Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Microsoft gives up on Windows 10 Mobile

bbc.co.uk

Post reply on HN
Print view View original

311–320 of 724 posts

Re: Microsoft gives up on Windows 10 Mobile

#311

Post by NiveaGeForce »

garganzol wrote: post #297
dagaci wrote: post #283

One of the biggest failures of windows mobile was constant abandonment and reset, + all the mobile developers and users from the pre-iphone days were simply dropped with no path to wp7, + the earliest and devs of wp7 were all dropped in favour of wp8, + MS execs then approved of a slightly less damaging strategy with 8 to Wp10 but at this time it was already too late Microsoft could not solve the app-gap problem beca…

This. But wait, in order to further harmonize "cloud first mobile first experiences" they may be deeply pleased to announce one more important transition from UWP to OneCore.

UWP is getting new features all the time, it's the way forward for Windows apps and will stay for many years to come. Even all the default Windows 10 apps and stuff like settings UI are UWP.

OneCore is an internal Windows refactor that won't affect users nor developers.

Re: Microsoft gives up on Windows 10 Mobile

#312

Post by pjmlp »

cjsuk wrote: post #280
pjc50 wrote: post #208

Earlier quoted context omitted.

Outlook+Exchange is really the thing keeping Microsoft alive.

For now. Some of the younger people aren't having any of that shit these days.

If they work at any big enterprise it is either Outlook or Lotus Notes.

The latter will make them love Outlook.

Re: Microsoft gives up on Windows 10 Mobile

#314

Post by WorldMaker »

garganzol wrote: post #284
WorldMaker wrote: post #159

Earlier quoted context omitted.

You can sideload applications in Windows 10 mobile easily enough. Like Android it's been on by default for some time and just a matter of downloading an .appx package file. The ability to sideload doesn't help if there aren't apps to sideload. I don't think the problem was ever the openness of the platform. An open platform doesn't guarantee developers either (look at decades of people not support Linux ports of thei…

The problem with sideloading is that it requires a lot of hops. So many hops nobody does this. Usually "a hop" is represented by some required certificate, a special "container" the app needs to follow, a special API voodoo call you should make to sideload the app. It quickly decays from being a reliable reproducible computer science, and turns into the joke of marketing greed you would never trust.

From a user experience standpoint: sideloading is just "install this app". Click a link on a website to an APK for Android or APPX for Windows and watch it install. It doesn't get much easier than that. Fewer hurdles and security concerns than installing some random EXE that may be a wrapper for an MSI and needs Administrator rights to your machine for who knows what reason. The install experience for an APPX link is really nice in Windows these days. It's a great user experience.

From a developer standpoint: those "hops" have been your job for decades. Complaining about them says more about whether or not you are good at your job than the platform itself is suitable to applications.

Windows has always encouraged executable signing, and downloading unsigned EXEs from random websites has always been a bad idea. (The SmartScreen warnings for them these days have become appropriately diresome.) Other platforms have signing requirements, too.

Containers have always existed: ZIP, CAB, MSI, etc. If I was forced to write a container by hand-formed scripts, the APPX format is much easier than MSI or most other installer packages for classic Windows desktop: it's a zip file with a pretty easy to read XML manifest.

(With the Desktop Bridge, no application developers have an excuse to use an EXE or MSI installer anymore if they are only supporting Windows 10. If you can build an EXE or MSI, you can build an APPX. Build an APPX.)

There's no special voodoo API calls needed to sideload an app on Windows. Users just double-click an APPX package and magic happens [since the Anniversary Update in November of last year], they get a simple installer dialog. (You can use PowerShell to automate installing APPX packages, but you don't have to. It's an advantage to power users, not voodoo to pass to normal users.)

As for "reproducible"? APPX installs are way more reproducible than EXE/MSI installs. It's a pretty slick system if you bother to look under the hood. It's also rather well documented in that case that you do.

I can't assuage any conspiracy theories about marketing greed, but looking at sideloading from a technical perspective, it certainly isn't "a lot of hops" and "nobody does this" is a matter of perspective; I've got several packages in development that currently are outside of the Windows Store.

Re: Microsoft gives up on Windows 10 Mobile

#315

Post by SyneRyder »

wmeredith wrote: post #124
ssully wrote: post #79

Earlier quoted context omitted.

I was going to chime in a little on it being a stretch. One of the promotions for students they did was near the summer of 2012 or 2013. The promotion was you got paid 100 dollars for every app published on the Microsoft store limit 5 for the mobile store and 5 for the regular store. So a total limit of $1000. My school actually had a Microsoft rep run a workshop over a weekend showing students how to publish an app…

Did everything they could... Ah, yes. $100 per app. I'm sure that's what most apps cost to produce. /s It seems like they did a few things, but never actually, you know, paid app developers to build out their ecosystem.

They tried that too. The Verge reported in 2013 that Microsoft was paying some developers $100,000 each to port their apps to Windows Mobile. The article hints that was how Pandora & Temple Run ended up on Windows Mobile.

https://www.theverge.com/2013/6/15/4433082/microsoft-paying-...

Re: Microsoft gives up on Windows 10 Mobile

#316

Post by pjmlp »

winphoneappdev wrote: post #165
itsmenotyou wrote: post #30

I was working in Microsoft about 5 years ago and Satya's not lying when they say they tried everything to incentivise app developers. It was a big focus of the company at the time. For keystone apps they tried to partner with developers doing most of the work for them. For more niche apps they ran promotions for students and independent developers giving away free phones etc. But nothing was enough to get over the pr…

Having done my fair share of windows phone app development (8.0, 8.1, 10) I can confirm. Microsoft incentivized app development, and paid top game development companies millions and gave them featured placements for their half-assed windows phone ports. And then as a regular joe you would actually put in the effort* to get things working well. Then you would watch these bug-plagued big name 3-star ports appear out of…

Give it a few more OS releases and Android will be the only one supporting GL.

Re: Microsoft gives up on Windows 10 Mobile

#317

Post by hitgeek »

"incentivize app devs". as an early adopter I felt abandoned.

I was all in on windows mobile around v7 and windows8. bought the flagship lumia phone and surface rt tablet. invested the time to develop some basic apps. 6 months later both platforms were basically discontinued with no upgrade path, so I abandoned it too.

if microsoft had just made an enterprise successor to blackberry, instead of trying to make a consumer phone, they probably would have been more successful. once the platform had traction in enterprise, maybe they could of made the consumer jump.

Re: Microsoft gives up on Windows 10 Mobile

#318

Post by taormina »

pjmlp wrote: post #312
cjsuk wrote: post #280

Earlier quoted context omitted.

For now. Some of the younger people aren't having any of that shit these days.

If they work at any big enterprise it is either Outlook or Lotus Notes. The latter will make them love Outlook.

Or Gmail. Or Inbox.

Re: Microsoft gives up on Windows 10 Mobile

#319

Post by wlesieutre »

zuminator wrote: post #306
wlesieutre wrote: post #152

Earlier quoted context omitted.

Hey, if you're optimizing for "number of apps on our store" I bet it worked great! Why spend $100,000 developing something one app when you could get 1000 for the same price?!

Instead of paying per app, would make more sense to let devs keep 90% of store revenue, which would incentivize the development of apps that are actually popular and make money.

Hell, let developers keep 100% of store revenue, at least until you have a customer base.

It's not much worse for MS than 90%, and it makes a much nicer marketing story if you're trying to convince developers to get on your platform.

Post reply on HN