NoiseRiver: Filters are not the only next big thing!

Phew! While I was sleeping, the web was buzzing about that NoiseRiver thingy I made... I sincerly never expected such a reaction! People are talking about the filters feature and how it can help improve users experience on FriendFeed, and some are even predicting that FriendFeed will acquire NoiseRiver! This, folks, is the biggest flattery! Thank you Duncan :)

I still think that user's based filters and highlighters are the thing to add to social applications, because that overwhelming "noise" should be handled and consummed on the user's preferences, not only on the popularity parameter. What might be popular for Scoble (for eg.) may be something I really don't care of, and vice versa. Popularity is not the only facade of value.

So filters are nice and all, BUT, they're not the only thing... -I should learn how to tease more :) - In fact, the real next big thing would the "behavior learning" systems. Or in simpler English, what would be smart is: algorithms that can learn from your behavior on a website and then tweak and improve your filters set.

That shouldn't be that hard to implement on NoiseRiver, and they even can be simply added to FriendFeed. How? For example, when I "like" something on FriendFeed, there must be an algorithm that interprets that "liking" as an act by which I clearly define my preferences, thus, a simpler way to set my filters. The same thing applies on when I comment on or "Like" often other people's entries. That should help me define my proximity preferences too.

I, humbly, suggest that FriendFeed adds a "I hate" link too... because our feeling, we human, are alas not all about "love". I suggest them to add tags to the entries, at least for blogs feeds (which provide keywords and categories, almost by default), I suggest, I suggest... so many things! That's to tell you how I LOVE FriendFeed. The guys behind it are really smart people. And, as I always repeat there, the smallest thing in friendFeed is a real technical feat!

The missing NoiseRiver's "about" page, by Louis Gray

Generally, you'd agree, every website has an "about" page that describes it. It's important, and noone can deny this. People at "A list apart" even wrote something about it.

I confess, I'm a little bit lazy when it comes to writing. First, because English, as much of you know by now, is my 4th learned language and sometimes it's hard for me to express myself in it. Really. Second, I told to myself that FriendFeed is by itself self-explanatory and I should invest more time in fixing bugs and developing the thing more than talking about it... I know I'm wrong :)

Anyway, I'd like simply to say that Louis Gray has written a great blog post about NoiseRiver that really should be read as the "about page" of NoiseRiver. He explained, maybe better than I could have, all the details and features that NoiseRiver has to offer. So if you still wonder what NoiseRiver is about, I really recommand reading Louis Gray blog —which, by the way, is one of my favorite blogs: simple, detailed, true and not "hypy and buzzy just for the buzz of it" —I'll explain this someday, maybe :)

And here's NoiseRiver! Because your interests deserve attention!

I am not a blogger. At least I don't define myself as such. Not that I have anything against blogging of course, but I'm more a developper, so sorry all for the long time without any post.

I've been among the people over friendfeed to request filters based on interests, because even if some people like the noise —all the noise— I myself don't. So I put my "money" ( fbrunel one's actually, whom I thank so much!) where my mouth is and made NoiseRiver.

NoiseRiver is a web application based on the friendfeed's API that aims to extend friendfeed with some notions like: interests and neigborhood. You still have all the flow that is in friendfeed but the flow is, from now, on colored. Green meaning that you'll probably like the entry and red meaning that you'll probably hate it. Yes it's almost like feedego :)

The first minutes of the launch were some of the most stressfull minutes I lived in my coder's life. I've had to fix bugs and issues so quickly with everyone complaining about "404 Not found" errors. Believe me that was... hard!

Finally, I can say that NoiseRiver is usable right now, and you can already give it a try with your friendfeed's nick and remote key.

Of course this is just the begining and NoiseRiver will, I hope, be allways a wok in progress. So your feedback, your suggestions, errors reporting, and features queries are all welcome ! Join us in the NoiseRiver's room! :)

Can I hire Robert Scoble for 1 dollar?

So I sent the script to her with a 20 dollar bill... Inside I wrote, 'I heard you're getting 20 a picture.' Then she said she'd do it.''
George Clooney on how he got Julia Roberts on ''Ocean's Eleven"

Like many of you readers may see, I'm addicted to that big thing called FriendFeed. I just can't tell you how much it's wonderful! You have a never ending flow of content, people discussing it, ideas exchange, no notion of karma... It simply and really rocks, and it's freaking awesome and fast! One of their best feature is the "Rooms Feature". You create a room for a topic and people exchange articles related to that topic, discuss them, and exchange opinions. Yeah, I hear you saying: "isn't that what forums are for?" Yes, you're right, but it's just nicer on friendfeed. Come and see by youreself :)

As a fervant believer in data portability, I created a room for this, and well... it's the begining, people over the data portability group on google are a little bit slow to join, but I hope that someday this room will be a great place, where we'll be able to exchange knowledge on this topic.

So, In a precedent article, I've told you about how popularity is perversing truth, and I still somehow belive it, even if Scoble came and told us about how he's "ranking" people on their "importance". Robert wasn't completly wrong... Yes, some people can handle things, evangelize ideas and new concept and make them widely used.

So here's my question: Robert, I'm not rich, I have only one dollar, can I hire you to promote data portability? :)

Feature: Import from del.icio.us (or when Folksonomy meets attention profiling)

We just added the ability to import your profile in feedego (the keywords and their values for you) from your activity on del.icio.us. This is an effort to make the creation and the use of your preferences portable and easy!

What's in fact nice about this, is how just folksonomy meets attention profiling to make a whole one puzzle in which pieces just stick together!

We're trying to make importing from other services available too, so be tuned, we're just starting the game! :)

Dear FriendFeed you're missing THE thing!

Today, the best ever social network I know, and in which is more addictive than I originally thought added a new feature called "personalized recommendations" I first was happy to hear this, but just after reading their blog entry and testing out the feature I was a little bit despointed.

Why? Let me explain this in a the clearest way I can. There's NO algorithm, and there will NEVER be any algorithm that can know what ME I like or what I hate without explicitely asking ME. Simple and very clear. diving in my recent interraction on their website, or worse, just looking at my friends activity is not "best" it's something other, somthing that I'll call "popular".

Believe me guys, popularity easily gets perversed and corrupted, by the human nature first and by the tools. Having Scoble promoting Zobzee is an example. That's why a per user profiling is the best strategy a program should adopt. Because if you respect me -the user- enough you should stop the babysitting and start asking me :)

Heck, with the users base, the means, the smart folks working on FriendFeed, why don't you guys adopt things like data portability, and apml especially?

What are your favorite feeds?

We're always looking to improve Feedego and give our users base -that is growing each day more- a better service. So many nice features are under developement, and you will be surprised about how much already easy things get easier!

In the meanwhile, we would like to have your opinion on the content we serve on feedego. We selected all the feeds you're seeing, but we can never be perfect on the various topics. So let me ask you to help us in giving you the best things to read. 

What are your favorite feeds?

We need your feedback to serve you better! Thanks for giving us a minute of your time :)

What's wrong with Socialmedia?

I'm getting all the time flamed, accused of trolling when I talk about Socialmedia. Maybe it's my mediterranean mood, or just me not expressing what I really mean in a clear language. So let me try to clear this.

When talking about Social networks, people often refer to others defined as "A-Listers", those that are famous enough to dict what's worth attention. IMHO it's not about A-Listers, it's about personal interests first. We people should be described by our interests (profiling data here, things like APML serve the purpose greatly). My interests, my profile, is my very thing! I can, and of course want to share it, but it's up to me to define this first. Then I connect to people, not by their status, but by interests and things that gather us.

Let me give you an example, and tell me if I'm wrong in thinking so. Back in the late 90's and the begining of 2000 we've seen many people blogrolling people like Zeldman, when they're actually not in web stuff, sometimes don't even know what are standards, what is CSS or even HTML, moreover they simply don't care of this techy stuff.

This is what I call "corruption" or "perversion", in that sense that the inner-person gets so corrupted by the environnement that it forgets what most imports to it. And this implied some strange behaviors in people's interractions that we've all seen. So, should we blame it on the people or on the tools?

I think that tools are guilty too.

I have a dream —yes, me too :) that we people will be seen by algorithms and programs as clouds of interests, first and then nicknames, names, pictures and all that stuff. I really believe that online, I am, what I like and my friends are those I physically know or those who share the same interests with me :)

P.S. Some of you will easily understand that I like this Coltrane guy

Feedego: How-to Screencast

Many people are wondering about how to use Feedego. It may seem obvious for me, since I created it, but I guess that it's not the case for everyone. So, ok folks, I created my very first screencast and it's about Feedego!

I hope you don't mind my poor English, and accent :)

Enjoy!

Comments and emails

It seems that I'm having problems with both Disqus comments and emails at feedego.com the last one being because google's checking MX records and the former because... well I don't know yet... I'm invetigating it.

So stay tuned and meanwhile, you can always reach me on FriendFeed directly or by joining the Feedego's room there :)

Edit: Comments are back! Sorry, it was all my fault, not disqus' one.

Is socialweb the end of the web?

Feedego didnt do it for me! looked pretty poor and didnt seem to really have any good features The third eye

I always wondered if the community is the end-point of the web. Doesn't the individual-web have any chance in this big brouhaha? I mean what if someone doesn't agree with what the crowd is shouting? Should we ban it from the village?

The real spirit of feedego is in fact this: YOU deserve to see/read what YOU think is good for you, be it "popular" or not. That's the essence of APML in fact. And Feedego is trying to reduce the ratio/information noise, that's why it's not a social thing. No karma, no popularity, no digg this, burry that. 

Am I wrong? 

Disqus and Friendfeed for Feedego

A little post to let you know that comments on this blog are from now on managed with Disqus and that I started a FriendFeed room for your suggestions, feature requestions, and critics :) 

You're all welcome, I'll be glad to be your friend on friendfeed too! I'm directeur there!