ant0ine's blog

Playing with D3 force #holidayshack

My aggregator takes many things into account to rank the stories, and of course among them, the inbound links from other posts and tweets. So far I never took the time to write a visualisation tool that actually shows this conversation. But today I did, it made me learn more about D3 and SVG, and I think the result is interesting.

Here are a few examples:

 Windows Phone is Superior; Why Hasn’t it Taken Off?

 

Apple to unveil two versions of next-generation iPad in January, sources claim

 

 

End of an Era: The Golden Age of Tech Blogging is Over

 

 

 

 

StoryFindr

 

Here is my new hack, well, more than a hack, this is a personal tool that I want to share with you.

Since I heard about RSS, long long time ago, I always had a personal homemade aggregator. It used to be very simple, a Perl script iterating over a list of feeds. Over the years, it gets more sophisticated, becoming something like a news discovery system.

I did a minimal frontend work last week end, and today, I’m happy to put the result page public at storyfindr.com.

This is what I read at breakfast, I hope you'll find this useful too.


PS: It’s configured to be only technology centric for the moment.
PS2: I tried to make it not too ugly on Safari Mobile. Not sure it works on other devices. Feedback welcome.

 

Storm over SF

Devel::NYTProf is THE perl profiler

Google has to realize that Devel::NYTProf is THE perl profiler. The best, I think, is just to quote Ranguard:

"So, I'd like to talk about Perl profiling, or even profiling Perl. Devel::NYTProf doesn't make the claim that it is THE Perl profiler, but it really does seem to be the best profiler for Perl that is currently available."

OK, I'm a fan. What a voice !

Maybe I'm the last to discover Florence and the Machine (It looks like they have been at the top of the UK charts a few months ago) but in case you don't know, go listen now ! The whole album is just great.

My big news to start 2010, I'm moving to San Francisco

After five years working at the Paris office, I'm moving to San Francisco. I'll still work at Six Apart, still on TypePad, but not remotely anymore. I had the chance to fly to San Francisco a number of times before. All stays cumulated, I think I already spent six months in the city, I'm not landing to an unknown territory, but still, I think it'll be a great adventure. I can't wait to be on the plane because "closing" my Parisian life is a lot of paperwork and is not very exciting. In the real world, you can't zip all your life and upload it on the other side of the planet in a few seconds, too bad. Anyway, I'm very excited by this new experience, and I'm looking forward to see my san franciscan friends and coworkers again ! 

I'm not sure about the move date yet, Visa process in progress, stay tuned.

Playing With Tokyo Cabinet

I'm still playing with RDF, and while I'm at it, I'm now experimenting with Tokyo Cabinet and Tokyo Tyrant.

I rewrote my experimental triplestore using the Mikio Hirabayashi's software. Right now, I've got the Subjects, the Predicats, and the Literals stored in a Tokyo Cabinet Hash, the Triplets in a Tokyo Cabinet Table, and my abstract Node object (which is basically a set of Triplet) in Memcached.

I tried Tokyo Cabinet mostly because of the benchmarks, and indeed, I confirm, it's fast !

But also because I don't need all the features of a traditional RDBMS, and it really feels good to get rid of this complexity.

Nice Sunday afternoon hacking.

Update your feeds !

I just moved my blog from MT to TypePad. The feeds have moved too, you should subscribe again if you want to follow me:

• This blog feed

• The photoblog feed

The design is very simple for now, I'll play with the custom CSS feature in the next few days.

Playing with RDF

Rainy week-end in Paris, the sun waits until Sunday afternoon to show up, anyway I had good time hacking with perl, memcache and mysql.

I started to play with RDF, it came up like this.

The web apps in general are slowly moving from the traditional relational DB model, to something more flexible, schema less, and to something more scalable, distributed and heavy cached.

There are a lot of different components already existing, memcached, couchDB, Hadoop, ...

This is exciting, lot of innovations.

Back to RDF, I think that one way to go can be to store the objects in a triplestore. Kind of Semantic Web Object Mapping I you want. Every object has already a URI inside the app, plus, you get the notion of graph for free.

I started hacking something about this on my free time, at least as a good way to learn more about these technologies. If I manage to have it working, I'll probably share it on CPAN or github.

I loved LeWeb08

It's a quiet Sunday afternoon, four days after the conference, and I find myself thinking about what Chris Anderson said comparing the internet and the human brain, about the Open Stack and its ability to compete with closed environments, about the opposition between the Marc Simoncini's "rules" and the regular web2.0 business, ....

Every year these two days are the opportunity for me to take some distance with the everyday life, get inspired by great people and think about the future.

Congratulations Loïc and Geraldine !