I'd never be able to go back to scrolling with "skeleton rows" that's the web default because it can't render fast enough..

I don't usually use windows, but my employer uses Office 365, so I've been using Outlook web for several years now. It felt much snappier and more pleasant to use than the native client. I understand there are some features missing, but since I'm a very basic user (the most advanced thing I use is combining messages into conversations), it never bothered me.

The new Outlook UI looks extremely similar to the web one, and the app feels snappier on my machine. The thing I dislike the most is the empty space at the top of the window, but I guess that's just the current fashion.

However, there may be additional reasons for building new products with web tech. Someone already needs to maintain the web versions of these applications. If you're going to offer web access to Outlook, a team needs to be responsible for maintaining that web application. By replacing the native desktop app with a wrapper around their existing web application, Microsoft reduces the need to separately build and maintain features in both platforms.

Companies (actually the information security teams) may be taking shortcuts based on Microsoft’s recommendations to reduce their work while imposing a huge cost on every worker.

It used to be that companies were locked in with Windows and MS Office. But there were some alternative options for those. IMO, now companies are heavily locked in to Microsoft by subscribing to Microsoft365. There is no migrating out or exit path for this. They just have to suffer through whatever Microsoft puts them through.

Modern commercial UI/UX design really has drifted far away from "good." Photoshop suffers the same problem - 10 years ago the UI was quick, responsive, and efficient. Now it feels like wading through mud.

  https://www.postbox-inc.com/

Won't be for everyone - it's not open source, not free and there's no Linux version (yet?). Personally I'm happy to pay for a professional / premium email client experience, since it's so mission-critical to my work.

Generally I worry about the creeping consumerization of Office apps and what it means for power users. We're still out here, and we need power tools that work for us. Have the big space eye candy if you like, but please give us the option to turn the bling off and get down to work!

Context menu key does nothing? That's weird -- I must need a new keyboard.

I think it's more about trapping you in the MS365 silo than anything. They want all your data in the cloud because it makes it makes it non-portable. I think the big push they'll make after the desktop apps are replaced will be about data governance because it let's them increase lock-in.

First they'll make a 100% online workflow a possibility. Next they'll convince businesses to abandon local files because it's "too risky". Then they'll start locking down everything that could be used for "data theft", including copy / paste. At that point they effectively own your data.

And it's not far fetched for them to lock down things like copy / paste. I think that's why there's been a big push for TPM and things like Passwordless. It's a solid base for authenticated / authorized actions. Yeah, they'll use it for authenticating to websites first, but there's no reason a similar system can't be used to support signed / authorized actions between apps.

So, when you copy text, it doesn't go straight onto the clipboard. Instead, the office app uses a TPM managed key to send a signed request to MS365. The TPM will have a key enrolled for office apps and will only encrypt/sign for trusted (think code signing) office apps, so Microsoft knows the request came from an official office app. You'll be logged in, so MS can check if you're authorized to copy the content to the clipboard. The public half of your TPM managed key will be stored in your MS365 account.

If you're authorized to copy content to the clipboard, MS will encrypt it with your public key and send it back. Your clipboard will have an encrypted copy of the text. When you go to paste, the receiving app will need to ask the TPM to decrypt the payload and (remember) the TPM will only let the key be used for a specific set of trusted apps, so the copied text will never touch an app that's not authorized.

Think of taking Passwordless, adding a TPM as a requirement, and using it for a workflow of app actions rather than auth actions. That's where we're headed IMO.

But, yeah, I wholeheartedly agree that they should either develop or use a native framework (Flutter perhaps?) rather than trying to shoehorn a technology that has never been very good for application development.

Go ahead, I'll wait.

Do Mac computers not encrypt their drives? Like Bitlocker on Windows?

Having sensitive information on your laptop should not be a worry if your IT team does their job right.

I'm personally using various free alternatives to Microsoft's offerings - Thunderbird for mail, LibreOffice for office work, something like Nextcloud for cloud file storage etc.

While things aren't always great, it's still surprising to me how viable this alternative path is nowadays, especially when it feels like control over the software we use is getting taken away from us (for a variety of reasons).

Either way, software that meets the users and their needs where they are is great! Someone will benefit from cloud based software, someone else with local software that's very popular, whereas others will just be happy to get by for free.

Of course, it's also great that we even have those options and alternatives in the first place!

There definitely ARE limits to it -- I've never tried to merge personal stuff into my work cal, as you seem to want to do -- but for what it does, it's the best I've seen.

I used the new Outlook just long enough to see it was horrible at using existing rules and making new ones and just didn't try anything else. But I can't imagine why they would fix one little thing like properly handling inline quoting when they wrecked the remainder of what had been a decent corporate email app.

That's OK though. If you look at this thread, you can see there's been a shift in what people mean when they say "native". It used to be well defined: Win32 on Windows, Cocoa on OS X/iOS, GTK or Qt on Linux, app written in a C-derived language. Nowadays people are using it to mean "not web".

Why has this occurred, well, most people don't really care what exact libraries or languages an app uses. What they care about is the user experience. People are using the word native to mean stuff like, works offline, works with files, isn't too laggy, consumes reasonable amounts of RAM, integrates well with the OS. And above all - uses a sophisticated GUI toolkit. A big part of the native vs web experience gap is that although HTML has a lot of styling functionality it's a seriously impoverished UI toolkit, barely reaching Win 3.1 standards out of the box. So people roll their own widget toolkits that they layer on top, but none of them are all that good and the modern culture of conflating widget toolkits with design systems means they are constantly being tossed out and redone from scratch (this isn't unique to the web, Jetpack Compose has the same issue).

The result is that many widgets taken for granted by Windows 95 users like splitters, menu bars, status bars, pervasive keyboard accelerators, virtualized and editable table views, tree views, customizable toolbars, draggable tabs, floating tool windows etc are just gone and power users are the first to notice this.

You can therefore satisfy people's desire for native apps by providing that sort of UX regardless of what actual tools you use to do so.

Before that was a neat Linux project Pyrodesktop (2007) which was an x11 window manager using Firefox guts to render. There was also a trend of trying to mate Javascript technologies to gnome back then, with efforts like gjs seeing some adoption. I don't know how popular it is, but a spinoff of css was/is used for styling in GNOME for a while.

These days there's tons of web desktop projects. https://github.com/syxanash/awesome-web-desktops . Only sort of in the spirit but i quite adore Greenfield, an html5 Wayland desktop/compositor. https://github.com/udevbe/greenfield

I absolutely think there is all-in potential for the web. I think we're missing a visible rallying point, where a stronger community of app makers can cooperatively advance beyond-the-web capabilities. Project Fugu & WICG help a lot but are also working with Safari (and sort of Firefox) handcuffed to them.

You may be thinking of Palm webOS which failed in the market, partly because it felt slower than Android, iirc.

Capyloon https://news.ycombinator.com/item?id=36262627

Kera https://news.ycombinator.com/item?id=36260589

https://en.m.wikipedia.org/wiki/Active_Desktop