Windows 8 is a pivotal moment for the company. Rumors point to an early-to-mid 2012 release of the new OS. The problem with that date is that it misses the 2011 holiday season, ceding another four quarters to Apple and Google as they continue to grow tablet mind and market share.

It feels now like we have to move faster than is possible, which is the point at which I unhinge myself from reality and enter a world of tech fantasy.

Information about Windows 8 has been under lock and key since the company announced its development back in 2009. The OS traditionally follows a three-year release cadence, making July 2012 roughly the right time to expect Windows 8′s release. As I said, this leaves the red meat to the competition for a good year–a dangerous prospect when one considers Apple’s recent record-shattering year on the back of iOS, and Android’s blistering ascent to the top of the mobile heap.

Now Steven Sinofsky, President of Windows and Windows Live, is no software development novice, nor is he unaware of market pressure. As the leader of Office 2007 and Windows 7, he has a track record of success. If any leader could pull the fantasy I’m about to outline from his hat, it’s Sinofsky.

Typically, when Microsoft releases a major product, it engages in extensive external testing. So far, that has yet to happen on a grand scale with Windows 8, which is interesting considering there’s less than a year left in its development cycle. By this point, builds and features have typically leaked because versions of the new OS are externally distributed. It’s not atypical to see screenshots plastered all over the Internet. The absence of this process has me wondering: could Windows 8 be on the hook for a holiday 2011 release after all?

Why would testing indicate anything about a release date, you ask? It hints at the work required to ship something as complex as an operating system. The primary reason for external testing is to subject the new software to as broad a range of inputs as possible prior to release, hoping to uncover bugs that customers would have likely run into otherwise. It’s crowd-sourced bug hunting, and it results in more polished products.

If the external test process is less intensive, perhaps it means the test barriers are lower this time. The major push in the OS thus far has been the new Metro UI, tailored for tablets and touch form factors. The new UI, however, is brand new and has no need for legacy testing. No existing apps are expected to work with it, namely because it’s built on a new platform: HTML5, JavaScript and CSS. From demos I’ve seen, the old Windows 7-style UI exists still, though little changed. If that’s the case, it’s possible, albeit unlikely, that there’s less need for legacy testing with this OS–nothing’s drastically changed with what exists, and what’s new is so new that nothing exists externally that needs to be tested.

From my experience, a new feature on average results in 10-20x the amount of work to implement than it took to think up, and anywhere from 10-100x the amount of work to test it, depending on the breadth and complexity of the feature. Accepting that, the team could shave off a huge amount of time in the development cycle by reducing the need for existing app developers and device makers to test out their products. The message is simple: what works will work, no need to test it.

Now back to earth. The reason this is out of touch with reality is that it’s unlikely the underlying Windows 7 OS has changed so little as to require zero app compatibility testing with existing software. There are ways around it–for instance, giving partners the OS with no visible changes and having them test the guts earlier on. If the rate of change in the old OS is less than the new one, active development on it could have ended and been ready for testing much earlier in the cycle. It’s improbable, though, albeit hopeful.

All of this of course is speculation spurred on by the belief that Microsoft has to act soon to have a chance at relevancy in the tablet and mobile markets. Windows Phone is quality software with a hard sales pitch in front of it. Windows is a dominant OS with high expectations for how it will resize to fill all the device needs as they rapidly evolve. Hopefully Sinofsky and his team can pull a rabbit from their hat and give Google and Apple a run for their money this holiday instead of next.

It hasn’t gotten better, that’s the lie here. We get older and wiser, but it doesn’t get better. We become jaded, stuck in our state of learned helplessness, and succumb to the heteronormative culture that surrounds us.

You, however, Mr. Vice President, and your boss, do have the power to ensure it gets better and you’ve done little since your election to significantly move our country forward on gay rights.

For AIDSVu, we built a locator for STD testing centers around the US. Using your IP coordinates as a starting point, the site looks up the nearest testing centers to you. If the IP coordinates aren’t right (which they likely won’t be, as they map to the ISP location), you can enter an address which is geo-coded using the Google Maps API. This was an impressive feature for our clients as it provided immediate value to their customers and users. It promoted content in a way that’s more personal, a continual goal of information technology.

With the medical society software, the locator is built to let patients search for physicians by term and location (e.g. Cardiologists in San Diego). The former isn’t that exciting–keyword-based search is as old as the Internet (not really, but it’s old). Combining keyword with location-based search, though, is new. For someone with a medical condition that needs the nearest help they can get, being able to find the closest doctor who specializes in treating their symptoms could be both a time and life saver.

Implementing both of these features was a blast. The GIS (Geospatial Information Systems) world is fun, complex, and deep. After a year dabbling in parsing geographic datasets, map plotting, data mash-ups and geo-coding, I’ve barely scratched the surface. But there’s a mountain of potential here and it’s progressively catching on.

With the emergence of mobile phones as replacements for computers when accessing information, the potential of GIS to change how we interact with the world around us is huge. Wherever you are, the phone knows your geospatial location. It can transmit that data to an endless supply of web services which return location-sensitive data, providing a new layer of interactivity to life.

Imagine this: you’re a bicyclist in a busy city. Part of every day involves subconsciously calculating the risk of an accident as you speed down the road alongside cars, trucks and public transportation. In your back pocket is your cellphone. As you roll along, your phone is using your location to query the city’s crime lookup service, searching for bike accident reports in your vicinity. As you near a particularly hazardous intersection, the phone begins vibrating in pulsations, warning you of the risk ahead. In this application, it could save your life. In others, it can simply, but powerfully, add a new dimension to how we process and apply the myriad information we consume daily.

Our primary challenge each day is to make the best decisions we can with limited information. But there’s far more information available to help us make decisions than we concurrently have access to. The issue is that it’s not easily accessible in a way that’s helpful to us right here, right now, wherever that may be. With the world of mobile computing on the rise, we’ll see more and more applications that mash-up data in ways that offer the chance to improve the quality of our decisions. This has monumental implications.

It’s going to be an exciting decade and beyond.

One rather large factor we overlooked in adopting Symfony, however, was its documentation. At first glance, the example-driven approach they take seems sufficient. Rather than invest effort in documenting all framework components at their lowest level of complexity, they chose to aggregate components and document them through a sample application, from start to finish. Even after only a month or so of use, this approach has fallen far short of what a good framework needs in terms of documentation, and more pointedly what Symfony deserves.

Most frameworks that I’ve worked with choose to document components at the simplest level, building examples around say specific function calls like $sf_user->setFlash(), outlining the complete list of ways such methods can be called, and providing tips for their use. This works well because it makes few assumptions about how any one developer is going to use this piece of the framework. Other popular SDKs and frameworks, from Zend to .NET, take this approach. It should, after all, be the end-developer’s job as a good programmer to stitch components together. High-level examples that do so for us have only limited use beyond a beginner level and, when the only form of documentation, and an anemic one at that, lead to much time spent searching through API docs (generated from short code comments) and around Google for answers, often to questions about simple usage.

To the latter point about searching Google, a lack of framework-provided documentation may not be a problem if developer community contributions fill that gap, and that contributed content is easy to find. That’s just not the case with Symfony. Most answers to searches on Google, in my experience, either refer to deprecated documentation on the Symfony site, or blog and forum posts, which require additional time to abstract away proper usage from the specific implementation of the original author (if their questions are even useful and timely to begin with). In the best possible scenario, this is a matter of SEO, which should make solving this even easier. But documentation discovery now is abysmal.

With Symfony 2.0 on the horizon, slated for release some time in 2010, I really hope they change their documentation story–I’ve even e-mailed the release manager for 1.3/1.4 requesting as much. It’ll be a rather large undertaking as there’s quite a bit of functionality, to the framework’s credit. But if this issue isn’t resolved, I could see it preventing adoption by anyone with realistic documentation expectations looking for a mature PHP framework.

Taken in the context of Goldman Sachs’ history, I don’t believe it’s possible to argue they were simply bookies. They knew what they were doing–they were in it for short-term profits, to get out before the melon hit the sidewalk, as RollingStone put it. That’s sleazy and, if not currently illegal, should be made so immediately. At the very least, GS and all those responsible should be stripped of their profits, because right now they’ve made off with billions of tax payer dollars, meanwhile the gulf between the rich and poor continues to grow. In the larger context, Mr. Zakaria, GS and other culpable firms are swindlers and robbers, not innocent victims of financial complexity.

I’m so looking forward to making this blog live up to its name. I strayed from my technology roots in favor of politics and now it’s time to blend the two together more. The tone will change, probably for the nerdier. Should be fun…stay tuned.