“WHAT are you on? The ‘fuck Windows’ strategy?” Back in the late 1990s, when Bill Gates was still Microsoft’s boss, any employee who had the temerity to suggest something that could possibly weaken the firm’s flagship operating system was sure to earn his wrath. Even after Steve Ballmer took over from Mr Gates in 2000, that remained the incontestable law at the company’s headquarters in Redmond, in Washington state. Everything Microsoft did had to strengthen Windows, to make it ever more crushingly dominant. Many of the company’s best innovations were killed because of this “strategy tax”, as it was known internally.
Today the rules are different in Redmond. The new boss who took over last year, Satya Nadella (pictured, centre, with Mr Gates to the left and Mr Ballmer on the right), recoils when he hears the term “strategy tax” and says he now tells his staff simply to “build stuff that people like”.
Microsoft seems to be making a lot of interesting moves lately that never would’ve happened under Gates/Balmer. Office on iOS/Android devices is great, and Windows 10 is shaping up to address everything that was wrong not just with Windows 8, but also with everything that came before. It’s clear Microsoft is finally embracing its new APIs and Metro environment properly, relegating the ‘classic’ Windows elements to the legacy bin.
The big question in that regard: Metro Explorer shell. It’s clear they’re working on it, but will it come in time for Windows 10, or will it be pushed to Windows 11?
Actually, I think it’s quite the opposite. Windows 8 is the one that truly left classic Windows (the desktop) in the legacy bin. The desktop was simply considered an ‘app’ of the Metro environment whose future would be no more than that. Windows 10 changed all that and is bringing all of the classic desktop features to the forefront — along with enhancements.
In fact, the whole question of them even working on a ‘Metro Explorer’ is the point here — they aren’t leaving the classic elements behind, they are bringing them forward (i.e., the Start Menu and Windowed applications, two very ‘classic’ elements).
Microsoft is currently in the process of replacing every non-Metro aspect of Windows with a Metro one – Metro ones that work on both desktop and tablet (and phone and Xbox and…). This is distinctly different from Windows 8, where Metro was clearly a tacked-on environment with crap applications that barely worked on tablets, let alone on desktops.
Windows 10 is different. Windows 10 makes it very clear that from here on out, whether it’s desktop, tablet, phone, or something else – it’s all Metro, all the time. In Windows 8, they could not relegate the classic environment to the legacy bin because it was a crucial part of the operating system; it had no replacement.
That’s changing. Finally.
You guys are still calling it Metro. That’s funny
http://www.osnews.com/story/28430/Microsoft_rebrands_Universal_apps…
Edited 2015-04-02 23:33 UTC
Well, to be clear, my use of the word ‘Metro’ is a reference to it’s meaning as it was generally used shortly after the Windows 8 announcement; a full-screen, touch-first application. As for Thom, well, as he said, he “will still, and will continue to, call them Metro applications” even for today’s Windows 10 applications. As I alluded to in my other post, I don’t think that quite works anymore.
To me, Metro means a little bit of both depending on the context.
Windows 8 came with a number of new technologies which IMO are better discussed separately, especially when using terms such as a “Metro Explorer”:
1. There is the WinRT runtime along with C++/CX. This was meant to replace the classical COM API approach that MS had been using since the late 90’s (popular called the Win32 API when also including the earlier non-COM stuff like CreateFile, CreateWindowEx, etc).
The main point of this move was to try unify the higher level .Net framework with Windows. In particular make it so that the API’s are directly digestable by the CLR while not being hell to produce from C++ (like old C++/CLI was).
2. Then there is the actual UI framework built on top. This is mostly just XAML with windows restricted to the Metro tablet window manager. Clearly they changed this part so they can now be hosted in the “classic” desktop as a normal window now.
3. Finally there are Windows Store sandboxed apps. This includes the two above, but now also restricts apps to the sandbox and all the walled garden requirements of deployment etc. When Microsoft says “Universal apps” they are talking about this type of metro apps.
I think it is important to point out that a “Metro explorer” most likely only includes step 2 at this point. They are probably dreaming of including step 1 here as well, but the technical challenges of doing so are very large. Essentially the problem is that “Classic explorer” has so many APIs evolving around HWND windows and GDI rendering that at best they will have to introduce new interfaces slowly iteratively one step at a time. And then they’d have two explorers in one for probably a decade until they can retire the old interfaces.
I’m not sure if Thom meant a rewrite of Explorer, but given the work required I find that very unlikely. The way I understand the changes we’ve seen in the Windows 10 beta so far, it is more that any new UI they do (like their new start menu) will be based on their newer “metro” (XAML) UI framework.
Yeah, I know exactly what you mean. I just think it’s hilarious that people are still calling them ‘Metro’ apps after Microsoft has renamed them about 5 times. As for me, I’ve moved onto ‘universal’ apps, so I’m still one generation behind
IMO, Metro apps = Windows 8/8.1. Universal apps = Windows 10. Windows apps = desktop.
Well, that’s the problem, isn’t it? They’ve been renamed so many times that nobody can remember what the current name is, so we just use whatever term we can remember. Apparently Metro was the name that sticks in the mind best…
No, what’s hilarious is that they renamed it n times. We call them as we call them because we all know what we mean when we do, instead of mixing n different names following every minor move their name-generator drones make.
Name changes & naming schemes often times make more sense when they’re the result of things going on in development. What may seem nonsensical to the outside world might make complete sense internally and to those actually involved in the development. You don’t really want to expose much of that if you can help it because of the confusion it can cause.
Name changes & naming schemes for marketing reasons is another story. Too much of that and people naturally start questioning if you know what the hell you’re doing, and that’s not a good thing.
As it stands (to the best of my knowledge) my issues with Universal apps are twofold:
1. Microsoft runs the store, and gets to decide what it is OK to do in app. Only games are allowed to include “adult” content – from nudity, to references to drugs and alcohol have to be left to be store-worthy. I don’t like software companies deciding what I can run on my PC.
2. While getting better, many APIs available in desktop apps have no equivalent as yet in the Windows Runtime libraries. In particular, the “sandboxing” of apps seems designed to shift the processing model to a more client-server type of architecture, that may not actually suit all applications.
Indeed, I agree. I suppose it’s a matter of what you are considering ‘classic elements’ in your original post then. Or for that matter, what you mean by ‘Metro’ in today’s terms. As it pertains to Windows 10, ‘classic elements’ are things like the Start Menu, the desktop interface/windowed applications, a taskbar, explorer, etc. Clearly, these elements are not being thrown away.
As you said, Windows 10 as much about the desktop as it is anything else. But back when Windows 8 was announced, a Metro application was exclusively a full-screen touch-first application built using the WinRT API. There was no concept or even plan to bring the WinRT API into the desktop, because as far as Steven Sinofsky was concerned, that was dead.
Microsoft is changing course so hard here that it actually makes sense what they are doing when they want to no longer call these applications ‘Metro’ — especially given the initial trajectory they were on. Microsoft is accepting the fact that the desktop is not replaceable and, as such, is NOT throwing the desktop to the trash. Instead, they are modernizing it and updating all of it’s elements for the future.
So perhaps it’s a matter of semantics, at this point. We’re both saying the same thing in the end. I just wouldn’t consider what Microsoft is doing to be “relegating the ‘classic’ Windows elements to the legacy bin.”
ansidotsys,
I agree, this is a clear shift back to the desktop, and not really relegating it to the legacy bin at all. Maybe, as you suggested, Thom is using different definitions than we are.
In any case, MS seems finally willing to address the problems that made the Metro experience so unpleasant for desktop users since the very first W8 previews. While it’s still too early to see the outcome, I like Satya Nadella’s strategy “build stuff that people like”. Customers were rightfully becoming fed up with microsoft’s former strategy “you’ll like what we give you”.
Oh Good Lord, Metro lost….get over it. Windows XP has more than THREE TIMES the users Win 8 has, it was hated THAT much, and Windows 10 is “desktop desktop and…oh yeah, desktop”.
Nadella can see the writing on the wall, hell even Apple can see it…tablets are dead mate. they are toys you give your kids to play Angry Birds on for car trips, look up the figures for yourself, even the iPad sales are going down.
The hilarious part is you look at any of the Win 10 blog of the major media sites, what parts do they hate of Win 10? Anything that reeks of Metro. Its Windows 2.0 designed by hipsters, with UIs designed by people that don’t understand how UIs fricking work with everything the same shade, no raised buttons, nothing to instantly tell the user what is a UI and what is background. Nothing I’ve ever been shown in Metro has been good design, its not intuitive, not discoverable, and I’m not the only one saying this as I can wallpaper this page with UI experts saying the same thing.
Will tablets disappear? Nope but neither did the netbook, but just like the netbook is just a blip, on the radar compared to what it was in 2010 so too is tablets gonna be just a teeny tiny blip on the collective radar. The design is too limiting, touch is a PITA to get real work done, and even the best chips from Intel designed to fit that space are a bad joke compared to their weak end desktops from 5 years ago….its over, Metro lost, get over it.
I am so glad I’m not the only old timer who thinks Metro and the “flat UI” look are a throwback to Windows 2.
You’re comment about “Windows 2.0 designed by hipsters” made my day. It’s 100% accurate, and there is nothing good about it.
Tablets are dead? Because sales numbers are down?
Not really. I read an analysis of that just today that I agree with. Tablet sales numbers are down because everybody that wants one has a tablet. Or two. Or three.
Most people agree that there isn’t much left to add to tablets at the moment.
Apple engineers say they need new breakthroughs in battery technology or CPU manufacturing before it makes sense to upgrade their current iPad designs.
The Retina screens are at the limits. The CPU/GPU is at the limit of speed vs power. The battery is at the limit of usage time vs weight.
The only tablet upgrades we’re going to be seeing is increasing RAM and Flash storage and faster WiFi and LTE protocols.
So the tablet market isn’t dead. It’s in a steady state where people will be buying replacement tablets as needed. Much like the desktop PC market right now.
That’s fine with me.
Microsoft can commit suicide if they want to.
I’m ok with that.
I really can’t wait for Server 201x to have Metro and touch apps.
Configuring Server 2012 Core can be a total PITA.
Then we found that some of the new versions of the business apps we use need a GUI so now even CORE is out of the question. So we are stuck with a mongrel environment.
Come on MS, just give us the Server 2008 R2 GUI on the server. Laave that Metro thing to Desktops, tablets and phones.
Ummmm…. Install ClassicShell on 2012 or 2012 R2…
*You’re welcome.
You can’t even begin to understand the approval process that we’d have to go through to get Classic Shell installed on Prod Servers.
IMHO, having all four wisdom teeth extracted at the same time would be a more pleasurable experience.
The new CEO is certainly more level-headed than Gates or Ballmer. Both of them ran obliterate the competition at all cost campaigns against competitors to maintain OS & Office monopolies. Product innovation and cost to customers were basically irrelevant.
Gates seems to be content with his life away from Microsoft. Is Ballmer? If he isn’t, the current CEO better be always looking over his shoulder.
Writing file manager “universal application” is not a big deal, but it is rather interesting whether they will replicate the infrastructure of explorer. Eg. there were some explorer extensions like Swish,¹ which were rather useful; on the other hand the idea of extensions flies in face of Microsoft’s policies on universal applications…
¹ http://www.swish-sftp.org/
ddc_,
You are right, extensions are still very useful for many things like backup utilities, archiving utilities, file transfer utilities, context menus, etc. If they drop support for these, it would be a functional regression in windows explorer itself, which would have to be implemented in separate applications. Sometimes that’s ok, but IMHO it’s the extensions that enabled windows explorer to be refined so well.
On the bright side, I’m sure there’s plenty of bad code lurking in there. Ditching legacy code should give MS a chance to fix bugs/limitations; like the one where files that are drag&dropped from apps to explorer have to be placed into a temporary directory under the hood only to be copied by windows to the destination. Not only is the delay very annoying when dealing with larger ISO files. It is also a problem with low disk space since it requires twice the space to complete a simple drag and drop operation.
What I’m happy about is the willingness by the current CEO to tell the Windows to sink or swim based on its own merits rather than castrating the growth and development of their services and Office simply to prop up the Windows monopoly. Office for 2016 for Mac is a great piece of software even at this early beta stage but I’d love to see Microsoft add OS X Extensions to improve integration in with Microsoft’s own services. Microsoft has the talent but the question is whether politics is going to undermine that.
Interesting side note regarding the new MacBook that is around the same performance as a A8X which has fueled the rumour of an ARM MacBook in the future. How that relates back to Microsoft is the fact that iOS applications have to be 64bit by 1 June 2015 if Microsoft want their updates accepted. On Arstechnica a Microsoft programmer was talking about the work they’re doing to meet that deadline which makes me wonder whether the move to ARM for Apple’s computer range would make life easier for Microsoft assuming they’re sharing large amounts of code.
When it comes to the Office code base I wonder whether long term they’re going to move it all to WinRT and eventually WinRT becomes a shim that sits on top of Windows, OS X, iOS and Android as a way of delivering a consistent experience across platforms. Which goes back to WinRT, is WinRT merely a means to an end with Windows playing a decreasingly important role in it all. Then add on top of that ChromeOS can run Android applications then are we going to see the rise of Chrome Books, Android on smartphones combined with the growth of OS X at the high end and iPhone that Microsoft is accepting that they’ve got a decreasing profile in most consumers minds.
Edited 2015-04-03 05:15 UTC
I don’t think this is particular likely as WinRT is intended to be the lowest level API for user space processes in Windows. Porting it to other platforms gives the same kinds of problems as Wine got.
Are they even using the same codebases for Windows and Mac? The Office 365 I got installed on my Mac looks very differently from the Windows version to the degree I suspect they could have only the name in common.
The problem is that WinRT very much re-uses code from Win32:
http://arstechnica.com/features/2012/10/windows-8-and-winrt-everyth…
I would love to believe that WinRT is a subsystem in its own right or at least is based on the low level parts of Windows but from what it appears it bases a lot on Win32 – warts and all.
From what I understand they’re gradually working to bring them together – well, that was what I read in the comments section on Arstechnica from a Microsoft programmer but I don’t know how close they are to achieving the goal since it is a pretty large and expansive codebase.
Windows 10 is making the WinRT portable across their devices, where the COM interfaces are the same, while the implementations vary across platforms, with possible stubs.
https://channel9.msdn.com/Series/Developers-Guide-to-Windows-10-Prev…
The key here is “their devices”. Which is another way of saying they are making WinRT portable across the NT kernel.
No different that those companies that make their code portable across UNIX systems.
This is because WinRT is really more about creating a replacement for COM than creating a new subsystem.
In the early 90’s Microsoft’s Platform SDK APIs were all done with classical C APIs (CreateFile, CreateSemaphore, CreateWindowEx, etc.). Then they invented OLE because they wanted a more object oriented way of defining interfaces and this eventually became COM. From approximately 1995 and onward all their new Platform SDK APIs were done with COM interfaces.
Then around year 2000 they introduced .NET which introduced the CLR, which I believe at least DevDiv wanted to be the replacement to COM. But this never happened and WinDiv kept on producing COM interfaces. Maybe they even tried and had to abandon the idea with the famous Vista reset – I can only speculate here as I’ve never worked for MS.
Anyway, the consequence of this has been that .NET developers had to resort to P/Invoke bindings and 3rd party libs to just access basic functionality such as Direct3D. WinRT is the intended solution to this problem.
You are kind of right.
There is a document about .NET origins when it was called Ext-VOS internally.
http://blogs.msdn.com/b/dsyme/archive/2012/07/05/more-c-net-generic…
You will see the design kind of resembles WinRT.
The WinRT came to be when the whole “Going Native” wave started at Microsft, after Longhorn’s failure. Hence why parallel to WinRT, .NET compiles to native code on WP8 and with .NET 4.6, NGEN will get a brother compiler campable of producing static binaries.
But the idea of having the exact same UI for both touch tablets and desktop PCs, simply put, stinks. And Microsoft is heavily pushing that idea:
http://i.imgur.com/yUSvKsO.png
jbauer,
I’d agree if we’re talking about the entire desktop interface. However if we’re talking about the applications themselves, I think it’s helpful to have the option of being able run the same application on different platforms. They can have tailored interfaces for specific devices when that makes sense. I can even see applications that dynamically adapt between mobile and desktop (ie when a mobile device is plugged into a desktop base station).
The problem with Metro apps going forward won’t be the underlying technology, but rather the walled garden environment. The one-store-for-all business model has already demonstrated it’s end-game for developers… it’s not pretty.
It’s not just you. Microsoft is replacing their desktop applications with metro-looking ones. They might be developing themselves out of the competition because these artsy-fartsy new apps will inevitably be crap for years, won’t they? Is it possible Microsoft can redesign everything and succeed at first shot? Take a look at the new calculator. Shrug.
So we keep the desktop, but we lose the Windows application design principles developed over decades of trial and error. Now we get lots of beautiful white space?!
http://www.winbeta.org/news/heres-look-new-alarms-calculator-and-ma…
http://www.neowin.net/news/windows-10-10049-battery-indicator-updat…
The new battery icon has less visual detail than the Palm Pilot one. Fashion…
The gigantic application space make sense though, when you consider that Microsoft clearly intended metro for phones and tablets that need larger areas of the screen and bigger text to display things and for you to fat-finger the touch screen.
On the desktop… yeah that’s just dumb.
But it isn’t the exact same, it switches form depending on orientation and size. Personally, I think it’s the right idea, as I often want to work on the same thing in different places. Take a photo, send it over WLAN to my tablet to adjust white balance, continue the work on my 27″ IPS monitor — why not attached to the same tablet computer? — without switching between applications that are based on completely different conventions and styles. Ideally, it should be possible to work at any task concurrently across all my devices: phone, tablet, laptop, desktop (notice: no wristwatch).
The way it is now, we have front ends to web services, which work well for simple apps like calendars, and segmented platforms for native apps, which work terribly together and are toys at best on mobile anyway.
Not at all a fan of these modern universal apps on the desktop. Most of the apps you find make very poor use of the available space compared to what would a traditional Win32 app. Example:
http://apps.microsoft.com/windows/en-us/app/8-zip-lite/73049c99-963…
I’m also not a fan of the visual appearance, which has killed expressiveness at the altar of modern minimalism.
The comments in this section, and even Thom’s reaction to the article proves one thing about the computing culture in general, and Windows’ in particular: It’s focused on bling & features, not the core of the OS itself.
Everyone hates Metro or whatever MS wants to call it today. That’s old news, MS got it on the chin.
Oh gee, they are adding yet another shell to the GUI, yet another feature. I don’t mind useful features, power shell is useful. The problem is there was already a command line shell in the system and some new hire decided he didn’t like it and refused to expand upon and correct any short comings. That’s not teamwork, that’s “my way or the highway, and look what I did! Isn’t it grand!”
For the past decade the problem has NOT been a lack of GUI features and pretty interfaces (that’s sarcasm, “pretty” is in the eye of the beholder).
It’s what’s underneath. The mess of APIs, the increasing abstractions from hardware and the resulting losses of performance associated with it. DirectX 12 is a step in the right direction, finally. But there’s far more problematic with Windows than what the user sees when playing the Game of the Week.
Computational power has lagged behind UNIX-like systems for years. NTFS is a horrific latency inducing beast even with certain functions disabled. Full screen programs that lock up can still blank the screen requiring a number of keyboard hoops to get rid of them and return the desktop to a useable state. These things shouldn’t be happening in 2015. Microsoft’s internal politicking and fiefdoms need to end for the good of its products, users, and developers.
I hope Mr. Nadella can induce the needed changes, but I’m not going to hold my breath that Windows 10 is going to fix all of that. Performance metrics between 7 & 8 were completely flat. Ultimately that’s why I stayed with 7 even when 8.1 released. There was no reason to upgrade other than “it’s new”. It wasn’t improved.
https://www.youtube.com/watch?v=HZnbGNtcyMc
Particularly the bit from https://www.youtube.com/watch?feature=player_detailpage&v=HZnbGNtcyM…
They should fix the REAL issues in Windows like the super annoying file locking problem.
Hm. I thought you missed April 1st, and the title means MS is open sourcing Windows %)