Is a Digg killer? is tiny start-up based in a small office an hour’s drive from London. But this “feature rich community-based feed reading system” is about to unleash a wholly original take on reading blogs and news feeds which could see it face-down even the social bookmarking giants like Digg and the newer kids like CoComment.

Favorit brings together blog reading and replying into one simple web application. Its innovative web interface is designed to allow users to let users read any kind of RSS feed, cut-up, mashed-up with other feeds or “sliced” in any kind of way.


It’s also is a classic Web 2.0 startup which will attempt to solve one of the web’s most frustrating issues, i.e. the separation of reading RSS feeds from being able to comment on the post. Admittedly any blog post is only a click away from a user being able to comment on it. But imagine being to comment, Twitter-like, under a feed and not even have to care about filling in your name, email, etc. Just comment, save and carry on reading. Having witnessed it myself at an exclusive demo, I can confirm that this is what Favorit is capable of.

Favorit will this week launch a private beta based on a submitted database of 10,000 blog feeds. (The site is exhibiting in London at FOWA this week and at mashup demo.

What’s the Backstory? Favorit is owned by Assembleron Limited, a one-man startup based in in the grounds of Reading University, UK. Founder Nick Halstead (and managing director) came from the games world, having worked at BSkyB interactive games, Empire and Argonaut prior to that. But over the last two years he has immersed himself in the world of blogs and Web 2.0 trends. For the last five months Halstead, working with a couple of freelancers, has been developing the Favorit core engine.

Although Favorit is being bootstrapped right now, Halstead is in conversation with a number of potential backers. To see why, read on.

Turning feeds into slices

Favorit approaches the issue of reading RSS feeds with the concept of ‘slices’. Each post in a feed is categorised and tagged. By choosing a category, tag or rank (or a combination of each) the user can filter what they are reading in a more efficient manner than the normal ‘hose’ effect of having to laboriously wade through hundreds of blog posts in hundreds of feeds. Favorit has dubbed every combination of tag, rank or category a ‘slice’.

Slices can be created from tags, categories, rank (0-5 stars), a person (all that persons feeds, or their posts, or a mashup of more than one person), one or more feeds and finally date (range, single dates, /when/yesterday).

If a post does not include any meta data for tagging Favorit automatically generates tags using its own home-grown system. It will also allow the community to edit tags based upon a simple voting system that would allow tags to be added or removed. When you read a feed or a slice on Favorit it will automatically generate a tag cloud based on that content. Pick one post and it lists the tags. Pick 5 feeds to read all in one slice and it adjusts the cloud again.

So for example, if you only want to read about the iPhone in one feed, you create an iPhone slice for that feed or tag. You can also combine several feeds, so “Scoble”, “MacWorld” and “Google News”, and create an intelligent ‘slice’ of all three which only picks out the word iPhone.

Now I know you are going to say you can do something vaguely similar with smart lists inside an RSS reader. But to do this requires a lot of copying and pasting of feeds and searches. On Favorit it’s just about clicking the feeds from which you want to create a slice. Anyone who uses Favorit will have an OPML file which they can also export.

Any slice can be shared and on top of that the slice can be instantly turned into a feed in its own right. Thus Favorit is going up against the business models of feed aggregators like FeedDigest.

Comment posting with an API

Currently, sites like Digg try to draw the conversation on blogs towards its own site. And they’ve done it very well. But surely blogs would be better off having those Digg comments appear on their own sites?

The problem right now is that not all the blogging platforms have open access to their commenting systems. Plus, RSS feeds for entries are generally separated from feeds for comments.

Based on PHP and the Zend Framework, Favorit will launch an API during the public beta enabling it to hook into many more blogging platforms to allow it to send comments back to the sites. Halstead hopes the API will create an ecosystem outside of Favorit.


Now of course there is a glaring issue here. Sites thrive on traffic. Take away the need for a user to go to a site and comment and you potentially rob the site of a great deal of traffic. What’s the answer?

Well, by removing barriers to commenting, Favorit potentially creates a faster turnaround of comments to blogs. Halstead, admits that aggregating comments may remove some traffic, but it could also build extra readership and feed subscription. The short answer is we won’t know until Favorit has been operating for a few weeks.

But it’s fair to say that Favorit will probably be of most benefit to ‘long tail’ bloggers currently sitting in the echo chamber.

Think about it. If your blog can be indexed and appear in a river of news alongside some of the biggest blogs, with commenting made a great deal easier, a smaller blog could easily be dragged further up the tail. On Favorit, every blog has a chance to be read by every user, and although users have the option to filter out lower ranking blogs, the system could encourage users to find new blogs outside of the top 1000. Unread feeds won’t be moved into the main database until people start adding them.

It must be noted that Favorit can only put comments on a blog with its permission. A blogger either creates an account on Favorit and supplies it with account details (via XMLRPC) or the blogger installs Favorit’s widget. The full list of supported blog platforms will appear in the next few weeks, although I can safely say that Blogger and any or hosted WordPress site will be included, though Typepad or LiveJournal just yet. They are aiming to support 80-90% of blogs out there.

Favorit will also pick up trackbacks and coments that appear on a post.

CoComment take note: Favorit also tracks any comments that you would happen to write directly onto other blogs, so long as it’s in the database.

What about comment spam?

Since the site won’t allow the posting of comments without a user being signed in, spammers will be severely restricted, unless they plan to hire hundreds of people to manually spam (which is unlikely given the economics). Favorit will also internally police the system and allow the community to report spamming from signed-up users.

But could Favorit pick up spam from external sites it trawls? Possibly, says Halstead. That is a possibility for which they will have to rely on their own anti-spam systems and the community to police.


It’s easier to figure put who Favorit is not competing against. In its sites are, for example, are elements of Blogger, Digg, Buzztracker, to name just three. Then there is Co.mments and Commentful.

And the recently launched Intense Debate is a souped-up blog commenting system that adds a lot of features for publishers and commenters alike, but unlike Favorit it requires a plug-in to be installed.

Affect on advertising?

This a glaring issue with Favorit. If everyone who currently reads commercial blogs (like TechCrunch) switches from the original web ages to the feeds forever, no-one would see the advertising. The answer of course is that advertising, to be effective, would need to switch to RSS feeds. This is possible since Favorit does not touch the content of the feed. But it is also controversial. Many publishers still only provide partial feeds, and many blogs hate putting ads in their feeds.

It’s a tricky topic. “I suspect partial feeds just won’t be read as much,” says Halstead. And therein lies the rub.

Tracking attention beats voting

Because Favorit uses Javascript it will gauge how long you read a post and what you did before during and after. This data is invaluable both for advertising targeting and for data mining, and its far more sutble than Digg’s voing system. Eventually the site hopes to rank as many as a million blogs in order of attention.

Because it will track what people actually read, Favorit will be a far more accurate reflection of what is popular online than Digg, which everyone knows is increasingly subject to gaming.

Although Halstead went to great lengths with me to emphasise that Favorit is a different animal than Digg, there is no getting away from the comparison. And it’s quite clear that capturing attention meta-data beats ‘voting’ hands down.

He does admit, however, that “We’ll have so much data it’s about working out what to do. We’ll be in the same category as as Facebook in terms of amount of data.”

A blogging platform as well?!

Favorit is not just going to be a feed reader. It is also a blogging platform. By creating a subdomain, such as ‘’ users will be able to write their own posts into the system. Using this, they can pull in their feed from their blog as well as post directly into Favorit. Any comments on the Favorit subdomain blog then appear back at the original blog.


Since it’s all widget based, users will be able to ‘pimp’ their Favorit blogs with a set of widgets – many form outside suppliers – which Favorit will build into the system. But you won’t be able to access the underlying HTML or CSS.

Here’s where the revenue comes in. Favorit plans to share advertising revenues with users who create these subdomain blogs.

Of course, it goes without saying that allowing users to create subdomains could open a can of worms. Favorit plans to police the creation of the sites. and will limit the number of sub-domain names one profile user can register. So to create a lot you’d have to create more than one profile by hand. But we’ll see how they manage that in due course. It could get interesting.

The ‘Long Tail of Techmemes’

And of course, this is worth money. If the Nike subdomain pulls in everything there is to know about Nike, Google could be among those knocking on the door.

And consider this. Favorit could also enable a ‘long-tail of Techmemes’. In theory anyone could create a sub-domain site which, say, tracks blog posts about football, or golf, or race horse breeding. You name it.

The future

Halstead is aiming to get 50,000 users inside the next 3 months. “We’re not aiming for Twitter growth” he tells me.

But with its geek-friendly layout and jargon, Favorit is not going to be viewed as a mass-mark consumer-friendly site any time, soon. But given that it will be indexing tens of thousands of blogs, being able to slice and dice what you read but rarely seeing the letters XML or RSS, it does have the potential to open up feed-reading to a wider audience than perhaps other aggregators have done so far.

At the end of the day, by uniting feeds with comments Favorit has an opportunity to put itself at the centre of the blogosphere. And that’s a good place to be. To re-work a well-worn phrase, if markets are conversations, then Favorit is money.

[Digg this story – ironically]