Previous: Kindle reader on Ubuntu
Michael Mace posted a very interesting assessment of Microsoft Windows 8 on his blog, Mobile Opportunity (see Fear and Loathing and Windows 8). I have a few observations to add, and since they are a bit verbose, I thought I shouldn't clutter his comments section with them.
I decided to post these observations on my Ubuntu blog, although the subject isn't Ubuntu, because some of Michael's readers are starting to ask about alternatives to Windows. They might find some of my "noob" experiences with Ubuntu useful. The short version is this: You won't find Ubuntu to be hard to install or use. If you're already accustomed to Windows and comfortable using it, then rest assured that any reasonable consumer-oriented alternative will be easier, not harder.
Learning curve worries
The tiled interface, Metro, is very different from previous Windows UIs. Michael writes that people have "spent a huge amount of time learning how to use the system, and the last thing they want to do is start learning all over again."
The statement really captured my attention. It serves as a reminder to me that "end users" approach computers very differently than programmers. All operating systems perform the same basic functions, such as locating files and starting programs. All user interfaces provide some means of accessing those functions. Any user interface is fairly straightforward, be it a command line, a menu-based UI, or a GUI of some sort with menus, a launcher/dock, or icons/tiles.
But even in view of different people's comfort levels with different UIs, I'm not convinced that the Metro interface will confuse the average user who has experience with Windows. In my work, I see lots of people who use Windows every day. Most of my clients have settled on Windows as their corporate desktop standard. Most of the people I see who are comfortable using Windows fill their desktops with icons. The icons are shortcuts to programs. In quite a few cases, I've seen the entire screen filled with icons. People tend to use the Start menu to locate a program just once, and the very next thing they do is create a shortcut and stick it on the desktop. Done and done.
If I take my Android phone out of my pocket and look at it, what do I see? A screen filled with icons. This is a tiled UI. It's a bunch of little rectangles, each of which invokes some program when touched. The program might live on the phone, or the tile might just be a gateway to a remote service. Nearly everyone in the industrialized world has a smart phone and/or a tablet device with a touch-screen, equipped with either iOS or Android.
The tiled interfaces on smart phones and tablet devices are essentially the same as Metro. People are already using tiled interfaces. Furthermore, the way in which people tend to use Windows looks like an attempt to get as close to a tiled interface as they can, using shortcuts and icons on the desktop. There is no learning curve problem. Windows users already know how to use Metro. They wish they were already using it. They are simulating it as best they can.
Michael uses a car analogy to illustrate the significance of the UI change. He compares Windows 8 to a new car that uses a joystick for steering as opposed to a steering wheel. The problem with the analogy is that Windows 7 is not a car with a steering wheel. It is a car that you have to steer by pulling on a rope. It is a car whose engine takes 5 minutes to start and 25 minutes to shut down. It is a car that goes slower and slower the longer you drive it, and that may just stop dead in the middle of the road for no apparent reason. Backward compatibility is a good thing when the old system worked well. When the old system wasn't so great in the first place, then, well...<shrug>good riddance</shrug>.
Corporate perspective
What current Windows users don't already know, maybe, is where all the system configuration goodies went. Michael notes that some configuration options are no longer available directly through the UI. They've either been moved (as there is no Start menu anymore) or eliminated altogether. People can still change user preferences, like background images and desktop themes and so forth.
What he doesn't say in his article, but must be true nonetheless, is that all those configuration options are still in "the system" somewhere. I mean, they would have to be in there somewhere, right? Otherwise, the bloody thing wouldn't work. They're just hidden from the casual user's view. I recall when Windows 7 came out there was no longer an easy way to change file associations; you had to do it by editing the Registry. When I installed my preferred office suite, I discovered that Windows was preconfigured to try and launch Microsoft Office no matter what, so I went a-Googling for solutions. Irritating. In Vista, you could get rid of of those annoying notification balloons by editing the Registry, too. So, where there's a will, there's a way to make Windows usable. But Michael's article is not aimed at people who edit the Registry.
Anyway, to get back on topic, the Metro interface seems well-suited to corporate environments. My guess is that Microsoft makes most of its money from corporate customers. The individual who wants to use social media and watch videos probably doesn't account for a large proportion of Microsoft's revenue. A single corporate customer could be worth 10,000 individual customers.
In corporate environments, companies typically preconfigure the desktop systems with the applications necessary to support each employee's job function. They don't want employees mucking about with the system configuration, plugging in memory sticks, downloading executables, installing programs, or having local admin rights. There are plenty of good reasons for this, including supportability, governance, security, cost control, managing network load, and license compliance, to name a few. Since most corporate computer users seem to prefer filling their desktops with icons anyway, which is a poor man's Metro interface already, the Metro interface will only give them what they want straight out of the box with no hassle. That can't be a bad thing.
Companies are busily implementing service-oriented architectures, outsourcing non-mission-critical support functions to external suppliers, and shifting some of their workload to the cloud. The end-user interface to all that stuff can be thin and simple — just a bunch of icons or tiles or what-not that invoke services. The services may reside locally on the PC, within the enterprise but behind the service bus, at an external supplier's IT facility, or somewhere out in the cloud. It doesn't matter.
The Metro interface is an excellent fit for that model. A company's central IT group can preconfigure machines with tiles representing the information services each employee requires to perform his/her job function. It's better than desktop shortcuts because the tiles can be remotely managed to point to new or moved services as needed, with no end-user intervention required. There is no need for those people to configure their own Windows instances. Therefore, the lack of convenience for doing so is not a problem in that context.
Consumer perspective
I'm not a close follower of Microsoft. I don't read much of their marketing material. Yet, based on what I've heard and read, they have strategy that involves providing a consistent user experience across all platforms. A person using a conventional computer, a tablet, a phone, a game console, or some device to be invented next week, should have a broadly-consistent general user experience. The Metro interface seems to be a step in that direction. From that point of view, it appears to be a step in the right direction.
The trend these days seems to be away from conventional PCs for general consumer use. Tablets, smart phones, and perhaps more innovative devices on the horizon will supplant the PC as the platform of choice for casual use. The Metro interface carries the UI metaphor from these newer devices to the PC. This is probably a smart move for Microsoft.
The notion that Windows 8 isn't really targeted to conventional PCs is supported by some of Michaels other observations. For instance, Windows 8 doesn't support all the peripheral devices found on PCs designed for Windows 7 and Windows Vista. In addition, it isn't laptop-friendly. It doesn't want to be turned off. It doesn't make power management options easy to find. These are features that suit corporate desktop platforms and non-PC devices such as tablets and phones. Corporations want to be able to push security updates and so forth off-hours, and it's more appropriate for tablets and phones to sleep instead of shut down, most of the time. It's possible the conventional PC has passed its market prime.
Support for "productivity"
Michael worries that Windows 8 is geared for information consumption rather than information creation, and will therefore be an impediment for people who create information in their work, like office workers who use office suites to prepare reports, presentations, and spreadsheets. But I've seen no indication that products such as Microsoft Office or Outlook will not be supported. A tile can just as easily invoke those programs as anything else. So, I'm not sure this is a valid worry.
Oddities
Michael notes that the Windows 8 preview release wipes your hard drive and irrevocably installs itself. He notes that Microsoft does not make this clear enough to people who are browsing the preview release site. It's possible some people will get burned.
I my case, I installed the preview release as a virtual machine and ran it under VMware. It wouldn't have occurred to me to try out a new operating system by actually installing it directly onto a "production" machine, even if it's an OS with a trustworthy history (i.e., not Windows). It was only when I read this in Michael's article that I realized some people might just do a normal install. If you want to try the thing out, be aware of this, er, feature.
Michael also discovered that there is an expiration date on the Windows 8 preview. All these preview copies will expire next year, and customers will be expected to purchase a copy of the real thing. I'll just delete the VM file. Come to think of it, why not do that right now? Just a sec, I'll be right back. There, done. It was interesting to explore Windows 8, and I can see the potential benefits for corporate customers, but I've got no real use for it. Although the tiled interface seems to be consistent with current industry trends, it still isn't a particularly good OS. I found it to be annoyingly clunky and slow. Have to bear in mind it was only a preview release and it was running as a VM. But still...really? Nah.
Another funny bit is the quote from Antoine Leblond, VP of Windows Web Services, in which he boasts that Microsoft have sold over 500 million licenses for Windows 7. I couldn't help thinking of the moment when I first powered up my new laptop recently. I plugged in a USB stick with a bootable Ubuntu iso first. I never brought up the pre-installed Windows 7 software at all. When I purchased the machine, I was purchasing the machine. The fact it came with an OS pre-installed was irrelevant. It makes me wonder how many of those 500 million licenses sold actually turned into licenses used.
And that brings us to another interesting tidbit. My nephew works in the game industry, where people have to be one step ahead of the general public when it comes to the details of new operating system releases, because those details affect their code. His understanding is that Microsoft intends to force OEMs to rig their hardware so that customers are unable to install any other operating system besides the pre-installed Windows one. I suspect this really means they would rig the BIOS. ("Force" means "if you want the official Windows 8 sticker on your machine, then you must do our bidding." Obviously, they can't physically force anyone to do anything, although their behavior sometimes suggests they would enjoy being able to do so.) If true, the strategy could create very interesting lawsuit opportunities against Microsoft, and possibly against OEMs who fail to provide a non-Microsoft alternative.
We live in interesting times.
Next: Quest for a good Twitter client