Google Store apps will have identities linked, making it harder to mass-produce and mass-publish scam apps
Enabling app installs outside of the Google store will have an additional barrier to combat scammers interactive pressure, maybe a cooldown of 24 hours or something like that
So they're addressing students and private hobbyists, but not open source and hobbyists willing to publish.
Sounds like it will be a kind of sideloading onto your own devices.
that allows experienced users to accept the risks of installing software that isn't verified
So for F-Droid, a vetting and curating publisher, users will have to go through this expert process. The announcement that activation under pressure will be prevented makes me thing of a time cooldown, like activate now, and it becomes active by tomorrow, 24 hours later.
Scamming is a real problem, and to a degree, it may end up being a good thing. As long as Google does not take this opportunity to push hidden agenda of increasing accessibility and choice, to seize more control not for security but as market and platform strategy.
F-Droid says they don't want to impersonate other projects in order to be able to publish their projects, arguably decreasing security, which is a valid concern. As long as there's a setting to allow this kind of sideloading and the use of F-Droid like before, I guess it is what it is, and may be acceptable.
If only they had started from where they are now. It's plainly obvious there's these kinds of users and use-cases. Did they really need "the community feedback" to learn about everything outside of their primary "linear" users?
Godot is certainly the easiest and simplest to install in terms of full engine and game dev IDE.
Whether they wanted to showcase or deliberately chose it for how it looks or not, I think the simple install onto a presentation desk/PC/Steam Machine may have been a reason as well.
On AniDB I can enter dd.MM.yyyy or yyyy-MM-dd (text input), which I like a lot. I often prefer reading and writing yyyy-MM-dd.
Some time ago I changed my Windows number format settings to show me yyyy-MM-dd formats. Unfortunately, that broke my webbrowsers date input / datepicker. :( So I had to go back to the standard culture format (de in my case).
The worst is when you work with dates and don't know what is what, or when the behavior is unexpected.
Probably everyone knows about the Excel shitshow of implicitly converted values.
In SQL Server, what do you think 0000-00-00 is when converted to a date, explicitly or implicitly? Well, unfortunately, yyyyMMdd is a safer format than yyyy-MM-dd.
Your question was very unspecific and broad, and despite that, now it goes into a direction I have not foreseen. Your question would have been much more useful and you would have received a lot better answers if you had provided some context, established a premise, been more specific about what you're asking.
You asked about PC. Given that Windows is the prevalent PC operating system, I'll answer for that.
While Windows has a Microsoft Store app store now, traditionally and still prevalent, most software and applications is installed and managed not through this "app store", but manually or with other non-OS-integrated software.
I feel like the premise of the question is from a very different understanding of how things work or are.
My customer gave the go-ahead to use LLM in our project very recently. We'll be trying it out. I'm interested to scope out its use and limitations especially. I'm skeptical it will increase efficiency for me overall. The project is too complex, my/our requirement on quality too high, and I'm thorough to the last var name and code formatting for readability and obviousness. I'm not sure whether I could find it acceptable to compromise on those.
Between customer communication, planning, review-prep, guiding and helping my team members, and doing reviews, and other tasks within the company, time for my own work can be reduced by a lot. Still, I have tasks I work on, and that includes coding.
Microsoft pushes cloud and AI with increasingly negative side-effects. Eventually, EU regulation steps in to require offline-capable OS with fair and obvious choice. Microsoft tries to argue security, but ultimately fails.
Microsoft continues to push and connect their services as one, with synergy effects. Eventually EU regulation and prosecution steps in, requiring a neutral OS that must not pre-install software or point to other products in OS settings and apps, etc. Integrations must be openly standardized first, before implementing their own.
Despite all this, and despite a move from EU and EU-national institutions to sovereignty through shared open source solutions, Microsoft retains their strong/prevalent market position because the market as a whole is not as strategic and concerned, and Microsoft products like office, onedrive, Teams, and their other business software and services remain a predominant and grab-first choice, and the security promise of big enterprise software, battle-tested, with strong established auth etc remains a big selling point for them.
At work, I set up convco for automated commit checks and changelog generation with custom/slightly adjusted configuration of conventional commits (types) and changelog template.
I'm sure this gets downvotes because AI. The post talks more about dotnet and webassembly though, and gains through it. Not about AI (beyond the marketing speak referencing the product itself).
How does JPlus handle compile time null checks against Java library interfaces? Does it consider them all nullable and to be handled as nullable?
If nullability information is a type metadata extension for compile-time checking, does that inevitably break on library interfaces when I create both library and consuming app with JPlus?
In my interpretation, the gains will be