I'm sure they intend to do something like that, but unfortunately time/funding is probably getting in the way. Making an app like that is not exactly cheap. You need either the right combination of skills or a lot of trial an error, which is either high rate times short time, or low rate times long time. So they probably released this intermediate work to try to gin up some interest to get more funding.
Good luck to them. Never really worked out that way for me in my efforts.
What if the keyboard was put in the user's off-hand and they typed on it by tapping their palm? Then the keyboard wouldn't be in a fixed position to correlate eye movement against it.
I know it's really tired, everyone knows it, and it is repeated constantly:
Drugs affect different people in different ways.
If 20% of ozempic patients have personality changes, that is very significant. Yet you would still have 80% of users going around saying they never had any issues with it and those who are might just be doing it wrong/misunderstanding/confused etc.
100% of patients who have taken ozempic have had a personality change. The change is that they're now on ozempic. This causes many of them to eat less. Which is also a personality change. So the more useful question is how does their personality change, how much does it change, and is it a desirable change, according to them or others.
There wasn't even a cleaning crew or custodian or security person that at least passed by?
I mean, this isn't some frothy AI startup that has approximately zero chance of having any impact on the world. This is one of the largest banks in the country. They should, at the very least, have someone checking every office and cube, every night, to make sure no unauthorized personnel are there after closing.
It's also not like they wouldn't have reasons to double/triple check in this specific case without needing a general patrol policy: It should have been a security concern at the end of Friday that someone had badged in and not left. That's among basic reasons for having these badge systems, in the first place. Someone failing to badge out is a security risk. (It's also a Fire Marshall risk because another reason companies require accurate badge out is that in the case of a fire emergency and they have a badge in process, they are then liable for knowing every employee inside the building in the event of that fire emergency and helping make sure everyone gets out safely.)
Even if the company encourages sleeping at your desk overnight "for productivity", because companies sure do love underpaid labor and accidental burnout, there's an increased OpSec risk for every extra hour someone has access to secure corporate resources. You'd think in a bank where security is supposed to be important at least one security software would be paranoid and presenting alarms about someone still badged in over a full weekend.
I've had a mixture of the two in my years. The growing "Fire Marshall liability" thing (which obviously depends on your state, and other factors) has been pushing companies to do both, even in cases where they weren't convinced for security reasons alone. (Though security reasons alone you would expect a bank to require both in and out.) (ETA: And the article implies that they did have a badge out.)
No, Microsoft's .NET only supports WinForms on Windows. They do have an official cross platform GUI toolkit in MAUI, but it strangely does not support Linux.
Last I knew it is also considered pretty lackluster. Every time I read up on it it feels like, even beyond the lack of Linux support people just don't care for it.
If I was building a cross platform native app with .NET I'd probably use Avalonia right now.
Yeah, the took an age delivering it, then it came out and most of the early reports were “It’s still not ready.” and then I think Microsoft just gave up.
I think not supporting Linux was a tactical error, though. Some people will put up with a lot for Linux GUI support, and some of those people are the types who can resolve problems with your half-baked GUzi framework.
Does it really need help? I struggle to imagine a scenario where one would consider MAUI not supporting Linux to be an issue (if we discard superficial bad faith concern) when Avalonia, Uno or, if you care about Linux as the main target, Gir.Core exist.
And, at the end of the day, you have a tool with an extremely rich FFI capability so whatever is available from C you can use as well.
Sorry I clearly was not clear enough. I mean specifically an issue with MAUI itself. I agree dotnet/c# have some solid UI options cross platform at this point. MAUI however seems to be at best a mess and at worst dead in the water.
> "The future is already here – it's just not evenly distributed."
Were I live and work (IT and consulting in central south-east Norway) it has been the year of the Linux Desktop on and off since 2009.
That was the first time I worked full time at a place that deployed Linux for everyone and everything that didn't have a verified reason for needing Windows.
I think we had one 3rd party trading software running on a Windows machine and maybe the CEO and someone in accounting got Windows.
Everyone else was upgraded to Linux and it worked beatifully. It was my job to support the sales department with desktop related issues and it was absolutely no problem to do it while also being a productive developer.
Since then I have not worked on a place that required Linux, but I think most of the places I have worked on since has had Linux as an option as long as you supported it yourself, and some places also have been very active writing how-tos and working with me to troubleshoot issues that were related to Linux, since many of them were also Linux users.
At the moment I use Mac, but at my current job I'm also allowed to use Linux.
Open Source Support reasons. If Linux developers want better MAUI support there is a "Community Repo" to contribute to and help move things further along. The impression is that if things were further along it might get formally "adopted" (by the Dotnet Foundation) for "official" out-of-the-box "support", but it isn't far enough along and doesn't seem to have enough contributors with enough momentum. It currently seems that the Venn Diagram of "Developers that say they want MAUI support for Linux" and "Developers that would contribute to Linux support for MAUI" has too small of an intersection.
Sure, Microsoft could pay more employees to work on it faster, but Linux loves and prefers open source from Linux devs "untainted by Microsoft", right?
Contribute to the Maui backend for GTK and/or Qt, nothing is stopping you
Alternatively, just because you're on .NET doesn't mean you need to use Microsoft sanctioned UI toolkits, just as C++ has no "official" UI toolkit. You're free to pick up some GTK or Qt bindings if you want a native feeling and your application is already architectures correctly. Alternatively, throw Imgui at it if you just need dev tooling, or maybe try other cross platform toolkits in the ecosystem like Avalonia or Uno
Latest Visual Studio update has made CoPilot much more prominent. I can't tell if it's actually running or not. I can't figure out how to turn it off completely.
LLM-based AI is technically banned at my work. For somewhat good reason: most of our work involves confidential, controlled, or classified data. Though I've seen a lot of people, especially the juniors, still using ChatGPT every day.
Also noticed the UI has gotten a lot slower. I'm guessing the two things are related.
If my company wasn't locked into "Microsoft everything" this would push me the last inch to ditch VS completely. I already did at home.
The VS one is an entirely different “Copilot” to this one. This article is talking about the Copilot that’s a RAG ChatGPT that integrates with Sharepoint etc.
I suspect it's like a lot of DoD-funded research. The work gets done, the report gets written, and then it goes on a shelf (or rather, some ancient SharePoint site on a classified network with no functional search ability). When a crisis occurs, the word goes out to everyone, the one lifer analyst spending too much time in windowless rooms at the Pentagon pipes up about something he remembers, and then everyone is satisfied that the work got done, "the problem is understood". Everything goes back to normal until it happens again.
I've been using Entity Framework for the last 5 years and have not encountered this issue, as long as I've got all my Includes specified correctly.
There is also the AutoIncludeAttribute that you can specify on entity fields directly to always include those fields for every query.
My main complaints with EF are that the scaffolding and migration commands for the CLI tool are nearly impossible to debug if they error during the run.
But when they run right, they save me a ton of time in managing schema changes. Honestly, I consider that part worth all the rest.
There can also be some difficulty getting queries right when there are cyclical references. Filtering "parent" entities based on "child" ones in a single query can also be difficult, and also can't be composed with Expression callbacks.
But in any difficult case, I can always fall back on ADO.NET with a manual query (there are also ways of injecting manual query bits into EF queries). Which is what we'd be doing without EF, so I don't get the complaints about EF "getting in the way".
Good luck to them. Never really worked out that way for me in my efforts.
reply