greyman

On output tokens, GLM-5.2 is less than a fifth the price of Opus.

Opus is most expensive model in pay as you go model, but IMO fair comparison should include subscription price as well. For example when one has $100 Claude Max and use it up through the month, it might not be more expensive than GLM, or at least not 5x.

From my login credentials Anthropic do not know I am non-US national. They could deduce it from my chats, but that would take some time to implemwnt.

Yes. I did not left but visit the site less often and kind of worry about its future. The engagement over what I post is just much lower, while the number of reported visitors seems to increase. I don't mind some good quality AI comment, sometimes one see it, but overall it is slowly becoming ghost town.

I need to do that anyway, but I can treat the agent I work with as a human, I do that for months and didn't encounter any big problem with that approach.

What if I WANT to anthropormorfise AI agents I work with?

From a moral perspective, I would argue that this is still theft of IP, even if it's a "clean room reimplementation". The code carries valuable information about what works and what doesn't — knowledge that Anthropic had to discover through real work and iteration. It's the same as a Chinese factory duplicating a product: they skipped the entire R&D phase and saved time and money.

But nobody says code is free(?). Certainly not Claude, that experimental compiler costs $20K to build. That openclaw author admitted in Lex Fridman talk that he spends $10k's on tokens each month.

I usually now just ask agent, for example Gemini in Antigravity to check certain article or a group of articles, like "check all AI-related article in tldr.tech and tell what is interesting"... I am already a bit lazy to browse myself, and in this process I dont care about ads.

Meta question: do you guys feel the adblockers will maybe not be that important in the future? As for myself, I ended up to use just a few websites, but those are reputable and I don't mind a few ads they provide. The only adblock which is still very much needed is one for Youtube.

Yes indeed, and also the title promise - I looked forward to read how the personal blogs are back, only to discover the author didnt provide any evidence, but not even examples. Maybe they are indeed back, if we count Substack newsletter archive as a "personal blog".

Why it is absurd? Conversation between me and ChatGPT can be read by a lawyer working for NYT, and that is what is absurd.

But that vault can contain conversation between me and chatgpt, which I willingly did, but with the expectation that only openai has access to it. Why should some lawyer working for NYT have access to it? OpenAI is precisely correct, no matter what other motives could be there.

I conditioned myself to not type too-revealing texts about myself into the computer. It isn't ideal but of course this is quite a big problem.

Keep going with the blog, I found several good topics there.

Another direction is that more and more content is directly put on social media. I found I am visiting websites less and less, I even dont care about ad blockers anymore (YT is an exception).

Because all that scepticism, non-support etc. are like small micro-defeats that will suck the life energy. Not to mention all those taxes which sucks money energy. :-) But the deeper reason I see in in Europe is that everything is biased towards big and old, and not new and small.

But what is lacking is some kind of large European tech journalism

Yes, even when tech media in europe would want to hype the original OP startup, they are not that influential in general. I follow tech news daily but all of them are U.S. media, there is nothing comparable here.

Question: I agree, but doesn't that other model need to be trained so it knows how to work with MCP servers? Or that isn't an issue?

Thank you for your post, it's very sensible setup. Does Cloudflare Pages offer unlimited bandwidth?

As a corporate senior eng, this does't look like a mess to me - just a few things, easily configurable in the matter of hours. My only concern would be if Cloudflare pages offer truly unlimited bandwidth, but so far the site is live. :-)

Yes this happens. Then I use "dumb" IDE like Goland... or, I didn't stop using it. My point is that I currently do not invest my time into learning "agentic IDE" like Cursor, since I am not sure this is something useful in the future.

I also stopped using Cline as well as Claude Desktop + MCPs. Gemini for example is rushing forward, Google surely is putting huge resources into developing it, and if in the matter of months AI will be able to implement additional feature itself in 0-shot, why bother with IDE?

I also stopped using AI code editors, but for different reasons. I realized, that with advances like Gemini 2.5 Pro, AI will soon be able to implement whole features, with correct prompt. So the real skills is how to prompt the AI a maintain the overall architecture of the project. I wonder if the IDEs like Cursor or Cline will ever be needed in the future; as for myself, I stopped investing into learning them. I currently use 2.5 Pro + Repo prompt app which prepares prompt and then apply the result to codebase semi-automatically.

1) It isn't a standard yet, but what else apart from filesystem-mcp can be used for prompts like "write me README.md for this repo" (like really produce the file)

2) For me it is not clear the future is smaller self-hosted LLMs. As of today, most useful for me is to use best models, and those are not self-hosted.

We do, but I am transitioning towards MCPs. One thing is that it is free, and secondly, Cline don't bring that much more to the table - in Claude desktop I can also ask to implement something, and then tell it to write solution to the source code files. This in effect does the same thing as Cline, but I feel I have bit more control.

For those already using this MCP, what does it bring more comparing to just assemble more specialised MCPs together, like filesystem, fetch, git, brave-search.

Or you can also use Gemini Code Assist extension for VS Code, which is basically free, but so far, the code it wrote almost never worked for me. So far I use only Claude 3.7 or Grok in chat mode. Almost no model, as of today, is good at coding.

OP, thank you for your work, I will continue to watch it.

I tried to built something similar, but what I didn't discover and think is crucial is the proper FE: yes, word segmenting is useful, but if I have to click on each word to see its meaning, for how I learn Chinese by reading texts, I still find Zhongwen Chrome extension to be more useful, since I see English meaning quicker, just by hover cursor over the word.

In my project, I was trying to display English translation under each Chinese word, which would I think require AI to determine the correct translation, since one cannot just put CC-CEDIT entry there.

P.S: I dont know how you built your dictionary, it translated 气功师 as "Aerosolist", which I am not sure what is exactly, but this should be actually two words, not one - correct segmentation and translation is 气功 师, "qigong master".

there's undoubtedly some societal good from having GPL or LGPL licensed products encouraging organisations to release their code under a similar license

How often does this happen, though? I remember at my previous employer, there was a rule to never use GPL code in our projects.

But you can sell Vscode if you want? -> yes you technically can but in practice it's almost impossible to be competitive, since for several major languages the best extension isn't open source. Author called it a "fracture".