victoryhb

It appears that Altman was fired for "not being consistently candid" by a board that is neither consistent nor candid.

Most people who sympathized with the Board prior to this would have assumed that the presumed culprit, the legendary Ilya, has thought through everything and is ready to sacrifice anything for a course he champions. It appears that is not the case.

The official name for the model has always been GPT-4. OpenAI has not used the term ChatGPT-4.

Why should we listen to someone who cannot even spell "ChatGPT" right for opinions on AI?

Super smart move for OpenAI to potentially monetize the existing infrastructure, which will make it easy for corporations to integrate GPT into their internal data and workflow. It also solves two fundamental bottlenecks in current versions of GPT: factuality and (limited) working memory. Google, with its lackluster Bard, will face new threat, now that everyone can build a New Bing clone in a matter of days.

When they released ChatGPT, they would have foreseen that the least it could do is forever pollute the Web and all human media with AI-generated content indistinguishable from human output. It would be naive to hand everyone a gun for next to nothing (on the pretext of helping them to hunt) and expect nothing bad will happen. What is the point of a warning if they have not been more judicious with the weapon distribution?

Excellent job. I can already see this being much more flexible than Numba and much more elegant/easy to use than Cython. Please keep it coming:)

If Codon becomes similar enough to Python, it will be trivial to port Python libs to it, thus opening Codon to the vast Python ecosystem.

The bottom line is Apple products post-Ive have improved dramatically in terms of practicality, efficiency, and even design. Without Jobs, the guy should be without jobs.

So you have nothing "new" to say after all, but repeating what "everyone" should know already. HN should now censor your comment.

So the reason for the censorship is they read the entire article and saw nothing "new" about something you "all know is true"? That's indeed a "new" way to discourage reasoned discussions.

I am also surprised that a reasoned discussion like this gets flagged while comments calling it a "shitpost" are not.

Why is HN flagging this article? Just because it exposes the (debatable) flaws of a language that the HN crowd love? Justifications must be provided.

I tried to learn Cython last year, but was thwarted by two issues: (1) its syntax was too ugly for my taste and support for the pure Python mode was immature; (2) performance bottlenecks were opaque and hard to profile (at least for beginners). I ended up picking up Nim, a language with Python-like syntax and C-like performance, and was productive within hours (literally). I never looked back.

I find R's strengths lie in its unmatched collection of statistical libraries, but I dislike R's syntax so much that, if forced to use it, would call an R package from Python (using RPy2), or just use a Python alternative (e.g. Plotnine).

The idea of bi-directional links sounds promising and exciting in theory, but after trying it out on Obsidian for a couple of weeks I have found following it overly laborious and have now reverted back to the good-old tree structure of Dynalist.

How do you define mental models in the first place? It seems the examples given by you and other commenters mostly fall into the category of sayings, as you rightly pointed out, which are (sometimes) useful but overly general and unfalsifiable claims.

No, it means they are using a standard web tech stack (HTML, CSS, JS/TS) inside a native shell (to utilize hardware features).