I'm really not into math and got really lost in the second half of "Adding points on a curve". Just don't understand what the author wants to tell me with the grouping and the role of the identity element, which is called infinity but is zero?
However, after looking at the next section and playing with the chart I immediately got the idea where the whole article is heading. Interesting to see how this works.
There is a slight bug on the interaction. When you set P=Q or for example you can't get the one P at the top and Q at the bottom. The lines disappear.
Basically you need the "infinite/zero" point to compensate for a situation when you have two points completely perpendicular to the x-axis. AKA it is not intersecting a third point. So it intersects this special "infinite" point.
And conceptually why you need this "infinite" point is that without it you can't add points together properly.
Say for counter argument instead of doing this "flip or mirror" across the x-axis (in the interaction it is the red dot appearing). And instead the red dot just appears on the same side as the two points being added on the curve - without the flipping.
If P1+P2 = Q instead of this Q' that is flipped.
And P2+Q = P1
If you try and add P1+P2+Q you would get either Q+Q or P1+P1 depending on if you did (P1+P2)+Q or adding up P1+(P2+Q) which are not equal.
so you need this red dot flipping thing happening in the interaction. However, if you have this flipping that means P1+P2 = Q' which is the mirror flip of Q.
So Q'+Q need to equal this special infinite/zero point to ensure associativity works.
Just to toss on some info you might already know, the mention about grouping is related to group theory. [0] If a set satisfies those 3 axioms, there's some assertions you can build off that are common to all group theory sets, and having an identity element is one of them. It's weird that it's NOT zero, but in this case, infinity behaves LIKE zero. (Imagine going infinitely along the curve on the x-axis towards the open part of the curve, so therefore going infinitely up/down the y-axis. At somepoint, you're essentially have a vertical line between the original point, and your infinitely far away point, which points at the exact opposite side of the curve, which reflects back to the original point.) For natural numbers, zero is the identity, since X + 0 = X, in the same way P + infintelyfarawaypoint = P in this set.
To use a dumb analogy, it's polymorphism where your interface is something like regular old natural numbers: as long as your class behaves like natural numbers in some key ways, you can pass them to any add()/subtract()/multiply() functions relying on that behaviour.
Unfortunately, I can second this, both as a developer and a user. His IMHO childish behavior has ruined his image for me, and is not a good lighthouse for the Fediverse itself. Also, as a OSS veteran myself, I see it extremely critical that he is starting new projects all along, denies to get proper help and build up a maintainer team, and leaves older projects in the dust. Pixelfed is the one product he might should focus on, jet it feels like the platform is in maintenance only mode. Pixelfed is a wonderful addition to the Fediverse and deserves to be on good hands.
Maybe, and this is a very personal opinion, his product success and the Kickstarter campaign raising over 100k made him feel like he's better than everybody else. And one can see the effects.
The sad part is, that Apple used to make somewhat stable, functional software. I started with the iPhone 3 and a bit later with Mac OS Snow Leopard.
It all started when Mr Cook decided to serve the shareholders, instead of focusing on Apple's core values. The software went downhill in such a speed in just a few years. And moving out of the ecosystem is a painful, if not unbearable, task that barely anyone loves to do. At least I can't even think about moving back to Android.
I recently tested Swiftkey after Typewise is sadly abandoned. It's sooooo much better than the stock keyboard. Not only is the auto-correct working incredibly well (garbage like witjoit is correctly transformed to without, which Apple Keyboard can't), Swiftkey also manages multi-language typing astonishingly well. Last but not least, I can customize it. I am also not signed in to my account, so no settings or whatever is stored on Microsoft servers.
I consider moving away from Github, but I need a solid CI solution, and ideally a container registry as well. Would totally pay for a solution that just works. Any good recommendations?
We can run a Forgejo instance for you with Firecracker VM runners on bare metal. We can also support it and provide an SLA. We're running it internally and it is very solid. We're running the runners on bare metal, with a whole lot of large CI/CD jobs (mostly Rust compilation).
The down side is that the starting price is kinda high, so the math probably only works out if you also have a number of other workloads to run on the same cluster. Or if you need to run a really huge Forgejo server!
I suspect my comment history will provide the best details and overview of what we do. We'll be offering the Firecracker runner back to the Forgejo community very soon in any case.
I actually went through some of the issue/pr stuff for the forgejo project after I asked you. It seems like things are moving along nicely and you seem to have found a welcoming environment in their repo. I will keep an eye on that progress. Thanks very much. I do not have a pressing need but firecracker runners would be pretty awesome to have.
Long time GitLab fan myself. The platform itself is quite solid, and GitLab CI is extremely straightforward but allows for a lot of complexity if you need it. They have registries as well, though admittedly the permission stuff around them is a bit wonky. But it definitely works and integrates nicely when you use everything all in one!
Should our repos be responsible for CI in the first place? Seems like we keep losing the idea of simple tools to do specific jobs well (unix-like) and keep growing tools to be larger while attempting to do more things much less well (microsoft-like).
I think most large platforms eventually split the tools out because you indeed can get MUCH better CI/CD, ticket management, documentation, etc from dedicated platforms for each. However when you're just starting out the cognitive overhead and cost of signing up and connecting multiple services is a lot higher than using all the tools bundled (initially for free) with your repo.
I have searched for a proper keyboard replacement. But there is not a single one that 1) works properly without major bugs, 2) adheres to privacy standards (because no way I am sending all keystrokes to Microslop) and 3) does not cost a fortune. Typewise came close, but seems abandoned now.
I really wish Apple would allow us to swap out the Finder with something else, so files open in that other app instead of the Finder. This works reasonable well on Windows, where I "replaced" the Explorer with Directory Opus.
This used to be possible, I remember that I replaced Finder with some other app many years ago. I strongly assume that this doesn't work any more, though.
I recall you used to be able to flip some bit somewhere to allow you to Quit the Finder, but I assume that's disappeared inside the encrypted and signed partition where Apple keeps all the things us stupid users shouldn't be allowed to touch.
But even then, you'd want more than just that, as when you tell the OS to "Reveal" a file or open a folder, that's the association I'd want to be able to change.
Honestly I'd really prefer the Windows XP File Explorer to the pile of crap the Finder has turned into.
> But even then, you'd want more than just that, as when you tell the OS to "Reveal" a file or open a folder, that's the association I'd want to be able to change.
yep, that should just be a normal setting like default browser (one thing i like about linux nowadays)
Ah yes, thank you for reminding me - of course, it was Path Finder! You could even have it respond to "Reveal". Not sure any more if it was by renaming Path Finder to /Applications/Finder, or by changing its Bundle id to com.apple.finder, or some other trick.
"of course, it was Path Finder! You could even have it respond to 'Reveal'"
This still works. I have been using macs since 1985 and have always hated the Finder. In the days of classic Mac OS, my go to for file management was a desk accessory called DiskTop, which was great. Super fast and easy to operate from the keyboard.
When I switched to OSX, I needed something better than the Finder, chose Path Finder, and have been using it ever since. I have my complaints about it but have not been able to find anything I like better.
I've also found many Actions that do other dodgy stuff, like pulling and executing unpinned scripts from external websites, or installing unpinned binaries from GitHub releases. Pinning an Action isn't enough, you have to audit it.
This may be better than Google, but I don't understand why a blog needs a font CDN at all. Just use standard fonts or host them yourself if you really can't do without your SuperCoolFont.
For real - computers come with a shitton of nice fonts these days. Plus like, font styling allows fallbacks -- choose a nice font stack where all the fallbacks look nice, and you're good to go.
The problem is that you also can't use the majority of local fonts people already have installed because advertisers started requesting those on hidden canvases to try to pinpoint segments of users based on which fonts you have installed, so `local('Some Font Name')` in CSS now sometimes just doesn't work in some browsers to prevent that privacy leak.
Advertising networks: the exact evil for why we can't have nice things on the web.
However, after looking at the next section and playing with the chart I immediately got the idea where the whole article is heading. Interesting to see how this works.
reply