3 ms·

Desktop operating systems as a differentiating feature started being a strategic failure because of the web. Ballmer doubled down on Windows at the expense of o

Desktop operating systems as a differentiating feature started being a strategic failure because of the web. Ballmer doubled down on Windows at the expense of other stuff. There is a common HN trope that Ballmer was doing everything perfectly and Nadella just continued it. I don’t buy it and neither did the market. (Neither did Microsoft’s board presumably) I think Ballmer’s entire strategic direction was long-term wrong and Nadella changed it. This isn’t to say no projects started under Ballmer (they definitely did) it’s about whether those were prioritized or would have ever shipped under him successfully. If the wrong things are considered important - it doesn’t really matter what projects are happening.

zeusk 5y ago
Google was the one pushing web relentlessly and has arguably made computing UX worse for everyone. From bad performance and overdone advertising+tracking to relatively smaller issues like bad accessibility. They're still trying very hard to put everything on web so Chrome can be what windows used to be.
- fossuser 5y ago
  The web won because it's cross platform, keeps state across everything and puts collaboration among users first. In a lot of ways that's easier for users - it's the closest thing to a universal OS. This is mostly a failure of desktop operating systems to be interoperable and perhaps most seriously fail to support good network first/collaborative thinking in design. I think they did this in part because they saw being distinct as a way to differentiate themselves, but also because they were created pre-internet before a lot of modern application use cases were really possible or relevant, in the end this made them less competitive and native applications less desirable (in most cases). I also wish it wasn't that way because the web has a lot of issues, it's slow, creates incentives towards megacorps that run centralized servers and ads, etc. All this is one of the reasons I'm hopeful Urbit might succeed against all odds - they're rebuilding the OS with network first in mind. Chrome isn't really comparable to windows because anyone can make a browser that reads the open web standards - while only Microsoft could make an OS that ran windows apps (outside of hacks). This reason is why web won out over native even though native apps have nice performance advantages. That plus limited amounts of high quality devs can target the web and ship to everyone without having to build multiple native applications for multiple platforms, and again native platforms still haven't solved collab really.
  - zeusk 5y ago
    Keeping state across and putting collab first is something native apps can do as well. You don't need a web interface to do that. I do however agree about the positive cross-platform support. From my experience, which is very limited - it is the decades of backward compat and complexity that makes native development so tiresome. It gets even worse when there are multiple platforms with not much in common. Thanks, I'll check out Urbit. And? How many non-Chrome browsers do we have? Even Microsoft gave in to Chromium. Can't wait for Google's next move to tie their protocols deeper into the browser just like they did with Android and playstore services. This was another eye opener to me, how limited high quality developers are.
NicoJuicy 5y ago
> I don’t buy it and neither did the market. I suppose you didn't check the stock...

zeusk 5y ago

Google was the one pushing web relentlessly and has arguably made computing UX worse for everyone. From bad performance and overdone advertising+tracking to relatively smaller issues like bad accessibility. They're still trying very hard to put everything on web so Chrome can be what windows used to be.

fossuser 5y ago
The web won because it's cross platform, keeps state across everything and puts collaboration among users first. In a lot of ways that's easier for users - it's the closest thing to a universal OS. This is mostly a failure of desktop operating systems to be interoperable and perhaps most seriously fail to support good network first/collaborative thinking in design. I think they did this in part because they saw being distinct as a way to differentiate themselves, but also because they were created pre-internet before a lot of modern application use cases were really possible or relevant, in the end this made them less competitive and native applications less desirable (in most cases). I also wish it wasn't that way because the web has a lot of issues, it's slow, creates incentives towards megacorps that run centralized servers and ads, etc. All this is one of the reasons I'm hopeful Urbit might succeed against all odds - they're rebuilding the OS with network first in mind. Chrome isn't really comparable to windows because anyone can make a browser that reads the open web standards - while only Microsoft could make an OS that ran windows apps (outside of hacks). This reason is why web won out over native even though native apps have nice performance advantages. That plus limited amounts of high quality devs can target the web and ship to everyone without having to build multiple native applications for multiple platforms, and again native platforms still haven't solved collab really.
- zeusk 5y ago
  Keeping state across and putting collab first is something native apps can do as well. You don't need a web interface to do that. I do however agree about the positive cross-platform support. From my experience, which is very limited - it is the decades of backward compat and complexity that makes native development so tiresome. It gets even worse when there are multiple platforms with not much in common. Thanks, I'll check out Urbit. And? How many non-Chrome browsers do we have? Even Microsoft gave in to Chromium. Can't wait for Google's next move to tie their protocols deeper into the browser just like they did with Android and playstore services. This was another eye opener to me, how limited high quality developers are.

NicoJuicy 5y ago

> I don’t buy it and neither did the market. I suppose you didn't check the stock...