So I set up Yunohost [0] on a small box, and now I install self hosted services whenever I need them. Installing a new service is a breeze–but more importantly, upgrading them is a breeze to.
For now I self host Mattermost, Nextcloud, Transmission.
1,093 karma · joined January 21, 2011
So I set up Yunohost [0] on a small box, and now I install self hosted services whenever I need them. Installing a new service is a breeze–but more importantly, upgrading them is a breeze to.
For now I self host Mattermost, Nextcloud, Transmission.
Even on my own projects, on a much smaller scale, I've never been able to go that fast.
Congrats for shipping, and to many good years to come!
The compression algorithm is said to have been written by Satoru Iwata himself, giving a hand to the overloaded GameFreak team when the development of Pokémon G/S was hitting size limits pretty badly [1].
There is a readable version of the decompression routines in the Pokémon G/S disassembly repo [2]. As the code was originally written in assembly, it is not decompiled–only re-documented.
Which means the code we read is the actual code Iwata wrote it 20 years ago.
I really like seeing the traces of actual human beings in a disassembly. Noticing different coding styles; seeing where routines were written, then patched over later… We read code, but we see people.
[1] https://retrocomputing.stackexchange.com/questions/11476/how...
[2] https://github.com/pret/pokered/blob/10289bf/home/pic.asm
https://mattermost.com/blog/how-uber-uses-mattermost-to-enha...
> Of all the browsers I tested, only Brave and Firefox currently disable it by default and do not appear to have any plans on enabling it in the future.
Oh, that's why.
If yes, it could be because Firefox started honoring the system font settings of Mojave [1].
Enabling font smoothing in System Preferences now results in slighlty bolder fonts, and grayscale font smoothing. Firefox and Chrome needed some time to adjust to these settings, but now have the same rendering than the rest of the system.
Just tested opening Firefox on iOS, and the "Open page in Firefox" pops up in the Dock of my Mac. Neat.
(It is advertised as also working with Mobile Safari.)
It seems that, after Ruby borrowed from many previous languages, a lot of rubyisms went into Rust (the nice functional-style enumerations, the awesome package manager inspired by Bundler). And I'm thrilled the best of the Ruby world found its way into other languages.
A future where the elegance of the Ruby language is available both as an interpreted-scriptable tool (Ruby) or a compiled-statically-typed language (Rust) looks like a nice place to me.
Still very much a work in progress – but one can already publish articles, comment them, and follow people using RSS or a Mastodon account (because ActivityPub rocks).
Not ready ret, but quite promising.
RSS lets me focus on what I actually want to read, makes me take long-break (instead of short bursts of Twitter updates), and gives me longer and more structured content.
I also use RSS for sharing articles, by giving the URL of your Pocket RSS feed to friends.
I'm glad that, in the age of algorithmically curated timelines, RSS still exist!
I remember reading in A study that feet warming up was the factor best correlates to falling asleep (I remember it from only a few years ago, maybe it was a follow-up to this 1999 study? [1])
All of this without breaking compatibility across instances running different versions; quite a nice piece of engineering IMHO.
From what I remember, Mastodon switched from OStatus to ActivityPub a few releases ago. In the past they had to extend the OStatus protocol for private messages, which meant some clients may not honor the private status of theses posts. ActivityPub has private messages defined properly in the spec.
It seems PeerTube (https://github.com/Chocobozzz/PeerTube) will also use ActivityPub.
So you can choose to be charged every month (and have ~2% of banking fees), or to provide money for the next 3 months (and the banking fee will be less). Money will be redistributed to creators once a week (or once a month) anyway.
Plus they only charge banking fees, nothing for themselves (tip is voluntary).
The process was surprisingly smooth and easy. We'll see how it works on the long run, but for now Liberapay looks like an excellent alternative. (Also they don't take automatic fees for their own finances; tip is on a voluntary basis).
Looking back, I feel like the short-messages limit is fueling flaming statements rather than good discussions. I also see how constant enabled-by-default notifications and dark patterns made me addicted to new tweets ; on Mastodon I feel less pressure to keep up with the feed constantly.
Also, the project grew from a few to 100+ contributors in a few weeks, and apparently managed this transition quite smoothly. Development is ongoing at a steady pace, including new features and major architectural changes. Congrats to the team and contributors!
The complete disassembly of Pokemon Red/Blue [1] is an outstanding piece of work. All the source code and data of the ROM have been figured out and commented.
For the Zelda DX disassembly I'm currently working on [2], devdri wrote a custom GameBoy disassembler named "awake" [3]. It produces a nice pseudo-code view with C-style statements, and also do some static codeflow analysis with variables propagation to know which memory banks are being switched in and out. Quite impressive IMO.
But damn, even with these tools disassembling is still a lengthy work…
[1] https://github.com/pret/pokered
We know it's possible to implement logic gates in Game of Life. So would it be possible to re-purpose a VHDL compiler to emit schemas for GoL-gates or GoL-transistors?
In this case, wouldn't programming a Tetris game in VHDL output a much smaller automata (avoiding the need to use huge OTCA metapixels, except maybe for display)?
It influenced early design choices (the "retweets" count is not proheminently displayed ; fine-grained options for muting/blocking/banning).
Plus, in the end, isn't per-community-moderation the only way to avoid the "one rule fits all" dictatorship?
Interesting. In my experience the clean separation of layers in Ember-Data (network, serialization, models) make it quite suited to non-standard APIs. I found writing a custom (de-)serializer for a non-standard API to be relatively easy.
Care to elaborate on what were your issues precisely?
Remember how much the Javascript ecosystem changed in 5 years? There is no way we could have maintained the same code base without Ember; we would have needed to rewrite the app once or twice by now. And with each rewrite small features, little subtleties would have been lost, replaced by some bike-shedding about how we should hand-compose an entirely new stack.
Instead, we upgraded quietly from time to time, and kept all our carefully tuned code and beloved UI tweaks with us. To be honest, the maintenance and version bumps took quite some work, and some early upgrades required heavy refactoring. Ember paradigms changed frequently during the pre-1.0 era, and later React clearly influenced the best practices recommended for Ember 2 (like Data-Down Actions-Up, etc).
But boy is it worth it. Ember-Data is neat, the testing tools are great, the Ember-Inspector add-on is a joy, and the CLI tools gets us so many free features. And we never could have maintained this product without a stable framework across all these years, and the friendly developers community.
Also, to answer your question about add-ons: they're great. Among the core team there is a nice commitment to make many new features available as add-ons (rather than baking them into the framework). And the community solves many common problems for you indeed.
That said, I see with this demo how these metrics could be useful to track, well, when reviews are taking too much time (needs more people? better repartition of reviewers?) – or when maintenance tasks are slowing new features (needs better tests? more maintainers?). I guess I'll have to try this out :)
Here with some animated gifs: http://kemenaran.winosx.com/posts/links-awakening-rendering-... (disclaimer: I'm the author).
Nuclear is a fossil fuel too (uranium has to be extracted and is limited), costs a tremendous amount of money in power-plant building and dismantling costs, and concentrate the production of energy in a few remote places (which makes transport costs higher).
Meanwhile, the number of places running on 100% renewable energies is surprisingly high (https://en.wikipedia.org/wiki/100%25_renewable_energy#Places...). Why would we need nuclear again?