Kate’s off to an interesting start. She has gone out of her way to differentiate her work from Tacit. She argues that this established player in expertise location is about finding solutions not people. Not sure if this is real or marketing but we will see. I have a lot of respect for Tacit and David Gilmour their founder. He built the software on the principle of privacy, i.e. you should have a right to keep the fact that you know something private. There must be a reason for this strong statement.

Now we are getting to the problem statement. A cute statement to start Its not what you know, its what you do that really matters. Looks like she is going on to expand on this so I will report (with comments).

She is arguing that its not just about just finding an expert, but when you find someone you want to communicate with them, but what if they don’t want to talk to you? By definition any expert location system is about communication with strangers

Its about who you know
The responsiveness is already built in, trust exists. The problem is that as we need to find more and more things our networks are not sufficient to the task. She is talking about something I call trust tagging, by which your network links you to other people. Interestingly she seems to be arguing that this is not responsive enough, and not good enough if you just want an answer not necessarily a connection.

Its about what you do
In an expert location system you have to find a way of representing people, it could be a profile. But this sort of thing is not context rich. I need to know how someone acquired the knowledge. Social Computing has the advantage in that its more authentic, its what people are volunteering about themselves. However there are problems on coverage and reliability. Two problems: How do you find the person who is not participating? How do you test the claims?

OK problem definition over and we are now moving over to a solution (I think) and she is talking about a new system called Small Blue, so called because Its about making IBM a small world

Its got a module called Small Blue Ego (I can’t see that working with some of the IBM Execs I used to know, small and ego rarely went together) which is about seeing your own connections
how do you get the data? People input it? they don’t and don’t update, the alternative is to mine and here we mine the outgoing email and its opt in – you choose to use it. Not looking a threads, only looking at content and stripping off message history. The mining is about keywords which is a limit, some of the visualisations look cool as you would expect. Speaking of which, whatever did IBM do with Babble? One of the greatest collaboration tools ever, but no one ever let it out of research.

Clever feature – if you search on a term it finds other people with expertise, and then shows you how you are connected to them (a sort of linked in type feature). People declare who they know and who and you can find ways to find people and link to them. Tells you about their forum participation, their recent blog posts and the like. No analysis just reporting as the user has to decide what is relevant.

Now I know the reason for the opening statement that this is very different from Tacit. IBM once tried to acquire it and failed. Given that Gilmour had patented the privacy concept IBM would have to be very careful not to infringe.

Small Blue seems all very explicit in terms of how it displays material and seems reliant on the profile’s being maintained. Not sure how that need is not going to fall vulnerable to the issue of keeping things up to date. They have about 25K people signing up which looks like a fairly standard early adopter profile in IBM. Their survey of 60 people showed that the participation of others mattered; participation signaled a willingness to work with people and is more important than content.

Summary now – what has she learnt in the last decade? Notion of expertise sufficiency, there is a huge spectrum of what counts. Back to her important starting point, namely you don’t want an answer, you want to have a conversation with someone, You may just want to build awareness of what is going on.

Questions – how far can we go? Personal networks don’t scale (well that is controversy) but can technology help. Also what happened to intermediaries (her initial IKM linked projects). Good questions to leave the group with.

I asked the privacy question (comparing with Tacit which allows you to be contacted for your expertise without the requester knowing that you exist). Answer was interesting, Small Blue does require you to be visible, but for the 25K earlier adopter community there has been no comment about privacy. Also you can have stop words which allows you to exclude expertise from your profile. That rather defeats the point for me. I want people to be able to find me, but not know who I am so that I choose who I help.

Key question from the floor – if this system is adopted how will it be gamed? Answer is use of email as source makes it difficult (sort of agree) and its an occasional use system

Kate made the very good point that any expert location system draws a fine line between functionality and privacy. A few more software developers should pay attention to this.

< Prev

On being what I am not

I am a novice blogger. Dave Snowden invited me to blog for a couple of ...

Category:

Further Posts

Next >

Skype

Nice post from Nicholas Carr on the eBay confession that they made a bad investment ...

Category:

Further Posts