In that case, it will be their own dang fault. Modern MacOS is a mess of update nags, unwanted notifications, and permissions prompts. We want sandboxing and security patches but the UX around it is abominable.
Which are these "world-class delivery organizations" and are they the exception or the rule?
We can acknowledge that political volatility is a main cause but it's not some exogenous factor. It's inherent to the federal structure of the country and it hamstrings trying to build social goods, whether that's transit or healthcare infrastructure.
There is also nowhere near a culture of developing and trusting institutional planning expertise. Infrastructure is done on a pork-barrel basis of which promises will get who elected and create which jobs and allocate which contracts. Or who complains the loudest about the design of any given plan.
Canada's 20th century social system was also based on maintaining social stability through mass property ownership, which is now breaking down as unrestrained property speculation is displacing any kind of productive investments (while also ending the possibility of that mass property ownership in the near future).
Sorry to bring the negativity but I feel as a whole that Canadians are much too tolerant of institutional dysfunction (in the manner of the classic "Canadian nice") and think our society is far more advanced than it actually is. It's a completely complacent and naive culture that is quickly being left in the dust by more functional systems.
Last weekend I was writing some quick notes in the Notes app, and I could not get it to stop performing nonsensical completions in blinking yellow text. Apple Intelligence disabled, predictive text disabled, various combinations of backspace escape etc. Nothing worked. How hard is it to code a Notes app that doesn't mess with you?
I really wish homebrew supported this feature. There are a bunch of terminal tools I have installed through it (and wouldn't make sense to sandbox) that are basically one-person projects, and I imagine subject to supply chain compromises. I don't urgently need the very latest versions for any of them, as they are unlikely to have exploitable security vulnerabilities on my machine (due to not being involved with any network facing service).
I do kind of think the solution to this issue lies at the OS level. It should provide a high degree of UI and workflow standardization (via first party apps, libraries and guidelines). Obviously it's an incredibly high bar to meet for volunteer efforts, but the user experience starts at the OS level. Instead of even installing a program like "Handbrake" or "Magicbrake" the OS should have a program called "Video Converter" which does what it says on the tin. There should also be a small on-device model which can parse commands like: "Convert a video so it can play on facebook" and deep link into the Video Converter app with the proper settings. Application-level branding should also basically not exist, it's too much noise. The user should have complete control over theming and typography. There has to be a standard interaction paradigm like the classic menubar but updated for modern needs. We need a sane discoverable default shell language with commands that map to GUI functionality within apps, and the user should never be troubled with the eccentricities of 1970s teletype machines.
> The real innovation of a lot of these alternative DVCS systems is that they free the state of the source from being dependent on the history that got you there. Such that applying patches A & B in that order is the same as applying B' & A' -- it results in the same tree. Git, on the other hand, hashes the actual list of changes to the state identifier, which is why rebasing results in a different git hash id.
Anybody who's wrestled with reordering/rebasing git history or has done git archaology is able to understand this benefit.
From Pijul's site:
> Pijul is the first distributed version control system to be based on a sound mathematical theory of changes
After years of grudgingly tolerating using a deployed prototype for a VCS, yes, I want the mathematically sound alternative.
All that being said, I do wish you the best, because truly, I am tired of git and JJ does seems like an improvement.
Agree, the feature of commutative patches just seems obviously superior? Not sure why there's a critical mass to adopt jj over Pijul apart from jj's git backend.
I can easily imagine a world where you could install an open source PWA from an archive file into its own security sandbox without any further hoops to jump through, and it continues to just work indefinitely because the web has very good backwards compatibility guarantees. Instead, we have to get licensed and notarized by the monopolies and they keep you on a constant treadmill of drudgery just to stay up to date. Or you install somebody else's monopoly-approved "legitimate business" app which steals or leaks your data. Sad!
I think this is amazing work and I'm looking forward to trying it out and maybe adopting long term, but I can't help but think they're selling themselves short with the default theme. Personally find the electric blue pretty grating. Would be nice to see more examples of "calm" theming with more neutral tones, or even a preset alternative default for that.
If you're an amateur and really want to spend this much money, get a good enough knife (e.g. the classic victorinox) and a chef's choice electric sharpener, and you won't have any more issues.
I think people are dismissing the possibility of universally accessible self-hosting too quickly. We really need to be ambitious as engineers and imagine a future where people have sovereignty over their own computing. In 1990, we wouldn't have accepted someone else taking custodianship of our personal documents, memories, books, music, and films, and yet it's normalized today. With the benefit of hindsight, we need to brutally simplify every single layer of the stack and optimize it for being usable by non-technical people.
The better comparison are those operating systems which tried for mobile-desktop convergence without adequate consideration of the differences between those platforms, thereby making the desktop experience worse and alienating users. In fact, it's not even as well considered as that, since mobile was a well established phenomenon at that point, whereas AR is still very niche.
The comparison to the evolution of iOS is misleading. With iOS, they introduced users to a new platform by using familiar and appropriate design language on that platform. With the current redesign, they're using design language that is really geared towards AR on non-AR devices. The design is in service of devices that most people don't use and haven't showed much interest in using.
This is an existing and somewhat nitpicky issue, but it's also annoying how they specifically insist on rounded corners "because that matches all modern devices" in the announcement. Pretty much all third party external monitors don't, and even their latest top line laptops only have them at the top of the screen. So we're stuck with these dumb little triangles of background peeking out. It's kind of the "charging port on the bottom of the magic mouse" of MacOS.
I agree, I think it extends to anybody who wants a calmer experience or has vision trouble or strain. I guess you can turn those options off but if the aesthetic appeal of the design is based on them then I assume we'll be getting a second-class version of it. I was already leaning towards switching to Linux for other reasons but I think this is the thing that finally pushes me there. I think optimizing for VisionOS is quite a bad idea from a UX POV, since they're two entirely different usecases. With augmented reality you need and want to see things in the background, whereas on other devices you don't. It's a fairly fundamental difference, and it's sad that they chose to go this way in my opinion.
I think the original Material 3 was more sensible. Coming from ios-land, its color palette formula was a nice break from the glassy apple standard. It looks like this new version goes overboard on the boldness and especially on the typography. Looks like it uses the same palette generator but is encouraging use of the more vibrant colors in it? Seems to draw inspiration from magazine design (content is ads or interspersed between ads, appropriately enough for an ad company). Indeed, the brief is written like a pitch to "brands" (not UX designers for useful apps). The reference to "dialing up the feeling" makes me think of the extremely uncomfortable feeling I get when watching TV or film ads for the first time in awhile -- that it's a manipulative assault on my lizard brain.
In my opinion, the main promise of M3 wasn't apps triggering "more feeling" in users, but in users' ability to personalize their color theme to their taste to make themselves feel more comfortable and at home. Instead of this iteration, the best direction for google would have been to push apps to implement dynamic color (themeability) as a matter of best practice, increasing user sovereignty over their experience.
The example touting the huge send button with the tiny email composition space is baffling. This is peak "call to actionism".
Dot cursor that doesn't let you select text properly and link highlights that aren't visible against the page background are not a great look.
Side note, surprised to see some commenters wishing for the return of Material 1. I simply couldn't get over the aspect that each google app chose an arbitrary garish primary color, and you were stuck with that as the user chrome for that app. I still cringe when I see docs formatted with (for instance) random bright purple chrome using a M1 theme.
Just the advantage of being available at convenient times, rather than in the middle of the day sandwiched between or immediately after work/school is huge.
That's funny cause in my experience, Elixir docs are some of the best I've used, and a major strength of the ecosystem which I miss when I'm not writing Elixir. HexDocs has been integrated and standardized from the start, leading to a consistent experience across all packages. The formatting is clear and visually pleasing. All libraries you would want to use have adequate docs. It also includes sections for guides and cheatsheets, avoiding a bifurcation between those an API docs. In general, it's really easy to find the function being called because Elixir mostly avoids duck typing, and you can just directly look up the module in question. All in all, I seriously miss the docs when programming for example in Ruby (slow, ugly, rdocs which many projects don't adopt or adopt half-heartedly), or especially Javascript. In the latter case, the trend of making flashy marketing sites instead of using a standardized tool is a serious pain. I really wish the community would settle on a high quality standard. Deno's JSR docs looks kind of promising in how it's positioned, but I don't think the formatting and usability is that great so far.
This is a great site, and thank you for the effort.
One suggestion for an addition to the section on FOSS: Related to the issue of not being able to modify the source of apps we use, we also can't verify that an "open source" app on iOS is built from its claimed source code. We just have to trust the developer. This blocks true auditing of iOS apps for data privacy practices, something we know is needed given that the "privacy labels" are often deceptive https://archive.ph/Ak6qU. As such, this is a data security issue as much as a user freedom issue.