Enterprising, but not enterprise.

Peter Rip of Leapfrog ventures has written a great article about the issues creating enterprise standard products using public APIs. Web services are generally released without a SLA (Service Level Agreement) and the terms and conditions basically don't offer very much security to the user regarding uptime and support. Service level problems are already evident in a range of online applications, and occur for a number of reasons. Every Saturday morning Spurl.net goes offline for backup, which might be appropriate for such maintainence in the States, but in the UK its a peak usage time. Even Googles own Gmail service has experienced outages that have desperately upset users, so I don't think there's currently much hope in using Web Services for mission critical applications.

Hopefully this situation will change over time, but the companies who host the APIs need to sort out what kind of revenue model is going to suit them best. Peter highlights that one of the major reasons that Google and Yahoo have not released business models for their web services is that their current model is based on 'direct sales' and not 'indirect sales'. A mashup incorporating the Google API is would be considered by Google to be an indirect sale, so there is no model, and companies like Google can't adopt a new channel strategy overnight.

One route Google could go down would be to move the distinction of the customer up a level from the end user to the developer or company that is using the service. There are also a number of financial models that could be adopted including bandwidth usage, subscription packages or simply pay per hits which could provide a healthy revenue stream for supporting 24/7 uptime for the services.

It's inevitable that the Mashup movement is going to move into an enterprise model and quickly in my opinion. If these first movers don't sort decent SLA's there will be a market opportunity for companies to supply enterprise standard web services to support the mashup movement.

Tags Mashup, Web 2.0, gmail, Peter Rip, Web services, Google, SLA

Police told to blog off

The Register today reported that the Metropolitan police are the latest organisation to 'encourage' it's staff to stop running blogs. The advice acknowledges that they cannot force officers to stop blogging, but requests that they: "consider the impact of expressing views and opinions that...bring the organisation into disrepute" which incidently makes for ironic reading on the day that Ian Blair is in the headlines for illegally recording conversations with the Attorney General.

This situation the Met have found themselves in is similar to that of a growing number of organisations trying cope with the level of conversation the web is enabling. Rather than trying to encourage officers to stop blogging altogether I would have preferred to see a clear and concise best practice for officers running personal blogs.

It's never a good policy to smother conversation, but most people are happy to moderate it to suit a purpose.

tags: conversation , blogging , metropolitan police,

Mash makes Smash

There are two kinds of mashup getting online attention at the moment. The first is the use of open API's like Flickr to create new online tools and the other is the latest Japanese video on YouTube showing that a potato can be peeled in just one second. Now even though I hate peeling potatoes (any other tips gratefully accepted!) I'm more interested in the Web 2.0 kind of mashups. Dion Hinchcliffe wrote today that MashUpFeed is posting 2.63 new projects every day, with Googlemaps being the most popular API used.



The range of web mashup topics is as imaginative as the 2005 mashup trend of mixing contradictory tracks together. In 2005 we were treated to DJ BC mixing the Beatles with the Beastie Boys, we've now got Virtual Places mixing Amazon, Flickr, Microsoft and FeedMap API's to create a brand new mapping/webservice, and with sites such as MyFirstKiss and Run London emerging from the crowd, it's obvious that there's a lot of creative work is going into the production of these new web services.

What I hope happens now is that the bar to entry is lowered, as it has been in many other technologies. Perhaps Garageband for web Mashups could emerge allowing people with modest skills to join in the fun.

Tags: Mashup, Web 2.0, API, Flickr, Googlemaps, MyFirstKiss, MashupFeed

Portal [2.0]

I got an email from a Web 2.0 Consultant this week claiming that Portals are dead and we shouldn't use the term any longer. His perception of a portal was of a dinosaur technology/terminology that didn't fit in with the new Web 2.0 way of thinking. He supported his argument by linking to a couple of news articles quoting Rupert Murdoch's acquisition of Myspace.com where Murdoch stated that "young internet users don't have to go and work their way through Yahoo's home page or MSN's or someone else's."

Now, whilst I agree with Murdoch's comment, I don't think that NewsCorp have gained enough credibility through buying MySpace stop me from using the term portal. I also don't believe that Web 2.0 is about throwing the 'baby out with the bathwater', it's simply another milestone on the road.

The major problem I have with ditching the term portal is that I believe portals have been instrumental in the conceptual development of Web 2.0 thinking, technology and interaction.So that left me needing to go back to the definition of portal to sense check this concept. This is some of the interesting stuff Wikipedia had to say on Portals:

'Web portals are sites on the World Wide Web that typically provide personalized capabilities to their visitors.'

Some features of enterprise portals are:

Single touch point - the portal becomes the delivery mechanism for all business information services.

Collaboration - portal members can communicate synchronously (through chat, or messaging) or asynchronously through threaded discussion and email digests (forums) and blogs.

Content and document management - services that support the full life cycle of document creation and provide mechanisms for authoring, approval, version control, scheduled publishing, indexing and searching.

Personalization - the ability for portal members to subscribe to specific types of content and services. Users can customize the look and feel of their environment.

Integration the connection of functions and data from multiple systems into new components/portlets.

For me that supported the concept that web portals developed in the late 90's created the user and technological foundation for the 'revolution' of Web 2.0. So what happened to them? I don't believe that when OReilly flicked the switch into the age of Web 2.0 all these great Portals just disappeared in a puff of blue smoke.

Well when you look at where the market is now it's obvious that the Portal didn't disappear, it evolved. Today's portals such as Google Personalised Homepage, Microsoft Live, Netvibes and many others still align themselves with the guiding principals of "single touch point, Collaboration, Content Management, Personalisation and integration" they just do it better.

So now I believe there is a real justification in the use of the term Portal. It's a term that the public are used to and have an anticipation of what functionality to expect from them. The internet is too grown up to simply adopt technical terminology such as 'Ajax homepage' for public consumption. The end user doesn't care about the hype or the technology only that that they understand what its for and how to use it.

So lets ditch the hype, the talk of revolution, and guide users through the ever increasing speed of innovation and development. It's just going to get faster and more intense, lets not lose our customers on the first bend.

Besides in the sprit of Folksonomy I can tag portals anyway I like!

Identity crisis?

Earlier this month, I stumbled on the excellent work that's currently in progress on online identity. At that point I suspected it would eventually come up in my work, and sure enough it has. I'm currently designing a community tool which will involve mashing up a number of different online applications. One issue that I'm concerned will hamper the user experience is the problem of maintaining a consistant identity over a number of different sites.

I've found a couple of interesting tools that are trying to overcome the problems we all encounter when working on multiple online applications. They both concentrate on managing user comments in blogs and are well worth taking a look at:



Cocomment is in a pretty good working state and is compatable with a lot of the main community tools. The other application that I'm interested in following is Sxore which is one of Dick Hardt's projects.

Friendly error messages

After reading this post on we break stuff, the first site I visited was down for maintainence. It got me wondering who designs these pages for an organisation, is it the witty designer, the IT dept armed with fonts and stock photos or the marketing dept trying desperatly to stop valuable users drifting away?

This Flickr group group archives these quirky, geeky error pages that make us feel ok about stuff being broken - I suspect they'll be a thing of the past in Web 3.0!

The Most Promising Web 2.0 Software of 2006

The Most Promising Web 2.0 Software of 2006 This is a good roundup of the companies who are starting to deliver on the promise of Web 2.0 marketing hype. I think its refreshing to see how investment is supporting the work of developers and creatives who have been pushing the boundaries of online user experience over recent years.