Coil. The Washington Post themselves experimented with cheap "day passes" a few years ago but I guess they didn't work well enough to keep. Arguably Medium's rev share program was another failed attempt. Heck no less a content middleman powerhouse than Apple tried and mostly failed to do a rev share / micropayments scheme with Apple News.
The trick is market share. I'm not going to subscribe to a NYT/WaPo/WSJ trio because sometimes I want to read Reuters and I'm not going to pay for a partial solution.
I was very happy with my Apple News subscription because it has every English-language newspaper I've come across.
I like Apple News too, though they are walking a bit of a tightrope with their publishing partners which is why you can't just click a link to wsj.com/blahblah and have it open in Apple News, and you can't really just browse a front page. (Also it notably does not have NYT.)
which had startup royalty behind it and a very slick web site that they didn't promote very well. (A friend of mine who is interested in this space didn't find out about it until it was announced it was shutting down.)
I can only guess that the New York Times, WaPo and such were too good to talk to these people because they only managed to sign up third-tier news sources.
I assume it was more about dollars and cents than pretensions. Scroll was plausible as a replacement for ad network revenue, but at $5/month it can't come close to robust subscription or premium ad revenue. NYT subscription is more than the entire $5.
I believe the main reason that Flattr failed was just bad execution. After a successful start and Flattr being used more and more, they just stopped iterating on it, leaving it dormant for years. Partnering with Adblock Plus was the final nail in the coffin.
reply