Pressflip is a new search engine which learns your preferences over time to deliver more relevant results. I signed up and tried it a few times. I can definitely see the potential, it worked especially well for niche queries like Humax. I wrote about the concept behind it when it was still Persai and I like that they have released it as a search engine. I said at the time that I don’t think competing with Google is impossible so long as you use a completely different method of indexing and searching the web. Attempting to beat Google at PageRank is clearly a waste of time.

Pressflip’s founders are well known for their antipathy to Web 2.0 generally and social features in particular. This appears to have arisen  from the number of terrible sites which lazily plaster on a social networking aspect to attract VC from me-too investors who missed out on facebook and want in on the action. I don’t think scepticism is necessarily a bad thing but you can take it too far. If anything will fell pressflip it will be this attitude.

My advice to them would be to ignore everything that happened on the web since 2003 and take a fresh look at the basic concept of building a network of users and the advantages it can bring. Pressflip relies on the user telling it what links are relevant, by doing this it builds up a picture of what I don’t want to see. Positive feedback is generated by clicking the links served – this was something I didn’t get initially so it may help to explain this somewhere to time pressed Internet users such as myself.

Networks matter

They need to apply the concept to other users. There may be someone else of a similar mind to me and I want to see their links served alongside mine. In the same way I could flip a user I didn’t like so Pressflip could learn who I approve of. Doing that would leverage all the work being done by the users flipping article and release them from the individual silos Pressflip currently traps people in. I originally wrote about the concept in relation to delicious:

The beauty of delicious is that it can look into the mind of other people with similar interests to me, see what they have tagged and return more relevant results. Another advantage it has over Google is that delicious can also assess the strength of a link between me and another user by seeing how many times we have tagged the same page with the same tag.

All Yahoo has to do is analyse these links between users in the same way Google analyses links between pages and it will have a search engine which thinks differently and gives people a real reason to use Yahoo search over Google.

Do this and you win. Don’t do it and, well we all know what Uncov’s favourite word was.

Driving around the UK I am assaulted by various billboards with assorted celebrities telling me how easy Sky+ is. They are essentially extolling the virtues of PVRs but the ad campaign implies the only service which supplies such functionality is Sky+ which, of course, requires a Sky subscription.

What amazes me is that Humax has not piggybacked on this by pointing out that you can have all the same features with no subscription by buying one of their boxes. They are equally easy to use. I bought one for my mum and she was up and running in no time. Sky is making money by making people pay a sub for something they can have for nothing. Astonishing.

Disclosure: I own a Humax box, so does my mum.

Roku has just released its Netflix Set Top Box. As you can tell from my previous posts I am an STB evangelist and I’m always glad to see new boxes out there. At the moment I am unmoved from the position that Humax will take this game but props to Roku for trying.

I am about to make my usual mistake of critiquing the thing without actually trying it but here goes. First the design. The box looks awful. Hasn’t anyone learnt anything from Apple’s success? When will consumer device manufacturers realise that design matters? The remote is ok, nice and simple, but it doesn’t look that great either. Seriously dudes, Pininfarina does a lot more than cars these days so just give them a call next time before sending another soulless brick our way.

Second the service. It’s called the Netflix Player and as far as I can tell that’s all it can do – stream films from Netflix. I’m not going to get as excited as Arrington over this for one simple reason: no tethered box will ever make it. The Roku doesn’t even have its own hard drive, it streams everything directly which means people will have to drag an ethernet cable to the living room to make it work. This is way too much effort for the consumer. There is a limited amount of space under people’s televisions and they want one box that does everything. Every service (Amazon, Apple, Netflix, Blockbuster, whoever) will have to be available to every box. Trying to tie everyhting down will simply delay the STB revolution and encourage piracy.

Although I have gone long on Humax I’d like a side bet on Myka, the most interesting STB currently out there. In BitTorrent they are harnessing the best underlying technology and they are pursuing an open platform. Now I’m not sure where they are with the content deals – that will be the hard part – but I think open boxes are definitely the way forward. This is why I’m so enthusiastic about Humax: one payment and you are away, no fees, no subscriptions. What people really want is freedom, not free.