It is not going well. Here is a user comment on Gizmodo that says a lot about the majority reaction to Windows 8:

Im still using a computer with win7 on it, so I had not had much experience with win8. That changed when I helped my girlfriend buy a new computer for her mom. In all honesty, we found the win8 interface a huge pain. It took forever to figure out how to use it, and in the end we did everything we could to get back to a normal non-tile setup. Heck, it takes like 3 menus just to get to the shut down screen! Its insane, and the overall experience for the 3 of us was negative (two of which are very experience computer users).

Spelling and grammar left as-is! Note a few features of this comment:

• The person who wants the new Windows machine is the parent. I notice this frequently: a significant part of the Windows market is older (I am not saying elderly!) people who are less interested in the latest shiny new thing and more interested in sticking with their existing trusted apps and familiar user interface. The changes in Windows 8 are not welcome to this kind of user.
• The new features in Windows 8 are perceived as obstructive.
• There is no mention of touch control or new-style apps. It is a guess, but since the largest market for Windows is low-end laptops without touch screens, it is a fair bet that this is what it was.

The business world is different, of course, and here things like Hyper-V virtualisation or Windows To Go secure deployment are more likely to be appreciated. People are not so different though, whether they are at work or at home, and given that most Windows users spend most of their time on the desktop (hard to avoid however much you like the "Modern UI") I've noticed similar reactions from business people trying out Windows 8.

That is, if they have tried it at all. Most businesses I encounter are sticking with Windows 7.

With PC sales in probably permanent decline, in favour of other computing form factors, is Windows now set to become a legacy workhorse operating system? Will it ever break through on tablets?

Microsoft's official position, as far as I can tell, is that everything is fine. Leaked builds of the next major update, known as Windows Blue, show only minor changes:

• New 50-50 Snap view for modern apps
• Easier and richer Start screen customization
• New tile sizes for the Start screen
• Internet Explorer 11

Of course there will be more to come; the recently announced Build conference in June is when we will find out more. The stage is set for Blue to be generally available by the end of the year.

I am guessing though that Microsoft does not intend to implement what many users would like, things like:

• Making the Modern UI easier to ignore, by booting straight to the desktop and having file associations set by default to desktop rather than Metro apps
• The ability to run Windows Store apps in a desktop window
• An option to use the old Start menu

I do see Microsoft's thinking here. There is not much point in making a brand new platform, and then designing it so users can easily ignore it.

It is also true that if Windows 8 had simply been like Windows 7, but a bit better, that would have done nothing to stem its decline.

On the other hand, Microsoft's attitude to the problems people have with Windows 8 seems to me denial. More could be done to help desktop (that is, most) users get to grips with Windows 8. For example, I would like to see small visual clues to the presence of menus and "Charms" features (the right-hand menu which hides many settings, search and sharing features), an easier way to raise the Charms menu with the mouse, and a visible Start button on the desktop.

Windows 8 usability is fine if you make the effort to learn it, but making that effort is hard to justify when the world of modern apps is uninspiring. This is the second and most important area that must be improved, if the world is ever going to want versions of Windows beyond 7. There are few delightful apps, and a large part of the reason is that the built-in controls available to developers for building apps tend to be blocky in appearance, inefficient in use of screen space, and short on important features.

Miguel de Icaza, of Mono, Gnome and Xamarin (C# for mobile) fame, said on The Register that "the new Windows basically has no style. The style is DOS with large fonts." A little unfair, but he has a point.

We also need to see further unification of Windows 8 and Windows Phone 8. The phone side does have some momentum now, and the app story is better. Why not enable Phone apps to run on Windows 8, as iPhone apps do on the iPad, as well as making it easier for developers to target both phone and tablet with new projects?

Finally, it is time Microsoft came up with inspiring examples of Windows Store apps that really are a joy to use. There are a few good ones. I still like the weather app, and Fresh Paint is a good effort though sadly has not made me into an artist. More is needed though, and I am guessing that Microsoft's own developers have the same problems that third parties have faced in trying to code for the new platform.

Yes, Microsoft does need to fine-tune the Windows 8 user interface to make it more enjoyable for upgraders; but what is even more important is that Windows "Blue" needs to improve the Windows Runtime platform. That is the thing to watch for at the forthcoming Build.

How are we doing at teaching our kids how to code? Until relatively recently, the answer was 'not that well'. Hopefully, things will improve now that the UK government has seen fit to include computer science as a key skill. From next January, it will be part of the English Baccalaureate, and will be counted as a science in school league tables. In the long run, the Baccalaureate certificates could replace some GCSEs.

For a long time, kids in the UK were taught ICT, rather than computer science, as a part of the science curriculum. ICT concentrated on teaching kids how to use technology, but didn't necessarily talk about how to understand it. It's one thing to learn how to create a Word document, and move things around in it. It's all very well practicing how to put it in an email attachment and send it to someone. But this is a far cry from understanding the underlying mechanics of technology.

In his book Program or be Programmed, Douglas Rushkoff talks about how, when any important information revolution comes along, power tends to be divided along the lines of those who consume, and those who control. When the alphabet was developed, those in power used it to articulate themselves in writing, while everyone else listened to them read. By the time the printing press was developed, more of the hearers were able to read, but publication was a privilege enjoyed by a select few.

Now, in the digital revolution, Rushkoff suggests that we face the same challenges. We are in danger of creating a nation of workers who use technology without really understanding it, and who therefore can't manipulate it outside of carefully designed parameters.. We can watch countless dogs on tightropes on YouTube, just so long as we're willing to work within the confines of Google's system. We can create Powerpoint presentations, as long as we're happy to accept what Microsoft gives us.

Coders move beyond those confines, and are able to create and manipulate, rather than blindly use. Coders don't have to be content with the same old mousetrap; they can build a better one

Now that computer science has been promoted to a first class citizen, we may see students getting more interested in programming. This will be a major boon to an industry that finds itself perennially short of coding skills.

There's just one problem: we're not starting early enough.

The computer science component that the government is including in the baccalaureate covers GCSEs, making it applicable to teens. What about kids in elementary school, where interests are first realized, and inspiration can take root at an early age?

We need to show children what technology can do at an early stage, rather than leaving it until later, not simply by exposing them to video games but by showing them how to make their own programs. 

The tools for this are already available. MIT produces a wonderful program called Scratch for young children, designed to show them the basics of programming structure through the use of graphical aids. I tested it on my kids, and they were creating a 'play' featuring two separate programming objects within half an hour.

If we don't start our kids early and get them truly fired up about coding, where will our next generation of web designers, coders, and system architects come from?  

1. Submit your proposal/abstract early. If not the first week, do it the second. You will get better feedback. Yes, I submitted mine the second week and got great feedback. I was able to get multiple rounds of feedback on all my sessions.
2. Try OneStartlingSentence as a way to write the first paragraph of the abstract. You can always write more paragraphs if you have room. This will allow you to grab the reviewers. 
3. If you have a chance to be a reviewer for that conference (or any other conference), grab that chance. You will learn from your reviewing. You will also grow your network, which is a great thing.
   

January was a dark month for Github. The collaborative source code management site was found to be sharing the private SSH keys of many members via its public search function.

The website, which helps programmers in far-flung locations to collaborate with each other, had just upgraded its search function with many new features. However, this caused several enterprising hackers to take another look at its search functionality.

Git hub works using a series of repositories. These are folders that hold the source code for software that a developer is working on as part of a collaborative project. A private repository on a developer's own machine is replicated with a public one on the Git hub site, enabling that developer to work on their own version of a piece of source code, before it is then incorporated into the main source code along with everyone else's changes.

Unfortunately, it turns out that many developers are not very conscientious when it comes to security. They copied over the entire contents of their UNIX machines' home directories into their private repositories, which were then copied up to the public folders. By default, UNIX stores SSH keys in the home folder.

SSH is a certificate system designed to make it easier to access remote computing services without continually re-entering passwords. When a user generates SSH keys on their own computers (which can be done with a single command line instruction), it creates a private and public key. The public key can be given to servers that the user wants to access transparently via different tools on their computer. The private key is supposed to stay with them, and never be distributed. 

If these private keys are made public, then an attacker has the keys to the kingdom, because they can access any online services that the user is logging into. What makes it worse is that the user's computer also keeps a list of these services on their machine in a 'known hosts' file.

So, until Github recognised what was happening, links to people's private SSH keys were popping up in its search results. This could have had far reaching ramifications. Developers' machines may already have been compromised without their knowledge. Their Github accounts could have been accessed, and malicious backdoor code could even have been inserted into their project code.

But who was at fault here? Was it Github, for making the search results available, or was it the developers themselves, for not understanding security well enough to protect their own private keys?

And if developers are making rookie security mistakes such as this, how much should we be trusting them to produce secure software?

This proved the case again at the Monki Gras recently, an unusual London event run by analysts Redmonk.

Craig Kersteins and Matt Thompson from Heroku asked a question: how often are you interrupted at work? Software development is partly about keeping a lot of information in your head so that you can see patterns and make connections, and avoid bugs by remembering exactly how the code you are working on fits into the application. Getting a summons to a meeting or a call from a colleague in the midst of that kind of concentration is costly. They even put a figure on it. 76% of the worst-performing engineers are frequently interrupted, they said.

How often is frequent? That was not stated; but they did suggest aiming for 4 hours of uninterrupted work each day. That still leaves plenty of time for meetings; and I have little doubt that 4 hours of good work counts for more than 8 hours of choppy work that leaves you feeling that you should not have bothered to turn up.

More human factors: Mazz Mosley and Nick Stenning from the UK Government Digital Service advised us not to recruit "rock star" developers who become a single point of failure, as everything stops if they become unavailable. A team with collective intelligence is better.

Ted Nyman from GitHub weighed into managers. They do not have any. I was reminded of a comment from Joyent's Bryan Cantrill at Monki Gras 2012: "it is very hard for middle management to add value".

Do I think that most companies remove all their managers? That is neither realistic nor likely to succeed. As another attendee observed, companies with managers generate a lot of revenue.

The point though is this: the way developers are managed impacts their productivity. That human factor matters more than whether they use Java or C#, or which tools they use.

I hear similar insights from the QCon conference in London each year. Coming up in March and recommended.

Shanley Kane from Basho spoke about honesty in software development. Roadmaps are a lie, she claims, because attempts to map features to a timeline will always fail. When roadmaps fail that erodes trust in the team. Interactive "what we're working on" documents work better, she said.

I will leave the last word to Cyndi Mitchell from Logscape and Thoughtworks, who remarked at Monki Gras that "Software is fundamentally a human, interactive activity. If you don't understand that, forget it."

Verizon's security team just released a bizarre case study, describing a company that approached it after seeing some strange network traffic. It was experiencing a VPN connection from China, for no reason, which often stayed up for a day at a time. Even weirder was the fact that the person at the other end of the VPN connection was using two-factor authentication to check into the account.

While investigating the problem, the security team decided to trawl the hard drive of the account's legitimate owner, an employee inside the company. They were looking for malware that may have been planted by an attack. Instead, they found dozens of invoices from Shenyang, China. It turned out that the employee, a software developer for the company, had been outsourcing pretty much all of his job to low-cost labour on the other side of the Pacific. They were able to access the system because he had Fedexed his RSA token to them. 

The developer, who was paid a six-figure salary, was paying a fifth of what he earned to the overseas contractor. That's smart if you're someone who wants to get your job done for your while still earning a hefty salary. It's smart if you don't care about ethics, or security, or the wellbeing of your employer or the broader community that they serve. 

It's also incredibly insecure. Many security attacks have been documented as coming from China. If someone else is accessing your systems and writing your code, then they will also have carte blanche access to your infrastructure, and potentially elements of the infrastructure beyond. This company was also part of the critical national infrastructure, said Verizon. Suddenly, stories of cyberattacks on oil and gas infrastructure and malware planted in the electricity grid seem far more plausible.  

What's irritating isn't just the guy's irresponsible actions; it's the fact that he wasn't even doing anything productive with his time. The Verizon team's blog on the subject outlined his average day thus:

9:00 a.m. - Arrive and surf Reddit for a couple of hours. Watch cat videos

11:30 a.m. - Take lunch

1:00 p.m. - Ebay time.

2:00 - ish p.m Facebook updates - LinkedIn

4:30 p.m. - End of day update e-mail to management (ironically, he got consistently excellent performance reports). 

5:00 p.m. - Go home

He was also lazy enough to have the Chinese works connect directly with their VPN, rather than running a proxy at his house and having them connect to it first. That's what got him caught - thankfully for the company concerned, which presumably would now want to audit its internal network and pull a full static analysis of its code, to see if there are any security holes.

This is the darker side of the extraction theory proposed in Tim Ferris's book The Four Hour Work Week. In that book, he advocates distancing yourself from your company and working from home where possible, so that you can be more efficient with your time and start your own lifestyle business. He also suggests using a virtual assistant to take on mundane tasks. It seems our enterprising developer skipped the lifestyle business part, and just got someone else to do his job. 

Or perhaps this was his version of a lifestyle business. The Verizon team said that evidence suggests he was pulling the same scam in multiple companies.

There are most certainly things you can do to make your job easier as a sysadmin or software developer. I know of one tech expert who took a job as a sysadmin for a US company, demanded to work from home, and then scripted 80% of his tasks. His job ran smoothly, there were no security risks, and he was free to get on with other things. But that takes real smarts. 

Lessons learned here? For companies: watch your system logs more closely (this had been going on for over six months). For employees: sure, work efficiently, do what you can to automate your job and make it easier - but never, ever step away from your work ethic or basic trustworthiness.

We have seen a kind of preview of BlackBerry 10 in the unsuccessful PlayBook tablet, released in April 2011. This is the first RIM product based on the QNX operating system. QNX Software Systems was acquired by RIM a year earlier, in April 2010. That said, the PlayBook runs the PlayBook OS, not (yet) BlackBerry 10. BlackBerry 10 SmartPhones will have a new user interface and many new features.

I spoke to William Vablais, Head of Developer Relations EMEA for RIM. "We've been very successful in changing the sentiment of developers," he claims. "The interest level has been rising significantly."

One would expect him to say nothing less. But what is distinctive about the BlackBerry 10 platform; what does the it give you that couldn't easily be done on iOS, Android or something else?

Vablais points first to the diversity of development approaches it supports.

"We have SDKs for C/C++, we have entry points for designers and developers for HTML and CSS, we have entry points for Adobe AIR," he says.

There is also an Android runtime which makes it possible to repackage Android apps. Vablais observes that it can pay to offer your app on a minority platform.

"There's a community out there that developed for Android who don't have any exposure or visibility in that world because it's such an overcrowded market," he says. "They can take their application, port it to our platform, and suddenly they get visibility, generating revenue."

Fair enough, but what does the BlackBerry 10 platform give you that cannot easily be done on some other platform?

Vablais points to two key BlackBerry 10 features that he believes will draw users to the platform. One is social netwok integration. "We have the social network capabilities built into the OS," he says, referring to BlackBerry Flow and BlackBerry Hub:

BlackBerry® Flow is a new user experience that allows seamless navigation across open applications and the BlackBerry® Hub. All messages, notifications, feeds, and calendar events come into the BlackBerry Hub and no matter what the user is doing with the device, with a simple gesture, they can peek into the Hub at any time.

says the press release,

More important to business users though is security. "What no-one else has is that the OS and the framework has been based on security. The user interface and some of the components allow you to split out work related data from your personal related data."

This is the feature called BlackBerry Balance. Again, here is the official description:

BlackBerry® Balance™ offers the most elegant way to satisfy both customer and corporate needs without compromising on either. With BlackBerry Balance, personal apps and information are kept separate from work data, and the customer can switch from their personal to work profile with a simple gesture.  The work profile is fully encrypted and secure, enabling organizations to protect their content and applications, while at the same time letting customers get the most out of their smartphone for their personal use.

In the era of BYOD (Bring Your Own Device), this does sound like a great feature. The industry is only just coming to terms with the idea that smart devices are personal; they do not live in the office and they will be used as home as well as at work. If BlackBerry 10 makes sense of maintaining work and personal data on a single device without compromising security or the user experience, then it could indeed be a game changer.

The success of the original BlackBerry phones was primarily based on its appeal to business users, and RIM already has tools for deploying internal apps in a managed and secure manner.

Even an excellent platform counts for nothing if you cannot market it successfully in a world now dominated by iOS and Android, as Microsoft has discovered with Windows Phone. Whether RIM has enough resources to establish yet another mobile ecosystem must be an open question.

At the same time, there is a lot to like. QNX has long been an excellent embedded operating system, and if the devices are excellent and the security lives up to its promise, BlackBerry 10 could be a significant platform for mobile buisiness apps.

Mark January 30th in your diaries and watch with interest.
 

1. Define your team norms. I like my definition. No surprise there! But if you don't have a cross-functional team that includes testers, and you are writing your own automation, decide what you are going to call done for your team.
2. Write your own automated tests and decide what level they will be at. If you are a developer, you need to write automated unit tests. If you have no testers on your team, someone has to write system level tests. Or, something that looks like system level tests.
   
3. Consider swarming around each feature as a team, so every person chooses a role for each feature. Trade off roles for a given feature, so you all share the love of writing code and tests equally. This is especially good if you have technical debt and you have to pay it off. If you know you have a feature that will touch on an area of large technical debt, the swarming will allow to work as a team to address the debt and pay some of it off.
4. If you don't have testers on your team, make sure you raise this as an impediment and have some management person address this. Remember, agile teams are cross-functional teams. Yes, people can play other roles, but if you are not, by nature, a great tester, it will show in the product. 

If, for some reason, you have a handoff to another group before the product can be released, step back and examine why. It's not a crime, but it does create another step. Is there a good reason for that step? If so, consider ways to integrate that step into your work. Is there a way to integrate?

The more you integrate all the work for a feature into your cross-functional team, the more agile you will be. That is, the more able you are to release an increment of product, get feedback, and accept change. And, that's the point of your definition of done.

So, when you think about done, think about a definition of done that works for your team and your environment. And, if your definition of done is not release-able, re-examine your definition every so often to see if it is still working for you. Maybe things have changed.

These days, learning to program needn't be quite so painful. A rising tide of online programming schools is making it easier for people to learn the basics of any language, and take things further, gaining a respectable, useful knowledge. They vary in format, running the gamut between interactive and simply informative. Here are a few to whet your appetite.

Codecademy

This site provides interactive browser-based lessons, explaining the basics of programming and explaining what to type and where, so that you can get a solid idea of programming rules and structures.

Udacity

This higher education site covers a range of beginner, intermediate, and advanced courses. It is broader than Codecademy, covering physics and statistics in addition to meaty programming topics such as how to build a web browser, and HTML 5 games development.

Code School

Code School offers a variety of in-browser programming courses covering Ruby, Git, CSS, and jQuery among others. Some of the courses are free, while some cost money. The service is aimed at both individuals and teams, the latter of which can be monitored by a manager. Code School also runs a couple of other sites: a free Interactive Ruby tutorial called TryRuby, and an applied, tongue in cheek Rails course called Rails for Zombies.

Treehouse

Aimed at new and experienced developers, Treehouse has two plans: Silver, at $25 per month, and Gold, at $49 per month. Both give you access to over 700 videos, but the premium plan includes access to interviews with industry pros, project feedback, and new ideas to inspire your programming skills. Like the other sites, it enables you to learn by doing, rather than simply watching the videos. If you are into iOS or Android development, Treehouse has special tracks for these.

O'Reilly

I'd be remiss not to mention this heavy hitter, which offers a range of novice, intermediate, and advanced courses covering everything from an introduction to C programming through to database administration, PHP, and Java. The price for these courses is a little steep, though; even a basic introduction to JavaScript will set you back $450.

Lynda.com

Lynda.com is heavily video based. Instead of providing in-browser interactive coding tutorials, it provides videos to watch, along with exercise files to play along with using your own software. This approach stems from its breadth of courses. Detailed programming tutorials for Java, Perl, ASP.net, and Ruby nestle alongside other courses ranging from negotiating your salary through to advanced photo shop CS5 tutorials. There is something for tech-heads of all kinds here. The site had 187 courses designed for developers at last look.

Google

Yes, Google is your friend, but I'm not fobbing you off by asking you to search for 'Python courses online'. Google runs the Google Code University, which contains a variety of educational materials covering programming, web security, Android, and distributed systems. Materials are scattered and noninteractive, but they can be useful pointers. The University offers discrete classes in Python and C++, along with online coding exercises. These classes aren't designed for glitz and glamour, but they are instructive for the already technically adept who want to extend their range.

Coursera

This site features courseware from a range of universities including Stamford, Princeton, and the University of Toronto. There are some applied practical courses, such as Functional Programming Principles in Scala, with specific programming exercises. Some courses, such as Introduction to Databases, are self-study, meaning that all of the online videos and materials are available at once. Others are released at a pace set by the University, and result in a certificate.

Some of these courses are more intensive than others, and aimed at a higher level, but that's one of the great things about learning online; there is something for people at every stage of knowledge. 

Of course, the gap in most of these online courses is real world experience. Learning how to code in a browser is one thing, but applying that in a commercial environment, which will have its own restraints and requirements, will present its own challenges. Still, in many ways it beats teaching yourself out of a book. And given that many of these are entirely free, or at least offer a free trial, it's worth a go.

Maybe I'll even have the confidence to try. Mr Eales told me that I was going to fail my A-levels, but I still turned out with a respectable B. So there. Not that I'm bitter, or anything.

The controversy comes as no surprise to anyone who has been paying attention. Windows 8 is two operating systems in one.

On the desktop side it is a slightly upgraded Windows 7: slightly faster and quicker to boot, revamped File Explorer, better Copy function, improved Task Manager, Hyper-V virtualisation, and a few other things, though a little less pretty thanks to the removal of the transparent Aero theme in favour of a plainer appearance.

These are nice improvements, but the real innovation is in the new Windows Runtime platform, in which apps are sandboxed, support touch control, run full screen, and are easily installed from the Windows Store or by distributing an application package, which is called sideloading. Sideloading is mainly restricted to developers or enterprise deployments.
Microsoft ensured that the tablet personality would not be ignored by replacing the Start menu with the Start screen, a new-style app. This also ensured that newcomers to Windows 8 would find it unfamiliar and difficult to navigate.

Another contentious issue is that Windows Runtimes apps - officially called Windows Store apps - look stupidly large and bold on large desktop displays and do not work at all across multiple monitors.

Underlying Windows 8, especially in the new tablet personality, is a design concept with a couple of notable features. One is "content before chrome", the idea that content rather than screen furniture should be foremost. This has led Microsoft to have many essential controls hidden by default; you have to right-click or swipe to see them. You can see the point, but unfortunately this is bad for learnability.

Another is a grid layout, giving rise to the tiled appearance which characterises Windows 8. It is neat and logical, but constantly looking at boxes can be wearisome.

That is the preamble. What is the business impact of Windows 8? Here are a few observations.

One is that the learnability issues in Windows 8 are sufficient to deter upgrades, both in the consumer and business market, even though it is an improvement on Windows 7 once learned. Bearing in mind that the desktop improvements are relatively minor, most IT administrators will conclude that Windows 8 is not worth the hassle.

Therefore, we can expect only a slow uptake by businesses.

Against that, Windows 8 does work much better than Windows 7 on a tablet without keyboard and mouse. What if you have a project which is suited to tablet deployment, or users who simply prefer to use a tablet, does Windows 8 make sense then?

In many ways it does, with easy app deployment and a touch-friendly user interface - provided you stay away from the desktop side. However, it is up against Apple iPad and Google Android which are more mature as tablets, and already well understood by users.
The availability of Office and compatibility, in the case of the x86 version of Windows 8, with existing applications may swing it for Windows 8 in some instances, though since existing applications are generally not designed for touch control that compatibility is less useful than it first appears, if what you want is a tablet.

Windows 8 strikes me as a decent version 1.0 of a new tablet platform, though with some of the weaknesses that you associate with a 1.0 platform. The modern-style Mail app is poor, for example, which is surprising given how critical it is.

Considered as a tablet OS, Windows 8 is already usable and has plenty of potential; but the hard question is whether Microsoft is able to establish a third tablet platform at this stage in the game. The company has the money and the resources, but has struggles to engage its developer community, OEM partners or retailers to fully understand and support Windows 8. In this context, the unexpected departure of the key executive driving the platform at Microsoft, Steven Sinofsky, looks like a disaster.

On the developer side, Microsoft lost substantial trust and goodwill over the repositioning of .NET in Windows 8. After telling developers for a decade that .NET was the future, the company now says it is one among three equals, with C++ and HTML/JavaScript the other two. The way this was communicated, with little flow of information especially during most of 2011, left developers unsure how to plan future client applications and some simply migrated to other platforms.

Microsoft's biggest problem may seem nebulous, but it is real. The Windows brand is strongly associated with desktop, keyboard and mouse; the name also brings to mind unwelcome associations like forced reboots, virus warnings and troublesome setups. The new Windows 8 tablet platform is designed to escape that legacy; but changing public perception is not easy. In addition, it has to support its legacy since much of the world's business runs upon it.

The task of making Windows a successful tablet platform may be as hard for Microsoft as for Polaroid or Kodak trying to make the transition from film to digital; and those stories do not end well.

This translates to a whole lot of uncertainty, which is not what businesses want to hear.
Personally I like Windows 8, but its success and in particular the success of its tablet aspect is by no means assured.

That brings me back to the current state of play. Windows 8 is out, and it is controversial. Where is the Start menu? Why are there so few decent apps?

Since these issues were well flagged in advance, you would think that Microsoft was prepared for them, and ready to suffer some difficult quarters while building up the app ecosystem, working with its partners to help them with the transition, and energetically preparing service pack 1 with performance and usability fixes.

The alternative scenario is that Microsoft allowed itself to believe that the world would be instantly won over and when disappointed will respond with a lurch in another direction which would be fatal.

It is all fascinating to watch, but from a business perspective the right word is caution.

The issue has been covered, of course. Microsoft was clear about this (with the exception of one aspect) from the first detailed unveiling of Windows 8 in September 2011. In the Windows 8 dual personality, the desktop side is an updated Windows 7 and apps are installed in the same way, from setup files obtained from anywhere, but on the touch-friendly side formerly called Metro, apps are, for most of us, only installed from the Windows Store.

The ARM version of Windows 8, called Windows RT, is where the lockdown really bites. No desktop applications can be installed at all, so it is Windows Store or nothing for consumers.

The rules are different for businesses deploying custom apps, and this is where Microsoft has not been quite so clear. Deploying apps without the Store is called sideloading, and is possible subject to certain limitations. Sideloading is described here. In summary:

• On Windows 8 Enterprise edition, you can sideload by setting a registry key and signing apps with a trusted certificate.
• On other versions of Windows 8, including Professional and the ARM edition, you need to "activate the sideloading product key" before sideloading will work. What is this? It is not clear, but normally a product key means the long serial number used to install Windows. This does not sound like something that most users will have access to.
• Developers can install apps using a developer license, but such licenses will expire so this is not a long-term deployment option.

The immediate conclusion then is this:

• The desktop side of Windows 8 on Intel is equally open as older versions of Windows
• The tablet side of Windows 8 is designed to be restricted to apps downloaded from the Store in most cases, other than for businesses with custom apps

Since Windows 8 on Intel is open on the desktop side, smart people will likely soon work out how to overcome app restrictions on the tablet side as well. Windows RT may be more challenging, but I imaging jailbreaks will come along for that as well. The rules still make a difference though, because only a minority is willing to hack their machines.

Is this the beginning of the end for Windows as an open platform? That is an interesting question, and I have seen predictions that the desktop will wither away, such as this from former Microsoft employee Hal Berenson:

The Start menu, and indeed the entire desktop, are legacies that will have to be removed from Windows over time.  While the desktop itself is probably with us for a couple of additional major Windows releases (though there may be truly desktop-free editions sooner than that) ...

Still, even if Berenson is right, desktop Windows on Intel will be with us at until Windows 10 in say four or five years time, which means maybe six or seven years before a potentially desktop-free Windows 11. None of us can see that far ahead though. If Windows RT takes off, we could be thinking of Windows as mainly a tablet OS much sooner than that.

On the other hand, if the tablet personality in Windows 8 fails to achieve its goals, desktop Windows may remain the main version forever until Windows fades away.

Nobody knows, though it is fun to speculate. However, it is interesting to examine why Windows is becoming less open and whether or not this is a bad thing.

Microsoft could not get away with this so easily without Apple, which came up with the "App store only" model for the iPhone and iPad. That makes them closed platforms, and makes Apple a ton of money from store fees, but users are happy because app store prices are generally low, plenty of apps are free, and there is no uncertainty about "where do I get apps for my iPad?", and the App Store is curated so that it should be malware-free, and if malware were to sneak through, it would soon be removed.

In a business context, where administrators want full control over which apps are installed for security and stability reasons, a closed platform is mostly a good thing. A closed platform in combination with the sandboxing of apps on the tablet side is a very good thing.

What about open source? What about consumers? There are fundamental objections to closed systems, which inherently put too much power in the hands of the platform owners. How can we be sure that Microsoft, for example, will not favour its own applications over competitors for things like Office? This is already the case with Windows RT, which comes with Office pre-installed and no way to install an alternative on the desktop side.

On the other hand, the openness of Windows, once a benefit as it encouraged a strong application ecosystem, has become a disadvantage. Users love the performance, stability and clean user interface they get on an iPad, versus Windows machines which more times than not are spoilt by apps running on startup that are not needed, mysterious toolbars that appear unasked in web browsers, along with adware and sometimes malware.

Anybody looking at Windows today would conclude that the system needs to be brought more under the user's control; and that is hard to do without lockdown. Windows Store apps in Windows 8 are easy to install and, more important, easy to remove.

There is every likelihood than that both businesses and consumers will see the lockdown in Windows 8 more as a benefit than a burden; which is actually a good reason for those who may in consequence lose out (like Valve which runs a games store for Windows) to shout loudly about the issue.

The case is not simple though, and on balance most users may well be better off with a less open system than they currently enjoy.

Finally, remember that Intel Windows 8 on the desktop side remains as open as ever, and will be for the foreseeable future.