Building Feedly

(Thank you Sean. It was great to watch you fly your helicopters!)

A neuroanatomist has a stroke and so eloquently expresses her revelations on the energy and matter of being human.

Today Youtube announces a new set of APIs to allow third parties to save, tag and search for videos. This is going to dramatically lower the barrier for the creation of video-enabled applications. Here are a couple of interesting links:

• http://code.google.com/apis/youtube/overview.html
• http://code.google.com/apis/youtube/chromeless_example_1.html

It is going to be interesting to see how Seesmic reacts to that because this announcement slices them right in the middle: they have to determine if they continue to invest in their server side technology or cut their losses early and focus on re-building Seesmic on top of YouTube. Not a trivial decision!

Here is a post from Marc Andreessen on a private discussion he had with Barack Obama. An interesting way to try to see through the distortion field created by the media and the mechanics of the campaign.

Here is the TED talk introducing the Microsoft WorldWide Telescope project. This is totally amazing…It is almost like having a new sense. Microsoft just needs a few projects like this to make people feel differently about their brand.

My friend Ludovic, co-founder and CEO of xwiki, is participating to a contest organized by the french Radio BFM. If you speak french, you can listen to the interview here. You can vote for him here. Ludovic has done an amazing job bootstrapping xwiki from scratch to 22 people. He is an open source lover.

So if you have a couple of minutes, please do cast your vote!

Thank you!!!

Here is a 47-minute video on What is Money.

Update: some people have mentioned that parts of this video are not factually correct so make sure that you do your own homework. The question of what is money and what is modern finance remain interesting.

(Thank you Michal)

Here is a 2-part article (part 1, part 2) from Chris Wetherell on the birth of Google Reader. Nice story!

(Thank you Loren)

one word: beautiful! - a great way to start the day.

Going Down Droste, originally uploaded by Pisco Bandito. (Thank you Thomas).

Here is an interesting post from Dave Winer on Google and Open Social. He concludes with a broad claim: “Google is good at doing AJAXy web UIs, they make users happy, but they’re not very good at APIs. This one is a loser, for sure.”

Dave is plain wrong on that: As part of streets/feeddo, we have been using Google APIs (gdata, search, auth and undocumented google reader API for about 11 months now) - and people who played with the preview of feeddo know that we go pretty deep in terms of integration.

The result: an integration that works without us having to ever meet or talk with anyone at Google. Never! Not a single time!

The secret sauce: under the thin Javascript layer developers see, Google has a set of very clean RESTful APIs based on JSON and ATOM. The very same API the Google teams use internally to build their AJAXy web UIs (no lipstick on the pig here).

@Dave: All APIs evolve over time but this is as good as it gets. If Open Social fails, it won’t be due to how the API is exposed but the lack of community and the issues around security, authentication and access control.

Update Dec 14th, 2007: Yesterday, we had our first integration hickup: Google changed one of the API from XML to JSON. The total time to fix the problem was less than a day (about 3 hours).

There has been a lot of noise around the Facebook platform in the last 6 hours. We had the opportunity to kick the facebook tires as part of the feeddo+facebook integration. I would like to address three common misrepresentation people might have regarding the facebook platform (especially given that in this case, it comes from Om, one of the bloggers I respect the most).

Mis-representation #1: Facebook is closed.

Facebook is not closed. They expose a rich set of API allowing thirds party applications to query, read and write the social graph. Any kind of third party application! web/Javascript application, desktop applications, server-based applications. I think that what people mean is that the API exposed by facebook are specific to facebook and not inter-operable with other back-ends (ie this is a problem for the developers building social applications not a problem for the end user).

Mis-representation #2: If you are building for facebook, you are stuck with FBML (as if it was a bad thing)

This is actually an area where facebook is very advanced. The reason why they require developers to use FBML and push for integration with the profile page, it that is allows for a more a more secure and faster user experience. More secure because it allows scripts coming from multiple origins to run in concert on a single page (which we all know is a hard and important problem). Faster because a) all the content is cached and b) the widgets are forced to play nice during load time (one widget can not start using resources before the user starts interacting with it). Those are all very nice and advanced feature. To my knowledge, the only social network which has this level of sophistication is Ning.

Said that competition is always good. It is going to force facebook to stay on their toes, continue to innovate by added new services and more importantly continue to remain open!

(Thank you Sandor!)

Here is an interesting video of Cory Doctorow on Boing Boing on things you can do as a blogger to improve it.

Overall theme:  “It is the attention economy, stupid!”
Lesson #1: make the title of the post as expressive/search friendly as possible.
Lesson #2: allow users to determine if they want to read of skip the article based on the first sentence.
Lesson #3: design for user convenience  not for page views.

(Thank you to Thomas Crampton for a great interview)

You can reach the top and keep your sense of humor!

(thank you Kireet)

I am writing this post to increase the awareness of a cause that is very dear to my heart: the C.A.R. children’s development program.

What is C.A.R?

C.A.R. is an organization which supports and promotes the achievements of people with developmental and other disabilities so that they can continue to be valued members of our communities. What they really are is an amazing group of people who help day-in and day-out small kids with development delays and disabilities. They are amazingly kind, patient and generous, to both the kids and their parents. They make a real and lasting difference in the life of the people they interact with, in ways which are difficult to describe using words.

What is this aquathon/fund raising about?

It is about helping them continue to do what they already do so well. It is a way to re-affirm to them that what they do is important and valuable. It is a way to participate in their cause. It is a way to say thank you. It is a way to help kids with disabilities.

How can you help?

Here is a link to Janel Astor’s Team, Janel is the director of the children’s development services. You will find there a link to a page where you can make a donation - a single page process.

Thank you for helping if you can afford it!

Great video put together by the Google Mail team to advertise their service. Something different and fun!

Note: The geek in me likes how this video is an assembly of message-driven parts/clips.

Robert Scoble interviews Mehran Sahami, head of Stanford University’s undergraduate computer science program. 45 minutes is a little long but they cover a lot of interesting points.

(Thank you Cyril!)

Disclaimer, I am a huge Apple fan and a proud owner of a MacPro, MacBookPro, AppleTV and an iPhone. But I have come to realize in the last month or so that there is something fundamentally broken with Apple: I do not think that they have extended their DNA to include the creation, deployment and management of large scale internet applications.

Why?

• I recently purchased a .mac subscription to try the iPhoto integration. I was really disappointed by how weak the homepage authoring tool and the email applications are (both in terms of user experience and performance). If you compare it Flick’r, WordPress or Typepad, Apple is probably 3 or 4 years behind.
• Apple does not have much experience in deploying and managing large website. The iTunes store which is probably their biggest infrastructure project is focusing on fairly static content and runs on a handful of racks (compare that with Google who is pushing the limits of storage to hexabytes through 10,000s of interconnected computers).
• The best application on the Apple TV is YouTube. The content on it is not great yet but it is the only one which does not require the user to go through iTunes and its local desktop. The iTunes model of replicating locally and sync’ing around is so broken: why can’t - like YouTube - my music and data be on the cloud and accessible directly by my devices? Is it because Apple is more comfortable building desktop applications than internet services?
• The best applications on the iPhone are YouTube and Google Maps (see previous point).
• The choice of promoting Safari over Firefox re-enforces the fact that Apple see the browser as just another application (and not an extensible platform).

It would be great if Apple could extend its DNA/culture from hardware design and client software to include large scale internet applications because that is where I think the next wave of innovation will come from.

What could they do?

1. Stop thinking that the desktop (and iTunes) is the center of the universe.
2. Hire a few of the infrastructure guys from Google and make them kings (the same way some of the Apple design people are considered kings).
3. Acquire 37 signals and a few of the other innovative startups and get them to fix .mac.
4. Replace Safari with Firefox - and invest in it to turn it into a shield against Silverlight and AIR.
5. Do it quickly because Google and Microsoft are not standing still.

It is just so easy to give free advices! Have a great week end!

Here is a very impressive video on a new technique for resizing images called seam carving. The software can detect what are the least important areas in an image and use that model to let you fluidly resize a picture while minimizing the loss of information. You have to see it to believe it. The end result is that you can resize or manipulate part of the picture while making sure that it continue to looks good [note: The presentation is also a good example of the power of story telling].

For more information:

• PDF document (20MB)
• Website of Dr. Ariel Shamir

(Thank you Olivier)

Great video of a wild life battle at the Kruger Parc in South Africa. A nice way to escape “normal life” for a few minutes (and a rather happy ending…)