"Windows was designed by a committee. The Mac, by contrast, often feels like it was designed by a single person."
The Mac was apparently designed by a person I can't freakin' stand. I recently spent a year with an OSX laptop, and it was a merciful relief when I was finally able to get the hell out of iLand. My first love is KDE because it bows to my will like nothing else, as a good window-manager/OS should. Win8 is also pretty malleable, even though it's not so good for those lacking the chops to bend it to their will. OSX, by comparison, is a constant freakin' PITA if you don't happen to be the "single person" it was designed by and for.
This might be a bit of a rant but I'm sick of people ragging on Windows for its default settings. Windows is eminently customizable. If you aren't a complete n00b you can bend Windows to suit yourself far more easily than OSX. KDE absolutely spanks them both, but still... I would wholeheartedly recommend OSX or iOS to my grandmother, because she's not so good at customizing things. This blog is aimed at people who probably aren't running their OS on 99% default settings. Customizability matters, and Windows fuckingowns OSX in that respect.
My experience has not been the same. Making Windows do reasonable stuff, like not lose focus for dumb reasons, close windows instead of minimizing them when you hit the close button, have virtual desktops at all, have a good fuzzy-search app launcher, and so on, has been a pretty rough experience. Lots of these problems have solutions that technically exist but are so cumbersome to use I'd rather not bother. A real solution called LiteStep allows you to create an environment that is reasonable to use if you put in a lot of effort up front and are okay with your shell crashing every couple days.
Beyond the UI stuff, some MS software is just not meant to be uninstalled. I wanted to test an app without the VC++ 2008 redistributable installed, to see which dlls a user would need. Easy, right? Just go to control panel -> add or remove programs -> click on VC++ 2008 redistributable, and it's gone... except that all of the dlls are still there, in a huge folder with hundreds of other dlls. Which ones are for the VC++ 2008 redistributable? I have no idea. On some platforms I could just rm -rf VC++2008.framework, but not here. I guess MS thinks it would be better for usability if people had to reinstall the OS and wait through hours of updates just to remove some shared libraries.
"I guess MS thinks it would be better for usability if people had to reinstall the OS and wait through hours of updates just to remove some shared libraries"
I would probably just use a snapshot in a VM in that case. After all, testing "package not installed" is different than "package installed, then removed" unfortunately (which is your point).
Early on, I was big into customization. I even baked my own widgets into VGA.DRV to make my Windows 3 look the way I wanted. My programs used CTL3D.DLL to avoid standard 3.x dialog boxes. Later on, I changed every icon in Windows Explorer to make them look like BeOS. Even on XP, I used Royale and Royale Noir themes instead of the original Luna. Then I moved on to Linux. I downloaded countless themes and icon sets. My Linux machines were unlike any other and were exactly the way I wanted them to be.
Then it ceased to matter. Now I use Gnome shell or Unity. I install software using package managers. The most radical thing I do it to customize the launcher and change stock Ubuntu background to a bluish one. Sometimes I even power up my AIX machine and work from CDE. The way it looks doesn't matter. I spend my days between browser, text editor and terminal. From the desktop, the feature I use the most is task switching followed closely by virtual desktops.
Making the OS look a certain way is doubly risky. It's a waste of time and, if you develop desktop apps, you risk making them consistent with your taste rather than usual conventions your users rely on.
Developers live in Terminal and text editors, and regular people don't care about customization? There's a reason why Linux on the desktop didn't pan out. People don't want choice. They want to be told what to do and get on with their life.
Why degrade people like that? You sound like those hard-core Android fanbois that never misses a chance to tell you they think iPhone users are sheep. :-(
No, they want the software and options to be thought through before released.
Why have 20 options for 100% of the users, when 16 good defaults and 4 options is the better way for 90+% of the users? In OS X, I bet most of those 16 options can be changed using a terminal command.
Unfortunately some of the defaults do not seem to have been thought through and cannot be changed from the terminal. For instance, fat-fingering the button right above backspace will pause your music and make you log back in to keep doing whatever you were doing, and full screening YouTube videos will sometimes trigger a soul-crushungly long transition to the needless removal of all the contents of your other monitors.
Hence the Gnome desktop's new direction. I actually quite like it (on good hardware with fully functional graphics drivers) but I know that some find the reduction in configuration and removal of options hard.
Could you list a few customizability examples which you could not do on OSX but are possible on KDE or Windows? I am a life long Windows user, used KDE quite a lot, and OSX for the last year, and I love OSX. I wouldn't claim it's more customizable, but to me it's about the same.
Bind keyboard shortcuts to maximize (vertically, horizontally, or both), minimize, restore size, tile/split, move to adjacent desktop. There's a lot I could learn to deal with on Mac, coming from Linux, but it amazes me that this doesn't seem possible without Applescript.
There are tools to do it as people allude to but I can't help but feel this is a square peg in a round whole solution.
When I use linux, I use a tiling window manager, when I use OSX, I use tabbed terminals and fullscreen mode and workspaces.
Also, I should note that the Command-~ functionality which doesn't exist on Windows and spotlight is enough to make me prefer the OSX desktop to the Windows one.
The problem with windows 8 is that from the start it wasn't designed to solve anyone's problems except Microsoft's. Windows 7 appeal was obvious: a modern version of windows that doesn't suck (ie: not vista). Windows 8 is supposed to fuse desktop and mobile, but, that is a problem for Microsoft, not their customers, outside of Redmond nobody needs that
Indeed, Microsoft seems to have this belief that we must have the same look and feel across our PCs, laptops, tablets, and phones. That is obviously the answer because, hey, it worked for Apple, right?
There's only a handful of people I've spoken to that like Windows 8 and hardly anyone seems to share Microsoft's belief (above).
Meanwhile, Canonical has came along and pretty much came to the same conclusion as Microsoft -- that a common experience across all of one's devices is the way forward. And, just like Microsoft, they apparently fail to realize that nobody wants this but them.
> That is obviously the answer because, hey, it worked for Apple, right?
The fact is, it really didn't work for Apple. Short of a few common design choices, iOS and OSX are very different, and for a good reason. I am not sure why it's not obvious to people. Mouse or keyboard driven UI can be much smaller and tighter, with a lot less chrome, to allow more room for data. Touch UI has to be chunkier by default. The simple truth is Metro UI works amazing on a touch device, I love it. Windows 7 works just fine on PC. What Microsoft really needed is 2 separate operating systems, just like apple. The Windows line could have been updated with a fresh skin, more stability improvements, and some core new features here and there. The Metro OS could have been a separate OS for the Phone, Tablet, and TV markets. Instead Microsoft now has 4 incompatible OS lines which all function very different, but look identical, only adding to the confusion:
1. Windows 8 - For PCs, Tablets, and God knows what else,
2. Windows RT - For Tablets which are not as nice as the Windows 8 tablets but can run some of the same software, but not all of it, and some apps which work on the RT actually do not run on Windows 8 tablets, and, so ..... yeah.
3. Windows Phone 8 - Exclusively for phones. For convenience, incompatible with Windows Phone 7 and 7.5.
4. XBox OS - Runs only on XBox, looks like Windows 8, but works very different.
So, instead of 2 visually distinct operating systems, Microsoft now has 4 visually similar operating systems. I can see their logic, they wanted to leverage their huge PC instal base to drag some users to their Tablet and Phone products. Much like Apple is using iPods to sell people on iPhones and iPads, and perhaps later, Macs. But the problem is, they are not really achieving that goal, while pissing of their core PC customers.
this is spot on. The only thing I would add is to point out how meticulously little has changed in the Apple OSX over the years. It's mostly carefully vetted polish & performance improvements. Microsoft tries to reinvent the wheel with every other release when what people really want is something stable, fast, simple & familiar, with tasteful updates so that nothing feels stale. The last few OSX releases, WinXP and 7 were in the sweet spot.
When Apple added touch-sized icons to the Application menu in dock and eventually Launchpad, Microsoft and Canonical saw it and just completely jumped the shark on device/OS convergence. Meanwhile, Apple has tip-toed with the changes, to the point that they're not really moving forward with touch UI on OSX because it seems the value-add is not really clear for the end-user.
But that's what didn't work with Windows 8. For majority of consumers UI is operating system. Hence the confusion with difference between Windows RT and Windows 8. Hence the perception that Windows 8 has two isolated operating systems in it.
I think what he means is more like iOS and OSX. Both run a very similar operating system but completely different UI. The best approach would be to have a common core with Windows and Metro interfaces, but separate them. Desktops and Laptops run Windows UI. Cellphones, Tablets, and TVs run Metro UI. This way you can maintain backwards compatibility on Windows side but allow the Metro side to be all new.
Microsoft seems to have this belief that we must have the same look and feel across our PCs, laptops, tablets, and phones.
Actually I think their key argument was that people should be able to develop apps across all these devices using the same APIs and technologies. Same look and feel is just a (unfortunate?) consequence of that.
The idea was that having a unified technology-platform, if people made an "app" for Windows, it would also work on Windows-based tablets and phones. This was Microsoft's attempt at trying to leverage their desktop-dominance into working for their mobile strategy.
If that worked out OK or not can clearly be argued, but the basic premise wasn't all bad. If they had made that work, developers would probably want to create apps for Windows first, instead of iOS or Android.
> Same look and feel is just a (unfortunate?) consequence of that.
It's actually a consequence of a strong coupling of GUI and the programs and the unwillingness of developers to refrain from pixel-perfect control. In order to make a program consistent with different interfaces, you need to write them against a "generic" layer. You should declare your screens, menus and visual elements in a GUI-neutral environment and let the specific device use the GUI consistent with it.
In order to make a multi-GUI app you either code for each and every GUI or you give up fine-grained control.
Actually, in order to write an app that runs on both desktop and mobile platforms, you shouldn't mess with visual elements at all. You should just tell the UI layer to display your data/commands/options, and let the UI layer decide how to display them.
Having worked with WPF and trying to write Android-apps it's easy to despair.
They've taken the worst of WPF (XML) without offering any of the goods (data-binding, rapid application development). You feel like you're back in the stone-age of computing, only this time with XML instead of objects.
This often means you have to do lots of UI-wiring in code indirectly, based on handles, and due to Java's lack of functional constructs, even wiring up a simple event-handler becomes 6 lines of tedious, annoymous class instantiation nonsense. It's horrible.
Had anything like WPF been available in any stable or unstable form, I would have jumped at it in an instant.
I think MS is conflating unified with Apple's end-to-end approach. You can have a great end to end experience with two completely different UI. The key take away is interoperability, not looking the same.
What I don't get is why architecture designers keep insisting that the look and feel must be the same, when the obvious solution needed here is a cross-plattform framework/SDK that allows maximum code similarity between the different platforms while taking care of _generating as much as possible_ of the natural look-and-feel of each platform. It would be the best for both users and developers: The developer writes an application that has different front-ends for mobile and desktop, all the while being helped by the tools, whereas the user on each plattform gets the interface that is intuitive on that plattform.
It's more like what I usually describe as "car crash architecture" - ramming together two things that really don't belong together in the hope that they still work and nobody notices the rather obvious gaping seams.
It's not even just the fusion of desktop and mobile in principle, but the failure to execute that concept. My mom took to her iPad like a fish to water, and quickly learned to send e-mails, surf the web, receive and send photos, etc. Meanwhile, I had a Surface RT for a bit, and while Metro was slick, I found myself constantly being switched to the desktop. Want to open a word document as an attachment? You have to flip to a completely different UI parallel and barely interoperable UI to do it. Even a lot of the basic settings were duplicated incompletely between the two worlds (I haven't tried Windows 8.1 myself, so I don't know). It's simply an unacceptable design feature for a consumer-level product.
I don't think the concept is flawed. I actually think Metro is the "right answer" for your typical user, with a hidden file system and tiled window management, even on the desktop. However for this to be plausible, you need to be able to ship without the desktop, which means Office needs to be Metro-friendly. The stop-gap of shipping with both Metro and the traditional desktop is a fatal execution flaw in what's otherwise a reasonably sound design.
A better, technically possible, solution would be to open Win32 apps as full screen apps within Metro, and allow the usual tiling and app switching. This is basically how OS X works, and it's very effective and also quite efficient if using a touchpad that allows gestures.
Yes, Microsoft forced Metro on PC users not because "it made sense", but because they needed that to happen, in order to get as many users on Metro as soon as possible and as many as possible, even if they had to force them into it. Metro in that way is a lot like Google+.
Windows 8 is everything wrong with Microsoft Corporation as an organization and how they go about developing product. I blame it on personas. When you have designers, PMs and devs creating products for imaginary people instead making something they'd be proud to use themselves, you're going to have problems. It's fine to develop products and features based on market research, user studies and metrics, but at the end of the day it should pass the "will I use it at home" test. Otherwise you'll have a company full of people that go home and use iPhones, Macs and Google Chrome.
The only group at MS I see right now making products for real people is Azure. They're actually developing features to remove friction and make its customer more productive and happier—even if it means supporting Linux, Git and Ruby.
Dog fooding is heavily encouraged within Microsoft and heavily practiced (source: Microsoft employee). Whatever you think is wrong or went wrong, its not because none of us used it. The criticism in the article, if I understand it correctly, is that Microsoft is trying to please too many people.
I currently work with Microsoft and what you're doing isn't really "dog fooding" as I understand it. Dog fooding should mean that people are making a product that they themselves would want to use, to scratch an itch they actually feel.
What Microsoft does is "dog fooding at gunpoint" even if the result is poisoning oneself. Microsofties carry Windows phones around... and hate it. Openly. This isn't good.
> Microsofties carry Windows phones around... and hate it. Openly. This isn't good.
Switched from my Android to WP, loving 95% of it, have a few gripes, but far less than on Android.
For one thing, my phone no longer leaks memory. My start screen doesn't randomly take 10+ seconds to load up. My dialer doesn't occasionally decide to just screw it and not appear.
(I had those same problems across 2 different android handsets, a 2.3 model from Motorola, and a 4.2 model from Samsung)
Couple features are annoying. Too many steps to turn off GPS is one, I liked having a quick pull down drawer. But that is a minor nitpick. Compared to swearing at my phone every single day for being a bloody POS like I was doing with Android? WP is a great experience!
> What Microsoft does is "dog fooding at gunpoint" even if the result is poisoning oneself.
I also disagree with this, my group uses a variety of handsets, WP, iOS and Android.
Have you tried Windows Phone? I'm a bit biased, I spent 3 years working on it, but I'll say that of the current crop of mobile OSs out there, it is by far the best designed.
From the underlying infrastructure used by the development team (one of the few teams at Microsoft I've ever been on that really believes every daily build should be stable and reliable, and has the test infrastructure to ensure it[0]) to the lovely UI to the outstanding performance requirements the team places on themselves.
Back when I was on the WP team, a list of user frustration levels with wait times was widely disseminated. It was something like " 50ms unnoticeable to user, 100ms+, somewhat noticeable pause or skip, 250ms, interface begins to feel slow, 500ms+, user satisfaction begins to drop, 5s+, users become overly irritated" with a lot more detail and gradations of course.
The team's internal requirements was that all UI elements in the OS respond in less than 1 second, and even then, you'd better have a damn good excuse as to why what you were doing took a whole second!
The UI flies. Animations are smooth and beautiful. Anything that does not please or delight the user does not get into the code base.
The bar in Windows Phone is not "do we have this feature", the bar is "does this feature make the user happy?"
All of this is of course not spoken as an MS employee. I'm just a lowly dev, albeit one now in charge of a UI dev team, and I can tell you that our product will sure as hell meet the same bar of smooth beautiful performance that Windows Phone does. It is damn well going to be a matter of pride if nothing else, and the entire management chain here is behind me on that!
[0] The Windows Phone division has over 4000 people in it, although I am not sure how many of them are devs. Let's say 500 or so devs total (likely an underestimate!) checking in code to what is a huge code base. Now imagine every single day, you can take the code that is committed and, without worry, put it on your phone and expect it to work. Daily's aren't always 100% stable, but they are dog-foodable and usable. One thing I miss about working there is coming in every morning and flashing the latest code to my phone and seeing the incremental progress that the team made on a daily basis.
Of course this ignores that many teams work off of branches which are then tested and integrated. The overall engineering effort to keep that platform moving is huge.
That's funny, I switched from WP8 to Symbian a few weeks ago. An OS that doesn't have an accessible file system isn't very useful to me. Might as well use an S40 phone.
When Nokia killed Symbian, they had 60% of the phone market here in Brazil, and their sales were RISING (not declining, as they were in many other countries).
Now Brazil is the dream land for chinese knock-offs, and OS fragmentation, my startup make iOS and Android apps, and I found it bizarrely hard to sell those here, when I ask a random person on the street what phone they use, they don't know, and when I take a look I see:
If the person is very rich, or very dumb, a iPhone (very dumb because we had instances of people without money to eat taking crazy debts to buy iPhones... Apple Reality distortion field is crazy).
If the person is part of the shrinking real middle class (real because the government says that if you earn 500 USD month you are middle class, obviously this is absurd), a Samsung Android.
But those are 5% of the population at most... The rest of the population I see:
S40, Symbian (specially on... SONY phones O.o Sony had once a great Symbian line, and they were really popular here, and those phones are hell sturdy, lots of people still have them), Android customized to look like Symbian (Sony phones frequently do that too, I own a Xperia Play, and my mom a Symbian Sony Phone, and for any tasks the two are almost the same), HiPhone (a popular clone of iPhone that use some weird OS), Symbian clones (that usually have Facebook and Youtube but you cannot install anything else and is incompatible with actual Symbian), S40 phones, some blackberry and old windows phone (I never saw a WP8 phone! and I see more old windows phone than WP7 phone!)
This proved to be a nightmare to my startup, I cannot nab a random person in the street and convince them to install my app, because most likely they have some crazy phone OS that is not iOS, not Android, not WP8, not Blackberry.
And if someone launched a good sturdy phone with Symbian, hardware buttons (god how I hate touch-screen, specially when you need to dial in the rain in a very rainy country) but with a good API, I would GLADLY pay iPhone levels of money to have it. And I guess most brazillians would buy it too (although to them it must be old Symbian levels of price, iPhone level is reserved to the top 5% richest)
Many managers are used to a world where customers have no choice but to buy the latest offering. I don't think they've fully awakened to the thought that Microsoft can be left entirely out of a burgeoning market.
Everyone at MS knows, but nothing changes--especially not changes to WP to win over their own people. There is no changing the course of the ship once it has sailed. Everyone is too busy working on their tasks for that sprint to step back and look around. And the backlog is so large that there is no way that tiny annoyance of yours will get done before the "priority 1" bugs, even though it might have more meaningful impact to end users.
That's one of the biggest reasons I left. The majority of people I met seemed career driven, not product quality driven, and were happy to march to orders set a year ago -- however misguided now -- if it meant a checkbox filled on their review.
I used to work at msft too. You completely missed the point.
Dog fooding at ms is eating a substandard product to provide feedback. You grudgingly use it to help the company. See internal drives to get people to use bing.
How about making a product you don't need to cajole employees into? Think Google has ads to get people to use google search, or apple forces employees to use iPhones? (Or do they call it caviar-ing?)
If dog fooding is the reason employees use your product you should not release it.
Dogfooding is all about eating a possibly substandard product to provide feedback.
And then, you are expected to work on the feedback, so the product is not substandard anymore. I guess Microsoft is lagging on that part.
Anyway, I fully agree that if your employees are only using your product because you require it, you shouldn't release that, and that you shouldn't eat your own dogfood all the time.
Inside of google? (That's what I'm referring to with dog fooding, to convince your employees to use a substandard product "for the good of the company").
Google probably has "google search rocks" internal ads. They probably don't have "hey guys could you stop using bing for a bit and try our search engine?".
Sorry, I got sidetracked from the dogfooding main point. I'm pretty sure google don't need ads to cajole their employees into using their search. I was referring to their public ads.
This does seem true for the google ads. An overriding theme in them does try to make Google appear more friendly and likeable, rather than to grab more user share.
Although reinforcing a brand is still a big part of attracting and keeping customers. If Coca-Cola stopped advertising, they'd definitely lose sales.
I actually think dog fooding is a net negative because it teaches people to ignore bugs and bad UX. Everyone at MS is encouraged to run the latest half finished version of some software. You assume it will be "fixed later", but never actually follow up. Or worse, you make excuses for bugs and bad design instead of fixing them.
Ultimately, this means no accountability because you are not developing products you yourself would use by choice. You use IE because you're supposed to, not because you prefer it over Chrome or Firefox.
Trying to please too many people = personas. Anyone who has worked at MS knows what personas are.
The only group at MS I see right now making products for real people is Azure. They're actually developing features to remove friction and make its customer more productive and happier—even if it means supporting Linux, Git and Ruby.
I think it is no coincidence this is where Mark Russinovich etc. works in now.
For those thinking "oh, another piece of Microsoft bashing from some Apple / Google / Linux fan" (my own initial thought), look at the byline.
Paul Thurrott's been covering Microsoft, almost always positively, for most of (if not more than) two decades. He's been labeled one of the "notorious Microsoft shills" by Daniel Eran Dilger (http://www.roughlydrafted.com/2009/01/18/dan-lyons-paul-thur...) -- and that's not far from my own impressions of Thurrott based on some of his earlier Linux commentary.
So -- yeah, the rats are, if not fleeing the ship, talking loudly about its soundness.
Thurrott has been bizarrely bipolar about Microsoft for a very long time. He is pro-MS "on average" but will careen wildly between unwarranted hyperbolic praise and unwarranted hyperbolic invective. Sometimes about the exact same things within the span of a week or two.
When your focus is sufficiently narrow, minor variations look like massive shifts, and perspective suffers.
Thurrott seems to look at the computing world through a magnifying glass pointed squarely at MS, overreacting to every little shift and bump there, while never being able to see what's going on elsewhere particularly clearly.
It's even in this article. The first sentence is largely at odds with the tone of the rest of the article: "I defended windows 8 from the outset" then "it's been a massive cockup from the outset". There seems to be little self-reflection that all the other commentators were right, and Thurrot was wrong.
The advice in the final paragraphs is just plain wrong as well. It should be "Don't use one OS GUI for two different user paradigms", not "abandon an entire market segment". Windows needs to go back to a traditional desktop, and Windows Mobile should continue on as is.
Given the rise of Nadella, I'm sure the new lockdown/home platform of Microsoft is going to be their cloud services.
Following the Google/Apple model of having cloud profiles maintain/inform/manage (depending on home/gaming/work platforms) all the Microsoft devices seems the best approach.
It's also what people want - they want unified logins (note: plural, Google!) so that they can get on with their work/personal/gaming needs and ensure that new devices that are platform compliant "just work" with minimal/no config.
Microsoft is in the same place with Windows 8.x that it was with Windows 2.x: trying to migrate its userbase onto a new computing paradigm (GUI/mobile touch apps) while still showing support for the old paradigm (DOS/traditional GUI). Yes, it's going to be a force fit. In part because the paradigms are so different, and in part because it's Microsoft and they don't know how to do anything with any grace, at least at first.
But eventually they'll get it kinda-sorta right. And when they do, they'll crush the competition. Especially when tablets and convertibles start to supplant traditional laptops as daily driver devices. Apple probably hasn't much to fear. The real loser here is Android. A light, cheap, powerful device that can run full Windows like the Dell Venue 8 Pro has insane added value above the equivalent Android device.
I think you're right. But that doesn't translate (IMO) lead to MS crushing competition. They have a lot of assets. Important software. Decades of user experience. Several war chests. An institutional ability to grind out a success. etc. etc.
But, "computers" has become a big industry. It seems strange to say something so obvious, but the number of "computers" in use, the importance of those computers and the amount of money businesses and individuals spend on them is still growing and growing. I don't think MS can keep ahold of "market share" when the market gets to that size. Market share has always been MS' biggest weapon. The difference between 95% and 80% is 3X as much competition, even if it is nothing in total sales.
Imagine an incumbent (say Amazon, Samsung, Google or whatever) company launching a new product that runs something inspired by Chrome OS & iOS on $250 hardware that can replace windows laptops for 10%-20% home users. That sounds a lot more plausible today than it did 10 years ago.
6-7 years ago Vista came out. It had issues and didn't run well on the low end hardware retailers were selling. They probably lost a percent or two of market share to Apple/OSX. Apple never really went after the low margin low-middle end. 6-7 years from now that could be 10%-20% to an unexpected incumbent.
The markets important to MS are still growing and expanding so I don't think they are in trouble. They could lose market share and still grow absolute sales.
tablets and convertibles starting to supplant traditional laptops as daily driver devices is an opening for new players.
Perhaps the problem is that the new paradigm is not so clearly superior to the old one. GUIs have huge advantages over text based computers for productivity. Does touch actually have that many advantages over a mouse and keyboard for someone sitting at a desk? I'd probably choose a touchscreen over a laptop touchpad, but I'd definitely choose a mouse over a touchscreen for most of the actual work I do.
I think if you're doing the same task a lot then a command line based workflow can be very productive. However, the big advantages GUIs have is that they allow the discovery of new functions. Reading the man page for a command line application is complex and often quite confusing, while a well designed GUI will often allow you to easily find the functionality you want with a tiny bit of trial and error.
You're correct, but the new target isn't touch. It's the cloud.
Take a look at who the CEO is now. The hub is the cloud. Apple and Google got it right, and Microsoft realizes this (I hope). If they have indeed learned this lesson, they will be competitive in the coming years, and perhaps come to dominate. Ignoring this lesson means Windows is going to be torn apart by Android, iOS and Steam.
Touch, Desktop, Search and Console should all be spokes on the hub - different ways to consume the same services.
It will be a challenge meeting this vision, but Microsoft can and will meet it. Question is, will it be in time to fight off Google/Apple?
Well, I don't really remember thigs that way. Microsoft didn't push people anywhere, people were already switching to GUI based systems (IBM OS, MAC OS, etc) when MS finally got there. When they finally moved, they crushed everybody, but you are claiming a kind of power that they simply didn't show.
Anyway, this time it quite different. GUIs are quite compatible with CLIs, but you can't have a touch-based small-screen hight input throughput big-screen GUI desktop, even the name is a contradiction.
I'm primarily a Windows user (but I run Linux on some of my devices as well). I'm also tend to be quite particular about my UIs getting things right. Saying all that, I've felt next to zero impact in moving from Windows 7 to Windows 8.1. Windows 8.1 feels a bit snappier. Some options have been moved around a bit. That's about it.
I just don't understand all the polemicising. Especially with statements like: "The specifics of what's wrong in Windows 8 don't really matter". They matter to me, because after reading all these articles about how terrible Windows 8 is, I still can't figure out what those specifics are! I would love to be enlightened about what I've been missing (or how I've been using Windows 8 wrong).
Want one example? Fine. It's Fuck Ugly. There. There's your reason.
You said yourself you are fickle about your UI of choice. And yet you can't see how many people would be just like you but unlike you they can't stand Win8 UI? It's just like a car, the design matters about as much as the performance of the vehicle. Sure, some people look at cars as tools and don't care what they look like as long as they have 4 wheels and can get them from a to b.
But for many people, the car they buy has to have a great design. It is about as much about lifestyle as function. And the same goes for a desktop OS.
And I for one find the Metro Tile fest that is Windows 8 to be just that. Fuck Ugly. No other way to sugar coat it. And that isn't just the Metro side. Nope, also the desktop with its sharp 90 degree corners and flat 8 Bit color pallet window blinds looks horrible. Like a complete step backwards looking at how the desktop used to look under Win7. And on Win7 you could even change the themes if you wanted to something more to your liking. Good luck trying to do that on Win8.
So, add that to the fact that Win7 filled a need the consumers had and Win8 just filled a need that Microsoft had and you have ample reasons why Win8 is where it is today. Even if there are people like you who apparently can't imagine other people might have a different take on what a great UI should look like.
Sounds like you're really fishing for a reason to hate it if that's all you can come up with. Swearing for dramatic effect doesn't sure up an argument as much as you think it does.
"The problem is that Microsoft is like a broken record. No matter what the question is their answer is always the same.
Windows.
Need a phone OS? - Windows
Need a Tablet OS? - Windows
Need a Desktop OS? - Windows
Need a Server OS? - Windows
Windows, Windows, Windows. It's supposed to be all things to all people. And it just doesn't work anymore. You're right, on the desktop nothing beats Windows. Even on the Server Windows is fine. But when they try to force Windows on to phones and tablets everything starts to fall apart. It isn't suited for mobile devices. It's too big, to bloated for mobile. And it's interface is terrible for mobile, but great for the desktop. So they come out with a new interface that's fine for phones but sucks on a desktop.
JUST STOP THE INSANITY.
You are exactly right. Focus Windows on productivity and the corporate market. If Microsoft feels that they have to get into mobile fine, come out with a mobile OS. But don't throw out the baby with the bath water. Leave the damn desktop alone."
In short, Windows 8 solved no real pressing problems for users - it solved - or was supposed to solve problems for Microsoft - which is to say a lack of penetration in what all the kids these days (and pundits) say is the environment of the future - mobile.
Microsoft has, and will have the enterprise and most of the desktop market locked up for the near future - for all the apps out there for whichever mobile platform, they pale in software for PC and the overall power and usability of the traditional desktop/laptop computer paradigm - and likely will for the foreseeable future.
I actually like Windows 8 more than any other version. I'll never use it full time (Linux user, right now SUSE with Gnome Shell) for reasons that have nothing to do with its design, but I really like the Metro interface.
Most of the naysayers are simply unwilling to accept change, the same naysayers that whine about modern Linux DEs.
Instead of relying in random comments, It would be great to see in an actual poll, how many people hate the new Windows 8 UI. In my personal opinion, the new UI is mostly fantastic. Perhaps I find it so, because I have strong visual memory, and usually adapt to new UIs in minutes. If it were available for Linux, I would consider it for my day to day use.
It is unfortunate, how it seems that so many people resist to UI advances. Even more so, on a place like HN. This kind of stubborn rejection seems pulled from something like a geeksquad forum. Another example is Ubuntu, which has collected massive hate, on considerable less groundbreaking UI changes.
Yup. Never understood the Ubuntu Unity hate either. While I prefer Gnome Shell, Unity is still an improvement on Gnome 2, and has some very interesting features of its own (like lenses and the HUD).
While I agree that the Windows team needs a singular vision about what they want their product to be, the advice given by the author is starkly reminiscent of the strategy adopted by Blackberry when they realised consumers weren't interested in using their products any more.
For Microsoft to abandon the consumer market and solely focus on a sub-section of 'doers' would be to set in motion an unstoppable slide into obscurity, as it did with Blackberry. Increasingly people only have one device and they're not going to choose a product which is only good for 50% of what they want. This mythical group of people who only care about using products for business just doesn't exist.
What they need to do is to make the product better so that people in general like using it again. It's that simple.
Even Jobs himself used the car/truck analogy to compare OS X to iOS. I think he's right. These are separate products. One is for work, the other one is more general purpose. There will always be more cars than trucks, but there will always be a need for trucks. You're not going to be hauling manure or towing a boat in a car.
The question for MS is will they be OK with selling trucks in a car world? Will they be satisfied with how much money they would make selling only trucks to farmers and construction companies?
I say MS needs to make cars and trucks. They need to make the best trucks out there without compromises. The best cars without compromises. Not this weird car/truck hybrid.
I agree. As with most bad products and failing companies, the problem here is focus.
Microsoft need to have a vision about what products they want to create and go about making them with a singular focus on producing something which customers will love. It's not about abandoning consumer software or business software, it's about making something great, regardless of it's purpose.
> What they need to do is to make the product better so that people in general like using it again. It's that simple.
Problem is that it's not so simple. They need to decide from which direction to make it "better", then drive that as a distinguishing factor. This can't be done if there isn't leadership from the top driving that priority incessantly (and demanding good reasons for exceptions).
E.g.: make an SSH built-in client so I don't have to install f*ing putty on every Win install to get decent work done. The cloud is here, and SSH is a core part of it's workings. Embrace and extend if you have to, but make it a native experience.
If you look at the evolution of OS X, the changes from version to version are minimal. In give or take 13 years it changed relatively little. It matured over 13 years to become a (love it or hate it) polished OS. However each Windows version has been completely different from the last version. Each release XP->Vista->7->8 Microsoft demands a lot of adapting from both their power users and their 'casual' users.
I've always felt that if they would have kept polishing XP or W7, diligently building on each version, they would've had a strong OS with a strong following. Maybe they should stop trying to do a "revolutionary redesign" every new release and just stick with one approach for a couple of years?
Totally agree. Between XP and W7 they moved a ton of settings around for absolutely no reason what so ever. Now they have done it all over again. There is no reason why "Add Remove Programs" should have been renamed to "Programs and Features." I am not saying that one name is batter then the other, just saying they should pick one damn name and stick with it, and polish the rest of the OS.
The change makes sense, considering no one actually adds programs from there anymore. To be honest, I don't remember a time one actually did. Still, I do miss it, even today it kinda trips me up not being able to look at the top left hand corner of the control panel.
7 was little different from Vista, to the point that some people were calling it Vista Service Pack 2. And the basic workflow between XP and Vista was pretty similar - they're both traditional desktops (though somehow vista takes 20GB to do what XP does in 600MB).
Maybe they should stop trying to do a "revolutionary redesign" every new release and just stick with one approach for a couple of years?
It's every second major release, not every release - 3.1 was alright; 95 big redesign, sucked; 98 serviceable; 2000 wasn't ready for primetime, but the redesign was adopting NT into the desktop OS; XP built on that to became the world's default OS; vista sucked; win7 was solid; win8 sucks; win9... ?
I completely agree with the "7 was little different from Vista" claim. I had purchased Vista Ultimate (stupid me) and then my only upgrade path was 7 Ultimate, which was so little different from 7 Pro that I completely regretted buying Vista. Especially when I found that the XP -> 7 upgrade price was the same as Vista -> 7 upgrade price.
In a sense, they did keep polishing. It's just that they had another, much larger team adding shiny features that became roadblocks to existing users.
I used to run Win7 in a form that looked very much like XP. The newer OS had many small improvements and refinements, not just the 'in-your-face' big ticket items. Windows 8 actually sounds the same - Microsoft published some interesting articles during development about lots of kernel-level changes, systems tweaks and so on (e.g. look at the windows process manager)
The problem has always been the massive forced-down-your-throat changes that MS insist upon, and Windows 8 took that even further.
Problem #1, where everything is starting to break down: Microsoft trying to put "Windows on everything". That may not be a huge problem if it was only the core base (although needing 16GB of space, something that's still the default for "high-end mobile devices" today, just for Windows alone, still sucks), but it gets much worse when they also want to have the "same UI on everything", which just doesn't work and never has. Even on the web, we still get to talk about "responsive design", because the same UI doesn't work on mobile devices, and it needs to be adapted for that form factor.
Problem #2: Having basically 2 operating systems in one, and forcing the user to switch from one to the other, by nagging him with all sorts of bad UX elements (charms on the desktop has always been a terrible idea to me, for example, and the UI looks completely out of place, too).
Problem #3: Price. I don't know if they're "feeling" this yet or not, but I think they are. If Windows for tablets cost $90 to license, then Windows tablets are simply not competitive from the get go. Yes, you may see some $300 Windows tablet, but it usually has much worse specs and quality than a $300 Android tablet. As I said, not competitive. This may be the reason why Microsoft may exist the consumer market eventually (3-5 years from now), and focus on enterprise, where they have much more lock-in and it's a lot more profitable/unit.
I recently bought a Windows 8.1 tablet for about $250. Specs-wise it's a lot better than my 2013 Nexus 7. 2 GiB RAM, a quad-core Atom, and 64 GiB of onboard storage.
And it runs full Windows.
Hell, I put VS on it and started writing C# apps with it. It's pretty nice. The two different GUIS are pretty jarring if you're running a desktop PC, but they make perfect sense with a touchscreen laptop or tablet.
The tablet game is about to change. Android may find itself crowded out of the mobile computer space.
Did it have a 1080p high-quality IPS screen, too? As I said, either the specs or the quality of the specs/materials suffer, to account for the Windows license.
Specs-wise it's a lot better than my 2013 Nexus 7. 2 GiB RAM, a quad-core Atom, and 64 GiB of onboard storage.
Followed by
It has a 1280x800 high-quality IPS screen.
So spec-wise, it has more storage, but apart from that, it has the same amount of RAM, and a display completely inferior to the 2013 Nexus 7.
I don't know about you, but I would trade storage for display anytime. You can always download/stream content from the cloud, but it's hard to stream a better display.
I think this clearly shows that different people have different needs, and that perhaps Microsoft needs to pick a crowd.
Most 'regular' users I know care a great deal more about a great display than storage. My mom would pay a premium for the former (although perhaps she specifically would actually not care and just want an intuitive interface so she can get to her games and email).
although needing 16GB of space, something that's still the default for "high-end mobile devices" today, just for Windows alone, still sucks
Not to mention virtual machines. Backing up and transferring around virtual machine images of vista and beyond really sucks hard. I know one ops guy who is keeping his XP install simply because it's base install size is about 3% the size of Vista an onward; much better for VMs.
Or you can just install Skype for Windows Desktop. Metro makes perfect sense when you consider a tablet formfactor like Surface, it just shouldn't be used for regular computers, which is where they screwed ujp.
Thanks, I didn't know that there were several versions.
I'll install it.
Metro and explorer are completely different worlds. There are even two configurations systems and I've yet to find how to reach the standard control panel without using search.
or open Explorer at "Computer". There is a big "Open Control Panel" button in the ribbon (reachable via Alt+C, C).
That being said, search brings me exactly where I want, while the control panel is not my end goal, so I didn't have to open it in quite a while (both on Win 8 and Win 7).
I thought that it was pretty well established that Windows is in an alternating cycle of good and bad versions. Everyone freaks out during the bad part of that cycle, as Microsoft takes a bunch of bold risks, many of which don't really work out, and then a few years later they prune many of the bad decisions and release something reasonably solid.
(Man, it feels weird leaving this comment as a Linux user and reformed Mac zealot)
I think Windows 9 (or whatever it's going to be called) will be a complete pull back into the classic desktop. By the time it ships there will be Modern UI only OS hybrid built with Office for the Modern UI finally caught up. They can all have the same underlying base OS (whatever it's going to be called) with different shells on top. But, I'm not sure how programming model is going to look like, unless they allow classic desktop Windows to run apps built for Modern UI in windowed mode (which seems to be the direction they are heading towards).
Posting this with a throwaway as I don't want it to be associated with my named account.
Thurrott is full of shit. I call for him to step down and shut down winsupersite and to generally shut up. He's not a windows user -- merely a critic. This is about spinning hits than a valid critique of windows. All he is doing is damaging the company and the industry by recycling this shit over and over again.
There are those of us who use windows both for our own personal use and business use. It suits us well simply because there is nothing else with the broad scope that Microsoft offer. Windows 8, whilst not a great success, is exactly what people asked for. The tech news sites spewed how the future was tablets and closed ecosystems powered by app models ao they dived into the consumer market with exactly what people expected. Unfortunately they discovered that the consumer market (which didn't really exist) is just a bunch of fashionista journos' wet dreams and fanboys.
Now I'm a Unix guy who saw exactly the same thing happen in the 1990s when Microsoft kicked in the Unix vendors. However they did it through actually providing products people wanted and providing long support lifecycles rater than disposable trinkets and rapid change.
Give them a break and look at your 2 year life throwaway trinket phone and tell me that's what you want for the future? That's what you asked for, that's what you got.
Thurrott is one of the biggest fanboys on the internet behind the crackberry forums. For him to backpedal like this maybe says theres something seriously wrong with the execution of Windows 8?
If you don't trust his opinions on MS because his is obviously not rational when it comes to them (often biased in Microsoft's favor), why would you trust it when he abruptly changes them?
That's the problem with fanboys and people obsessed with these things, they tend to obsess with other things as well.
I usually read them with my filters set on maximum level.
"...That's mostly business users, but even when you look at the consumers who will use Windows, that usage is almost entirely productivity related. Windows should focus on that. On getting work done. On an audience of doers. Job one should be productivity."
Won't that imply support for legacy applications (which Windows is good at) going back some years? Hence maintenance of features required by those applications?
Microsoft should start from scratch and make Windows2 Desktop, Windows2 Mobile and an OpenBSD-based Windows2 Server.
Cram thousands of users into a hall using all sorts of versions and customizations of Windows along with A/B testing, heat maps etc. to figure out the fastest and easiest way to get a thing done. This way users won't have to spend hours trying to customize their Windows (read: disable all the new stuff).
Also maybe throw a world-wide design contest to find the best suited man / woman to be responsible for a new Windows2 design.
Like they say, "perfection is achieved, not when there's nothing left to add, but when there's nothing left to take away".
Interesting and apt use of Steve Jobs's truck vs. car analogy at the end:
"Maybe Windows needs to be more like GMC, the part of GM that only makes trucks (and truck-based SUVs). After all, while many people choose to use a truck for basic transportation, they're really designed and optimized for work. You know, as should be Windows."
It's not just Windows. Microsoft has a virus design team. Look what they did to visual studio. The new visual studio flat design (from 2012) is really awful. All caps in menu. No color in icons. Everybody hates those changes.
Well at least the VS desgin team has to listen when there's enough complaining: Visual Studio 2013 has been available for a while and it fixes all problems you list by undoing those changes.
Of course, everything in a Microsoft product can be customised by changing or adding a registry key somewhere. The all-caps menus in Visual Studio are no exception.
The Mac was apparently designed by a person I can't freakin' stand. I recently spent a year with an OSX laptop, and it was a merciful relief when I was finally able to get the hell out of iLand. My first love is KDE because it bows to my will like nothing else, as a good window-manager/OS should. Win8 is also pretty malleable, even though it's not so good for those lacking the chops to bend it to their will. OSX, by comparison, is a constant freakin' PITA if you don't happen to be the "single person" it was designed by and for.
This might be a bit of a rant but I'm sick of people ragging on Windows for its default settings. Windows is eminently customizable. If you aren't a complete n00b you can bend Windows to suit yourself far more easily than OSX. KDE absolutely spanks them both, but still... I would wholeheartedly recommend OSX or iOS to my grandmother, because she's not so good at customizing things. This blog is aimed at people who probably aren't running their OS on 99% default settings. Customizability matters, and Windows fucking owns OSX in that respect.