Today we added four new ways to navigate the content on fav.or.it this is as part of our ongoing updates to the service to make the content as accessible to everyone as possible. The different ways are Popular Stories, Tags, Feeds and full-text search.
Tag Page
The first new way to navigate content is via the ‘tag’ page, this takes an individual tag/keyword and produces a homepage of lots of interesting related content. This includes.
- What is popular – The top stories that make mention of this tag.
- Feeds that mention – If you want to find a feed that talks about a topic of interest – this is where to look.
- Related – A list of the tags that are related to the one you supplied.
- Comments – Lists the most recent comments that are about the tag/keyword (.e.g you can find all the comments most recently made about Barrack Obama)
The tag page is reach whenever you click on any tag (mostly from our trends sections) – Here are a few pages we like the most.
Feed Page
The second new way to navigate is the feed page. If you want to keep track of a particular feed and see what is most popular, what comments are being made on which topics then this is a quick way to see it all.
Here is a quick run down of what you get on the page.
- Most popular – A list of what posts have had the most attention/votes within the feed.
- Attention Data – Whenever someone views any topics on fav.or.it we track how much time they spend looking at the article, the graph on the feed page shows the overall attention the feed is getting.
- Latest comments – Quick view of who and what people are commenting on within the feed (newest at the top)
- Trends – It selects the common keywords that the feed talks about and shows the trends of those keywords.
A few feeds which nicely demonstrate what this is all about.
Popular
We wanted to create a focused page for each category that that just covered what is currently popular, this now clearly shows number of votes each story has got. You can quickly navigate to the story, read it and then decide to vote or not. Even if you do not vote we will still have tracked the time you spent reading (or not). That attention is what we base our ranking on (including some magic formula, cough).
Better Search
We have implemented ‘Sphinx’ the Open Source full-text search engine which immediately has brought the benefit of much better general search. This will be shortly extended out into searching within comments as well. We have implemented a range of ways to order the content relevance, popularity and other combinations, this is early days but we will continue to tweak and take feedback.
A big shout out to the developers on this, Sphinx has been easily to implement and really is amazingly fast! It will be able to cope even as we scale up the number of feeds (we will be growing our index of feeds by an factor of 10 in the coming months)
Contextual Tags
As part of the new tag focused page we made fantastic improvements around the auto-tagging of content, we are using a number of services (including Open Calais, thanks guys!) – and we mash up the results by picking a strong combination which is giving us some very strong connectivity between posts. Included in the Open Calais data we receive is a lot of information about the ‘type’ of the tags, e.g. it knows that ‘Barrack Obama’ is a person and that ‘London’ is a place, we will be making much more use of this information in the future.
Categories
We have also added a new category sitemap – this just shows all our current categories, please get in touch if you think there some we are missing.
As always please get back to us with any feedback either as comments on here, on our page on getsatisfaction or send me a tweet on twitter.
Leave a Reply