More Surface Devices (Part 2)

Posted on February 3, 2013 by halberenson

In Part 1 of this topic I mentioned that Microsoft is going after “areas where they can identify scenarios and user requirements that are going unmet by both competitors and the OEM community”.  So what are some of those areas and how might Microsoft address them?

Let’s start small.  I’ve long seen a gap in Microsoft’s strategy around a portable gaming console which I last wrote about back in November when Xbox Surface rumors appeared.  At the same time it is clear that Microsoft needs to have something in the 7″ tablet space, and an XBox Surface meets the criteria I described in Part 1.  Moreover, it is a design center that is more amenable to a subsidization strategy than the 10.6″ Surface.  Take a look at my subsidy argument back in August 2012 when rumors that the Surface would be priced at $199 were rampant.  It turns out the Surface positioning wasn’t as entertainment-focused as I’d then expected, which likely explains the lack of a subsidy model for it.  But the proposition for a 7″ device isn’t going to be around a keyboard and MS Office (though both could be offered), it is going to be around entertainment.  Microsoft’s unique asset in this case is Xbox.  And that is a business where subsidizing the hardware has been part of the strategy since 2001.  There isn’t much more to say here as you can read the previous postings for my thoughts.

Going a little smaller is the possibility of a Surface Phone.  I’ve been a skeptic of this idea unless it is part of a larger reset of their mobile efforts.  I covered the latter in discussing a Plan B.  Microsoft will definitely continue with reference design work, as they’ve been doing since WP7, but I’m not holding my breath for a Surface Phone.  First, Microsoft has ramped up direct marketing of Windows Phone eliminating one of the purposes of having its own device.  Second, Windows Phone sales are indeed accelerating though it is still a little early to make a call on the success of Windows Phone 8.  And third, if they were going to pursue Plan B I would have expected to see them combine the Windows Phone and Skype divisions into a single business unit by now.

If Microsoft does do its own phone this year than what I expect is to see a Skype Surface phone from the Skype division, not the Windows Phone division.  Basically Skype will become an OEM (and carrier).  It won’t strategically be the Plan B I talked about, but it will look a lot like what I described in that blog posting.  The important thing to point out here is that it is Skype and video calling that represents the differentiation that Microsoft would use in a phone.  OEMs won’t go this route because they build devices to meet carrier requirements and be sold by the carriers.  Oh they love Skype as a feature, but “Skype as Carrier” doesn’t fit their business models.

Jumping to the other end of the spectrum is a large screen device that uses the Perceptive Pixel technology and is aimed at the telepresence market.  Microsoft has long seen the need and potential for enhancing how people work together from remote locations.  The Lync product has been at the center of its software efforts, and of course they acquired Skype.  They’ve also made attempts at innovative hardware such as the Roundtable (now the Polycom CX5000).  The Roundtable was an attempt to move traditional videoconferencing towards an IP-based telepresence experience at a cost more than two orders of magnitude less than traditional telepresence systems from Cisco and HP.   Even the cameras in the Surface/Surface Pro are optimized for video calling/conferencing and not (as is the case with rear cameras on other tablets) for taking pictures.

Microsoft Research has done tremendous work in telepresence/telecommuting while individuals and groups within Microsoft have also been working on this problem in order to meet their own needs.  When Distinguished Engineer Kim Cameron moved to Paris he created his own telepresence solution between his home office there and his office in Redmond.  Whenever Kim is in his Paris office he is in his Redmond office.  You can walk by, stick your head in, and chat while you sip your morning coffee.  When you meet with Kim it is pretty close to him actually being there; the main thing missing is you can’t shake his hand.  The Xbox team built (with MSR) a customized solution for individuals to have a face to face conversation between two of their facilities.  The SQL Server team put together a solution optimized for discussions between their teams in different geographies.  I made heavy use of Lync/Roundtable/etc. between my Denver office and my teams in Israel and Redmond, and was considering cloning the SQL Server setup at the time I left Microsoft.  Microsoft IT has more recently deployed a corporate telepresence system in various major facilities, using technology from HP (I believe).  Like other corporate telepresence systems this is a limited resource that must be scheduled, requires travel to special telepresence rooms, etc.

Telepresence, and the collaboration it brings, is the next huge advance in the information worker experience.  Microsoft knows this from both its own internal needs and customers.  I expect we are going to see one or more devices from Microsoft aimed at taking telepresence to the next level, both in terms of collaboration capabilities and price point.  Microsoft wants to bring telepresence to every conference room and every Information Worker’s office.  So something like a 60″ Perceptive Pixel-based device seems like a good bet in 2013.

While we’ll very likely see updated Surface and Surface Pro devices in the next twelve months, will we see other devices in the tablet/tablet crossover/Convertible/Ultrabook/Tablet categories?  A 12-13″ Surface Pro makes a lot of sense to me as an addition to the lineup.  It would have the same general scenario positioning as the current Surface Pro.  The Surface Pro has great specs, and its biggest limit is that 10.6″ screen (and keyboard dimensions to match).  Going up one size class would retain much of the Surface Pro’s portability while offering a screen size more appropriate for heavy Information Worker activities.  And it could allow for a true full-sized/full travel keyboard cover as well.

I’m not convinced we’ll see anything else in these portable device categories in the near term.  While a true Ultrabook or Notebook kind of device that combines Kinect-based gesture control and telepresence features would be sufficient differentiation to justify a Surface device, it might also be too much of a shot at the heart of Microsoft’s OEM’s business.  So I think they’ll pass on this area unless and until OEMs fail to sufficiently participate in Microsoft initiatives to spread this technology.  I have to say I won’t be shocked if I’m wrong and Microsoft brings a 13-15″ Information Worker-focused clamshell Ultrabook-type device to market in the next twelve months.  But I’m just not seeing the need in the short run.

That leaves one real market segment to consider, the desktop or All-in-1 market.  While it seems like Microsoft needs to put something into this segment in order to fill out its product line I’m not so sure.  Again this is a strong area for OEMs, and actually may represent one of the areas where they are doing their best work.  Microsoft will want to see Kinect-based gesture support in future models, but OEMs are probably more than willing to accommodate them.    So where would a Surface All-In-1 be targeted?  Is it the uber IW telepresence/collaboration desktop?  Or is it an attempt to revitalize the market for home desktops through a combination of video call capabilities, unique photography and videography capabilities, participation in the gaming ecosystem, etc.?  My problem thinking through this one is that I’m not seeing the “ah hah” scenario.  But Microsoft might have one in mind.

Finally, I do expect both Xbox “720” and a related streaming media device to appear this year.  So we probably will see 5-6 new devices out of Microsoft in 2013, with a more speculative possibility for the family to grow to 7 or 8.  That’s a lot of growth, perhaps too much growth in an 18-month overall window, for what was previously a software company with some niche hardware.  And I didn’t even touch on the possibility of someone wanting to do more server appliances.  I have no inkling that such a thing is in the works, though I’m sure somewhere teams who had proposals rejected years ago have reconsidered the option in light of the company’s “Devices and Services” refocus.

Let me close by saying that I feel for Microsoft CFO Peter Klein.  Every few weeks someone must come into his office with a proposal for some new hardware device.  I’m sure he puts his head in his hands for a moment, takes a deep breath, then proceeds to listen to them explain just how they are going to make money in a business known for low margins.  As Dell, HP, HTC, other OEMs, and hundreds of companies who are no longer with us have learned this is not easy.  Hardware is a much more complex business than software, with grand opportunities for losses that balloon out of control and far greater probability anything you do will be commoditized in short order.

To Klein, every hardware proposal carries the risk of a significant and prolonged hit to Microsoft’s margins.  So while we can all speculate on how Microsoft could do a cool Z device or Y device I’m sure Klein (and Ballmer) are trying to keep things focused on opportunities with both unique (often disruptive) strategic value and the potential for eventually achieving good margins.  And that will argue against some of the devices that you or I might otherwise hope to see out of Microsoft.

 

 

Posted in Computer and Internet, Microsoft, Mobile | Tagged , , , , , | 8 Comments

Surface Compression

Posted on January 30, 2013 by halberenson

Yesterday I Tweeted the suggestion that one could use NTFS compression to gain more storage space on the Surface or other Windows RT tablet.  People quickly pointed out there are a number of questions around, and potential issues associated with, doing this.  I’m not going to fully explore this subject, but did want to get the conversation going.  Perhaps someone into doing this for a living will fully explore the subject.

The only Surface I have is my go to machine, so I’m not going to muck around with it much.  But I did do a simple test just to see if exploring the use of compression might make sense.  I found a large folder (2.38GB) and compressed it.  The directory was C:\Users\xxxxxx\AppData\Local\Packages.  It compressed down to 1.67GB, a 30% savings that in concrete terms gave me an additional 700MB of space.  A not insubstantial improvement, and that is with just one folder compressed.

Now my Surface has 128GB in it, 64GB internally and I added a 64GB SD card, so I’m not exactly in need of compression.  I uncompressed the folder and started to write this blog post.  I do have my old W500 sitting unused at home, so I could dig it out and do more testing (including actual performance tests) on an older Atom/SSD/Windows 8 system.  But I don’t know if that will be interesting enough to suggest what Surface owners might want to do.  Hence my call for someone more into doing this for a living to give I a try.

Now to address some of the things that were raised over on Twitter.

Is using compression bad on an SSD?  I don’t think so, but I haven’t really studied the issue.  Michael Fortin wrote a blog post back when Windows 7 came out talking about SSD support in which he recommended use of compression for “infrequently modified directories and files” and against using it for actively written files.  Tom’s Hardware  did an extensive review on the topic a little over a year ago and also recommended use of compression on an SSD.   The real question around using NTFS compression on an SSD isn’t is it good or bad, but rather what directories would it be good for and which would it be a bad idea for.

Compression is basically a tradeoff in which you use more CPU time in order to take up less disk space.  If you have files that are infrequently accessed and an abundance of free CPU time than you have a perfect situation for using compression.  If you are trying to maximize CPU efficiency and/or client responsiveness then compression is counter-indicated.  That is why, for example, Microsoft wouldn’t simply ship Surface with compression turned on.  They would rather trade away some storage space than risk degrading the user experience.  In compressing the directory mentioned above (which is where apps store their local data) I probably impacted application startup times a bit, I just don’t know how much.  In previous experience on earlier versions of Windows I was never able to detect performance differences with compression enabled, even when I compressed the entire system disk.  But Windows RT systems are very carefully balanced to provide their user experience, so it is possible that misuse of compression would lead to noticeable degradation.  We need a guinea pig….

A related issue is around one of the more worrisome characteristics of SSDs, that they wear out from write activity.  If compression leads to more writes, or a write pattern that isn’t balanced across the cells of the device, then you could have wear issues.  On the surface (no pun intended) I don’t see it as an issue.  Generally compression reduces the number of writes by compressing the data.  But what it might do is increase the number of cells written on random updates as compression changes the update from as small as a single sector into chunks as large as 64KB.  So on a file getting a high volume of very small updates you could see a theoretically bad behavior.  But I doubt this is a relevant problem for today’s tablets, particularly given how SSD’s caching and internal file system behavior actually work.  If you are planning to use your Surface as a transaction processing machine updating a database hundreds or thousands of times per second then contact me and we can discuss the wisdom of compressing the database file.  But otherwise the wear issue is a non-issue over the expected lifetime of the device.

There is another point that was made in two ways, which is basically what if there is some other form of compression already at work?  Some file types like JPEG or MP3 already compress the data, so adding more compression isn’t really going to help.  True.  Also, some SSD controllers do compression at the controller level (which, if the controller is fast enough, is all gain for no performance pain).  But, as my test showed, not the Surface.

The largest amount of user data on a client device is likely to be already compressed files, so the purpose of using NTFS compression would be to reduce the amount of space taken up by Windows and applications rather than user data.  If you need more space for your photos, videos, and music then use an SD card.  They are cheap!  The one in my Surface has my entire music collection on it, and when I travel that is where movies and TV programs I want to watch are stored.  It is also where my camera gets backed up every  night during a vacation.  I just wish that Microsoft’s related Windows Store apps had better Library support (which is another topic and one of my few real complaints about the Surface).

I’d love it if someone took a deep look at this question and came up with a set of recommendations (and preferably a script) of what to compress on a Surface or other Windows RT device.  I suspect one could easily free up 1-2GB with no noticeable impact on user experience.  But I have other fish to fry so I won’t be the one to do it.

Posted in Computer and Internet, Microsoft, Mobile, Windows | Tagged , , , , | 16 Comments

Outlook RT for Windows RT

Posted on January 29, 2013 by halberenson

A few days ago Mary Jo Foley  reported that Microsoft was testing an Outlook client for Windows RT.  She also reported on some of the internal controversy around this move, and I wanted to add some analysis to this apparent situation.

I believe there are two Outlook clients for Windows RT floating around inside Microsoft.  The first is the existing Outlook 2013 desktop client that, along with other parts of Office 2013 Pro, have been ported to ARM.  That such a build exists should not come as a surprise as this is merely a matter of recompiling the existing applications and addressing any platform-specific dependencies or bugs.  This was probably done for engineering cleanliness purposes many months ago, even if Microsoft had no intention of making the build available to customers.

I believe one of the debates that is going on inside of Microsoft is how, and if, they should release Office 2013 Professional, Enterprise, or Office 365 for Windows RT.  This is a far more complicated question than most observers would give it credit for.  It’s not that it is hard to do, it is that it may not meet customer expectations while creating a long-term support (and migration) headache.  Just on the support front it commits Microsoft to up to a decade of support for what is likely an interim offering (more on that later).  And it commits them to new releases of desktop Office for a while on Windows RT as well.  It also raises questions about more generally opening the desktop on Windows RT, which is something that is counter-strategic.

But I think the bigger question mark is around customer satisfaction with the offering.  Let’s focus on Outlook since that is where most of the user interest lies.  Imagine this scenario:  Surface GM Panos Panay is giving a briefing to a major pharmaceutical company’s CIO at Microsoft’s Executive Briefing Center (EBC) and hears that they would buy Surfaces for all 10,000 field reps if only they could run Outlook on it.  Panos is on the shuttle back to his building and calls the Director of Program Management over in Office who is driving the Outlook on Windows RT decision.  The Director runs over to the EBC to have coffee with the CIO staff member driving the Surface investigation to understand their requirement.  This might be the case he uses to convince Kurt Delbene they should go ahead and release the port!  Unfortunately he returns to his office dejected as the ported Outlook 2013 won’t work for this customer.  They need it to support the “Siebel Desktop for Microsoft Outlook”, and that uses an extensibility model that they can’t expose on Windows RT.

In fact, every major CRM package integrates with Microsoft Outlook using the same mechanisms as Siebel used.  You see, Outlook was never intended to be simply a mail client or PIM, it was intended to be an application platform.   And while CRM is perhaps the most obvious case of where Microsoft succeeded in that regard, many enterprises use other third-party or home-grown apps built on the Outlook platform.  So while bringing Outlook 2013 to Windows RT would certainly bring a first class email app to the platform, it actually wouldn’t meet the needs of the customers who would really pay for it and justify the long-term downside to the approach.

Meanwhile Microsoft is working on Metro/Windows Store/Modern versions of many of the Office apps, and I believe they are trying to ship those within the next year.  This brings up two conflicts.  The first is, if Outlook 2013 is simply a better Mail/PIM than the built-in Windows apps and not the app platform it is on X86, and if “Outlook RT” will be available within a year, how could the short-term gain be worth the long-term pain?  Shipping both makes little sense if the time frame is really that short.

The second is a topic that Mary Jo explored, which is the relationship between Office RT and the built-in Windows Mail, People, and Calendar apps.  I am sure that the Windows (Live) team is actively working to bring the built-in apps up to a more usable standard.  But in this case, how much of a difference will there be between them and Office RT?  In the old days when we had Outlook and Outlook Express there were significant differences.  Outlook supported Exchange, Outlook Express didn’t.  Outlook had full calendaring, Outlook Express didn’t.  Outlook Express supported Hotmail, Outlook didn’t without a kludgy add-in.  Etc.  But over the years, particularly with Hotmail’s adoption of the EAS protocol, the justification for two email clients has diminished.  So how will Microsoft differentiate a major upgrade to the built-in Windows Mail/People/Calendar and Outlook RT?  They might not be able to.  And thus they are probably debating if Outlook RT should just replace the built-in apps.

A year from now the mail client situation on Windows RT will look very different from today.  Hopefully we’ll have only one client with the richness (if not extensibility) of Outlook.  One that is a full Windows Store app.

 

Posted in Computer and Internet, Microsoft, Mobile, Windows | Tagged , , , , | 27 Comments

More Surface Devices (Part 1)

Posted on January 27, 2013 by halberenson

I listened in on Microsoft’s Q2 conference call, and when I started seeing articles about what was said regarding lower priced Windows 8 devices I really started scratching my head.  It seemed to me that reporters and bloggers were trying to read what they wanted to hear into what was said, rather than taking what was said at face value.  Before I get into my own take on this here is the exact exchange as taken from the conference call transcript:

PHILLIP WINSLOW, Credit Suisse:  Hi.  Thanks guys and congrats on a good quarter in a pretty tough macro environment.

I just want to spend a moment on Surface.  You obviously talked about ramping up production and distribution of that.  Just what are some of your goals as you look at Surface RT and Pro for this year?  And within that context how should we think about the profitability of the Surfaces?  Thanks.

PETER KLEIN:  Thanks, Phil.  As we said, we think of Surface as one part of the overall Windows 8 story.  Certainly this quarter it was a contributing factor to the revenue growth in the Windows business.  And what it does is it sort of highlights some interesting innovation that can happen to sort of demonstrate the power of Windows 8 when tightly integrated with hardware and software and some new categories of devices.  And we’ve obviously had some limited distribution this quarter in our stores and as you know we’re excited about expanding that.  So, you know our goal is to continue to build that business, to highlight the incredible power of Windows 8 in an interesting set of devices.  We’re going to expand geographically.  We’re going to expand the product lineup.  We’re going to expand retail distribution and capacity.  So, we look forward to sort of continuing the growth of that business.

WALTER PRITCHARD, Citi:  Hi, thanks.  I’m just wondering, Peter, if you could talk about, obviously one of the big differences here between your devices in the market and some of the competing devices, I think the price point of touch machines, Windows devices are much higher?  And I’m wondering if you could just talk about what you’ve learned here in the first three or four months of the Windows launch, how important price is to the customer base in terms of driving units and what do you think the outlook is in terms of getting price points down the devices in aggregate, in order to potentially drive some demand?

PETER KLEIN:  Thanks, Walter.  We learned a lot this quarter.  We learned a lot about the types of experiences and scenarios and to some extent the price points customers are looking for from their devices.  We saw some really great demand for some of the touch devices that we’ve brought to market.  In some cases we didn’t have the supply that we needed to satisfy that demand.  I think from a price point we learned I think what we’ve always suspected, which is there’s segmentation and differentiation.  One of the powers of the Windows ecosystem, obviously, is the variety of devices and form factors and experiences at a variety of price points.

And I think we learned that that continues to be important.  And as I said, we’re working very closely with both our chip partners, as well as the OEMs, to bring the right mix of devices, which means, to your point, the right set of touch devices at the right price point, depending on the unique needs of the individual.  I think we learned a lot about that and one of the things you’ll see is a greater variety of devices at a bigger variety of price points that kind of meet the differentiated needs of our consumers.

Somehow these two answers to two different questions got sploshed together in many people’s interpretation of the conference call results.  In one case this got interpreted as Microsoft working with OEMs to produce more Surface devices.  I can’t see that being implied in any way in these exchanges.  In others it was reported that Microsoft said they’d be producing a lower cost (Surface) device, which while a perfectly good assumption was never actually said or implied.  Others read into this that Microsoft was going to leave the lower price points to OEMs, something else that you really can’t take away from what was actually said.

Peter said two independent though not unrelated things.  First he acknowledged that there would be further expansion of Microsoft’s own family of devices that highlight Windows 8-related innovation.  And then he talked about working with partners to hit the
“right set of touch devices at the right price point”, where partners are the chip partners and OEMs.  But that was part of the response to the second question, not the first.  Microsoft certainly is working with OEMs to get the OEMs to offer the right set of devices and the right price points.  The are also working with chip partners to make sure that chip sets suitable for all the price points are available.  But nowhere did he say or imply that Microsoft was working with OEMs to create Surface devices nor that Microsoft itself was going to address all the price points.

Having lower cost chip sets that meet the Windows 8 and Windows RT requirements is something needed by Microsoft’s OEMs if they are to address lower price points.  And they are something needed by Microsoft should it choose to address lower price points with a member of the Surface family.

Microsoft’s approach with the Surface is not to try to cover the spectrum of devices that customers might want and that OEMs already address.  They are very specifically looking for areas where they can identify scenarios and user requirements that are going unmet by both competitors and the OEM community.  Scenarios that Microsoft sees as being uniquely addressable by Windows 8 and the rest of their software.

The Surface is optimized for users who primarily want a tablet, but need to do enough content creation that they find the iPad and other tablets frustratingly insufficient.  The OEMs are addressing this segment with devices where you need to decide in advance if you might be doing content creation, and if so turn the tablet into a clamshell notebook that you carry around with you.  Microsoft addressed this by having a tablet with a cover that gives you a keyboard and trackpad “for free”.  The Surface is optimized for unique scenarios in other ways.  The rear camera, for example, isn’t for taking pictures so much as for conferencing.

The Surface Pro is optimized for users who primarily need a notebook, but don’t want to carry both a notebook and a tablet around with them.  OEMs address this, again, by having clamshell notebooks that allow you to convert to tablet usage or by detaching the tablet when you don’t need the keyboard.  But again you are carrying around a clamshell notebook rather than a tablet unless, in the case of detachable devices, you know you won’t need the keyboard and leave it behind.  Microsoft went for a design that would allow you to have a good, but not great, keyboard with you at all times without the size and weight penalty of clamshells.

If you really want a pure tablet and don’t care about having a keyboard always at your beck and call, the ASUS VivoTab RT is a better tablet optimization than the Surface.  If you really want a notebook with touch support, or that can be used as a tablet on occasion, then numerous really cool offerings are available from the OEM community.  Offerings that may be far more optimized for your usage pattern than the Surface Pro.

So what can we expect from future devices in the Surface family?  I’ll save speculating on device specifics for Part 2, but one thing we can be sure of is that they won’t be “me too” devices.  Microsoft will choose areas where it believes it can address user requirements that are going unmet by both competitors and its OEMs.

 

Posted in Computer and Internet, Microsoft, Mobile, Windows | Tagged , , , | 15 Comments

What to expect in today’s Microsoft earnings report

Posted on January 24, 2013 by halberenson

Seriously I have no idea if Microsoft will come in below, at, or above expectations.  It’s pretty clear that Windows 8 hasn’t done something “magical” to earnings, but that doesn’t mean the Windows and Windows Live Division’s revenue won’t be relatively healthy.  So, just a few observations on Microsoft’s various business segments.

Let’s start with STB.  For over a decade STB has usually been the star of the show as it’s revenue grew from being a distant third to about equal to Microsoft’s other two big divisions.  It’s less profitable than either Windows or MBD (nee, Office), but quite healthy.  And it likely did well this last quarter.  SQL Server 2012 is hitting its stride.  The reaction to Windows Server 2012 is almost the exact opposite of Windows 8, with WS2012 getting near universal praise.  The stream of announcements around great new Azure capabilities has been dizzying, and I keep hearing that it’s finally hit its stride on the business front as well.  So I expect STB results will be good.

Entertainment and Devices is now a rather confusing business segment because it doesn’t correspond to an organization.  But we know that XBox spent last quarter in the number one position in the gaming console market, despite Nintendo having a next generation console on the market.  Since R&D expenses are high, with work on the next generation console (and a more narrowly focused home entertainment device) around their peak, and no growth in the market for current consoles, one can’t expect much from this business.  Except for no negative surprises.

Windows Phone rolls up into E&D, but is still financially immaterial.  What will start to be confirmed is that WP is beginning to turn away from being a significant financial drain to something more neutral.  We know handset sales are way up.  And Nokia just confirmed that their royalty payments to Microsoft now exceeds the support payments that Microsoft was kicking back.  Again, probably not yet material to Microsoft’s bottom line but a very good trend.  And it likely means that overall E&D will come in around expectations.

Microsoft Business Division is likely to be weak, but no weaker than expected.  The weakness comes from Office 2013 and the corresponding version of Office 365 not yet being in the market last quarter.

Online Services Division will be interesting to see, particularly given my post last week about its losses and strategy.  If I’m right then we should continue to see expenses drop in year over year comparisons.  So the only question is how fast did revenue grow.  I don’t have a clue.  While differences in growth rates (in either direction) won’t have a material impact on overall earnings per share, OSD’s performance will be closely watches as a bellweather of how Microsoft is doing.

With all other businesses probably reporting in line with expectations, or not so out of line that the impact is material, that leaves Windows and Windows Live Division as the wildcard.  My guess is that with information that has already been disclosed, such as Windows 8 ramp being in line with Windows 7’s ramp, that the division’s performance will also be in line with Microsoft’s previous guidance.  So financially it won’t be much below or above expectations.  But this is the business where analysts and investors will focus their attention, hanging on every word as they try to predict Microsoft’s future.

Everyone wants to know how many Surfaces were sold, but Microsoft will not disclose this. Surface sales are not yet material to the business so Microsoft won’t be under any obligation to report details.  When and how they start reporting on details of the hardware business is something we’ll have to wait a few quarters to really find out.

My own bottom line on today’s announcement is that Microsoft is likely to report numbers in line with their previous guidance, and give us little real news.  Expect much Microsoft-bashing in the press as a result.

Posted in Computer and Internet, Microsoft | Tagged , | 6 Comments

Having a versatile Surface

Posted on January 22, 2013 by halberenson

On Sunday night I was sitting on the couch watching TV with my wife when I decided on a Windows 8 app I wanted to write.  So I picked up my Microsoft Surface and installed Visual Studio 2012 Express for Windows 8, then started to play around with it.  Given the lack of a table, I had it on my lap typing away on the Type Cover keyboard.  The next day I found myself grabbing a sandwich at Subway, and continuing to play with Visual Studio on my Surface.  Ok, so now I know what you are going to ask.  “Do you have a Surface Pro?”  Nope.  “Wow, you got your hands on a VS2012 for Windows RT?”  Sorry, no to that too.

There have been a lot of articles this last week bashing Microsoft, and bashing Windows RT, the Surface, and even the yet to be launched Surface Pro.   Most are way off base, some are so outright stupid that the authors appear to be legally incompetent.  I’m not going to try to go point by point in addressing their comments, nor even link to their articles (which might be likened to disability harassment).  But I am going to talk more about my own Surface experience these last three months and comment on a few topics that have come up about the product and form factor.

By now I’m sure many of you realized exactly how I was able to install and use Visual Studio with my Windows RT-based Surface, I made a Remote Desktop Connection (RDC) to a system in my home office.  That machine is a five-year old (Vista-era) Dell (Red) all-in-one that I upgraded to Windows 8 a couple of months ago.  Working on my home WiFi network I forgot that I was using RDC.  Using my Nokia Lumia 900 as a hotspot the fact I was working remotely was more noticeable, particularly since I was in an area not yet covered by AT&T LTE and was relying on their 4G-labeled HSPA+ network.  But it was certainly usable.

Despite not being able to run traditional Windows “Desktop” apps I’ve found the Surface to be an extremely versatile and pleasant system to use.  So much so that I not only don’t miss my iPad, I actually get frustrated when I try to use my wife’s.  I’ve talked about this in previous blog postings so I don’t want to be completely redundant, but having a device that meets both my consumption-oriented tablet needs and yet is far more useful for any task requiring substantial data input or content creation is just a blessing.

That brings me to one of my more recent observations from Surface usage, what happens when you have a Type Cover rather than the more common Touch Cover.  When Microsoft first disclosed the Surface the Touch Cover was positioned as the companion for the Surface while the Type Cover was positioned as the Surface Pro’s companion.  They are, however, interchangeable and I purchased both covers.  For the first two months, other than a brief experiment, I carried my Surface with the Touch Cover attached.  That included a 3.5 week trip in which the Surface was my only computing device.  Just before Christmas I wanted to see what carrying the Surface with the Type Cover would be like.

The Type Cover is a little thicker and heavier than the Touch Cover.  Basically think of the Touch Cover as a way to always have a keyboard capability with you on a tablet at virtually no compromise in weight or size.  The Type Cover gives you a much more traditional keyboard but demands small compromises in “tabletness”.  They are much smaller than the compromises required to add a keyboard to an iPad, and far more useful given Windows RT is actually is designed to work with keyboards and pointing devices.  I almost immediately forgot about the extra weight and thickness, and it didn’t change my habit of having the Surface with me at all times.

The one compromise that did take some getting used to was feeling keys move when holding the Surface w/Type Cover as a tablet.   At first you are disturbed by the experience because you wonder what weird things you must be doing when hitting those keys.  Fortunately they are disabled when the cover is flipped back so pressing them has no effect.  Later you either stop noticing them or actually find it fun to play with them.  It’s still a little weird, but I’ve adapted.

The biggest change you find with the Type Cover is that using the keyboard becomes so inviting that you spend more time using the device like a notebook than as a tablet.  For example, in restaurants I used to primarily use the hing with the Touch Cover flipped under the device and touch the screen (in other words, use it as a tablet).  Now in that same environment I find I set up the Surface with the Type Cover in front and make a lot of use of the keyboard and trackpad (although I still touch the screen quite a bit).

I also find that with the Type Cover I’m more comfortable using the Surface propped up on my lap than I was with the Touch Cover.  It’s a little sturdier and the tactile feedback makes it feel more natural.  So just changing the cover moves the Surface a little more towards being a true convertible than being a tablet with a physical keyboard capability.  That gives you hints about what the Surface Pro, arguably really designed to be used more as a notebook replacement with tablet capabilities, experience should be like when it appears in a few weeks.

One article floating around questions the portability of the upcoming Surface Pro vs current Ultrabooks by basically comparing it to the well proven clamshell design.  I’m sure buggy manufacturers used similar arguments to address why automobiles were inferior transportation devices to horse-drawn carriages.  One argument this article makes is that the footprint of a clamshell device is actually smaller than the Surface Pro with its hings open and Type Cover in front and thus clamshell’s are better for trains and planes.  Note to author, in coach seats on planes there usually isn’t enough room to use either a clamshell device or a Surface with the hing open and keyboard cover extended.  At least with the Surface Pro you can fold the cover back under the hing, or hold it in your hands, and use the touchscreen.  With a clamshell you can try to use it in coach, at least until the passenger in front of you puts their seat back and breaks your clamshell’s screen.  Yes, that has happened to me.

That article also dismisses the weight advantage the Surface Pro will have over a clamshell Ultrabook by pretty much arguing that users won’t notice the difference in portability between a 2lb Surface Pro and a 4lb clamshell Ultrabook.  Umm, should I really dignify that with a response?  Sadly I must.  In a world in which users are making purchasing decisions based on a 1-2 OUNCE difference in the weights of devices they sure as heck are going to make a “federal case” out of a 1-2 pound difference.  Even with the extra 8 ounces that a Surface Pro carries over my Surface it is likely I’d carry it everywhere.  But a four pound clamshell?  It will live in my briefcase until needed.  And when I travel?  With the Surface Pro I need one device.  With a clamshell I’d also be carrying around a Surface, iPad Mini, or Kindle Fire HD to meet my full computing needs, adding a bunch more weight and complexity.

I’m not arguing that the Surface or Surface Pro is a substitute for clamshell devices in every, or even most, scenarios.  I’m saying that it is the versatility of these devices that makes them so attractive, particularly in the portability dimensions.  It’s like having a handyman who can do all kinds of repairs and minor construction projects.  You want to move an electrical box over a few inches?  He can do that then repair and repaint the wall.  Building an extension to your house?  The handyman might be able to do it, but you’d be better off bringing in a general contractor (and her electrician, plumber, drywaller, painter, etc.) instead.

Back to the Surface, their has been some criticism that a few power-features are missing from the version of Microsoft Office that comes with Windows RT.  When we were doing PredictableIT  my co-founder created our business model in the most sophisticated Excel workbook I’d ever seen.  We could tweak any aspect of the business, from product mix to specific operating costs to pricing to commissions to you name it and see exactly what our bottom line projected out several years would look like.  It was based on a workbook that he’d created for a VC to use with their portfolio companies.  So I asked him if it would work on the Windows RT version of Excel.  His response: “I may have (unnecessarily) used a Macro but it’s use was limited and I am sure I could have worked around not having them.”  Now I personally do not recall seeing any macros in the workbook, and I did a fair amount of modifications to it.  But the point is, Office on Windows RT is not some super weakened subset of the product.  It is a fairly complete port of Office Home and Student.  In my own use of it to date I have not noticed any limitations, and I’ve benefited mightily both from being able to view and manipulate others’ documents and creating some of my own.

The Surface (and Windows RT in general) is somewhat held hostage to the immaturity of the Windows Store and Windows Store apps.  The RDC solution I opened this blog post with is one way (and sometimes a better way than having a local app) this can be addressed.  Another is the completeness of IE10 and that it allows Windows RT to access the full functionality of nearly all of the world’s websites.    Especially when using the Type or Touch Cover’s trackpad, website’s are often far more useful (and full featured) than the apps available on IOS or Android.  So while they don’t provide the full touch experience, full platform integration, or offline capabilities of good apps,  I find IE10 does mitigate the Windows RT app shortage while the Windows Store content grows.  Remember that we are only 3 months in to the life of Windows 8 and Windows RT.  A year from now we likely won’t be talking about a lack (or immaturity) of Windows Store apps.

Another sometimes claim about Windows RT is that you could just as easily use an Intel Atom processor instead of ARM and run full Windows 8.  That would give you the ability to run arbitrary desktop apps, such as Visual Studio, with no compromise to device price, battery life, size, weight, etc.  Besides running more non-touch first applications being a bad thing on devices intended for very heavy use of touch, and that desktop apps are not at all battery life or resource friendly, it misses a general point about configurations.  Yes you could install Visual Studio on an Atom with 2GB and either 32GB or 64GB of storage, but would you want to?  It isn’t going to perform well.  And you are going to significantly eat into that moderate amount of storage.  In fact you likely wouldn’t find it viable on a 32GB system unless that was all you were going to use the system for.  The rest of the configuration doesn’t scream development machine either.

This is why the Surface Pro comes with a Core i5, 4GB of memory, 128GB of storage, and a Displayport so you can use a large monitor when you are in an office.  The SD card slot is easily accessible, rather than hidden under the hinge, to make use of additional storage more mainstream.  It is a configuration much more amenable to use for traditional Windows notebook or desktop scenarios.  I could in fact see making it my primary computing device when coupled with desktop peripherals in the office.

In my experience the Surface does exactly what it was designed to do, without trying to stretch into territory where it would be deemed an inadequate answer.  It is those who want it to be something it wasn’t designed to be that are dissatisfied.  There is a valid concern that Microsoft hasn’t done a great job on positioning the Surface so that it is blatantly obvious when and why you’d want a Surface (or other Windows RT) machine rather than either an iPad or a Windows 8 system.  Hopefully they will soon tweak their messaging to address that.  But I believe no matter how good their message the critics of Windows RT would still be critics of Windows RT.  They just don’t like that they can’t have their cake and eat it to.  I’m happy eating my cake.

Posted in Computer and Internet, Microsoft, Windows | Tagged , | 32 Comments

Better thinking about Surface sales

Posted on January 22, 2013 by halberenson

We don’t have actual sales numbers for Surface yet, but let’s put this one in a little perspective anyway. The most thrown around number is that Microsoft sold 1 Million Surfaces during the holiday season. The Surface was only available for about two out of three months. And until well into December it was only available at the 23 Microsoft Store locations, 32 “pop-up” holiday stores (many of which were just small kiosks in shopping malls), and online at the little known http://www.microsoftstore.com. Meanwhile in the previous quarter Apple sold around 14 million iPads at its approximately 400 Apple Stores, and tens of thousands of other retailers around the world (Best Buy alone has 1400 locations).

Let’s ignore both online and 3rd party retail outlets and assume all sales of both products went through company-owned store (nearly true for Surface). One could do more precise numbers (actual selling days for example), but I’m just going to ballpark this. Assume Surface sold 1 Million devices over 65 days or 15385 Surfaces per day. They had 55 stores and kiosks, so that works out to 280 Surfaces per store per day. Apple sold 14 million over 90 days or 155556 iPads per day, or 389 per Apple Store per day. Not a bad performance by Microsoft actually, but this is the best case for Apple!

Apple gets about 11% of its revenue through the Apple retail stores. So obviously most sales do not come through that channel. Apply any rational number to spread iPad sales over the entire distribution system and the Microsoft Surface numbers start to look outstanding. If iPad sales were split evenly between Apple retail stores and other outlets then Apple was selling 195 iPads per store per day, or far less than the Surface sales rate through the equivalent (and in its case only) channel. Again that’s being very generous to Apple retail stores.

My point here isn’t to claim that somehow Surface sold better than iPad. That’s total nonsense. My point is that when you put the Surface numbers in the kind of context that most products sold through retail are evaluated on the doomsayers are full of bull excrement. Given a distribution channel of only 55 outlets and availability for only part of the quarter, hitting 1 million units sold would actually be an outstanding achievement! I didn’t even take into account that not all 55 outlets were in operation on the first day of Surface availability or that their were some initial product shortages.

It will be interesting to see what Microsoft says when they announce last quarters financial results this week, but if they actually hit 1 million Surface devices sold I’d consider it a Home Run (though not a Grand Slam). Microsoft has a long way to go to be successful with the Surface family, Windows RT, or even Windows 8. But put in proper context the Surface may be off to a great start, not the dismal start that many would have us believe.

Posted in Computer and Internet, Microsoft, Mobile, Windows | Tagged , , , | 11 Comments

About that $2B annual loss from Microsoft’s search business

Posted on January 18, 2013 by halberenson

A few days ago Scott Cleland published a thought-provoking piece on what would happen if Microsoft exited the search business.  It was all about what that would do to Google, but might just as well have been titled “Microsoft should risk mortally wounding itself just to spite Google”.  So I wanted to address the “$2B” annual loss as well as the strategic benefits of Microsoft’s participation in Search and the implications of Microsoft exiting the search business.

Let’s start with the financials behind Microsoft’s Online Services Division (OSD).  This division covers Bing, Advertising, and MSN.   These days MSN is primarily a search-based portal, so I think we can just lump it together with Bing.  So OSD is the Search and Advertising business.  When Microsoft got serious about going after the Search and Advertising space it got really serious and basically started writing blank checks.  It wrote checks to PC makers to make Bing the search default and install the Bing toolbar.  It wrote checks to acquire companies, at somewhat insane prices.  It began a large-scale layoff program in its traditional businesses, one much larger than necessary to deal with the recession, so it could massively increase OSD (and a few other areas’) staffing.  And it did a deal with Yahoo that sent Yahoo cash in excess of what it would get from actual advertising revenue to ease Yahoo’s transition from its own platform to Bing.

Why did Microsoft go so crazy on Search and Advertising?  Well for one thing Search had become core to many usage scenarios, both online and on computer systems.  With Google’s ambitions to compete with Microsoft across the board (operating systems and productivity tools being two major examples) Microsoft needed to get Search from someone else.  And everyone else was failing.  So strategically Microsoft needed to step up in some way.  You could imagine Microsoft having found an alternate way to do this, but in every case Microsoft would have in effect been outsourcing its future.

Microsoft had also taken note that Google’s search-based advertising business model was as outstanding as Microsoft’s own model had been in the 90s.  Microsoft wanted a piece of that pie.  Finally, every dollar of ad revenue that goes to Microsoft, and every bit of extra R&D or marketing spend that Google has to devote to defending its Search and Advertising business against a strong competitor, is a dollar that Google can’t spend going after Microsoft’s other businesses.  So for Microsoft, being a strong number two in Search and Advertising should be a win/win/win.

The reason Google’s business model works so well is that it’s based on a bidding system.  The more eyeballs you have the more advertisers want to get to those eyeballs.  The more advertisers bid for those eyeballs the higher price you get (either for clicks in Search or impressions in Display advertising).  As your user and advertiser base grows your costs grow linearly while your advertising revenue increases exponentially.

In theory all Microsoft had to do was spend a huge amount up front to reach a magic point on market share and from that point on it would benefit from Google-like economics with linear cost growth but exponential revenue growth.  The Yahoo deal was supposed to get it across that threshold, but you don’t see it in the numbers.  Microsoft is seeing revenue growth, but not the explosive growth one would expect at this point.

Will Microsoft ever see the Google-like economics that were so tempting back in the mid-2000s?  If it does then it likely won’t be from classic web search.  The search battle has moved on to other areas, and I’ll get to that later.  As is usual in these kinds of battles, the head-on attack against a dominant competitor yields at best fractional gains (unless the competitor screws things up and creates a big opening).  So Microsoft can continue to gain market share with Bing web search, but at such a slow pace that it hardly matters.  Those who over-focus on this come to the conclusion that Microsoft should exit the search business.  They are simply focused on the wrong thing.

On the cost side OSD is suffering from both ongoing startup costs as well as the cost of pursuing numerous end-runs around Google.  But if you look at the numbers you can see Microsoft has been working hard to get the startup costs behind them.  They are almost done with the payments to cover Yahoo’s switching costs.  And other expensive deals to acquire traffic also seem to be expiring, or entering a phase where guarantees beyond sharing of actual revenue are expiring.  The high initial spend to establish the Bing brand also appear to be behind them, with Sales and Marketing expenses down 23% in Q1.  R&D spend continues to grow, but at a slower rate than ad revenue is increasing.  Several more quarters like this and OSD will  break even.  Yes OSD lost $2B in FY12, but that loss will shrink significantly in FY13.

But is focussing on these losses important?  Maybe not.  Let me call your attention to this sentence from Microsoft’s most recent (at least until next week) 10-Q: Due to the integrated structure of our business, certain revenue earned and costs incurred by one segment may benefit other segments.  This allocation problem is one that all businesses face, but it is a big one at Microsoft.   Revenue allocation is simple compared to cost allocation.  For example, when Microsoft Access was owned by Developer Division but shipped in Office Professional you had to figure out how much of an Office Pro license to allocate to Access.  You could simply say that Access deserved to be allocated all of the price difference since it was the only difference between Standard and Pro, or you could argue that since Enterprises tended to buy one edition for the entire company that you should base the allocation on actual usage of Access.  Either way, Access was allocated revenue out of an Office sale.  The same thing happens today with Core CAL and ECAL.

But what happens if, for example, SQL Server spends a few man-years implementing a feature that Windows Live, or Office, or Bing, or XBox request?  Do those costs get allocated to the requesting business?  Generally not.  They are absorbed by the SQL Server business and appear in the Server and Tools segment of the financial statements.  Even during something as massive as the WinFS effort it isn’t clear how much of the R&D costs accrued to which business.  While it would have been possible to roll up the explicit WinFS team spend to Windows, the SQL Server team itself was doing a lot of work in support of WinFS and those costs would have accrued to the SQL Server business.

Bing is central to the XBox experience.  Bing is central to Windows Phone.  And Bing is central to Windows 8.  Take the set of free Bing-based apps that appear on and make Windows 8 attractive, like News.  Bing wrote the apps and incurred the R&D expense.  They incur the expense of operating the back-end.  They incur the expense of curating the content.  And that expense is part of the OSD segment in the financial reports.  How can I be sure they aren’t allocating it to the Windows and Windows Live segment?  Each app has some small amount of advertising in it.  Scroll through a level of stories and you’ll find a single display ad at the end.  It likely won’t generate enough revenue to make a dent in costs associated with the apps, but it does make it crystal clear where financial responsibility lies.  Even without the ads the costs probably would accrue to OSD, but the ads leave no doubt.

A significant amount of the costs associated with OSD are the result of spending that ultimately brings revenue to other business segments.  Is there method to this apparent madness?  Well yes, in multiple ways.  Just looking at it from OSD’s business perspective these activities are part of its ongoing efforts to find end-runs around Google’s search dominance.  Its Windows 8 apps represent many of the top categories that people typically do with web search.  Someone using those apps is reducing their use of web search, and given Google’s dominance in web search is likely reducing searches done on Google.  If you are using Bing to find movies on your XBox, Bing to find restaurants on your Windows Phone, and Bing apps to read the news, research travel, and follow stocks on Windows 8 how long until you consider yourself a Bing rather than Google user?  Oh, and throw in that you are using Bing as part of your searches on Facebook.

Indeed most users  in the world are probably occasional users of Bing, even though most of their explicit searching is done with Google.  And forget all those monthly measurements of Search market share, they probably don’t capture all this indirect activity.  The Microsoft strategy will move a lot of search traffic off of Google and onto Bing.  The trick for OSD is to figure out how to monetize it.

While OSD hasn’t yet turned all this cost into an exponentially growing revenue stream Microsoft as a whole is benefiting mightily by having Bing as a strategic asset.  A few weeks ago I caught an episode of a TV show I hadn’t seen and decided I wanted to watch the series from the pilot forward.  A Bing search on my Xbox 360 found the first season was available on Netflix but the current season would have to be paid for on Xbox Video or Vudu.  I’m a few shows from being caught up on season two.  Bing made the Xbox experience great, enhanced the value to Netflix of being on the Xbox, and brought the Xbox business some direct video rental revenue.  Did OSD see any revenue out of this?  I doubt it.  Though they probably did gain some personalization information that will allow more effective advertising to me on the Xbox 360 Dashboard, and thus eventually some revenue.

Bing News is my single most heavily used app on Windows 8.  It was the first thing to give me that “wow” experience on the platform.  I think they are carefully keeping advertising unobtrusive right now, but eventually OSD could find ways to get real ad revenue from this and other apps.

So where is Microsoft on Bing?  I think the top positioning of Bing at the CEO, Board, and Senior Leadership level is that it is a strategic asset that is to be used as a core part of the user experience across all Microsoft products and services.  It is also something that can generate significant profits in the long-term, but that is no longer as central to Microsoft’s growth story as it was believed to be back in the mid-2000s.  As such OSD is now more focused on reaching break-even within several quarters than on spending extravagantly to buy market share.  Market share, or rather better opportunities to sell advertising, will come with the increased usage through the overall Microsoft user experience and not from web search.  And eventually shareholders will find themselves happy with Microsoft’s investment in Bing.

Now what if Microsoft were to exit Search as proposed by Scott Cleland and others.  The first thing is that it is silly to think that Microsoft could just announce one day that Bing was shutting down.  They have contracts with people like Yahoo and Facebook that would need to be honored.  They’ve integrated Bing into all those user experiences, and by all reasoning they are working on even greater use of Bing as a central part of future experiences.  Before they could shut Bing down they’d have to find an alternative, and there are none!   Just shutting down Bing might put Google in a regulatory hell equivalent to what Microsoft experienced in the 90s, but it would also seriously and perhaps mortally wound Microsoft itself.  So a Microsoft exit from Search would not be via shutdown, it would have to be via spinoff into an entity that they could rely on.  Forget Cleland’s scenario, it is nonsense.

A scenario that could work would be to spin Bing off into a joint venture that Microsoft co-owned with Facebook, Yahoo, and perhaps others (e.g., Apple).  Microsoft would retain sufficient ownership to protect its strategic interest in the technology, but trade away the future ability for advertising to become a huge profit generator in exchange for reducing short-term costs.  There are only two problems with this scenario.  One is that JVs are notoriously difficult to make work and putting one in the middle of something as strategic as Microsoft’s overall user experience is suicidal.  Second, if Microsoft is already on a path to get OSD to break-even what would really be the benefit of such a JV?  Maybe if it brought Apple on board it could change industry dynamics enough to make a JV interesting.  But otherwise I think a JV or other spinoff of Bing doesn’t make strategic or even financial sense.

Bing remains important to Microsoft’s future, though perhaps more for its value as a strategic technology asset than for any expectations of explosive revenue and profit growth.  Microsoft will get OSD to break even in the next several quarters, which will at least make it less of a whipping boy.  And they continue to think that some day it will produce substantial profits.  But anyone who is focused on if and when that happens is, in my opinion, missing the point.

And with that I end what very well may be my longest blog post to date!

Posted in Computer and Internet, Google, Microsoft, Search | Tagged , | 26 Comments

Crapware, and how we got here

Posted on January 15, 2013 by halberenson

Crapware is in the news in a big way this week thanks to Oracle’s recent Java “security” update and a blog post by Long Zheng that exposes the economic incentives for software developers to bundle crapware installers.  He also points out that respected Silicon Valley VCs are funding Crapware-installer InstallMonetizer.  Ed Bott pulls the story together in his ZDNet piece.

One of the things coming out in all of this is that Microsoft, usually seen as one of the victims in the spread of crapware, is not completely innocent in this matter.  For example, Microsoft’s Online Services Division (OSD) at one point used InstallMonetizer to distribute its Bing and MSN toolbars.  And for a period (before Sun was acquired by Oracle) OSD had a deal to distribute the Bing toolbar with Sun’s Java distribution.  OSD wasn’t alone in this as Developer Division used this technique (and often piggybacks on OSD’s deals) to distribute Silverlight.  So how did we get here?

You probably know the general story behind crapware on PCs, but I’ll do a quick refresh.  Competition in a commodity market left PC makers unable to price their products to make a profit.  On top of that, gaps in the completeness of Windows (e.g., no antivirus or inability to create a DVD) left openings that OEMs had to fill.  Finally Microsoft’s antitrust settlements forced them to allow OEMs to replace the parts of Windows considered “middleware” such as Internet Explorer.  Numerous software vendors than stepped in to offer OEMs payments in exchange for equipping their PCs with trial or lite editions of the software vendor’s product.  Soon every PC came with a dizzying array of these third-party products, most unwanted by the consumer.  Besides that, the PC usage experience degraded as each PC offered a completely different experience.  And basically, there was nothing Microsoft could do about it.

Eventually other players in the industry recognized that they could get in on the game.  In particular Adobe (with Reader and Flash), Sun/Oracle (Java), and HP (printers) figured they could make money off of their free software by doing crapware distribution deals just like PC vendors.

Microsoft’s own foray into the world of crapware came as the result of the intersection between its antitrust woes and the business practices of Google and Yahoo in the search space.  In order to drive search traffic to their engines Google and Yahoo engaged in signing up PC vendors to make their search engines the default on new PCs and to install their toolbars.  These were huge deals.  The original deal that Dell got to make Google the search default on its PCs may have brought them revenues (almost all pure profit) in excess of $1 Billion (yes, with a B)!  I have no idea what the total value of this deal has been over the years, but obviously it has been huge.  Yahoo had a similar deal with HP, and Google and Yahoo pretty much split the rest of the PC vendors.

Now Microsoft comes along and decides it wants (nay needs) to get into the search game, but it has no control over search defaults.  Sure Microsoft ships Windows Internet Explorer set to use MSN Search/Live Search/Bing, but all the OEMs have been paid to change this before actually shipping their systems.  Microsoft can’t block (or make this difficult) because of antitrust agreements.  And even their ability to include toolbars or other software from separate Microsoft products/services in Windows is subject to scrutiny.  The only option they really have is to use the same tactics as Google and Yahoo, so they start competing to get deals with PC and free software vendors to make Bing the default search engine and install its toolbar.

One of the big early deals was for OSD to win Lenovo away from Google.  Later they would win HP (first printers, then PCs) away from Yahoo.  Microsoft also had a deal with Sun (believe it or not) to distribute the toolbar with Java!  With Java on most of the world’s PCs this was an amazing way to reach the broad user base with Bing.

On a side note, I may have been the person who first brought up approaching HP to do a deal to distribute the (then Live Search) toolbar with HP’s printer software.  I was just thinking about who had really broad reach that wasn’t already locked up and that’s what came to mind.  One of my friends was running Live Search marketing and I suggested it to him.  The actual deal happened so much later that perhaps my suggestion had nothing to do with it.  But still, I accept it as my own little contribution to the crapware problem.

Of course Google would go on to so dominate search that I’m not really sure how much of a difference these deals made.  Or maybe a better way to think about it is that by the time Microsoft decided it had to win these deals it was too late.  In particular it seems like the Dell deal was the last one to really make a difference in that phase of competition for search market share.  Soon thereafter every Internet user had their favorite search engine and search defaults and toolbars could make only fractional differences.

Silverlight had a similar problem in going up against Adobe Flash.  Flash had near universal presence on client PCs, meaning that web developers felt they could safely build sites using it.  To solve the chicken and egg problem (clients didn’t need to install it since no web sites used it, web sites wouldn’t use it because clients didn’t have the bits installed) Developer Division pursued a number of tracks.  Signing up high-profile sporting events to “broadcast” with Silverlight, for example, gave clients incentive to install Silverlight.  Working with those who had Bing toolbar distribution deals to also distribute Silverlight was another tactic.  I don’t know how far Silverlight installs got before it became apparent that the age of Flash and Silverlight were over, but Microsoft did make a substantial dent in getting Silverlight onto client PCs.

One of the differences between what I recall seeing for both Bing and Silverlight and what we see Adobe and Oracle doing today is that I belive all of the Microsoft deals were Opt In.  So the user was offered the Bing Toolbar and defaults, but they had to explicitly check a box to get it.  What is annoying about the current deals that Oracle has for crapware in the Java distribution and Adobe has in the Reader distribution is that they are Opt Out.  Unless you explicitly uncheck a box (or two) you are going to get whatever crapware they are offering.  And the fact that you need to go though this every time you try to install an update (e.g., to address a security problem) makes it totally unforgivable.

With systems increasingly moving towards locked down app stores and plug-in free browsers it would seem that the market for crapware has peaked.  Even Microsoft’s foray into retail has been accompanied by PCs configured with “Microsoft Signature“, thus doing away with crapware.  And yet we see VCs continuing to fund crapware installers like InstallMonetizer.  It makes one wonder what they see that the rest of us are missing.  It makes me hope they lose their investors’ money.

What can Microsoft do to further address the crapware problem?  Well for one they could be very careful about any deals they do for search defaults and toolbars, if they are still doing them at all.  Personally I’d vote for limiting the deals to search defaults in new PCs and drop installing toolbars (unless a user explicitly downloads them).  And I’d kill off use of third-party software distributions to try to install Microsoft offerings (ads and links to Microsoft web pages or the Windows/Windows Phone Stores are a separate and more acceptable solution).

The other things they could do is declare what Oracle and Adobe do with requiring the user to Opt Out of the installation of crapware unacceptable and either block the installers or force the user through a special confirmation process.  Besides some technical difficulty determining what is installation of crapware from installation of dependencies (e.g., a library that the primary software is dependent on) legal issues probably stand in the way.  Would regulatory authorities consider the user benefit of blocking crapware, or just see this as Microsoft returning to 1990s-type practices that they opposed?  I would put money on the latter, which could explain Microsoft inaction on this particular front.

While I hope crapware is on the way out, I think it will be a rather slow demise.  It’s a two-step forward one step back kind of movement.  Windows 8 introduces an app store, directly distributes Flash integrated into IE, and includes a PDF reader (so you don’t even need to run the Adobe Reader installation and thus risk crapware).  But it drops DVD writing support so OEMs will include one from the highest bidder.  Repeated security issues lead to recommendations to remove Java from PCs, but for those who don’t remove it every Java security fix also incorporates a crapware distribution.   And no doubt OEMs will continue to ship PCs with Java pre-installed because Oracle pays them to do it.  As Microsoft grows its own retail outlets it brings Microsoft Signature to more buyers, but OEMs don’t seem to be taking the bait and producing cleaner systems for distribution through other channels.

I hope in five years we’ll be talking about crapware in the context of computing history.  But I have my doubts.

Posted in Computer and Internet, Microsoft, Search, Windows | Tagged , , , , , , , | 48 Comments

The impact of “cheap” on PC buying psychology

Posted on January 6, 2013 by halberenson

Paul Thurrot wrote an excellent analysis of what may have held back Windows 8 sales during this past holiday season.  I think when you couple consumer’s expectations that notebook PCs should sell for $300-$500 with the absence of any of the new tablet/convertible devices being available at retail you get a pretty good picture of what happened.  Throw in consumer caution about Windows 8, driven by pundits who knocked it, and you’re close to 100% of the story.

The good news is that Windows 8 will run on pretty low-cost systems.  Paul has another post with Best Buy’s current deals that helps explain the situation.  The $369.99 Toshiba Satellite in the ad looks pretty sweet for a mainstream PC buyer.  Want a touchscreen?  The cheapest option with a similar configuration is the $579.99 Acer Aspire.  That’s a 57% markup primarily to get a touchscreen.  Or to put it in another perspective, you could buy the Acer or you could get the Toshiba and an Amazon Kindle Fire, and still have enough left over to cover your Starbucks bill for a week.  Want something in the heavily promoted Ultrabook camp and you are talking $700 and up.

The Toshiba isn’t a bad system for running Windows 8.  It has a 1366×768 display so you can use Windows 8’s Snap feature to have two Metro apps on the screen at once.  And it has a multi-touch trackpad so you can use gestures for navigating around Windows 8 even though you don’t have a touchscreen.  These are both valuable, but not near enough to get anyone to throw away the two, or three, or four-year old notebook that is working fine for them.

Overall when you look at the classic PC business it is now largely a replacement market.  And without a compelling reason to replace a notebook, and with consumer’s dollars now in search of Smartphones and Tablets, the replacement cycle for PCs keeps stretching out.  It’s not simply a pricing problem, though clearly that plays a role.

What’s at the root of the Vendor/Consumer pricing discrepancy?  It is partially the dynamic that Paul wrote about in which Netbooks dragged the average selling price of PCs into unprofitable territory while setting consumer expectations about future pricing.  Vendors are hoping that new technologies will both let them increase their average selling price (ASP) and accelerate the replacement cycle.  So far they haven’t found the formula that does either.

The second factor is a classic demand/supply problem.  Touchscreens are in very short supply with Apple and a few other players having locked up the entire supply chain.  That is keeping supply of touchscreens for notebooks low and prices high.   Just keep in mind that we’ve seen this before, particularly with display technologies.  The heavy demand and high prices will drive a lot of manufacturing expansion and within a year or two supply will catch up to demand.  At that point prices will drop dramatically, both for touchscreens and for the systems that incorporate them.

But some of this misses the point.  A $449 touchscreen notebook would slightly improve ASP but would have little impact on the PC replacement cycle nor lead to new business.  A $349 touchscreen notebook would harm ASP but only minimally shorten the replacement cycle.  To shorten the replacement cycle in a meaningful way and generate new business you need more dramatic change.  That’s where Windows Tablets and Convertibles come in.

To accelerate the replacement cycle, and expand the overall base of PC systems, you need form factors that meet needs that existing form factors can’t meet.  Windows 8 targeted those with its support for tablets and convertibles, but those devices were largely absent from retail this past holiday season.  These form factors can accelerate the replacement cycle by allowing the consumer to carry one less device around.  It allows a scenario where the customer prematurely replaces an existing notebook with a convertible rather than add a tablet to their computing mix. They also help with ASP expansion because they are replacing two devices, allowing a consumer to justify spending more.     The fact that all Windows 8 tablets introduced to date either come with, or have available, a keyboard dock (or cover) also testifies to how the PC ecosystem sees their ability to be both tablet and notebook as the key to revitalizing the PC business.

What we saw this past holiday shopping season should not have been a surprise as there was nothing going on in the stores to convince consumers they should replace PCs that were working just fine for them.  Touch will become a ubiquitous part of the computing experience, and help keep PCs relevant, but touch alone will not shorten the consumer PC replacement cycle.   For that to happen we need more dramatic form factor expansion.  Those convertibles and tablets are just now appearing at retail, so we won’t get our first really good read of the impact they are having until companies start posting first quarter 2013 results.

Posted in Computer and Internet, Microsoft, Mobile, Windows | Tagged , , | 11 Comments