I was invited to speak at Sage Publishing this week. My talk was pretty much the same one as the UKSG one I did last year but I threw in a couple of slides about web2 and the 'scarcity vs scale' topic I've written about recently on Electronic Museum.

One thing struck me very, very forcibly during the talk which I'd like to share here.

It's this: I am not normal.

My wife and friends of course know this. But it's not my dodgy tuna omelettes (don't ask), inability to do one thing at once or love of crumpets that I'm interested in here. It's the love, fascination and exposure to technology which define much of my life, both day job and personal time.

The reason that this came to light is that I was presenting at Sage to a biggish, youngish, coolish, intelligent London audience. In short, it's a supposed key internet demographic. These people are 20-40 somethings, very switched on, and interested and engaged in what I had to say.

You know the interesting thing? These guys for the most part didn't know what web2 was. All but one hadn't used Google Docs. The notion of the Semantic Web was a million miles away. They heard about an “API” for the first time during my talk.

I often say to people that I'm not 'deep tech'. By that, I mean that I consider myself to be a bit of a generalist. I 'dip': I roughly know what XML, OAI, RDF, GRIDDL, AJAX, SQL, .Net, etc - do. I don't, however have the brain (or inclination) to be a deep tech like many of those around me. It’s not the way I work. But I do for the most part consider myself to be connected to “real people” and able to voice the concerns and needs of the technology-requiring masses.

What I realised at the Sage talk is that I’m out of touch. I haven’t spent enough time recently talking and walking with those who actually matter: the mass market – the users - represented by the attendees at my talk. It got me thinking about how us geeks actually miss huge opportunities by not understanding how people *really* work and think. We don’t ask. We don’t test. We don’t mingle enough.

There was also a fascinating moment when I talked about tech systems often being incompatible with each other, hard to use, far from user-centric. Spontaneously, the room sniggered and nodded heads. It’s not a wholly original perspective, but I realised again how technology is a blocker and not an enabler for so many people. Whereas I wallow in this stuff, most people probably open up Word or IE with a sense of dread rather than a sense of opportunity. Of course I get frustrated too – bad user interfaces annoy anyone – but I’m in a different place from those who see these tools as just getting in the way.

One of my slides shows how I believe Facebook has pushed the Hype Curve and (luckily) this was backed up by a quick “hands up” confirmation – yes, pretty much everyone in the room had a Facebook account. This teaches us something very important, too. It’s hard to put a finger on what exactly, but it’s something about usability, user need, the “real” social web. It’ll have to be another blog post once I’ve thought it through, but needless to say Facebook has been very, very important in this equation. I’d always assumed that Google Docs were “mainstream”, for example, but compared to Facebook, they’re still geekville, by a long, long way.

I’ve blogged endlessly about how tech fails people. I’ve struggled with why and how systems in organisations are usually procured by people who haven’t asked for user input. I can see how these things happen – it’s the organic growth of requirement and technology – but I’m also determined that we geeks have a huge responsibility in educating, listening, being modest, open. We have SO much work to do - and I don't believe the vast majority of it is technical. It's "soft stuff" - asking friends, family, users, colleagues about what they use, why they use it and how it could be better.

So I’d like to thank Sage for having me and for providing some normal people to bring me back down to earth and remind me what it's all about .

I was invited to speak at a UKSG event a couple of weeks ago (mentioned in a previous post) and have finally got around to posting about it now.

First off, here's my presentation:

I went a little weird for this presentation - I wanted to do something a bit faster paced than the normal "slide / bullet, bullet, bullet : slide / bullet, bullet...." but probably not as ambitious as Simon Wardley or Dick Hardt. I think it went well - I've had a few people who were there contacting me and asking me to do the same presentation at their gaff, so it can't have been too bad...

In general I wanted to move away from the point I've been pressing for some time now ("web 2 - don't be scared") to something a bit more practical ("how to make the most of distributed services") - the main point being that it makes good sense to provide as well as consume distributed data. By doing so, a virtuous cycle is encouraged whereby more and more data becomes available to all.

There were some good speakers at the day and I made pages and pages of notes. I thought rather than publishing them as some kind of stream (and them meaning nothing to anyone..) instead I'd distill them down into three things that I felt kept coming up or were particularly pertinent:

1. Libraries and publishers are doing interesting things too

I think I gathered this from the reading I did before the event (see previous post linked above) but in particular I was impressed with the work that Talis are doing around their "Semantic Web Application Platform" and I fully intend to follow this up with an email or two and some more reading. Richard Wallis ("Technology Evangelist" - MY ALL TIME JOB TITLE...) from Talis did a good job of explaining the basics behind the distributed web.

Here's a couple more links I found when researching...

REST: http://www.daveyp.com/blog/index.php/archives/59/ , PATrest: http://www.blyberg.net/downloads/patrest_1.1_overview.pdf , AADL keyword REST search: http://www.aadl.org/rest/search/keyword/dog , Mashing the library: http://blogs.open.ac.uk/Maths/ajh59/007391.html
 

2. Blogging is sometimes being used because it's easier, not because it is necessarily the right thing

I already had a half-formed notion of this fact (I think I picked it up from Andy from the Eduserv Foundation?) at a completely different event when he'd described that he uses variously hacked blogging platforms to deliver content on his local primary school website. At the time I'd thought this was a little weird but never got round to asking him about it. Again at the UKSG event there were two occasions when it was very clear that people were writing and distributing content using blogs because the other systems were too hard to use / jealously protected / horrible legacy systems.

In one sense this is great: blogging tools are used because they are easy to set up, easy to use, easy to keep fresh with content. But in another this is a disaster: it speaks volumes about the situation whereby people at institutions around the country are turning to blogging as a way of doing what their CMS and IT departments should be doing for them already.

(IT departments: note the first part of the paragraph above: EASY is the key here, not FULL FEATURED ) 

3. People still seem convinced that the Semantic Web = Web 3.0

Even taking into consideration Simon's hatred of pointless debates I'm still going to embark on this one, 'cos I think it's just not correct that "the next web" (whatever that may be) is going to be the semantic one. It's always risky doing any kind of future-looking but I'll put my neck on the line and continue to reinforce my view: the next web is going to be about ubiquity, immersion, "everyware". Susan Wu agrees with me. So that's ok.  

Web apps are enormous news. If you haven't heard of Google's vast suite of applications: docs, stats, rss reader or used SalesForce, Basecamp... then you've probably been under a rock for a couple of years, and frankly it might well be the best place for you

It's not often, however, that you come across one which genuinely makes you think you're working on the desktop. That, I reckon, is when a web application has truly come of age: when it is so slick, so polished, so fast that you forget all about the browser running behind it. 

To date only a couple of applications have done this for me: first off (blowing an Eduserv trumpet, but it's true..) - Sitecore - the CMS with a GUI which not only looks but also works like Windows. Google Analytics is of course another.

One more which I've been aware of for some time (I think I was on the beta test) is MindMeister, an online MindMap tool. I was impressed when I first saw it, but I've just come back to it and it has got even better. They've done incredible things with the interface: it's a lovely, lovely example of well-used AJAX and an extraordinary awareness of user interaction. The synchronous nature of the way the application updates in the background is so slick it's untrue. The functionality is awesome, even without any kind of collaboration. Chuck in the fact that you can share mindmaps with colleagues and you've truly got the hallmarks of a killer app. They've gone even further: integration with Google Gears means you can get at your mindmaps when you're offline, too. How cool is that?

It's so refreshing to see functionality and usability at this level. More importantly, and worryingly, is that when you see it you realise how rare it is. We've all got a lot to learn about building applications of this standard. I know only too well how easy it is to get carried away with the stuff behind the scenes and forget about the most important bit in the equation: the user.

 

Next up (post - very good - lunch)...it's Sean Meahan from UHI talking about Extending Institutional Identity into External Services and Communities.

He's starting off talking about the takeup of external services, a topic very close to my heart.

Opportunity 1: Serve our ("our": for instance HE) users better by giving them an externally useable ID

Opportunity 2: Give an integration point via ID to merge best of breed services into the institutional framework

(Sean is talking about the notion of an ID pipe which I like - the applications built at the end of the pipe are unimportant: the pipe just provides the means to identify...)

Opportunity 3: Allows for the holistic approach: life-long learning associations - services for students, but also gives opportunities for SSO for everything from infrastructural apps to web2.0 mashups

So what does this require: well, first and foremost a change in mindset. Insitutions need to stop being "bunkered" - a similar challenge which I've blogged about endlessly in terms of museums and Web2.0. 

What are the risks in this approach?

> id theft: well, says Sean - this stuff is happening already. Nothing new here

> external service goes bust: back it up!

> Yet Another Protocol emerges...: ....

> Users change their digital id: institutions already lose users. Nothing new here...

Sean finishes by saying that this approach fits a more user-centric Web2 world...

 
Final speaker of the day: Scott Wilson from JISC CETIS

Scott starts off talking about how education works as a system - how resource is allocated across students, how collaboration is brought into the equation. Ultimately his point seems to be that a vast amount of education happens outside the formal structures provided by the institution.

Where does that leave identity? 

Previously the idea that we (HE) had the technology and students didn't is of course different now: the technology balance has changed. Identity in elearning, Scott argues, is focussed around Individualization, Self-Organisation and Activity and not around more "formal" educational activity.

He also points out that technology is user owned now. It isn't threatening, but instead offers an escape route from escalating costs and liabilities. 

So where does OpenID fit in this landscape?

Scott talks about the fact that OpenID is a proxy: doesn't really verify a user: all it really does is asserts a relationship between an agent and a URL. This, Scott says, makes it a potential boundary mechanism which provides an axis of coordination between formal and informal systems and connects things together that do use identity.

 

Gavin Bell from Nature just gave an interesting talk about how this OpenID stuff applies to educational institutions. To be honest I wasn't listening terribly hard as I was still pondering the conversation I had with Paul Walk during the coffee break...

Anyway. Here's Nicole Harris from JISC. She's not going to talk about OpenID at all, apparently

Nicole is making some very interesting points around the facets of identity: it's not only who we are but also what we like and what we are allowed to do. The way in which people are connected to resources is an important part of identity, but disconnecting identity from resources is important as well. 

The question about whether users are actually responsible enough to look after their own identities is an interesting one. We've all clicked through that screen (the one that asks about how much access to your data you allow) when adding Facebook apps. Ditto with the standard "certificate expired": none of us are really savvy enough to make these decisions about our data. This then brings in the question of how much responsibility the institution has in this space...

I must say, I struggle with the notion that people actually care that much about exposing (bits of) their identities online. I'm pretty sure that Facebook wouldn't be anywhere near as popular if we did. Of course, privacy is hugely important - and defining the edges of what is acceptable exposure for each and every one of us is crucial. But I'd be fascinated to see any research which defined where these edges usually occur. I'm happy, for example, to have my name, date of birth, blog address, job title, employer, country, town - exposed as much as anyone cares to expose it. I'd draw the line however at home address, email (mainly because of spam), phone number, etc.

But then, thinking about it: within the context of Facebook, I'm happy to post up pictures of my kids, my wife, etc. This implies some kind of absolute trust with the site: a trust which most FB users will also share. Ultimately, this is an unfounded trust: I have no idea who Facebook is or what they actually plan to do with my data...

More importantly, it's lunchtime... 

I'm at the Eduserv conference OpenID - online identity for the social network generation of learners and researchers, hoping to find out more about, er, OpenID. I've got a kind of passing knowledge about how it works, but I'm particularly interested in how (if) it relates to "Identity 2.0", particularly in the light of OpenSocial.

 I'm late (obviously) but Andy Powell from Eduserv Foundation is giving an introduction to the landscape of the web within the context of his family. His kids use this tool in different ways: Bebo, MSN, Facebook: they are true "Digital Natives". Andy, his wife and sister all use online tools in different ways. Andy makes the point that being on the web is a lifelong activity: it crosses all aspects of our lives, and spans all phases of education. People present different identities at different times, and no longer do we live in a world where institutions can (or should) "control" the online services used by their members.

This all means, simply, that to consider these things in one space, against one institution, is just simply not a useful exercise. There is a question about holisticness (holistic-tivity?) which underpins online identity.

Andy is making an interesting point about email, and how institutions such as universities used to give out email addresses but are now starting to realise that users have their own already - and asking whether they should still be giving out addresses

Andy talks about how Athens is migrating to Shibboleth and asks what OpenID means to this space. Athens was UK Centric; Shibboleth is institution-centric; OpenID is agnostic...

Next up is David Recordon from blog platform company Six Apart

2007: a good year for OpenID adoption. Microsoft, TechCrunch, Sun, ReadWriteWeb, 37Signals - all saying positive things about the technology.

So what is OpenID?

• signle sign-on for the web (that doesn't suck...)
• simple and lightweight
• easy to use and deploy
• built upon proven existing technologies
• decentralized: don't have to ask anyone permission to use it
• free...

An OpenID is a URI: globally unique and ubiquitous.

So what is the problem that OpenID is trying to solve?

> Simply, too many usernames, too many passwords, sign-up is too hard (consider TechCrunch new startups as an example..) - you're not likely to create an identity at each one.

> In the enterprise, directories are used, but they're too hard...  

 Overall, the web lacks identity. Email is often used, but if you get this hacked, you're stuffed: all the services you've signed up to are instantly available via "send me a new password" functionality...

David makes the point, as Andy did that Identity is not just one thing. Depending on where you are, who you are with, how you collaborate, you are different people. At the end of the day, this entire conversation is about trust. With OpenID, you get to choose who you trust - and even choose to change your mind again later. 

David gave a quick demo of OpenID: it's pretty straightforward; still open to improvement from a usability perspective, but nonetheless easy to use as well as implement for developers. He also then showed a graph of "Total Relying Parties" which shows adopting sites from 05-07. The graph is impressive, with a steep rise up to about 6,000 sites at present day. Obviously, given that OpenID is distributed, this isn't an accurate figure: no one actually knows how many people are using OpenID at the present time. Web2 companies are on the bleeding edge, as are some enterprises.

David also asked about OpenID in the .edu space. There are only a few, including MIT and Brigham Young University. It is starting to show adoption, but still in it's infancy...  

 

 

Google have obviously read my post on Social Graphing and decided to act

News out today on TechCrunch that Google will be launching a new set of API's called OpenSocial (URL not yet working) on Thursday this week. The basic premise is that Google isn't creating another social network but that they'll be releasing API's which talk to all participating existing social networks. They become, essentially, a node which lets developers hook into (and out of) these networks. As Michael Arrington says:

"Google wants to create an easy way for developers to create an application that works on all social networks. And if they pull it off, they’ll be in the center, controlling the network"

Orkut, Salesforce, LinkedIn, Ning, Hi5, Plaxo, Friendster, Viadeo and Oracle are among those participating. And it's obvious, really, why they would want to - the Goliath they have their sites on is Facebook, and Google support (ensuring with some certainty that the crowd will follow..) means much increased exposure.

Notable in its absence is of course Facebook, and I'm guessing that'll be the next big bit of blogosphere speculation...

 

The news of a few days ago that Microsoft has bought a stake in Facebook has not suprisingly got the blogosphere talking. Mainly because it's a HUGE sum of money (yes, that really was £117 million quid for a 1.6% stake..) but also because anything involving Facebook gets media attention right now. Here's a Google Trends snapshot of how Facebook has grabbed both search and media attention over the past 3 years:

 

I've blogged a lot over on Electronic Museum about the opportunities opened to us by these kind of social networking experiences but also about what it means for the various companies involved to be open with their data.

Facebook's Platform seemingly answered a lot of these issues, but at close inspection it isn't all that it appears to be. Yes, Facebook provides data to the Platform so that developers can build on it, BUT they don't provide data OUT. There's a few RSS feeds which can be grabbed, but not a whole lot more. Meanwhile, of course, LinkedIn and MySpace have both announced similar developer-esque platforms with as yet undefined edges: the extent to which these sites will provide data to the outside world (or not) remains a point of much discussion.

Where does the social graph go from here?

Whatever the end result, there's a much bigger issue afoot, and it rests on usability, privacy, attention data and a whole host of web2ish concepts around identity. It is about taking Facebook's notion of the Social Graph: the "..network of connections and relationships between people on the service" and asking where does this go now?

In a world where social networks becomes the norm (and yes, we're in it..), we can make some pretty easy assumptions:

1. We're likely (thank God) to see an internet with an extended and complex Social Graph (read: many, many social network sites)

2. We like and use some of these more than others

3. We are really bored of putting in the same details on more than one site

4. Just as in real life, our context changes how we use these. If I'm 14 (I'm not..) then I'll probably spend time on Bebo. If I'm 35 (I am) then I spend time on Facebook. If I'm 35 and freelancing or running my own business (I'm not) then I spend time on LinkedIn. If I'm in a band (I'm not at the moment) then I spend time on MySpace. If I'm looking for a band to be in (I am) then I spend time on MusoFinder.

You get the picture: lots and lots of sites do great social type things really well, but they all require investment in inputting data that is constant and consistent across all of them. No matter which of these sites I want to spend time on right now, I'm still 35 and not 14. My name is still Mike. I'm still married, and have kids. I still live in Bath. And I'm still really irritated that I have to change these details across multiple sites should anything change.

Right now, Facebook is the social app of choice. So as users we're all inputting data like crazy, building our own social graph, uploading and tagging photos, sharing with friends. But what about when Microsoft buys the other 98.4%? What if Google does instead? What happens when my mates start moving over to www.anotherdamnsocialnetworksite.com and I'm left high and dry with no-one in Facebook? What if the Facebook servers turn out to be horribly virus-ridden and all melt, deleting my social graph, images and wall posts since the dawn of time? What if Facebook's CEO turns out to be an alien trying to take over the world and I want to remove all my data from his green, slime-covered servers?

I'm being slightly fatuous (although have you seen Facebook's CEO - I'm sure that's a tentacle..) but the point is a serious one:

The social graph shouldn't be owned by just one company.

The only possible solution is a distributed social graph: a way of managing this complex set of relationships which is outside any one company, not owned or distributed by any one entity: an internet-like social graph which maintains itself, has complete clarity of data input and output and a hugely strong security model. 

Brad Fitzpatrick explains the whole issue really well and suggests a range of ways in which we could move ahead. It makes for fascinating reading. 

As he says, this isn't about getting rid of Facebook - it's a great, usable social networking resource. But these issues should also make people a bit wary about putting all their social eggs into one - albeit large - basket.

This week I've been doing some research on "Library 2.0" - I'm presenting at a UKSG seminar called "Caught up in Web 2.0? Practical implementations and creative solutions for librarians and publishers" in November.

One of the better papers - or at least one which seems to have been well researched, and covers lots of interesting ground - is by a chap called Jack Maness and is called "Library 2.0 Theory: Web 2.0 and Its Implications for Libraries". It's a great introduction to what libraries are doing in the brave "new" world of Web 2.0, why they're doing it and what the particular challenges are. He doesn't unfortunately mention one of my favourite YouTube clips - an oldie but goodie: "Introducing the book" which I link to whenever given half a chance...

I've got a fair amount of experience of what the issues are facing museums in this space, and it's fascinating to spend some time looking at a parallel sector. Not only do the similarities come shining through, but I've found that looking at this from a different angle also sparks off some thinking which I haven't had before. It's refreshing to "know about Web 2.0" but to be naive about a particular application of the various technologies and techniques. My experience of libraries online has until recently been almost entirely limited to OPAC searches at university, which was a frighteningly long time ago, and the integration with Unicorn which we carried out as part of the NMSI Ingenious project.

So many of the challenges which we face as museums - challenges which I've frequently tried to articulate at various conferences - are echoed in the library world: What about authority? Can we cope with the resource implications? What does it cost? Why should we "do Web2.0"? What does the future hold?

Mr Maness does a great job in articulating the various approaches to Web2.0 in the library sector. I'd recommend reading the paper, especially if you're new to this sector. Here's a few choice snippets to tempt you..

"Library 2.0 is...communally innovative and rests on the foundation of libraries as a community service, but understands that as communities change, libraries must not only change with them, they must allow users to change the library..."

"Library 2.0 blurs the line between library and patron, creator and consumer, authority and novice"

He also focusses on the vital nature of the cross-over between physical and real - something which readers of my Electronic Museum blog know I bang on about a fair amount:

"Library 2.0 will show no distinction between or among formats and the point at which they may be accessed"

Once I've done my presentation at UKSG I'll of course post it to the web and link to it from here...

Today I read three stories, all loosely connected which could make for interesting time for social networking in the not too distant future.

Firstly theres social stream, which seems to offer the potential for users to share their socal data to many different applications, an approach discussed at the recent FOWA conference by Yahoo and their Fireeagle application, which is also designed to share the (location based) information of a subscriber to multiple applications ...but i digress (told you so).

Social stream, suggests that the guys at Carnegi Mellon University are working towards a solution which will allow you to share social data across multiple sites, if you, like me have mulitple profiles on multiple socal networks, then this Unified Social Network as they describe it, could be a godsend. One place for all my virtual friends, for me to discuss the finer points of life with them and ......they with each other? wow. ok thats freaky...but 'super' cool.

Then there comes the news that Ibm and Linden Labs are working ona project that would allow users to share avatars across virual worlds, ok so now i can not only create my USN but i can create a persona that can literaly walk from one virtual environment to another. Socially thats impressive, economically thats opportunity, considering the value of some of this vitual property how long before the first virtual bureax de change is set up ;) virtual tour operators selling virtual holidays and renting space in other virtual networks, what next, world of warcraft tie ins, dwarfs and orcs running riot in second life...no wait ill get back to the point.

In his article social networks and groups formed Shiv Singh described (amongst other things) how, in social networks, 'isolated regions' became part of the more active middle region by nodes (users) in the group joining the larger middle region. Thus the larger middle region grows organicaly and new isolated regions are born, is this going to happen to all social social networks, ultimately will they become part of the collective, the single unified social network, web 4.0?

Then along comes Google(again), who its seems have been busy, it would appear that they have decided to out do facebook in the TechCrunch article, google to out open facebook Michael Arrington discusses how google are to open up the API to their social graph data, big deal, who uses orkut anway right? But wait, go back to the the begining, and add into the equation the work Google are doing with the social stream bods and there Unified Social Network stuff.

disclaimer, ok heres where you put on those rose tinted specs, this is the intraweb we are talking about

So now we have everyone moving across every (ok maybe not every) social site, and the API is and 100% open....with avatars. now that really is super cool and possibly a little scary, will there emerge a brave new world in cyberspace with an entirly new set of rules, a new reality and should I stop taking the Matrixtoo literally?

meh

maybe Ferris Bueller had a point I'm off to trash a ferrari, virtually of course ;)