It’s never been so easy to download music. Soulseek, yt-dlp, torrents for older stuff, spotify downloader websites etc. I still have spotify but I started growing my local music collection a few years ago. Considering canceling my spotify subscription.

For small-scale stuff like that it will surely work. It’s unclear if it scales to youtube volumes. Maybe it doesn’t have to though, small scale stuff is valuable too.

Hosting video requires a lot more resources than hosting text, hyperlinks, or even pictures. It might be too much for individuals to self host video on a scale that could even distantly resemble how we use youtube today.

Then again, maybe there are ways to make that burden smaller. IIRC Peertube does do some p2p stuff to try and share the burden a bit but I’ve also heard that it’s not really feasible to rely on that to scale.

I have it installed on a few of my machines but don’t really find it that useful. But then again that’s specific to my needs and usecases.

Potentially unpopular opinion: a bunch of rust replacements for the common terminal utilities: eza, bat, dust, fd, helix. Also fish and nushell, yt-dlp, and some of my favorite programming languages.

elementary os in 2016. I still use eos on my desktop machine, mainly because it’s kinda ubuntu but not quite. Running Fedora on one of my laptops, the rest are running macos

soulseek

I don’t really care, both are pretty fucking bad

AFAIK, this is the Google translate model that you can deploy locally https://huggingface.co/google/mt5-large

EDIT: seems like Whisper might be your best bet, it does speech-to-text + translation via a CLI that can be installed with pip: https://github.com/openai/whisper

Huggingface has loads of models for local speech to text and translation, see https://huggingface.co/docs/transformers/model_doc/speech_to_text

There is probably a nicer UI to use them than python coding, but I don’t know one.

I mean you can post and comment on the programmer humor community pretty long before stumbling on the topic of communism. But yeah I don’t think we disagree here

Even if you do, the worst thing that can happen is them banning you on a bunch of their communities

Ok? Why does that matter if you’re not on their server

Microblogging is definitely useful for many things, short and quick thoughts, links to news articles, jokes, memes etc. You can also comment and share things easily. Microblogging actually resembles instant messaging in a lot of ways, just with an undefined ’group chat’ size.

I find it kinda funny that Twitter has become so toxic that people start thinking there must be something wrong with the format.

Also RSS clearly can’t replicate a big chunk of the desirable properties of microblogging (eg. easy sharing and commenting).

Oh so they are not new users coming in? Well that paints a pretty different picture then

You talk as if you don’t understand how plots work. If the change looks like a 90% drop, that’s how it’s going to be perceived.

Moreover, if you do start from 0 you instinctively see the 5% drop, and can make the conclusion that it’s big or small yourself. If you don’t do that, you need to calculate numbers. People don’t do that. They see line go down, and get the impression from that.

Any drop would look the same on the initial plot. 5%, drop 50% drop, 0.005% drop. The ’start your y axis at 0’ rule has a lot of exeptions, but this is not one of them. In fact, it’s the quintessential example of lying with a plot.

If you want to see the actual values for each timestep, there are better tools for that. Such as a table.

Explaining all of this feels bizarre. You are de facto trolling by this weird contrarianism.

deleted by creator

If you start at 0, you see exactly what you’re supposed to: there is a rather negligible trend in the given timeframe.

That’s the point. The number of users has very slightly declined in the past few months. Under the original plot, you have a lot of people (rightly) misinterpreting the data, and saying that a lot of users are leaving the site.

That’s why you start at 0. So that people interpret the data correctly.

Let’s put that to a test

The same plot with a more reasonable y-axis:

Active users (monthly is what you should be looking at) is very slowly declining, however we are still above the level that we were before the most recent influx.