We’ve learned quite a bit in our first round of beta, and we’re happy to roll out a big set of improvements in these three areas:
1) Improving visibility
2) Improving clicks with machine learning
3) Growing your Network
Read More
Liveflows uses anonymous cookies to always show readers posts they havn’t read yet.
But when you visit your own blog (and you are logged in as the administrator of that Liveflows widget) we automatically turn off the “New to Me” functionality, so you can see some of the general stats.
You’ll know you are an administrator if you see the Edit controls: Read More
Users love using aggregators such as digg, reddit, techmeme, friendfeed, stumbleupon and the like to find great content. And in general, they work pretty well.
A typical experience of aggregator users is simply to go to the aggregator, click though to a recommended site, read that one page, then go back to the aggregator to find more content. Rinse and repeat.
But this behavior is considered to be “bad” traffic because visitors didn’t make any additional clicks on the destination blog. Visitors come, and they may even really like the content , but then they often leave without making another click. And when they don’t make any additional clicks, the blog’s bounce rate increases and blog owners talk about getting bad traffic from that domain.
Perhaps bad traffic could often be better characterized as lazy traffic. Read More
In a few days, we’ll start the beta testing of our new Liveflows service. [update: Learn more and sign up now!
Just follow our free sign up, and simply put our javascript snippet on your blog. Customize it with a graphic and color to match your site’s branding. When your page loads, anonymous cookies are sent to our distributed database. We process and analyze that data in real time to show each user what other people are reading now, and what they should click next.
Read More
Liveflows is a free tool we’re developing for bloggers and their readers.
It shows each person the most enticing posts for them: New readers see your top content, search visitors see the best pages to view next, and dedicated fans see the posts they haven’t read yet.
Liveflows isn’t a spider or a robot, but instead processes user attention and activity in real time to learn which paths users find most interesting, and what people are enjoying right now. Anonymous cookies allow Liveflows to personalize the results to show each visitor your best posts they haven’t read yet!
live: actually being performed at time of viewing;
alive; possessing life;abounding with life and energy
flows: to move or progress freely; continuous progression; to proceed smoothly and readily; to have a smooth continuity
This was something I wrote a few months ago, but it’s quite appropriate for this blog, now.
Catching up on an old Economist (yes, I still love magazines and happily subscribe to many) I came across a great article on counting:
How do humans develop our sense of numbers? What faculties do babies start with, and how does it develop? I learned that Babylonians used base 60 math, and that in the 1940’s a German animal behaviorist named Otto Köhler trained ravens to open boxes with the same number of dots on the lid as a card held by the scientist. Animals dig numbers too.
But what caught my attention was this beautiful nugget:
Babies are born with many ways of making sense of what they see and hear. When shown the same things repeatedly, babies eyes wander; when the scene changes, their gaze return.
I already knew this is how babies behave.
But this time it finally hit me: NEW. It’s all about NEW. Read More
This is excerpted from our about us page where we we explain a little about what we did before Liveflows and where we’re we want to go.
After Juniper Networks bought our last company [network switching / loadbalacing] in 2005, we turned our attention to that database layer. What would a really big database look like? Could you scale a db to web-size? If so, what could you do with it?
Personalization was always interesting to us, but what would web-scale personalization look like? What exactly would it do? How would it be deployed?
We read an NYT/ IHT article which talked about online news headlines, and how much editorial oversight was required to keep big stories around long enough to catch the infrequent users, but not so long that the site became boring to the super fans.
“Excuse me, but why can’t everyone just get the news they haven’t read yet? You know, the NEW stuff?
It was obvious this problem was begging for personalization, but wasn’t obvious to us how to tackle it. Read More