Neville Hobson's views on OpenID struck a chord:
It seems to me that OpenID is still a very early-adoption technology, the domain of serious geeks and tech enthusiasts.
Well, I’m as enthusiastic as the next geek but I just don’t really get OpenID yet.
Since reading Neville's piece, I've been trying to contextualize the problem from an enterprise perspective but it's not been easy. Jan Miksovsky's analysis, while technical in nature might be summarized in one sentence: It's too damned hard for users and there's no real context as to why, as a user I should use OpenID. Even Dave Winer's discussion didn't really help me except to the extent that he says:
There are enormous economic incentives for companies that run social networks to not let users of other networks access their services. Shareholder value is a function of how many users they have, how they are "monetized" and how hard it is to switch. The harder it is to switch, the more money each user is worth. Any exec that did anything to decrease the number of users they control would probably be fired. So anything that depends on this isn't very likely to happen, in existing networks.
Too true. In enterprise land, try stringing SAP to Oracle to Microsoft to...and see what happens. On that basis alone, it is hard to see enterprise software vendors making it a priority, something Brad Fitzpatrick alludes to in his paper:
Uncooperative sites, on the other hand, are the ones that are already huge and either see value in their ownership of the [social] graph or are just large enough to be apathetic on this topic. Please note that "uncooperative" doesn't mean "actively fighting it", but rather that they might just not prioritize supporting this. In any case, it must (and will) work with both types of sites over time.
I then listened to Dave Winer's podcast where he contextualizes in terms of Twitter and Facebook. I can understand why as these are the applications du jour and I take his point that someone from left field may come up with the answer but somehow I don't quite see it. This led me to consider Plaxo and Susan Scrupski's critique:
The only major issue I had with assigning categories to my existing contacts is there are so darn many of them after these years. I emailed Joseph Smarr (the Plaxo Architect from the video on my blog) and asked him if there was any way to group categorize contacts. He said, “We’re working on it…”
To my mind Plaxo's is the wrong approach for business.
While categories are necessary, fluidity and flexibility are crucial. I need my own taxonomy but it has to be portable and readily understood by others in very much the same way that Workday is looking to use Worktags to help different constituent groups down to the individual level. But even then I cannot see how Plaxo alone can become the supplier of choice for business, despite its brave attempt to mix business style mail/calendar access to popular, consumer services like flickr.
Business is driven by WIIFE (What's In It for Enterprise) thinking. When you drill down just a few inches it becomes WIIFM (What's In It For Me) so where might OpenID or some derivative work? The immediate place is in the extended supply/demand chain. Back in 2005, AMR reckoned that $3 trillion is tied up in global inventory. For years, vendors have talked about collaboration among suppliers but it has rarely worked except for the benefit of the channel masters. Wal-Mart is the stand out example but even there most people would say my referencing Wal-Mart in this context is a stretch. Ford Motor Company might be another where design, manufacture and assembly is highly distributed. The traditional answer to holding these networks in place has been integration, often through clunky EDI links or messaging kludges.
I'm thinking that OpenID (or something similar) might be the starting point for the simple reason that once I can access data through signing on, I no longer have to worry about integration in the traditional sense. I could even take data through a secure RSS feed. That's what I do with Basecamp so why not with R/3, Oracle Apps and the pantheon of other enterprise systems out there? Provided I can tag the data I'm accessing according to the taxonomy with which I am familiar, then I'm good to go for making ad hoc decisions in my own apps space.
Aaah - but what about process I hear the SAPpers cry? The reason I'm raising this is not to disrupt process thinking but because real business value is not about the stuff that can be automated but about managing exceptions. That's almost always done by people outside the process flow.
Who might provide an OpenID server? Following Neville's piece, I created an OpenID with Verisign. Now I have a ridiculously long PIP address but...I can sign in with my chosen username and password in exactly the same way as I do in any other online service. Adding the Seatbelt Firefox plug-in means I can easily use the Verisign 'version' of my OpenID on any OpenID enabled site. It detects whether I am already logged in to the Verisign service and if not, offers me the option to do so. I tested Seatbelt on several sites that offer OpenID and it just works. Now we're getting somewhere because Verisign is a company I trust as do many businesses.
Where do we go from here? I don't know though search would be top of my list. The whole thing should be ridiculously simple yet I can imagine a myriad of objections. I doubt the application vendors will lead the charge though it would be a lost opportunity to do nothing. They could if they see that ease of access via OpenID might hold the key to unlocking value for customers and for themselves but without the crippling cost of large scale integration. Here's how.
If I am at the center of a supply chain but want to unlock value, the only question I have to ask is how much am I willing to pay to allow third party supplier access via OpenID. $1 a month? $10? If I have 10,000 suppliers and $100 million in inventory, then the interest carrying cost alone is (roughly) $6 million per annum. Paying my top rate of $10 per month for 10,000 suppliers only costs $1.2million per annum. On that crude metric alone, ideal payback is achieved in less than three months; realistically 12 months. I can imagine supply chain participants would be willing to contribute to cost because their value achieved is equally compelling. I can't imagine any CXO not getting that. It fulfills the WIIFE criteria in spades.
I wonder what Eddie Herrmann thinks. He represents a SAP customer, Colgate-Palmolive, and has written enthusiastically about OpenID in the past.