ever worked in IT support? letting people customize their environment both increases the amount of support that users require, and increases the difficulty of providing that support.
a laptop in a stock configuration can be swapped out for a new one when it breaks. a laptop that has three years of accumulated customizations installed on it means that the employee wants their laptop back when it breaks, and they want it fixed ASAP.
when you're supporting a user who doesn't know how to type a URL into their web browser, it's a whole lot easier if you don't have to start that call with asking what web browser they're using.
It mattered in the sense that it provided valuable grist for the mill as they attempted to figure out if it could work as a Reels/TikTok alternative for companies to eventually deluge with ads.
But why not Bluetooth or NFC? I can’t imagine any way those could be annoyances, or even why websites would want them outside of some extremely specialized applications.
I'm personally a WebUSB, WebBT etc hater but I totally get why PWA developers want those features. For example, let's say you're manufacturing some sort of USB device and you need a way to flash drivers. The idea of being able to just make a webpage that can update your drivers is so appealing compared to having to ship apps on Windows, Mac, Linux, iOS and Android.
Similarly, if my bank website could do NFC tap-to-pay securely, that would be pretty cool. I can imagine lots of interesting opt-in uses for NFC in a webapp.
Arguments that these features are held back by Apple specifically in order to keep apps on the app store where they can control things and take 30% at least hold water, I think, even if that reasoning doesn't apply to Mozilla rejecting features.
> The idea of being able to just make a webpage that can update your drivers is so appealing compared to having to ship apps on Windows, Mac, Linux, iOS and Android.
I suspect like many here, at $work we use a shit-ton of Flexoptix SFPs.
Flexoptix are not a $megacorp, they are a (very) small German company.
They manage to ship cross-platform apps to flash the SFPs. So its really not that difficult.
I would think a web app would be more of a pain the the butt to maintain because you have to deal with CSS reactive UI etc.
For little utility apps where you don’t care to deviate from UI default appearance and behavior (and, as a user, it’s much better if you don’t anyway, though it’s very trendy to make UX worse by customizing everything) iOS and Android both are dead simple, very easy to write and maintain a utility app for either of them.
An enormous amount of the cost of developing a lot of native apps is customizing the appearance and behavior, to match some slide deck mockup or to make it “on-brand” or whatever. It’s better for the user, and way cheaper, if you just… don’t do that. Hell a lot of common UI elements are easier in native than web if you just don’t try to customize them a ton (data-backed tables and list views and such are sooooo nice)
I like WebUSB in Chrome to update my Meshtastic radios. I also like that I have to go out of my way to launch Chrome for that, and other websites can’t request permission to access local hardware in my normal browser.
Nonsense. Firefox on Android (and I'm sure everywhere else) allows endless redirects without prompting me about the redirect, sending me to sites I never wanted to go to.
And even then, where does one draw the line on "privacy"? Especially given that every other app on the user's phone is granted every permission under the sun and feeding on as much data as possible.
The core of the problem isn't supporting web bluetooth etc or not...the core of the problem is that dumbass humans will go "yes use all the permissions" because their hands are already shaking from tiktok withdrawal.
does anybody know what level this change happens on? is this change going to affect ubuntu desktop users on any system they ssh into, or will it affect all users of a ubuntu server who have ssh'd in?
It's the sudo binary installed on the host -- if you're SSH'd to a 26.04 host, you'll see stars; if you're running 26.04 and SSH to a different OS, you won't (unless the remote system is also 26.04 or otherwise using rs-sudo)
It’s wild to me how people are so fixated on this. Yes, obviously winter driving has challenges. And also obviously, the leading self-driving company has thought about that.
They’re preparing to launch and have already been testing in Chicago, detroit, Minneapolis, Denver, Philadelphia, Boston, NYC, and London. I think it’s safe to assume they’ve considered winter driving.
"protocol" is just an agreement to communicate in a standardized way. this is a protocol. a tool call exposed to the agent is a protocol - the act of "exposing it to the agent" means you're defining a protocol.
there's nothing wrong with calling this a protocol. the problem is in hyping it up as though every protocol is going to be world-changing on the level of TCP.
>Name one bad outcome a reasonable parent would care about that's prohibited under these bills.
the bad outcomes don't need to be prohibited under these bills. it's already illegal to, for example, distribute pornography to minors. which i think is something that a reasonable parent would have a problem with.
but if there is no way to determine who is a minor and who isn't, then it's impossible to determine the difference between "willful negligence" and regular old negligence and enforce any consequences for breaking that law. age verification laws are about mechanisms to make other, already existing laws actually enforceable.
It’s completely inconsistent for me, and any time I start to think it is amazing, I quickly am proven wrong. It definitely has done some useful things for me, but as it stands any sort of “one shot” or vibecoding where I expect the ai to complete a whole task autonomously is still a long ways off.
Copilot completions are amazingly useful. chatting with the chatbot is a super useful debugging tool. Giving it a function or database query and asking the ai to optimize it works great. But true vibe coding is still, imho, more of a party trick than an actual productivity multiplier. It can do things that look useful, and it can do things that solve immediate self-contained problems. but it can’t create launchable products that serve the needs of multiple users.
reply