Brian Prentice

socialism (sō-sha-li-zom) noun: - any of various economic and political theories advocating collective ownership and administration of the means of production and distribution of goods

Socialist systems are usually marked by a centralized group responsible for deciding the most socially advantageous way of allocating resources. Normally, we think of socialism in a political context. Here it is the government which acts as the centralized embodiment of socially-beneficial decision making. But socialist thinking and organizational patterns permeate many parts of society.

Software production is one of them.

Software, boiled down to its essence, is a codified solution to a problem. But the Byzantine nature of modern software development results in an inaccessibility to the means of its production. That puts software developers in a uniquely powerful position. They become the conduit through which problems are solved. As society increasingly relies on software there is an increasing dependency on the software development community.

People involved in allocating resources within socialist systems usually argue that their efforts are both benevolent and that the criteria used are transparent and open to influence. But while the decision process may be flexible, the underlying belief that society is best served when those decisions flow through an elite body is unwavering.

This is largely the pattern I see amongst much of the software development community. Particularly within enterprise IT organizations. As a generalized group they welcome suggestions on where they should focus their efforts. Equally, they actively seek input “from the business” on how to prioritize their work and how to quantify its value.

But when their hegemony over the creation and selection of software is challenged then the true colors start to emerge. I often speak to clients about the growing trend of user autonomy and federated decision making in software and the general response I get is “how do I stop this from happening.”

Usually this is framed as well intentioned compliance, sourcing or data management concerns. But seldom are these questions framed as “how do I manage compliance, vendors and data in a way that both embraces and accelerates the trend toward autonomy in our organization.” The focus, instead, is squarely on re-exerting centralized decision making under the auspices that it’s the software elite that can best manage these decisions for the organization as a whole.

Keep in mind that the most salient critiques of socialism are not bound to the efficacy of centralized decision making but to the corrosive nature of the dependency the system requires of those outside the governing elite. Alternative approaches to socialist frameworks often focus on the concept of “spontaneous order” - the idea that a harmonious, evolving order arises from the interaction of a decentralized, heterogeneous group of self-seeking agents with limited knowledge. I would strongly argue that the growth in IT autonomy is an expression of spontaneous order. Unfortunately, spontaneous order is an anathema to the central planner.

So for all you in the software development elite - particular in enterprise IT organizations - next time you’re confronted with issues related to business acquisition of SaaS, user adoption of social software, unofficial development projects using aPaaS or mashups, pause and check your reaction. Are you excited about the possibility of supporting spontaneous order? Or do you feel that the central management of the organization’s software is being threatened?

Then ask yourself a question. “Am I a software socialist?”

Form follows function.

We’ve heard it often and many of us are probably aware of it’s historic connection with the Bauhaus movement of the early through mid-20th century.

One of the key perspectives of the Bauhaus movement was a rejection of the superficial ornamentation so commonly found in the arts, crafts and architecture of the 19th century. In its place emerged streamlined, minimalist design forms meant to capture the essence of an objects purpose.

But what happens when functionality becomes superficial ornamentation? By extension, form must also ultimately fail.That is exactly the dilemma facing the software industry.

Conventional wisdom states that the value of software is correlated to requirements and users. That equation is linear - the more features the more potential users, the more potential users the more value. This “total is greater than sum of its parts” perspective is pervasive in the software engineering community and even amongst many designers. However, this view is self-referential - it sees the value of software through software.

But what we’re increasingly understanding is that for users, software is merely a conduit. Their focus has always been squarely on its purpose. However, as a software solution evolves to meet as many users “requirements” as possible, it actually ends up obfuscating value to its constituents rather than increasing it. This is my point. For the end user, every additional capability beyond what serves their direct purpose is superfluous ornamentation.

In the true spirit of the Bauhaus, users are clearly seeking ways to rid themselves of this ornamentation. That message is largely lost on those responsible for creating these solutions for them.

Form following function worked for Walter Gropius but it won’t be good enough for the software industry. We need a movement that re-engages us with the concept of purpose. Purpose which has clear human dimension. Purpose whose distinctions are deemed as important as its definition. Purpose which serves to actively engage its user spurring creative applications unimagined by its designer.

When function follows clear purpose, form can follow function.

Recently, I reached out to both Intel’s Chief Patent Counsel, David Simon, and Google’s Head of Patents and Patent Strategy, Michelle Lee with two specific questions.

1. Does your organization currently license any of its patents to other organizations?
2. Can you categorically state, for the record, that all patents your organization current holds, or is planning on seeking, are being, or will be directly applied to a product or service that you make?

Both Mr. Simon and Ms. Lee have been outspoken critics - both in blogs and in testimony to the U.S. Congress - of patent aggregators (or, to use their terminology, “non-practicing entities”).

The reason I’m asking these questions is to discern whether they are engaging in exactly the same behaviour as the patent aggregation companies they’re railing against. If the answer to question one is yes, and question two is no then I was hoping to explore the logic of why they feel the law should change to make it as difficult as possible to express these very activities as a discrete business model.

Intel was so kind as to respond in the affirmative to the first question. However, they demurred on the second. Google, on the other hand, ignored the questions outright.

I would be my sincerest hope that both Mr. Simon and Ms. Lee could find the time to respond to these questions. Ideally they could do that on their own blogs and link back to me.

Since both Mr. Simon and Ms. Lee have been on the front foot in advocating legislative changes that will have significant ramifications to the IT industry I would hope they’d be open to engaging in a broader dialogue on the thinking behind their positions.

I have surfed twice in my life.

Actually, calling what I did “surfing” is generous. It could more accurately be described as a rapid transition from lying prone on a piece of polyurethane foam to swimming in the Pacific Ocean.

Even with this minimal, and quite unimpressive experience, there’s an immediate lesson you learn about surfing. It requires the combined ability to read the conditions with a mastery of timing. First you have to be positioned correctly for the break. Then you have to set out at the right time. Fall behind and you miss the wave. Get too far out front and you’ll be dumped.

If you see an analogy approaching you’re right.

I am personally of the view that with Wave, Google is paddling furiously for a break that isn’t there. Sure, there’s a swell developing - but it’s still a couple miles offshore.

Why am I so hesitant? It’s not for lack of respect for the vision and engineering achievement of the Google Wave team. Instead, it’s based on personal experience.

Back in the mid-90’s I was a regional product manager for Novell GroupWise. 15 years ago GroupWise was a pretty cutting edge application outside of the mainframe world. But while we were pushing the concept of a universal inbox it became rapidly apparent that people were not using the integrated time management capabilities.

What electronic calendar power users quickly realized in 1995 was that the feature was useless unless everyone else used it too. Without near universal participation, busy searches yielded false availability results. Even with a clear solution to a problem that bothered everyone we still ended up calling around on the phone so that everyone invited to a meeting could check their paper diary.

The electronic calendaring phenomena is something I looked at in some detail in a research report I wrote titled “Corporate Social Systems: The History of Electronic Calendars Offers Key Adoptions Insights” (a Gartner subscription is required). One of the findings was that it took 5-7 years after the widespread availability of electronic calendars for it to become broadly adopted. It was only through the confluence of a number of external factors that electronic calendars transitioned from being an unused feature in an email program to being a mandated corporate social courtesy.

I can see with Google Wave a number of such electronic calendar scenarios. At a minimum, people will need to grok onto the concept of threaded interactions in a single, multi-dimensional environment rather than specific collaboration patterns each being represented as it’s own discrete application. In the real world, outside of conferences attended by software engineers, people don’t just change the way they interact with one another because I think there’s a cooler way of doing something. Like the electronic calendaring situation a decade and a half ago, just because I can see how I can apply Google Wave to my collaborative requirements doesn’t mean I can get other people to buy into the same approach.

The inertia of the status quo is a very real impediment to Wave’s success. It reminds me of a meeting I had a couple of years ago with the CIO of a law firm who spent an hour elaborating on their highly successful document management implementation. When two of the firm’s partners showed up they made it clear that they avoided the document management system at all costs choosing instead to send files around as email attachments. While the IT guy could grasp the concept and value of centralized document storage with check-in, and check-out capabilities the actual users saw the sharing of documents of being a simple matter of distribution. That job, as they saw it, worked best in email.

Nor is it just the new collaboration paradigm that Google Wave represents. Working within the environment will require a whole new set of social norms, the lack of which can create frustration that will drive users right back to the tools they’re already comfortable with. For example, concurrent editing is very sexy but it seems to me this can become an unwieldy free-for-all without an agreement by everyone to abide by some type of sequential contribution rule. Playback? I love it. But will some content additions in wavelet’s require commenting in order for there to be a meaningful context for playback? Is there the potential for Wave proliferation to the point where it creates digital clutter? All of these, and many more, will need to be resolved, agreed upon and broadly propagated.

There’s one other comparison with electronic calendars that I think is critical here. There is a lot of talk about Google’s ability to permeate the enterprise through the user. But it doesn’t always work in that direction.

How many of us actually carry around a paper-based diary any more? More to the point, how many of us in knowledge work type of roles can unilaterally avoid maintaining our calendar electronically? Not many I’m guessing. But while the electronic calendar has largely become a mandated corporate courtesy it has not become a mandated personal courtesy. With all the buzz around social software like Facebook, YouTube, and Twitter we can often forget that some types of collaboration tools are mostly of value in a work setting.

If that’s the case then Google has some additional hurdles they’ll need to jump based on their current ability to execute in the enterprise market. This is a point my colleague Tom Austin makes in his observation of Wave.

Let me finish on an observation that Google’s own Lars Rasmussen made. He pointed out that email, an application built on the paradigm of traditional mail, actually emerged over 40 years ago. What that means is that it took roughly 30 years before email was broadly embraced. While I am greatly impressed with what the Google team has put together I think we all need to entertain the very real possibility that it could be another couple of decades before the impact of what they’re proposing will be realized.

Today, TechCrunch columunist MG Siegler was conjecturing on the potential of a patent lawsuit between Apple and Microsoft. This, he reasoned, could result from Microsoft using their multi-touch technology in a potential Windows Mobile 7 device which was reported today by Mary-Jo Foley.

Both Apple and Microsoft have done quite a lot to advance the state of haptic interfaces. And, as any self-respecting modern vendor would do, both have sought patents on their innovations. In the case of Apple it’s the patents they’ve filed in relationship to the gesture recognition system currently in the iPhone and iPod touch and with Microsoft it’s the surface computing technology currently associated with Microsoft Surface and soon to be seen with Windows 7.

But I’m not particularly interested in whether a courtroom battle between Apple and Microsoft is likely. Instead, I’m wondering how this would be impacted by an extension of the Bilski particular machine test.

Back in October of last year, for those of you not in the know, the Court of Appeals for the Federal Circuit ruled In re Bilski that a business method could only be patented if it was implemented on a “particular machine” or it transformed a particular article into a different state or thing.

Now there are two open questions with the Bilski decision. First, will these same tests end up being applied to software patents? Based on recent rulings by the Board of Patent Appeals and Interferences it appears things are trending in that direction. Second, does a general purpose computer constitute a particular machine? While the court dodged that question in the Bilski ruling it appears that the US Patent & Trademark Office would prefer it the answer to that questions was no.

So here’s the question I’m struggling with - is the iPhone and/or the iPod Touch a particular machine? I certainly wouldn’t classify it as a general purpose computer so I’m assuming the answer to that question is yes. Therefore, if the Bilski particular machine test applies to software it would seem that Apple’s patents would be safe.

But would the same apply to Microsoft’s multi-touch patents? While Microsoft Surface can probably be argued to be a particular machine (it is a specific-built solution requiring custom motherboards) that would clearly not be the case in relationship to the implementation of this technology in Windows 7 or even with a Windows Mobile device.

If this turns out to be the case I think the ramifications are pretty serious.

The issue here is not whether Apple is innovating and Microsoft is not. On face value both are innovating. But Apple’s business model of tightly binding their software to their own manufactured products seems to help them overcome the emerging particular machine hurdle. Microsoft, with their business model of licensing software to OEMs, appears to be more exposed to having their patents invalidated.

Like Microsoft or not, there is no denying that their business model has been instrumental in lowering software costs and driving the prevalence of personal computing. If that business model ends up making it more difficult for Microsoft to protect their growing patent portfolio it would be very interesting to see how Microsoft responds.

I am not a big fan of the Bilski ruling. But if it’s impact ends up being more acutely felt on the business models necessary to commercialize innovation than on innovation itself it, that would seem to me yet another reason to find fault with it.

I was browsing around Amazon.com recently and I ran across an interesting book titled “Democratizing Innovation” by Eric Von Hippel. As the book description highlights Von Hippel asserts:

“Users, aided by improvements in computer and communications technology, increasingly can develop their own new products and services. …product and service development is concentrated among “lead users,” who are ahead on marketplace trends and whose innovations are often commercially attractive.”

I agree that this is not only a valid observation but is something that should be encouraged. As the book description also highlights Von Hippel view of “user-centered innovation” means that:

“…manufacturers should redesign their innovation processes and that they should systematically seek out innovations developed by users.”

Again, no disagreement from me.

But I wonder then; does user-centered innovation also requires some degree of user-centered IP management?

That means that the lead user innovators Von Hippel describes should be expected to combine their innovation work with a cursory prior art examination. To start, that sets an expectation that the property rights of other innovators need to be respected. Just as important, it forces lead user innovators to understand the difference between actual innovation and idiosyncratic rejigs of existing solutions.

[note - any one in an IT department that has spent months needlessly customizing a business application when a standardized process would have sufficed knows why this is important].

But here’s the problem. In addition to checking through business journals and textbooks, a prior art search should consider existing patents. And while tools like Google Patent Search and Patent Storm have made patent identification available to the masses, the gobbledygook that passes for English in a patent claim makes patent examination all but impossible to those outside the high priesthood of patent attorneys.

Let me give you an example. Here is the text for a single point used in describing Microsoft’s patent for grouping and manipulating windows. The idea seems simple enough - but check this out:

A computer-implemented method for operating upon windows as an aggregate, the computer system having a display device and a window system for displaying displayable windows on the display device, each of a plurality of the displayable windows associated with a program for implementing what is displayed in the window and having associated viewing states including an open state and a closed state, and having a current state, wherein the current state represents a viewing state, wherein each viewing state provides information for setting the appearance of the associated window and for setting an execution state of the associated program, the method comprising the steps of:

grouping a plurality of windows from amongst the plurality of displayable windows into a project group, the project group having associated information regarding which windows belong to the project group; upon receiving a request to close the project group, for each window belonging to the project group, saving the current state of the window and setting the current state of the window to the closed state, thereby setting the execution state of the associated program to a no longer executing state; and upon receiving a request to open the project group, for each window belonging to the project group, setting the current state of the window to the saved state.

Get that? Me neither. And this particular patent has 46 more mind-numbingly obtuse points to make. Nor is this patent unique in its complex language.

Let’s remember that patents are supposed to be “non-obvious” to someone skilled in the arts. As someone who’s been in the IT industry for 23 years I think I qualify as someone skilled in the arts. So why am I confronted with either a migraine headache or a whole pile of expensive six minute billing increments in order to comprehend an idea being patented?

True, patents do have an abstract. Unfortunately these go the other direction - they’re too superficial to be useful. That yawning gap between the superficial patent abstract and the dense, legal claim description leads to two outcomes. Either the legal profession has to be interjected into the innovation process way too early or patent examination is ignored. Option one is too costly and option two is too risky.

Now, I understand that patents need to framed in a way which is suitable for a legal environment. So I’m not suggesting that this complex wording should be eliminated.

What I am saying is that patents need to be framed in a way which is suitable for an innovation environment. And, as Von Hippel points out, that is increasingly happening in a wide-open, highly democratized fashion. What that means is that a simple “plain English” claim descriptions should be part of a patent filing.

Patent reform is critical to the future of our modern, globalized economy. It’s so important that we can’t waste precious time entertaining myopic shilling from self-proclaimed industry spokespeople bent on reallocating the costs of a broken system rather than fixing the system itself.

On that basis, the U.S. Congress - which is currently considering the Patent Reform Act of 2009 - should kindly but firmly request that the IT industry sit down and shut up.

I have been following the software and business method patent debate for over five years and I have come to conclusion that the vast majority of the IT industry can not see past their own narrow commercial self-interest when it comes to patents.

I include in this the segment of the “open source community” (whoever they actually are) which is vehemently anti-patent. What they position as a highly principled “open access benefits society” argument for the elimination of software and business method patents can rightly be seen as a convenience for those advocating an industry realignment towards service-based business models. One can’t help but think that they just don’t want 3rd party ownership issues to interfere with their ability to make money. Noticeably absent from this perspective, though, is a broader debate on whether these arguments are valid without scrapping the patent system as a whole.

Then we have the traditional vendor community which wants to retain the right to patent software and business methods and license these for commercial benefit but want to deny those same rights to organizations and individuals that aren’t like them. Namely those nasty “non-practicing entities” that keep suing them for infringement.

That perspective was again on display in the testimony of Intel’s Chief Patent Counsel, David Simon, in his testimony to House Committee on the Judiciary. Channeling Michelle Lee, Google’s Head of Patents and Patent Strategy, Mr. Simon appears to hold the view that a “true innovator” and a “product manufacturer” are one in the same. This is an attempt to frame the argument in the context of good and bad patent holders - Intel, of course, being a good guy. I personally consider this a little to convenient to be taken seriously.

Patent reform does not need to devolve into the type of industry lobbying that is polluting the legislative process. I point specifically to the testimony of Phillip Johnson, Chief Intellectual Property Counsel for Johnson & Johnson. Clearly Johnson & Johnson has an interest in leaving the patent system largely unchanged and that is reflected in his testimony.

But at the same time Mr. Johnson makes his argument from a broader perspective of the overall role of the patent system and backs his position up with facts and statistics. He even advocates changes that I would think would not necessarily be in the direct interests of his employer - things like post-grant reviews of patents. Importantly he puts strong emphasis, first and foremost, on addressing the financial and efficacy issues facing the USPTO. While I’m not sure I agree with everything he advocates I respect the spirit of compromise infused in his testimony.

Unfortunately the overriding spirit infusing the IT industry’s contribution to the patent reform debate appears to be narrow self-interest. Not to say that there aren’t balanced voices out there - but too often they’re drowned out.

That’s a shame. As an industry we have so much more to offer. We are better positioned than most to comment on the evolving nature of innovation and how that impacts centuries-old assumptions of the patent system. We can and should be more active in helping PTOs grant higher and higher quality patents. As an industry we’ve been at the forefront of driving process efficiency - why aren’t we collectively doing a little pro bono work to help a public agency that has such an impact on the future of our industry?

It’s time for our industry to pick up its game in these types of national policy debates. If not we should rightly expect to be ignored.

Since patent reform is back on the political agenda and since I think the current proposed legislation is little more than paid-for lawmaking, I thought I’d be proactive in suggesting areas to reform that I believe would benefit everyone. I want to start with the patent review process.

I am by no means an expert on the operation of the patent & trademark office. But it seems to me the that the concept of the patent reviewer harkens back to a by-gone era of Renaissance men. These wise men of years gone by used their luxury of time to become conversant in a broad range of technological domains. That feat was manageable because a century or two ago the breadth and depth of knowledge in each topic was limited and largely assessable.

But those days are long gone. Now we live in a world of domain expertise and every year the definition of “domain” narrows and fragments as the body of associate knowledge grows. So, is it really practical to hinge a review system on the presumption that a small cadre of patent examiners can attain, and retain, an amazing spectrum of technical and legal expertise? An expertise which must be sharp enough to assure high quality review given an average 18 hours spent per patent application?

This is particularly illogical given the Peer-to-Patent Review project (developed by the New York Law School and run by both the US PTO and the UK Intellectual Property Office) and private efforts such as Article One Partners are demonstrating that broad community support strengthens the review process. It extends domain expertise and helps in surfacing prior art.

It seems to me that we’ve reached the point where federated, community-supported review processes need to be established as the norm. Patent reviewers should transition from being conduits to coordinators.The alchemy in an effective review process would no longer be hinged to the breadth and depth of an individual’s technical expertise but rather their skills in attracting and retaining a group of skilled and committed collaborators. That means that patent reviewers should be valued as much on who they know as what they know.

From my observation one of the biggest challenges facing the patent system is patent quality. And in my area of interest, information technology, poor patent quality increases litigation, results in distorting business practices and acts as a governor on innovation.

This should not be used as an excuse to change damage assessments, invention accreditation and reexamination rules. Instead it should be a wake up call to address the systemic problems we have in the review process.

I’m all for reforming the patent system. But that doesn’t mean I’m all for the Patent Reform Act of 2009.

While I think some of the proposals have merit the proposed legislation as a whole looks like it’s been custom designed to serve the needs of a specific set of businesses. And when some of those businesses, like Google, Cisco, and Intel are also big political donors we enter the area of political patronage that guys like Larry Lessig have been demanding gets changed.

The result is likely to be a standoff as those interests adversely affected oppose this vision of “patent reform.” It’s lobbyists at twenty paces! And when the dust settles we’ll still have a deeply flawed patent system. The only question really being debated is who incurs the most pain.

This isn’t even like swapping desk chairs on the Titanic. It’s like demanding an upgrade to a suite when the ship is starting to sink.

The patent system is failing to keep up with a the rapid changes in technology and business models and the resulting new industries emerging in a globalized economy. These problems affect everyone. I’d appreciate it if President Obama - in the interest of “changing the old way of doing business in Washington” - would tell the bickering industry groups to sit down in the same room and determine a set of issues that everyone agrees need to be addressed.

President Obama should be promising a swift veto of any patent reform legislation that doesn’t represent a cross-industry consensus.

Now that’s change I can believe in!

Call it a sickness if you like but I actually enjoy reading amicus curiae! These are documents filed in a legal proceeding by an interested party who is not directly part of the case. Not only are they usually far more readable than court rulings they also offer a fascinating window into the thinking of the people and organizations that help shape the opinions of those in the judiciary and the legislature.

Last year’s landmark Bilski case saw a number of amicus curiae submitted to the CAFC most of which were largely in favour of doing something to curtail, if not eliminate, business methods as patentable material. Well, those parties got what they wanted. But I don’t think most were expecting the solution that the court ultimately came up with.

Rather than rule one way or the other on the patentability of business methods the court established a new “particular machine or transformation” test . That decision, in my opinion, has only made the whole issue an ambiguous mess. Especially since they ducked the question of whether a general purpose computer is a “particular machine” and the extent to which the Patent & Trademark Office will interpret the new test seems to be a work in progress.

Now Bilski is appealing this decision to the Supreme Court. In support of that appeal, a new set of amicus curiae have been submitted. Not only do these provide some powerful arguments against the CAFC’s decision but they provide a more balanced view of business method patents that were largely absent in the Bilski amicus curiae submissions.

As I read through these I found a number of consistent themes. The first of these is that Bilski introduces rigidity into a patent system always intended to be flexible. Consider these comments from Accenture and Pitney Bowes:

Throughout its patent law decisions, this Court has favored flexible, common-sense approaches over rigid, unbending rules….Bilski is yet another example of the Federal Circuit departing from this Court’s established, flexible approach in favor of a rigid, bright-line rule.

Furthermore, that rigidity is bound to a period where innovation largely had a physical manifestation. This issue, in my opinion, is of paramount importance to those of us in the IT industry. This is best expressed in the comments by the American Intellectual Property Law Association:

As technology thus ventures from the recognized into the unknown, innovation should be no less protectable than in previous eras of
transition….Regrettably, this test is derived from and tied to the vocabulary of technologies developed in earlier ages, and thus is backward-looking and ill-fitted to future discoveries and technologies as yet unimagined.

This “machine-era” approach also has significant consequences across a range of different industries. As Philips states:

The scope and fallout of the Bilski decision are not limited to methods of doing business….Processes claims are often seen as the only means to achieve effective patent protection for important inventions in medical, diagnostic, environmental control and information science technologies. The holding below thus inherently discriminates against industry sectors.

Or, more specifically, here are comments from biotech company Medistream:

Bilski casts a cloud of uncertainty as to whether Medistem and other biotech companies can continue to protect with patents their inventions relating to methods of diagnosing causes of diseases and methods of selecting beneficial treatment protocols … The question of whether potentially life-saving diagnostic methods or treatments should be ineligible for patent protection if they are not tied to a particular machine or apparatus, or do not transform a particular article into a different state or thing, has great importance to our citizens.

For those who thought that Bilski closed the door on business method and software patents I’d suggest that there still some way to go on the issue.

[disclaimer- much to my Mother's disappointment I am not an attorney. These are my personal views and should not be considered legal opinion]