paul walk’s weblog

On the subject of the e-Framework for Education and Research, Andy Powell, in eFoundations, asks:

…does the e-Framework support the Web (again, read Web 2.0 if you like) mindset, does it fight against it, or is it neutral?

Although the e-Framework is careful to position itself as technology-neutral it is, nonetheless, somewhat associated with WS-* in some quarters. This may be partly because the e-Framework was born partly out of a perceived need for our communities to engage with SOA.

The arguments between proponents of WS-* and REST respectively are, by now, well-rehearsed. I’ve been a little startled by the strong positions I’ve heard people take on this issue. There are use-cases for both, but the fact that the REST approach has gained a great deal more traction recently is interesting and shows a healthy interest in opening up ‘resources’ for simple web-based operations.

In a post entitled ” REST vs. WS-*: War is Over (If You Want It)”, David Chappell asserts:

To anybody who’s paying attention and who’s not a hopeless partisan, the war between REST and WS-* is over. The war ended in a truce rather than crushing victory for one side–it’s Korea, not World War II. The now-obvious truth is that both technologies have value, and both will be used going forward.

David cites impending support for REST in Microsoft developer products and similar support in ‘official’ Java frameworks. He suggests that the argument was caused by the use of the phrase ‘web services’ to describe a set of standards, protocols and technologies which have only a passing connection with the Web and that, by contrast, REST is entirely rooted in the Web and its architecture.

Interestingly, Elliotte Rusty Harold extends David’s ‘Korean War’ metaphor, to suggest that “WS-* is North Korea and REST is South Korea”. It therefore follows that:

WS-* fails because it believes massive central planning works better than the individual decisions of millions of web sites. It’s no coincidence that the WS-* community constantly churns out volume after volume of specification and one tool after another. [...]

By contrast you don’t see a lot of complicated REST frameworks or specifications. [...] REST/HTTP sets up a simple economic system based on a few clear rules, and then pretty much gets out of the way to let people do their own thing. It doesn’t even get too upset when people break the rules[..] they will be dealt with by the RESTful market.

Of course Elliotte has changed the central message of David’s original piece - Elliotte believes that WS-* is, essentially, doomed. The important point for him is the superiority of the market-place versus the inadequacies of central control.

So, back to the e-Framework, and Andy’s question. Well, the political answer might be to say that the e-Framework can’t afford to be seen to be fighting against the Web/Web 2.0. If the e-Framework confined itself to describing SOA within the education/research enterprise, then it might avoid the Web 2.0 market-place. But the e-Framework doesn’t want to be North Korea, it wants to be, if not global, then international at least.

The pragmatic answer is, I suspect, that so long as we allow the e-Framework to be shaped by our communities, we will see the answer emerge. Over time, we’ll be able to re-ask Andy’s question as “do our communities support the Web mindset….?”, using the e-Framework as evidence. If, on the other hand, we do not allow the e-Framework to be shaped by the market-place of our communities, then it will matter less and less what it supports or fights against….

Technorati Tags: Andy Powell, David Chappell, e-Framework, Elliotte Rusty Harold, REST, SOA, WS-*

My parents recently bought my son a sandpit and a couple of sacks of sand to go in it. He loves it (although his enjoyment has been curtailed by the UK recently developing a monsoon-like climate of continuous rain). He’ll generally sit outside of it, playing with the toys and sand inside, or sometimes he’ll just climb right in. Either way, the rule is the same. The sandpit is a semi-closed system - no sand is allowed to escape the confines of the sandpit, and only some compatible (i.e. sand-proof) toys are allowed in.

Because I’m to old to play with a real sandpit (well…often anyway) I’ve been looking at the facebook platform (API) which allows developers to build new widgets which they can deploy within facebook. A basic use of the API might be to surface an existing, external application within facebook in order to reach the users inside this system. This is the web-portal model, such as used in Netvibes. A more sophisticated use would exploit the social-networking capabilities, which are the real point of facebook.

There’s a lot of hype about facebook right now, and the opening up of its API has been warmly received. The way I see it, this now makes it possible for us to bring more toys into the sandpit, making the whole sandpit experience a more rewarding one. But it’s still the sandpit, walled-off and separate from the rest of the world.

Jason Kottke says:

you do know that Facebook is AOL 2.0, right?

The world is still adjusting to the “web is the platform”, core to the Web 2.0 meme. But is this already being usurped by a new reality, the re-appearance of walled gardens, as in the early days of AOL?

In a very interesting post about ‘platforms’ and lessons not learned in the web-platform era, Marc Andreessen points out that:

You can layer new code and functionality on top of what Facebook’s own programmers have built, but you cannot change the Facebook system itself at any level.

I just can’t get all that excited about facebook as a platform. From my point of view, in an exciting era of mashups, facebook is only seriously mashable in one direction, and it’s the wrong direction. If facebook’s social networks were exposed to the web, ‘mine-able’ and mashable - now that would be exciting. But as Jon Udell points out, that would be risky and with “no obvious benefit to facebook”.

Mike Ellis says:

…the mashup environment is about playing with technology - it is therefore partially technology driven (a bad thing) but also understands and build on content and data from disparate sources in the hope that the thing which pops out at the end is useful (a good thing). It relies on a Darwinian process to determine what works and what doesn’t: if your users like it, they’ll take to it and it’ll succeed.

I agree. And facebook’s viral aspect gives the Darwinian process a shot of adrenaline. It’s a pity that facebook’s social-network, it’s viral power can’t be applied to mashups, ‘out there’ on the web.

I think that facebook is a sandpit. I’ve had a little fun in there, playing with a few toys. I’ll probably play in there from time to time - I like a sandpit as much as the next kid. But then I’ll get bored and wander off, leaving my toys lying half buried in the sand, looking for something better to play with.
Having said that, people whom I respect have a higher opinion of facebook. Perhaps I’m missing the point?

Technorati Tags: AOL, GoogleMap, Jason Kottke, Jon Udell, Marc Andreesen, platform, sandpit, walled garden, web 2.0

At the Glastonbury Festival in the UK this year, the big story was not so much the performances as the weather. In particular it was the mud which featured heavily, slowing down the processions of bedraggled festival goers as they migrated from one stage to the next.

On Saturday, I visited Secondfest after seeing the publicity in the Guardian newspaper. This was a virtual festival, hosted in SecondLife and sponsored by the Guardian and by Intel. Virtual festivals have their own kind of mud. It’s called lag. Attempts to organise large ‘gatherings’ in SecondLife really do tend to show up the platform’s limitations. With a practical limit of about 40 ‘avatars’ per ’sim’, after which the whole experience degrades rapidly to a crawl, SecondLife really can’t deliver an experience which more than 40 people can enjoy simultaneously. In true festival style, this was mediated a little by having more than one stage, with other areas between.

According to one blogger (hosted by the Guardian), the festival ’site’ had attracted 3,000 unique avatars by the end of the first day. That’s about 10% of the average population of SecondLife logged in at any one time, which seems quite impressive. I logged into the festival twice, and the most I saw in front of the ‘main stage’ was about 20 avatars (dancing!).

It seems to be a typical experience of SecondLife that you find your avatar walking through very cleverly realised settings, alone. Secondfest was no exception once my avatar had left the vicinity of the main stage. But for the complete lack of other people, the ‘camping area’ was very well done - the tents were cleverly rendered and it was even possible to crawl inside one and have a rest:

I think that the provision of ‘portaloos’ for avatars might have been overdoing it a bit though:

I might pop in again today to see if anything’s changed, but so far it feels like most of my SecondLife encounters: it sounds promising, it looks good, but it’s ultimately a fairly unsatisfying experience. It’s a shame, because there is clearly a strong desire to used a platform like SecondLife to host large, virtual gatherings. Currently, the platform really just isn’t up to it. But at least the avatars that had waded through the virtual mud to the Secondfest main stage and were maniacally dancing, seemed to be having ‘fun’….

Technorati Tags: festivals, Glastonbury, Guardian, lag, mud, Second Life, Secondfest

Good to see that 37signals have adopted (optional) OpenID support for Basecamp. I was already using Basecamp with the older, local user account credentials, but the system allowed me to swap to using my OpenID very easily. Good work, as ever, by 37signals.

However, I can’t help thinking that is only half the story. The ‘ID’ in OpenID is, at one level, about identity as in ‘Identity Card’. But my OpenID is also an identifier, which is fundamental in the context of the web. This is being lost in the system - knowing my OpenID does not help you to locate my projects in Basecamp for example. (This is not intended as a criticism of Basecamp or 37signals - I’m still enjoying the convenience of having one user account fewer to remember after all!)

I’ve thought for a while that the introduction of URIs for people was the often overlooked yet potentially most interesting aspect of OpenID. In a resource-oriented-architecture, it would seem plausible to suppose that a reliable pointer to a representation of a person would be a useful thing. But when I try to sketch out a useful application for this, I struggle, partly because I don’t want to admit that the semantic-web people might have the answer, and it might be called RDF

Technorati Tags: 37 Signals, Basecamp, identifiers, OpenID, RDF, resource-oriented-architecture, semantic web

Today I briefly attended a follow up to the Eduserv Symposium, Virtual Worlds, Real Learning, Revisited which was an entirely ‘virtual’ event conducted in Second Life (SL). My attendance was cut short by an incident involving a laptop and a three-year old upon which I won’t elaborate…. I did manage to rejoin the event towards the end.

The event today was, essentially, a seminar, with a 5 or 6 ‘panelists’ and a varying number of audience members (maybe peaked at around 25 - difficult for me to say). Before the event Eduserv had given out full instructions about a ‘queueing’ system, designed to prevent everyone talking at once. When I joined the seminar, around 15 minutes late, this had been abandoned and a free-flowing chat was in progress. At times the chat system hit the same problem all such applications seem to do, which is that there were several threads or conversations mixed in together and it became difficult to follow.

At one point, one of the audience asked a question which made me sit up and take notice. The questioner wanted to know if people were looking at the avatars and virtual world surrounding them or if they were just reading the chat ‘history’. By this point I had enlarged the chat history window to the point where it obscured the rest of the virtual world entirely. This was the moment I wished I was using a better chat system - something more like Skype for example. To be fair to SL, it did attempt to cope with considerably more than Skype’s maximum of 10 participants in one chat session. Having said that, I found that the chat tool in SL is woefully slow if you’re in the middle of a flowing discussion - on my pretty powerful MacBook I found myself watching my keystrokes crawling across the input box. And my enlarged chat window was semi-transparent so that I could still see avatars doing that comical ‘typing in air’ thing which indicates they are ‘chatting’ (I imagine I can change this behaviour, but in the heat of the moment I didn’t want to go hunting for the switch).

I’m still very skeptical about the value of SL in education or e-Learning. Nothing in the discussion has made me feel differently and today’s experience, my first ’seminar’ in SL, just made me think back to the days when the early promise of ‘virtual learning environments’ began to pall and we realised that the integrated VLE was not generally better than the sum of its second-rate component tools, and that we might be better off just selecting better tools.

If I hadn’t run into an unfortunate child/laptop proximity event, I would have asked what the panel thought of the significance or otherwise of SL for distance learning. As I blogged before, the guys at Eduserv have already proved to me that SL can definitely play a part in delivering conferences - allowing a worthwhile degree of participation for remote delegates. I think SL has some real potential as a virtual meeting space - perhaps when the text chat gives way to voice (in development in SL) then we’ll start to see something really interesting.

Technorati Tags: chat, distance learning, Eduserv, efsym2007, learning, Second Life, Skype, Virtual worlds, VLE

An interesting post by Mike Neuenschwander on the Burton Group Identity Blog. I’m not certain I agree entirely with the main thrust of Mike’s argument, which he offers as an axiom:

There are no identifiers, only attributes

That is to say, things are identified by their existence as a collection of attributes in a given context. Some of Mike’s claims, such as “most people have [...] several dozen nicknames” seem a little exaggerated. However, his concluding remarks are interesting:

I understand why from a programmer’s perspective, it would be so much more convenient if everybody could simply have one globally unique, unambiguous, resolvable name. But such a quaint design constitutes a wanton disregard for reality.

The tech industry is adolescently ID-fixated.

I am professionally interested in identifiers, and am aware of serious interest in the notion of the creation of ‘authoritative’ resolution services for names, in the domain of the publication and use of scholarly work. The ability to reliably cite authors of scholarly works would surely be a good thing. But this requirement has been around for a long time now, and is still not satisfied.

Is the “adolescence” of the industry a barrier to pursuing a solution which doesn’t show a “wanton disregard for reality”?

Technorati Tags: burton group, identifiers, openid, Mike Neuenschwander

Someone has set a server somewhere in China to receive a URL and then act as a web client to check whether or not this URL is accessible. The implication is that some URLs are being blocked by some process in some parts of China. Maybe. It’s difficult to distinguish between technical problems and political interventions from here (to be fair, the creators of this site to make a disclaimer to this effect)

The site has a funky animation showing the HTTP request/response crossing a map of the world - reminiscent of so many movies where the ‘hacker’ does something clever.

The URL to the home page of UKOLN where I work is unavailable to that server in China. I also checked this blog’s address and it is currently available, but perhaps it will be blocked now that I have blogged about this! I must remember to check in a few days….

To check whether or not you are being censored somewhere in the world, go to www.greatfirewallofchina.org and enter your URL.

Technorati Tags: Apple, censorship, greatfirewallofchina.org, UKOLN

I’ll be giving a talk at The Shock of the Old 2007: Shock of the Social event in Oxford later this month, entitled: Does Web 2.0 Herald The End Of In-House Development And Provision Of IT Services? In this I am collaborating with my colleague, Brian Kelly, who came up with the original idea.

This is in the context of Universities especially. However, this sort of debate has been had before in industry - the ‘Software as a Service’ concept is certainly not new. But the collaborative and social aspects of Web 2.0 give a new spin to these discussions. For example, I have to give a presentation. Traditionally, I leave this until a few days before, fire up Keynote on the MacBook, spare a thought for the poor unfortunates wrestling with PowerPoint, and churn out a presentation. On the day I’ll give the presentation, and hope to spark some discussion. Then, once I have given the presentation, if it went down OK, I might upload it to Slideshare, and then blog about it. Maybe someone will comment and the presentation will continue to provoke/encourage further conversation in the blogosphere…. But wait! Why does the presentation have to start the conversation? Why not start it now, here on this blog? Get a conversation going before the event, create a better informed presentation, carry on the conversation afterwards.

So, does Web 2.0 mean the end of in-house development and provision of IT services in universities? If, for example, university staff and students can set up blogs & wikis on any number of hosted services, do they need a local IT department to do this for them anymore?

What do you think?

Technorati Tags: university, slideshare, Web 2.0, software as a service, web services, shock2007

Well, I finally got around to sorting out my own OpenID (paulwalk.net), following the excellent instructions provided by Simon Willison.

As I find myself signing up for more and more remote services, nearly all of which ask me to create yet another user account, the potential value of a user-controlled, decentralised identity system becomes clearly apparent. Like many others, I have been interested by Yahoo Pipes, enough to create a Yahoo account for the purpose of trying it out. I estimate this is the third or forth Yahoo user-account I have created over the years…. sigh, if only I could use my OpenID! Apparently, Steve Gillmor has the same problem.

I hope, as the year progresses, that I’ll be able to use ‘paulwalk.net’ in a steadily increasing number of places - see The OpenID Directory and MyOpenID’s directory for examples.

Another good resource for all things OpenID is at OpenID Enabled.

Technorati Tags: web services, OpenID