I love this. Here's another interesting thing I encountered. It's a way of organizing chromatic subsets by brightness

https://www.reddit.com/r/musictheory/comments/1etydas/i_made...

That is cool and different. Glancing at the bottom several rows I tend to agree with the classification, as a trained musician. I wonder though, what is (the mathematical principle) behind it that is causing the brightness/ darkness in the sound of these chords?

Don't say "dissonance" (or explain what dissonance is)- that much is obvious, looking for something a bit more detailed, e.g. why 1-2-5 sounds brighter than 1-4-5

It is going to be the relative amplitudes of the overtones. Pure tone (like flute) is bright, many overtones present with the appropriate mix can sound dark (french horn). Next, you are going to ask me for the coefficients, but I don’t know that. Break into the nearest church with a proper pipe organ and start pulling and pushing stops to see what happens. (Or ask your friend the organist to take you so that you don’t get arrested)

It's a reference to this particular chart, not the general idea of brightness of a sound. And it's regarding chords or scales, not timbres.

I don't really understand this chart at all, but I think it's based on the idea that an upward movement of a fifth is bright, and a downward movement is dark. 1-2-5 can be built on two upward movements of a fifth, whereas 1-4-5 can be built be one upwards movement, and one downward (from the tonic)

All this means is that you poorly curated your follows.

I’m not logged in. This is based on what I see as a potential user checking out the site for the first time.

Pepperidge Farms remembers that Indiana Jones was the good guy.

Censorship isn’t absent from Twitter. Quite the opposite.

This was a great little article. Very fair summary of the situation.

After the demise of Twitter, I first tried Post.news — which had great branding but failed to get the everything-is-a-tweet model right (comments were 2nd class citizens to posts).

Then I moved to Mastodon, which I enjoy. Mastodon’s biggest issue is the enormous UX hurdle to pick an instance before even signing up, though. That and the lack of a unified view (mentioned in the article) will probably keep it niche. Also lack of quote-tweeting, a deliberate choice.

BlueSky is the first truly worthy successor. It’s better than Twitter in its prime, before it went algorithmic. It allows quote-tweeting but gives the quoted party control over the scenarios that Mastodon was trying to prevent by avoiding the feature entirely.

Agreed about that they nailed the UX for BlueSky in terms of “best possible Twitterlike”.

That said, my biggest concern is how they plan to make money. With BlueSky being completely open at the moment, there is effectively no possible monetization strategy for them without making major breaking changes to the protocol.

For example, if they added ads to the app, someone could just write a client that didn’t have ads. Donations are another possibility, but probably wouldn’t be enough to pay for the servers.

Twitter had third party apps with no ads for over a decade without the sky falling. Few people use anything other than the default.

Fair point! Ditto for Reddit, I suppose.

Though, wasn’t Twitter’s financial situation starting to become somewhat dire toward the end of its pre-Musk life?

I mean, define dire. It was not what you’d call the world’s most profitable company, and had negative income for two years, but it had been cash flow positive every year for years. Its valuation was… optimistic, but it’s not like it was going to collapse or anything.

In fairness, Mastodon apparently is working on quote-tweeting. AIUI to do it properly (ie, like Bluesky, with the ability to remove quote-tweets etc) requires changes to ActivityPub.

Bluesky, at least for now, has the advantage that they essentially control the protocol, and are the only significant user; they can move a lot quicker. It’ll be interesting to see how this evolves.

I vastly prefer comments as second class citizens anyway. Twitter felt like such a rat's nest sometimes.

I was pleasantly surprised to discover that Bluesky behaves like the Twitter of yore where you see only who you follow.

There's another narrative that I've heard about the recent flood of former-X users into Bluesky: a new Terms of Service document goes into effect in about 4 days, giving X ownership of your content with respect to its use in training LLMs.

I thought Grok has been using Twitter content this whole time, are they just getting around to updating their ToS?

I can see some amount of protest, people who are anti-AI/pro-copyright might find a better niche on bsky, but everything you post there is public so it's legal for anyone to scrape and train with

Norman D Cook's work is interesting in this context. This is a great talk, if anyone is curious:

https://www.youtube.com/watch?v=CrmnaiyS5EE

This thread reminded me of the "Conclusions" slide.

See also the very recent "Symptom Of Life" by Willow Smith, a cluster chord bonanza.

To each their own, I guess. I find the various modern ML-influenced languages to be among the best looking.

I agree with you. The best looking ones are Gleam, ReScript and F#

I always miss pipeline operators in languages without them after F# exposed me to them.