The redesign itself is very nice; it’s more minimalist than any recent effort, and gives big emphasis to discovery of content, with a frankly enormous tag cloud and a selection of three videos, blogs and music albums which are deemed to be of the moment. Visually it maintains the obligatory rounded corners, so it must be a success.  The interface throughout is tidier and more effective than previously.

An interesting change (in that I don’t remember it last time I used Technorati) is in the search engine results pages (SERPs). They are manfully attempting to do something fairly similar to Google’s recently announced Universal Search, in which a single search will result in the most relevant content, whatever the form. On Google this can include anything from text to images, maps to books. On Technorati this would include video content, podcasts and blog posts, for example.

Technorati is, however, less successful than Google. Their attempt revolves around a ‘featured’ tab on the search, which is far less clever than Google, because all they do is search various mediums and then plonk it in a relevant box. Video results are shown in the video box, blogs in the blog box. You get the idea. This easier to do because it doesn’t require ranking algorithm, and all it really amounts to is a metasearch of the different content-type searches. Whilst a nice UI touch, it isn’t really much more than that. They continue to push using Technorati tags; if you want to show up on that ‘featured’ page, my initial look would suggest that you must use them.

So what’s that change in emphasis that I mentioned then? CEO Dave Sifry alludes to it in his post on the Technorati blog:

Whereas folks using Technorati a couple of years ago were predominantly coming to us to search the blogosphere to surface the conversations that were most interesting to them, today they are increasingly coming to our site to get the 360 degree context of the Live Web – blogs of course, but also user-generated video, photos, podcasts, music, games and more.

I think what he’s talking about (the emergence of many different forms of new media) only really scratches the surface of what Technorati is becoming, or could become. I’ve talked before about the problems of content discoverability in the long tail (explanation of what that means), and Technorati is well placed to provide a completely different solution to the problem from what’s being done by the current leaders in the area (Digg, Del.icio.us and StumbleUpon).

Through some sophisticated, well developed algorithms, it should be possible to leverage the vast number of blogs which they search to come up with personalised recommendations for a person based on their OPML file and/or web history, by analysing past reading patterns and then ranking possible content recommendations by similarity and discussion on other blogs. At the moment all they do is come up with the one-size-fits-all suggestions that you see on the homepage, which is possibly useful but of fairly limited value.

All in all, an update that goes in the right direction but isn’t massively exciting. The UI is now impressive, and they appear to have recognised their importance in facilitating the discovery of content. They’re not doing such a great job in that, or the universal search idea, at the moment but that will come. They face the ever-present challenge of Google Blogsearch, but seem to be growing well and crucially they are out-innovating Google in the space at the moment.

Formerly The Venice Project, Joost is an on-demand IPTV (television over the internet) startup from the founders of Skype, and it has has never been far from the blogosphere’s attention. Indeed everyone in media seems to have decided that it is the ‘next big thing’. It looks like the whole effort is going well – they’ve done deals with many key content providers which will ensure that they will have TV franchises such as CSI available, and last week I got yet another press release telling me that they have secured advertising from the likes of Coca-Cola and Nike. I’ve been part of the closed beta since it was called The Venice Project, and I wanted to take the opportunity to do a look at the product, the ideas behind it and its chances for success. This is an obscenely long post, for which I apologise, but I think it does cover most of the important aspects of the product.

The Application

The downloadable application and the content itself is completely free, and it is ad supported. It is on demand in that the user chooses what programme to watch and what time they want to watch it, but the programmes are organised into conventional-style channels which play their content sequentially. A widget platform is perhaps the only nod to social networking, as it could potentially allow users to chat to other people watching the same thing (as possible with YouTube), rate content and have content recommended. If features such as this were well developed and robust, it would go a long way to dispel people’s concerns about the lack of community.

The Philosophy

There are three areas of innovation with Joost: the use of peer-to-peer, the on-demand channel model and advertising.

Joost, like the founders’ previous endeavours, makes use of p2p technology to enable the streaming of reasonable quality TV over the internet without ridiculous bandwidth costs. There is an excellent page on Skype site giving the technical lowdown on the advantages of P2P, but simply put, if a centralised system were used then every time a new user joined the network costs would grow, because that centralised system has to support one more user. However, with P2P, as users join Joost they become part of the distribution system which means that Joost’s costs do not spiral with growth. That has to be healthy from a business point of view, but it also means that the speed of distribution is likely to be greater, and obviously there is far greater redundancy. The downside is that Joost has a significant bandwidth footprint on a user’s computer; according to their own help, “in one hour of viewing, approximately 320Mb data will be downloaded and 105Mb uploaded”. That will be enough to cause problems for many UK internet users, burdened by restrictive fair use policies and bandwidth caps.

The previously mentioned on-demand model, in which shows are organised into channels but a user can choose what to watch and when to watch it, is unique. In some ways this provides the best of both worlds, as you don’t always have to know what you want to watch when you sit down in front of it thanks to the channels, but you have the flexibility of on demand. Others (including our very own Chris) argue that the lack of any attempt to build a social network around the content is ignoring the success of YouTube, which has made use of its network effect to enhance discoverability of content and provide a two way experience.

Advertising is probably the most interesting part of Joost; it is the raison d’etre, after all. Whilst one might expect the advertising model adopted to be very similar to television, the unique aspect of the platform makes it much more interesting, in two ways. First of all, Joost is not limited to having only boring 30 second and 10 second video adverts inserted into their shows. From last week’s press release:

Creative executions, which will be added to Joost on an ongoing basis, include both conventional advertising units, similar to those found on TV and the Internet, such as stills, spots, overlays and widgets; and innovative types of advertising, such as branded entertainment and channel sponsorship and promotion. Users will be able to learn more about the advertised brands by clicking through the interface.

That gives advertisers many more options than previously available. Agencies like to be seen to be innovative, and Joost gives them an ideal way to be seen to do this.

However, it isn’t the most exciting part. Joost is part of the attention economy; it knows what you watch, how much you watch, when you watch and quite literally where you live and thus is privy to a significant amount of information on you. This, whilst scary, is obviously of immense value to advertisers, who are constantly looking to maximise their return on investment by targeting their ads at the right people. In some ways, the goals of the audience and the advertisers are quite similar; the audience doesn’t want to see ads that aren’t relevant to them, and the advertisers don’t want to show such ads because they are a waste of money.

Conclusion

Joost stands every chance of being successful. It appears to be getting great content deals, fuelled by a genuine belief in the mainstream media that Joost may be the future method for distribution of TV programmes. That content, combined with the on-demand aspect, will bring massive audiences who want flexibility currently not available easily or cheaply. And the massive audiences will bring advertisers, attracted by more ways to advertise and engage, and crucially better ways to target. It sounds like one big virtuous circle to me; Joost could well be the next, well, Big Thing.

There are already a host of services our there such as StumbleUpon which uses your browsing behavior to recommend sites you might like to visit. Even Last.fm uses your ‘Attention Information’ on your iTunes playlist (how many times you’ve listed to various songs) to create a personalised radio station which is tailored to your particular taste. Particls, which Huw will be reviewing shortly, is another application which is using attention data in a fascinating new way. What all of these services have in common is their use of past attention to predict what you will be interested in now and in the future. This could be extremely powerful in solving problems like information overload and content discovery in the long tail media economy (previously discussed here).

Such information is extremely valuable, and as such there is an effort to standardise the way in which it is stored, therefore making it portable. As a result, the APML (Attention Profiling mark-up language) has sprung up to try to standardise the way that “Attention Information” is formatted. This is certainly a good thing as I feel that the ‘Attention Economy’ really shines when attention information can be shared between services in a meaningful way. However, agreeing on standards and persuading companies to adopt them is notoriously difficult. It will be interesting to see how the effort develops in the coming months, and you can be assured that GizBuzz will be there to cover it!