78% of websites where it can be deduced from client side code what server side tech is used. W3tech published this statistic, and the whole community picked it up while ignoring that very important caveat.
It is almost certain that none of these bills represent wasted money. The piece of paper and the money are not identically the same thing. The paper is a document that is made to represent the money. At some point, the document is made to cease to represent the money. The Federal Reserve routinely acquires and destroys old worn bills, replacing them with freshly printed ones. This, by the way, has little or nothing to do with "how much money exists".
My understanding is that those 2,000 represent some very large and enterprise-y contracts. The GitHub itself has almost 2,000 contributors: https://github.com/ClickHouse/ClickHouse
I don't think that's the solution. Proprietary LLMs will just keep growing, and it doesn't seem like the open source alternatives are gaining much traction. I guess it's because you need a lot of money to train high-quality LLMs (tons of energy, maybe?). Besides, as stated in the title of this post, us, as software engineers, the collective, don't seem to mind much the current state of things as far as I can tell.
Strictly speaking - we don't really know if this is true. There is no study proving AI gets smarter up to a certain point. It might keep scaling forever, or one day we might unknowingly reach the soft limit of LLM intelligence. I think assertions like the one you're making require specific evidence.
For comparison's sake, proprietary models like GPT-3 now pale in comparison to the results you get from a 7b Open Source LLM. The Open Source stuff really does move along, if not at the pace everyone would prefer.
I have an agent that creates new tools here: https://github.com/kordless/gnosis-evolve. I use it with Claude Desktop for a lot of different things, including browsing or searching for content, with the various crawlers that are out now. There's a crawl4ai tool that is pretty useful.
Yes, absolutely. Can be done without it and fairly easy to even make it call remotely. The value is a framework that a client can use to make calls, however. And if I want to use Claude Desktop, much easier to use MCP to deal with the interconnect. Consider VSCode integration there is much there for me to learn. Here it's easy to use the MCP stuff but it certainly isn't necessary in a server deployment, for example. Although you may want to write a MCP server for your endpoints, if you make money off that stuff.
