FAQ: What you need to know about cloud computing's hidden tax hit

KPMG LLP tax expert Reid Okimoto speaks to state tax burden of cloud computing and questions you should ask before buying

Cloud computing services, both software as a service (SaaS) and infrastructure as a service (IaaS), are subject to taxes, whether your cloud provider tells you or not when you purchase them. Reid Okimoto, senior manager in the state and local tax practice at KPMG, shares tips to help you understand the real cost of cloud computing.

Q: What questions should companies buying cloud services be asking about state and local taxes?
Reid Okimoto
Reid Okimoto, senior manager, state and local tax practice, KPMG

A: One major question is, "Is it subject to sales tax?" If sales tax isn't clearly charged by the cloud provider, the customer may still be subject to 'use tax.' IT-focused professionals and their enterprises are consuming and purchasing cloud services and they're normally dealing with sellers and vendors of cloud services, not necessarily those familiar with the "taxability of services." Questions to ask the seller of the cloud services: "Are you charging sales tax or not?" If the answer is no, the next question is, "Why not?" It could be either that the provider does not have nexus or that the service is not taxable. This answer makes a difference to the consumer.

Q: So what happens with taxation and "nexus," which means a connection or link?
A: The concept of "nexus" determines who has the obligation of collecting and remitting a sales tax. Nexus is not synonymous with taxability. If the vendor has nexus in the state and the cloud service is taxable in the state, it should collect the sales tax. If not, the consumer will be responsible for self-accruing and remitting.

Q: State taxes vary, with some at 8% or even more. You note that state policies regarding taxation of cloud services also vary. Would you give us some examples?
A: New York imposes sales tax on many cloud services. California, in the majority of cases, does not tax cloud services. By contrast, a state like Washington shifts the burden of self-accruing and remitting the use tax from the cloud service provider to the consumer. States that offer multiple points of use exemptions for multi-state users of cloud services are preferred to those that impose sales tax of 100% of the purchase price based on the billing address. Imposing sales tax on 100% of the purchase price based on a billing address makes it likely that more than 100% will be taxed or nothing will be taxed. Taxpayers are always concerned about getting "whipsawed" by two competing states' taxing policies.

Q: If you don't go into this fully understanding the tax situation with cloud services, what might come back to haunt you later?
A: A state auditor does periodic audits of companies and they will require you to prove that sales tax was charged and paid appropriately. You may need to prove this by showing them your invoices. Also, your financial statement auditors may inquire about contingent sales or use tax liabilities on the sale or purchase of cloud services. This could result in a contingent liability being placed on the balance sheet. Another reason to keep straight on the cloud-tax issue is that when companies are sold or recapitalized, they typically go through a "due diligence" process with the buyer to "scrub" companies for all liabilities, including tax liability. If there's a lot of liability, there will be a "failed process" question.

Q: What more do business professionals need to think about here?
A: Clear guidance on taxability and sourcing of cloud services creates predictability from a business investment standpoint. States that do not provide clear guidance put cloud service providers in an awkward position, stuck between a possible state tax audit or False Claims Act allegation for under-collecting and a possible class action lawsuit for over-collecting.

---

Best Microsoft MCTS Certification, Microsoft MCITP Training at certkingdom.com

Microsoft promises more Windows Embedded Compact 7 updates

Microsoft promises more Windows Embedded Compact 7 updates

Microsoft has revealed several Windows Embedded Compact 7 updates, one planned for the fourth quarter of this year and one for the second half of 2012. Next year's version will get an updated kernel, faster file system, and broader hardware support, according to an EE Times report.
A 9:30 a.m. keynote was delivered Oct. 26 at the ARM TechCon show in Santa Clara, Calif. by Microsoft's Dan Javnozon, group product manager for the Windows Embedded marketing group. At the time, we were up north in our Palo Alto batcave getting other news stories out, so we're grateful to EE Times for reporting on what transpired.

According to writer Rick Merritt, Javnozon spilled the beans regarding two pending updates to Windows Embedded Compact 7. Building on an "Windows Embedded Compact 7 Update 3" version that was released last month -- see later -- the revisions suggest that the Windows CE-based operating system won't be left forgotten in the wake of an ARM-powered Windows 8.

Microsoft's Dan Javnozon announcing Windows Embedded Compact 7 updates

Source: EE Times
Javnozon, pictured above, is said to have promised a Compact 7 update for the fourth quarter of this year, though apparently no details were provided. In addition, Merritt writes, he promised "Compact v.Next" for the second half of 2012 -- with an updated kernel, faster file system, and "broader hardware support."

Compact v.Next will also get boosted real-time capabilities, EE Times reports. But in a brief post-keynote interview, Javnozon declined to provide further specifics, the story added.

Microsoft's most recent revision to Windows Embedded Compact 7 operating system was announced on Oct. 17. "Windows Embedded Compact 7 Update 3" includes approximately 125 code defect fixes, several new tools for automating testing, and available Silverlight source code for the operating system's media player, according to the company.

Windows Embedded Compact 7 was first announced in June 2010 as a significant upgrade to the previous Windows Embedded CE 6.0 R3. New features included multicore support, an upgraded Internet Explorer web browser, Adobe Flash support, user interface (UI) development via Silverlight, and the ability to share and manage content across DLNA (digital living network alliance) devices.

The operating system runs not only on x86 processors like its big brother Windows 7, but also on other architectures such as ARM -- including the multicore Cortex-A9 -- and MIPS. (However, Microsoft notes, Hitachi's SH4 is no longer supported by this particular Windows CE variant, and ARMv5 is the earliest supported ARM architecture.)

According to an Oct. 17 blog entry by Olivier Bloch, chief software architect for Windows Embedded, Windows Embedded Compact 7 Update 3 is now freely downloadable. He wrote that the new release contains "approximately 125 code defect fixes" for the Compact 7 operating system, Platform Builder tools, and the Compact Test Kit (CTK).


The installer for Microsoft's Windows Embedded Compact 7
The CTK has two new tools, Bloch adds: The Compact Automation Tool Solution (CATS) for automating test scenarios, and The Compact Stress Tool for automating stress tests. Also now included is new Silverlight for Windows Embedded (SWE) sample code for the Compact 7 Media Player, which was previously provided only in binary format. A previous dependency on the compositor in the sample code has been removed, so Media Player performance should be improved across all hardware configurations, according to Microsoft.

Microsoft originally promoted Windows Embedded Compact 7 as "bringing the power of Windows 7 across ... specialized devices such as slates, portable media players, and others." Indeed, the operating system was shown off last year on an early version of the Asus Eee Pad EP101TC (below), a tablet that was later revamped to run Android instead.


The Asus Eee Pad EP101TC originally ran Windows Embedded Compact 7
Since then, both the progress of Android devices and the announcement of a pending, ARM-based version of Windows 8 has caused Redmond to lower its sights -- or so it would appear. Thanks to its low cost, simpler hardware requirements, modularity, and real-time characteristics, however, Windows Embedded Compact 7 will continue to find customers, or so its supporters argue.

---

Best Microsoft MCTS Certification, Microsoft MCITP Training at certkingdom.com

Why you should take hacked sites’ password assurances with a grain of salt

Beware of e-mails that play down the ease of cracking your leaked passcode.

Reputation.com, a service that helps people and companies manage negative search results, has suffered a security breach that has exposed user names, e-mail and physical addresses, and in some cases, password data.

In an e-mail sent to users on Tuesday, officials with the Redwood City, California-based company said the passwords were "highly encrypted ('salted' and 'hashed')," a highly vague description that can mean different things to different people. "Although it was highly unlikely that these passwords could ever be decrypted, we immediately changed the password of every user to prevent any possible unauthorized account access," the e-mail added unconvincingly.

It's unfortunate that companies make such assurances, because they may give users a false sense of security. As Ars has been reporting for nine months, gains in cracking techniques means the average password has never been weaker, allowing attackers to decipher even long passwords with numbers, letters, and symbols in them. Even Ars' own Nate Anderson—a self-described newbie to password cracking—was able to crack more than 45 percent of a 17,000-hash list using software and dictionaries he downloaded online.

Jeremi Gosney, a password cracking expert with Stricture Consulting Group recently explained in an Ars forum post that it's highly unusual for a leaked password list to go uncracked, as suggested by the Reputation.com e-mail.

"It definitely depends on the specific leak we're talking about, but generally speaking, your average security expert/penetration tester/casual password cracker is probably only going to be able to recover at most 50-60% of passwords in any given leak," he wrote. "Seasoned password crackers will likely recover 70-75%; and truly exceptional password crackers will recover 80% or more."

Adding cryptographic salt to passwords is crucial to the safe storage of passwords because it forces password cracking programs to guess the plaintext for each individual hash, rather than guessing passwords for thousands or millions of hashes all at once. (Yes, it also thwarts rainbow-table attacks, but no one uses this method anymore.) But it's easy to overstate the benefits of salting. It in no way slows down the cracking of a single hash, so if an attacker locates the hash belonging to a particular high-value Reputation.com user, the measure does nothing to thwart the cracking of that hash. The security value of salting alone only slows down cracking of large lists by a multiple of the number of unique salts, so that value decreases with each hash that is decoded.

A far more meaningful security measure is the type of algorithm that's used to convert plaintext passwords into cryptographic hashes. If the company used SHA1, SHA3, MD5, or any number of other "fast" hashes, it's extremely likely that at least some of the leaked password data has already been cracked. If, on the other hand, the company used bcrypt, scrypt, PBKDF2 or another "slow" algorithm specifically designed to hash passwords, the chances are significantly lower. Reputation.com makes no mention of the algorithm it used, so users should presume the worst. Anyone who used their Reputation.com password to protect one or more accounts on other sites should change those passcodes immediately. Passwords should be randomly generated by a password-manager, contain a minimum length of 11 characters, and include numbers, letters, and symbols. They should also be unique to each site.

For a deeper dive into the benefits of salting and hashing, see last Saturday's story about the password breach that hit LivingSocial.com. Some of the user comments are especially illuminating.

---

Best Microsoft MCTS Certification, Microsoft MCITP Training at certkingdom.com

Companies explore self-detonating data as security control

Self-detonating data would put expiration dates on sensitive documents

The popular Snapchat photo-messaging app used mainly by Android and iOS mobile device owners to share images that then self-destruct after 10 seconds is the sort of security idea that businesses say can help them secure online transactions with business partners.

“It puts controls on what people see, and I can put expiration dates on sensitive documents,” says Marc McDonald, owner of Chicago-based Midland Metal Products that a few months ago started using the software-as-a-service called VIA from Intralinks Holdings that now lets the maker of store fixtures share computer-aided design files for custom manufacturing with business partners. Midland Metal Products restricts download of sensitive information and also sets a time for the files to self-destruct. McDonald says the password-controlled VIA option is simpler and has more security controls than the Dropbox option he’d previously used.

While Intralinks sometimes casually refers to the collaboration service, which is priced at $25 per user per month as a “Snapchat for the enterprise,” it’s not related to the real Snapchat, which was launched in September 2011 by Stanford students Evan Spiegel and Bobby Murphy as a way to share “impermanent photos” taken on mobile devices through their Snapchat app.
We've been getting a lot of inquires about Snapchat apps."
— Jason Novak, assistant director of digital forensics, Stroz Friedberg

After a short period of time, each Snapchat image is said to be deleted from the devices and the Snapchat servers. The still-evolving Snapchat service, which has started to receive venture-capital funding, is proving popular with teens and young adults that now send millions of Snapchat photos and videos each day. Snapchat is also starting to be noticed in business circles in connection with questions about whether unauthorized photos and images of sensitive business information are being sent via mobile devices.

“We’ve been getting a lot of inquires about Snapchat apps,” says Jason Novak, assistant director of digital forensics at Stroz Friedberg, the New York-based firm which focuses on cybercrime issues and providing digital evidence that will stand up in court, if need be.
One big question is whether Snapchat does leave any trail of evidence of use on a mobile device. Stroz Friedberg says its forensics analysis can detect a trail of use of Snapchat for the Apple iPhone, though not evidence of specifically what photos or videos were sent. It hasn’t yet completed forensics for Snapchat on an Android device.

In its digital forensics tests it did with Snapchat for the iPhone, Stroz Friedberg found Snapchat maintains what’s called the user.plist file which is not encrypted. The file is a way to identify, preserve and analyze that the user of the iPhone did send something to a recipient via Snapchat. Novak says it’s possible to clear the Snapchat plist file on the device if the user knows how.

He points out that other Snapchat-like services oriented toward mobile have sprung up -- such as Facebook’s Poke, as well as Wickr and Silent Circle which take advantage of encryption as well. These type of services are presenting digital forensics with new challenges, Novak notes. Mobile devices such as smartphones and tablets that use these type of services remains a new and evolving field beyond traditional computer-based forensics which is now more automated.

---

Best CCNA Training and CCNA Certification and more Cisco exams log in to examkingdom.com

Google Fiber divides users into 'the fast' and 'the furious'

Google's fiber push is making the 'have nots' mad. That's a good thing.

Every day is a beautiful day in the fiberhood.

The chosen ones in Kanas City, Austin and Provo are getting Internet connections that are 100 times faster than average at very low prices, thanks to Google's Fiber project.

Unfortunately, you and I don't live there. So we're stuck in a bandwidth backwater.

As Internet trolls like to say: U mad, bro?

If you don't live in one of these cities, you should be mad. Google's Fiber project demonstrates that very high Internet speeds are possible and nobody except Google has the vision or courage to make it happen.

One Internet bandwidth provider has admitted the ability to provide much faster speeds to consumers, but has decided not to. Time Warner Cable CFO Irene Esteves said in February that Time Warner is perfectly capable of "delivering 1 gigabit, 10 gigabit-per-second" Internet connectivity to consumers, but that the company just doesn't "see the need of delivering that to consumers." I believe Esteves' statement accurately represents the thinking of most existing Internet providers.

Now are you mad?

The issue isn't really that consumers don't want faster Internet speeds. And it's not that cable providers don't care.

It's really a chicken-and-egg problem.
Why your Internet is so slow

Average U.S. Internet speeds rank 12th or 13th in the world, which is pathetic for the country that invented the Internet and contains Silicon Valley, Hollywood and data-hungry Wall Street and a $15 trillion annual GDP.

Other countries are pulling away. A Sony-backed service recently announced 2Gbps download speeds in Tokyo for $51 per month -- twice the speed of Google Fiber and 200 times faster than the U.S. average and at a lower price than Google Fiber.

Now are you mad?

Gigabit fiber Internet access is affordable, but only if everybody gets it. But everybody isn't going to get it unless it's affordable.

And that's why we can't have nice things.

At least, that's what Esteves really means when she says that users don't want faster speeds. Providing consumers with the faster speeds Time Warner currently provides to some business customers is very expensive because only a few customers pay for it all.

It's not that Time Warner Cable's customers don't want fast Internet. They don't want Time Warner Cable's price for fast Internet.

However, if you lay fiber to every home in a city, and if a majority of homes sign up to use it, the cost can come way down. And that's what Google Fiber is all about. It's about making a bet on the future and investing heavily to bootstrap widespread use and high demand.

Google's Fiber project involves the actual digging of trenches and the actual laying of fiber optic cables all the way to homes. There are innumerable logistical and legal hurdles to overcome for each city.

Google is already providing the service in Kansas City, and is still expanding into new neighborhoods there. The company recently announced that it would roll it out to Austin, Texas, then Provo, Utah.

Google offers consumers three "plans." The first is Internet comparable in speed to ordinary broadband, and it's free. The second is 1Gbps speeds and 1 TB of Google Drive space for $70 per month. The third adds TV plus a 2TB DVR box for a total of $120 per month.

Getting Google Fiber service is just like getting cable Internet service (except 100 times faster). You get a Wi-Fi capable router, and you plug your PC into it via Ethernet for the full-speed experience.

Google is spending $84 million to build the infrastructure necessary to serve 149,000 Kanas City customers. That's $563.75 per customer, for you math majors. (If that sounds like a lot of money, consider that the infrastructure gives you 100 times faster Internet for the rest of your life for the price of an iPad. Still, customers don't have to pay for it up front -- Google is doing that.) And it gets cheaper per customer with each new person that signs up.

Goldman Sachs estimates that it would cost $140 billion to deploy Google Fiber nationwide.

To put that in perspective, that one-time investment would give entrepreneurs in every state of the union a radical advantage globally, ignite an economic boom comparable to the nationwide deployment of electricity 100 years ago and enable incredible new services -- all for less than what the U.S. loses each year in offshore tax havens.

Now are you mad?
Why mad users are the best thing about Google Fiber

It's unlikely that Google will lay fiber to every city in the US, and less likely still that Google will do that Internationally. And it doesn't need to.

Google Fiber is already "inspiring" ISPs to boost speeds and investment. Google may be triggering an arms race for high-speed Internet connectivity, because it's re-setting expectations about how fast the Internet should be.

This increasingly matters as HD movies and TV becomes more mainstream. Right now, Netflix alone consumes one-third of all the download bandwidth in the U.S. at peak times.

Hollywood and other movies-on-demand services had better get busy offering compelling services. More than half the upload bandwidth in the U.S. is consumed by BitTorrent.

I think the minority of providers who figure out how to offer vastly higher speeds at very low cost will survive, and the Time Warners will get out of the ISP business for good.

No, AT&T didn't announce gigabit fiber in Austin

Hours after Google announced that Austin would get the Google Fiber treatment, AT&T (which is headquartered in Dallas) announced that it would build a gigabit fiber network of its own in Austin.

Or, at least that's what the news reports would have you believe. But if you look at the press release, it was really a passive-aggressive bit of whining about Google getting special treatment from Austin authorities.

Instead of announcing a plan to build fiber optic connectivity in Austin, AT&T actually announced that "it is prepared to build an advanced fiber optic infrastructure in Austin," according to the announcement press release.

"Prepared to build" does not mean "plans to build."

Then the whining began: AT&T's plans "anticipate it will be granted the same terms and conditions as Google on issues such as geographic scope of offerings, rights of way, permitting, state licenses and any investment incentives."

The release ended with this zinger: "Our potential capital investment will depend on the extent we can reach satisfactory agreements."

In other words, the whole reason for AT&T's press release was not to announce the intention to build fiber optic gigabit Internet connectivity, but instead to complain about preferential treatment of Google by local authorities.

AT&T has a point. Local, state and government regulations and restrictions are a big part of why our Internet speeds are so slow. And that's yet another reason why Google Fiber is so brilliant.
Google is simply smarter than AT&T

Rather than approaching individual cities and begging them for permission to lay fiber, Google held a big contest and said, in effect: "OK, we're going to pick a city to gain a massive economic boost. You want it? What are you going to do for us?"

Then they started choosing from among the 1,100 applicant cities based on which ones were most serious about making Google Fiber possible.

In fact, Google Fiber triggered a gold rush of entrepreneurial investment and activity.

One enterprising local even rents their Google Fiber-connected home at a premium on AirBnB, and calls it "Hacker House."

Google Fiber is creating a lot of hype and attention. It's making people realize that affordable, ultra high-speed Internet connectivity is possible.

It's making people look at their local governments and ISPs and ask: Why can't I have this?

But mostly, Google Fiber is making people mad. And that's the right emotion in the face of the incredible waste of time and money and opportunity that takes place every day that goes by while we're held back by yesterday's Internet speeds.

But let's not just get mad. Let's get fiber.

---

Best Microsoft MCTS Certification, Microsoft MCITP Training at certkingdom.com

Windows 8 Update: transition from Android to Windows Phone made easier

Windows 8 Update: transition from Android to Windows Phone made easier
Also, iPad keyboard/cover to rival Surface, 8-inch Windows 8 tablet

UPDATE: Microsoft has delayed availability of its Switch to Windows Phone app until sometime next week.

Microsoft figures customers will be more likely to switch from Android smartphones to Windows 8 phones if it makes it easier to find the same or similar apps for their new phone as were on their old phones.

Microsoft is introducing Switch to Windows Phone, an application that finds identical or replacement applications for Windows Phone 8 in the Windows Store to replace their old Android apps.

The new application, which is being released today, is not available for iPhones.
Switch to Windows 8 inventories all the applications on the Android phone and sends that inventory to the Microsoft SkyDrive cloud. When customers log in to the same SkyDrive account from the Switch to Windows 8 app loaded on their Windows Phone 8, the app finds the same set of applications. If there are no exact replacements, the app recommends similar ones, according to Guru Gowrappan, executive vice president for products at application search firm Quixey.

Quixey supplies the apps-search engine within Switch to Windows 8. The engine uses descriptions of apps, reviews of apps, trouble reports about apps and other metadata it gathers from the Web to recommend substitute applications to users, Gowrappan says. The goal is to make them as close as possible to matching the app on the Android phone.

In the case of Switch to Windows 8, the Quixey search engine goes through the 135,000 Windows Phone 8 applications in the Windows Store seeking direct matches – such as the Windows Phone 8 Facebook app to replace the Android Facebook app – or to find applications that perform as close to the same function as possible, he says.

The search engine can also look for applications based on what customers want to do. So a customer could enter “cook Italian food” into the engine and would get a list of apps such as Tuscan Chef and Italian Video Recipes.

Sprint Zone and Sprint Digital Lounge use Quixey’s engine to find apps as does ask.com for searching Android, iOS, Windows Phone and Blackberry applications.

Keyboard for iPads mimics Surface
Logitech is selling a thin, fabric-covered keyboard/cover for iPads that give the Apple tablets similar functionality to Microsoft’s Windows 8 Surface tablet/laptops.

Called FabricSkin Keyboard Folio, the keyboards attach magnetically to iPads and flop down to convert from being a cover to being a keyboard. The device includes a prop to hold the screen at a slant for better viewing when typing. It can also fold over to allow use of the iPad as a tablet.

At $150, that puts iPads with keyboards on a price par with some models of Surface tablets with keyboards.

Some differences: Surface supports Office applications and a file system, something iPads lack. Surface draws power from the computer battery; FabricSkin Keyboard Folio has its own rechargeable battery. The Surface keyboard communicates with the tablet via direct electrical connection; FabricSkin Keyboard Folio uses Bluetooth.

8-inch Windows 8 tablet
Acer is coming out with an 8-inch tablet running Windows 8 if a leaked photo is to be believed.

Windows 8
The photo here was posted by the site minimachines.net but taken down at Acer’s request.

---

Best Microsoft MCTS Certification, Microsoft MCITP Training at certkingdom.com

Microsoft may backtrack on Start button in Windows 8

Analysts urge company to recant design ideology as nod to customer complaints

Microsoft may recant its Windows 8 design theology, bloggers reported Tuesday, by offering Windows 8 users an option to bypass the "Modern" UI and by restoring the Start button and menu to the beleaguered operating system.

A pair of longtime Microsoft hands, Mary Jo Foley of ZDNet and Tom Warren of The Verge, citing unnamed sources and messages on Windows discussion forums, said Microsoft was considering those tweaks for an upcoming update, called "Windows Blue" by some and "Windows 8.1" by others. The upgrade, the first of a planned faster development and release tempo, is allegedly slated for an October debut.

Warren pointed to evidence that Microsoft might allow boot-to-desktop with Windows 8.1. Foley added that the Redmond, Wash., developer was also pondering a return of the Windows Start button and associated menu.

Analysts welcomed the news, assuming it's accurate.

"I don't see this as a defeat but as a good thing," said Patrick Moorhead of Moor Insights & Strategy. "It's shows you're willing to make changes based on customer feedback."

The tweaks would be a concession for Microsoft. Publicly, the company has repeatedly maintained that its design decisions were correct and its executives have suggested that users would, in time, learn to live without a Start button and grow to appreciate the Start screen.

Today, Microsoft declined to comment on the reports.

But contrary to Microsoft's assertions that the dual user interfaces (UIs) in Windows 8 were "fast and fluid," customers have barraged the company's blogs and the Web in general for more than a year with complaints.

They were most upset about the disappearance of the iconic 17-year-old Start button and menu, but also griped that they weren't able to boot right to the "Classic" user interface (UI), or desktop, rather than first hitting the tile-style Start screen. Both issues have been sores spots among longtime Windows users, and at the top of virtually everyone's most-hated lists.

Even Microsoft co-founder Paul Allen took Windows 8 to task, calling it "puzzling" and "confusing" when last year he urged the company that made him a billionaire to offer an option that set the desktop as the default mode on boot.

And they voted with their wallets, either by staying away from Windows 8 -- and shying from any new PC purchases -- or if forced to the new OS, by supporting a cottage industry of third-party add-ons that restored both boot-to-desktop and the Start button. StarDock, for example, claimed earlier this year that its $5 Start8 add-on had been downloaded 3 million times, with thousands of people trying it daily.

Even with that on the line, StarDock CEO Brad Wardell applauded Microsoft's presumed move. "I hope Microsoft adds back the Start button and a boot to desktop option," said Wardell in an email Tuesday. "While we would miss the short-term revenue boost of Start8, it is important to keep the Windows software ecosystem healthy and growing."

The talk today suggests that Microsoft has rethought not only the design of Windows 8, but also its strategy.

"The feedback they've had should tell them that people are not ready to live in the Modern UI, so they need to make [Windows 8's desktop] as good as, if not better, than Windows 7," said J.P. Gownder, an analyst with Forrester Research. "When the tipping point happens, perhaps in a couple of years as the Windows Store fills up, when all the key apps are there, then they can rethink."

And withdraw the Start button yet again, Gownder meant.

Most outside Microsoft believe the company's decision stemmed from a misguided touch-first doctrine, fueled by the belief that only if customers were forced to run apps would they buy apps, and that only by coercing them could Microsoft quickly create a pool of users large enough to attract app developers to the new platform.

Gownder understood that thinking, even appreciated it, but still said it had been wrong.

"I understand Microsoft wanting to drive charms," Gownder said, referring to the set of persistent icons for chores such as searching, sharing content or accessing the OS settings. "There is an argument toward design purity, to reimagine Windows, and that people must become comfortable with the charms. That's legitimate. But the overwhelming feedback was that perhaps the train was taking off a little too early."

Moorhead argued that backpedaling wouldn't significantly hurt Microsoft's push toward an app ecosystem.

"This is very positive, because it doesn't take away from the experience of 'Metro,' " he said, using the older term for the Modern UI. " It just gives users a way to get back to Metro that's obvious. It doesn't say anything about Metro, doesn't say it's good or bad. It doesn't change that argument at all."

Gownder urged Microsoft to backtrack on the boot-to-desktop and Start button controversies, noting in a longer blog post Tuesday that the horse had left the barn -- users were already adopting Start button emulators -- and that the company should accept the inevitable, if only to keep its enterprise customers happy.

"Microsoft needs to step back and do this," Gownder said. "Enterprises are not about to support one of these workarounds. For them, this [functionality] needs to be in the OS layer."

Redmond has done 180-degree turns before. When customers howled about Windows Vista's intrusive User Account Control (UAC), the prompts designed to warn of risk when installing and running software, Microsoft dramatically reduced UAC's impact in Windows 7 three years later.

Now it has an advantage, as it's committed to a faster release cycle -- one executive called it "continuous" -- and assuming the leaks are correct, can modify Windows 8 in a third of the time.

"Microsoft misstepped a number of ways with Vista," Gownder said. "But they did change it. They have an established market and a lot to offer, and [Windows Blue] is, by no means, the last chance for Windows 8."

What a reversal will not do is magically turn around depressed PC sales, on which Microsoft is reliant for Windows 8 sales. Offering options to boot to the desktop or restore Start functionality won't change the dynamics of the industry, where consumers in particular are buying less expensive touch-enabled tablets rather than replacing older Windows computers.

But what if? What if Microsoft's design ideology had been more flexible before it shipped Windows 8? Would it have made a difference? Would Windows 8 devices be flying off shelves?

Not likely.

"Had Microsoft added the option of restoring the Start button and boot-to-desktop, they would be in a slightly better position than they are today, but not much," said Moorhead. "In fact, Metro app development would be behind the curve had they added the options."

The UI mistakes, Moorhead added, were secondary to a more fundamental misreading of the market and the available technologies. "In retrospect, Microsoft should have marketed and built a more pervasive and high quality touch pad experience. "They misjudged the number of touch-based devices that would be out, and under-emphasized the quality experience of a good touch pad."

Apple, for instance, has ignored touch-based computers thus far, instead depending on larger touch pads built into their notebooks and on the gesture support they've integrated with OS X.

Had Microsoft taken that approach for Windows 8, it could have avoided the entire touch screen issue -- shortages caused by low yields, and corresponding high prices -- Moorhead asserted.

"Unlike touch display functionality, which can add $100 to the [bill of materials], a quality touch pad may cost as little as an incremental $5," Moorhead said.

---

Best Microsoft MCTS Certification, Microsoft MCITP Training at certkingdom.com