Addition through subtraction, turning software installation on its head: Q&A with Joshua Shapiro

Addition through subtraction, turning software installation on its head: Q&A with Joshua Shapiro
Michael McManus, DIGITIMES, Taipei [Wednesday 5 September 2012]

One of the key issues historically on the Wintel (Windows-based PCs) side of the PC industry has been the lack of cooperation between software providers and PC OEMs (vendors). According to Joshua Shapiro, a managing partner at Princeton Analytic and former member of IBM’s management team, one challenge has been to get software pre-loaded onto PCs. To address the issue, Shapiro has developed a software loading technology (SLT) called Imbue that turns the problem on its head. Instead of installing software applications one at a time (additively) onto a device, a huge master image with as many as 1000 (or more) applications is created and then applications not chosen by a customer are deleted in one fell swoop. According to Shapiro, this type of subtractive installation is more efficient, faster and safer than previous methods of pre-installation.

However, in developing the technology, Shapiro realized that hardware vendors tended to have a myopic view when it came to software, which surprised him considering the state of the industry and the success of players like Apple who have looked to use software as a way to differentiate their products. Shapiro recently spoke to Digitimes about his findings.

Q: What do you see as being some of the key issues facing the PC industry?

A: If you look at the current PC market situation, OEM sales are flat and there has been a lack of optimism over the past several years. Some vendors have exited the market, while others like Dell and Hewlett Packard (HP) are downplaying the personal computing portion of their business. This is shocking because many of these companies are the innovators that built up the industry from its infancy.

On the other hand, the trend is not surprising if you look at historical sales growth, which was once exponential and then became linear, but is now flat or on the verge of decline. In addition, margins are approaching 0% for hardware sales and it seems that vendors make their money on ancillary areas such as services.

The problem in the market is commoditization. There is generally a lack of innovation on the part of the vendors, who have left it to Intel, Microsoft and the ODMs to advance the industry. Once upon a time this didn’t matter because computers were slower and users always looked to the latest models for more horsepower, more storage capacity and more pixels in the display. However, the traditional focus on increased speed or better hardware specifications is no longer vital for users, and without differentiation, vendors increasingly compete on price, leaving them to face diminishing returns over the past 10 years.

The exception has been Apple, which has seen its computing resurgence due to its ability to differentiate. But what if Apple lowers its prices and the price umbrella that other PC vendors have been working under goes away. Or what if Microsoft decides that OEMs are irrelevant and decides to design and market PC devices itself; and we have already seen that shoe drop. This could be a huge change in the industry, where your biggest supplier is also your competitor.

But apart from Apple, PC vendors are in denial.

Q: What do you mean by that?

A: Just look at what is going on with PC vendors. There is a lack of profit, a lack of innovation and their customers are bored. These vendors think they are in the hardware industry, but they need to realize it is not hardware they need to sell, but solutions. It sounds cliche, but if you look at Apple and what it is doing, you can see that companies can be profitable and differentiated, by delivering systems that are relevant to consumers.

Apple has been rewarded with high margins and high customer satisfaction. Other OEMs see this and think that if they develop hardware that looks like the Mac Air, customers are going to be happy; but they are kidding themselves.

The key to differentiation for these OEMs is to look for innovation rather than standardization, and they need to understand that it is through software that customers can take advantage of the hardware they provide. To make an analogy, you can’t use a PC without good software any more than you can use a good car without gasoline. But PC vendors seem to be solely focusing on hardware designs. Everything else is the customer’s problem.

And the industry also needs to recognize that when it comes to stagnation, it’s not only in hardware. There is also stagnation in the development community. Sure there is custom development for the enterprise, but nobody sees that in the mass market and it certainly doesn’t inspire anyone to buy a PC.

Q: That doesn’t really jive with Microsoft’s boasts about the number of developers working on its platform.

A: When I talk about cultivating a relationship between software developers and PC hardware, I am not speaking about developing for something like Metro. I am speaking about developing software specifically to take advantage of the hardware features of PCs. System OEMs have basically abandoned their relationship with developers and left it all to Microsoft.

There needs to be a whole new category of software that takes advantage of what the PC has to offer. For example, OEMs have recently been focusing on the trend of cloud computing, but that focuses on the network, rather than on the computing device. Whole new categories of software need to be developed that take advantage of the incredible lack of latency a PC has, rather than facilitating things through a cloud. Hardware makers need to enable software that use every bit of storage and manipulate every pixel on the screen. The only category that comes close at this point is gaming programs that do simulation.

There is potential with visualization programs, simulation programs, etc, that provide immediate and rapid interaction between a PC user and the device. Stock traders are another example. They need to manipulate enormous amounts of information – visually not numerically – and a PC can do that locally. The user simply downloads huge chunks from a database and then manipulates it all locally.

If new and exciting software is developed that can take advantage of the strong suits of PC, meaning they are immediate and can access all the processing power, it will drive users to seek better systems that can best run these programs. That is what used to drive the market.

Q: Your contention seems to be that hardware OEMs are not really interested in developing this kind of relationship with software makers. Why do you think this is so? Is it a problem of getting the software to the users of that OEM’s hardware?

A: For sure a new software acquisition model needs to be developed and that is what I am working on. The goal for hardware manufacturers should be to deliver a working computer with all the software a user wants and none of the software the user doesn’t want.

Historically though, this has been an issue for PC vendors. If you look at the manufacturer logistic model, the makers want the shortest manufacturing time and a predictable cycle time. Installing software has always made this model difficult. On the one hand, software could be installed using an image file (ISO) onto a new machine. Now, let’s say a vendor was offering a user six different application choices. The vendor would then need to prepare two raised to the sixth power (or 64 in total) ISO images to manage all the permutations of applications the user may choose.

The alternative to that would be for a worker to manually install each specific application then cook it (post processing) to optimize that specific machine. This can take hours and it means that vendors have no interest in offering users a variety of software choices.

Instead, vendors have historically chosen a business model of offering software based on how much they get paid by the software vendor rather than offering applications that the user chooses. For example, if a vendor gets paid a few dollars for every PC that ships with the Google Toolbar included, the customer can expect to get just that. A vendor may then ship millions of units and all of that money goes directly to the bottom line. It’s pure profit.

So vendors have been programmed to look to cut a deal – make one image and then don’t worry about anything. The vendors are hooked on immediate profitability, never mind serving the needs of the customer.

Q: So what is the value proposition you are offering?

A: What I have been developing is called Imbue. It is a software loading technology (SLT). The technology consists of two parts. It preconfigures a disc with a selection of programs or soft assets such as music or video. A master image is then built and when a customer makes an order, instead of loading what the customer wants additively, you remove what is not chosen from the original master image (that includes everything). You start with a huge image and you strip off what you don’t want. The reason I call this Imbue is it infuses the hardware with software. Selling software raises the average selling price and gross margin for every machine sold.

But the key issue I believe is not so much the software loading technology, but it is really the mindset of the OEMs, who simply think they build hardware and can’t be bothered with anything else.

Q: Why do you think that is?

A: Part of it is a huge dependency on Microsoft and part of it is vision. A platform like Imbue is not only about making it easier to install six programs on a PC. Vendors should think of every sector on the hard drive as a seat on an airplane that can be sold to a customer. An empty sector is equivalent to an unsold seat, a wasted revenue opportunity.Given the size of hard drives today, 1000 programs can easily be pre-installed and customers can be given unlimited selection.

Vendors need to think of a PC not as a piece of hardware but as a truck that delivers software to the customer. And then that truck transforms into the system to use the software.

It would change the entire dynamic of the industry because it goes back to the root problem of selling software that customers want – software that is relevant to them. This would in turn allow the PC vendors to be more profitable and to differentiate from others.

Q: But the choice to have the software would only be there when you buy the PC.

A: There is part of the Imbue technology that would allow for certain applications to remain locked until purchased, but that is not really the point I am making here. The issue in the PC industry is that hardware vendors lack a strong relationship with software vendors and developers. By going down this road, PC vendors would have a much stronger relationship with the players on the software side and so it is easy to imagine users being able to access more software options through their device – much like the app store for the iPhone.

Q: What has been your feedback from the industry so far?

A: Some think the idea is genius while others say they are too busy to look at it. However, up to this point I have been focused on getting the patents ready. By the end of the year I will have nine patents on the technology and I will also have code working for proof of concept. For now, I am just getting more people aware of the opportunities the technology can provide.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s