sudotstar

I think that’s the rub, in my theoretical scenario, Apple is not blocking the distribution or sale of iOS applications through third-party means, they’d enforce their existing restrictions on and power over building iOS applications in the first place. Developers would absolutely still be able to distribute unsigned applications - end user iOS devices would just be unable to install them.

It sounds ridiculous to me, and as I wrote earlier, it would be a clear violation of the spirit of the DMA, but I don’t see any reason why this scenario would not be technically possible for Apple to pull off.

I’m not too sure that these actions violate the letter of the law here, even though I agree that they’re 100% in violation of the spirit of the law.

It’s been some years since I’ve put the mobile development world behind me, in no small part because of Apple’s shenanigans, but the way I understand how this might work - Apple may be required to allow “iOS software” to be installed from third party stores, but software that runs on iOS must either be signed using a certificate that only allows installation in a developer or enterprise context (which require explicit and obvious user consent to that specific use case, and come with other restrictions such as the installation only lasting for a limited period of time), or through an “appstore” certificate that allows installation on any device, but the actual application package will need to go through Apple’s pipeline (where I believe it gets re-signed before final distribution on the App Store). All certificates, not just the appstore ones, are centrally managed by Apple and they do have the power to revoke, or refuse to renew, any of those certificates at-will.

If my understanding is correct (I’d appreciate if any up-to-date iOS devs could fact-check me), then Apple could introduce or maintain any restrictions they please on handling this final signing step, even if at the end of the day the resulting software is being handed back to developers to self-distribute, they can just refuse to sign the package at all, preventing installation on most consumer iOS devices, and to refuse to re-issue certificates to specific Apple developer accounts they deem in violation of their expected behavior. I haven’t read the implementation of the DMA in detail, nor am I a lawyer, so I’m not sure if there are provisions in place that would block either of these actions from Apple, but I do expect that there will be a long game of cat and mouse here as Apple and the EU continue to try and one-up the other’s actions.

My Deck and Linux desktop regularly have shader cache updates every few days, but they’re generally tiny and finish near-instantly. I’ve never seen the behavior here of needing to download multiple gigabytes of shaders daily (and I’m thankful for it, with the frustrating data cap I have from my ISP).

I can see the use case for someone who wants a single OS install on the Steam Deck that does gaming just about as well as SteamOS does, but has a more fleshed out desktop experience (or even just a different one). The linked article goes into detail on various desktop-focused developments within Bazzite that wouldn’t really make sense for Valve to prioritize in SteamOS itself.

I expect this is simply a case of “Valve Time” on that effort. Perhaps there’s a long-term path towards more “official” SteamOS on these devices, but if there’s any area where HoloISO diverges technically from SteamOS in a way that’s not reconcilable, that’ll be problematic for offering users a “seamless upgrade”.

Long-term, I think it would be slightly more harmful to Valve’s efforts if more manufacturers started standardizing around HoloISO, so I expect that this might be a motivating factor to speed up their efforts to bring official SteamOS to third-party devices.

I am very interested in the success of this device. I have, use, and love my Steam Deck, but my biggest hopes for this form factor in the future is it using generational CPU improvements to create a more diverse set of devices, rather than just chasing higher performance.

I don’t actually play many games on my Deck that toe the line on its performance limits, I prefer to play 2D and lighter 3D games on it, while leaving the “spectacle” games for a more powerful system outputting to a much larger display at a higher resolution. I would love long-term to have a more smaller, lightweight device for portable PC gaming, and I hope that increased diversity in the market, running Linux-based systems (even if it’s all just SteamOS) will help drive towards that. I think that the pipedreams of running x86 games on Linux on ARM on a really power-efficient device, even as unrealistic as they are, are far more likely to occur if there’s a healthy market of Linux based systems, than they would on Windows handhelds given the state of Windows on ARM, and on these devices in general.

If Valve is working with Ayaneo to get SteamOS shipped on these devices, then I imagine Valve would have some level of involvement on at least the software support side, even for things specific to the device. If Ayaneo is just like shipping by using one of the existing 3rd party SteamOS installers and not working with Valve at all, then yeah I expect things to be not as smooth sailing as the Deck.

I love the DualShock 4 and DualSense controllers’ support on Linux, but I’m not a huge fan of the controllers themselves despite exclusively using the DS4 as my PC controller. I’m perfectly okay with the layout since I grew up on the PlayStation, and in fact prefer it to the mainstream Xbox/Nintendo options due to being the only controller to have a touchpad, and both gyro and analog triggers, but the abysmal battery life on the controllers has been a frustration for my couch PC gaming setup, my fairly old DS4 controllers barely last for more than 30 minutes on battery now. The biggest thing holding me back from buying a new DualSense to replace those controllers is the fact that it, too, has terrible battery life.

I’m hopeful that Valve’s desire to make a Steam Controller 2 pans out, as I expect that such a device will also provide stellar Linux support (or perhaps already does if it ends up reusing as much of the Steam Deck’s input setup as it can), and would hopefully offer much better battery life than Sony’s attempts.

I was aesthetically a fan of the Fossil watches, and was using a Fossil Sport (1st gen) for quite a while. Unfortunately the layers of proprietary-Fossil required software/watchfaces on top of the layers of proprietary-Google WearOS hampered the software experience a tiny bit, and the frankly poor hardware quality marred the experience significantly. My charging band coil in the watch completely dislodged itself (it appeared to be held in with glue), rendering the watch unusable.

Fossil’s customer support was excellent, replacing the device fully when this happened, though that was when that model was still on store shelves. I recently inquired about getting a replacement battery and was told I can just trade it in for 50% off a current-gen model, which while being far more generous an offer than I expected, still leaves me hesitant to upgrade to another device that suffers from the same problems and is in danger of being outright discontinued.

At this point I don’t really need/want a WearOS device specifically, and would actually prefer something that’s less tied to Google’s whims, the hardware OEM’s whims, and whatever the interplay is between those two companies. I’ve been eyeing more hobby-oriented projects like bangle.js or the PineTime smartwatch, but the fact that I’m even looking in that space shows that it’s become a device I would get for tinkering, not one I strictly “need”.

In this case it’s referring to the fact that the OS is built upon the same containerization technology used on cloud platforms such as Kubernetes. As a marketing tool it’s a bit buzzwordy, but it’s not about running the core OS components outside of the physical machine here.

I’m curious to hear about yours and others’ experiences with containerizing Java applications in such environments. I used to work in a place that traditionally had such restrictions on JDK versions, but after the internal IT environment moved towards running applications within containers, either on Kubernetes or on public cloud platforms’ container runtimes, that restriction became unnecessary since the application would be shipped to production alongside its compatible JDK.

While there were still restrictions on exactly what JDK you could run for other reasons, such as security/stability, common developer experience, etc, it at least allowed teams to immediately adopt the newest LTS release (17 at the time I left) with little restriction.

I’d probably pick something esoteric and then just stop programming, tbh. I enjoy being a polyglot programmer, and learning many languages and learning from many ecosystems is incredibly interesting to me, far more than hyper-specializing in a single language would be.

It’s unfortunate, but it’s understandable if effort needs to be focused on a single good UI widget ecosystem fully under Mozilla’s control, rather than living by the whims of the three major desktop UI toolkits they have to support, as well as the hundreds of thousands of web pages that are exclusively designed and tested against Chrome which already has been using non-native widgets across desktop platforms for a very long time. I’m not in the web dev space anymore, but I’d constantly see sites built that were incredibly dependent on the exact pixel sizes of widgets as they would render in Chrome, and would visually fall apart on Firefox, or with other zoom/text size settings.

UI design across Windows, macOS, and Linux GNOME/KDE have converged enough that it’s probably good-enough if Firefox continues down the path of just theming their own widgets with the OS/user’s color scheme where applicable, and calling it a day.

If I’m not literally touching the content I’m trying to scroll, I’ll stick to the default orientation (scrolling down moves content down). Wayland touchscreen input handling seems to handle this just fine and not couple touchscreen scroll direction to trackpad/mousewheel scroll direction.

It might be somewhat controversial of a take, but to me an awesome-performing Proton version of a game is far better than a Linux version that may be native, but has severe deficiencies and/or lags behind its Windows version.

To me, my favorite native Linux games would be ones that do things on Linux that are not possible on other platforms. Generally, this would be an “unfair” advantage, as games should strive for feature parity on all platforms within reason, but so often we end up being on the wrong side of that equation that seeing some of the perks of the platform is nice.

To my knowledge, the only major game I can think of that does this to a certain extent is Factorio, which enables non-blocking game saves on Linux and macOS and not Windows. It’s not a Linux-exclusive feature, but it’s nice that the developers went through the effort to implement the feature on Linux even though it’s not possible on Windows.

I wasn’t a fan of the built-in weather widget so I immediately replaced it, so I have no experience with it being nonfunctional, sorry.

Yes

It’s not like Nova Launcher, but I’ve been really enjoying Kvaesitso. It does support icon transformations (better than Nova does IMO) but provides a very different interface that’s focused on search.

The reality is that the number of games, even AAA ones, that are releasing at that high a “minimum” performance requirement is incredibly small compared to other games that do release with more modest system requirements. Games that are “just good enough” graphically to go along with their gameplay tend to be the norm, I think, with the few games that really go for pushing visual fidelity being respectable in their own right but not frequent enough to fret about. What will matter the most is what games you want to play and what their requirements are, and that’s basically impossible to project out 1, 3, 5 years out or however long you expect the hardware to last.

For what it’s worth, I have a Steam Deck and spend a lot of time playing on it, but pretty much every “AAA, big budget => big graphics” game I want to play I’d exclusively do so on my gaming desktop (or remote play on Deck if I want to play it there at all), while sticking to 2D and lighter 3D games on the portable device directly. This is mostly due to what kinds of games I enjoy playing on what form factor, as for example my decision on what to play docked vs portable on the Switch is much the same way, and for about a year after buying the Deck, my desktop hardware was so out of date it was getting generally worse performance than the Deck yet I’d still use the desktop for “spectacle” games, but the necessary graphical quality to go along with that tends to correlate well.

Additionally, it’s devices like these, that have proven successful in the market, that incentivize Valve to continue Proton’s development. It’s hard to see given the already insane trajectory Proton’s development was on before the Steam Deck, but now that getting games running on Linux (in at least some form) is desirable by many game developers in order to gain Steam Deck support, Proton compatibility guarantees, and the corresponding development to make that happen, have shifted to before the releases of many major AAA games, and that compatibility work has cascading effects for many other games as well.