:::: MENU ::::


My secret to appreciate Facebook: NO TRUST !

The structure of part of a DNA double helixImage via Wikipedia

They are a lot of discussions these days around how Facebook handle it’s user’s data privacy.
I’ve talked about it in a previous post.

As far as I’m concerned, I’ve absolutely no issue with this.
Do you know why ?
Because I do not trust Facebook. I do not trust it at all !

As a consequence,  all the datas I put there can be public, I’ve no issue.
No secrets.
All my privacy settings are set to share everything, like this, I cannot be disappointed by Facebook if they change their mind.

I think everybody should do like me and this is why :

  1. Facebook is a very very young company that has grown very very fast.
    In these conditions, errors are always possible

  2. Facebook is inventing something totally new.
    Who could imagine this very big social network just 10 years ago ?
    Who can predict what it will be in 5 years ?
    The system is totally new, the rules need to be written, so for sure they will change.
  3. Mark Zuckerberg has never been very keen to commit on user privacy.
    In fact it’s the opposite.
    Facebook commit to provide its users a system to share more and more datas with more and more people, by default, datas are shared.
    To build a system that is really sure in term of security and privacy it has to be designed for this since the beginning, it has to be part of it’s DNA. This is not the case for Facebook.

Of course not trusting Facebook limit a lot the added value…
Of course I hope that something new will come in future with less functionalities, but with privacy commitment written in it’s DNA.
Maybe it’s Diaspora ?

Reblog this post [with Zemanta]


Facebook user’s data privacy management : What are they talking about ?

These days we see a lot of discussion around the way Facebook (does not) take care of it’s user’s data privacy.

Here’re the 2 best sources I’ve found that describe the “problem”.
In fact I don’t think there’s a problem, I’ll explain why in a coming post, but it’s interesting to look at these 2 articles to understand what people are talking about.

In this diagram The New York times give a view on the 50 settings with more than 170 options that are available to manage privacy on facebook.

In this series of images Matt McKeon shows the evolution of privacy default settings over the years (from 2005 to now).
It’s amazing to see how user’s data are more and more public by default everyday…
Reblog this post [with Zemanta]


Ipad is coming to France, but price too high !

I’ve already explained it few times on this blog, I want to buy an Ipad as soon as possible.

It’s now official, the Ipad will be available in France on the 28rh of may.
The only big issue is the price.

In fact, the price is a same in France and in US (for example, 499 for a 16Gb, WIFI version) but… it’s 499 USD in US and 499 Euro in France !
Which means 30 % more !

The other issue is the price of the 3G plans with phone companies.
Orange proposes 39Euros /month for a 2GB limited connexion. This is huge !
For my Iphone, I pay the same price for unlimited datas and 1 hour of voice.

I know IPad is selling very well in U.S. now, but honestly, I’m not sure I’ll buy one, because of the price…

Reblog this post [with Zemanta]


Robots on twitter are a big pain. An example with “Jamie” alias @HoldemTalkRadio

Spam is an issue on twitter, and I’m tired about it.
Since a long time third party application like twhirl or seesmic have implemented functions to report spam and help twitter to ban senders (basically, by sending a tweet to a special user @spam created by twitter team for this).
Other services like topify enhances twitter new follower email notification in order to help users to know who follow (based on bio, last tweets, number of followers, …) them and decide if they should follow back.

In a recent post, twitter have reported that % of spam tweets has drop from a 9% in aug’2009 to 2% in march 2010.

This is a big achievement and I hope it will go even lower in the future.
The issue I’m facing now on twitter is what I’d call “intelligent” spam.
Because people always want to have “more” than others, because people think that having more followers is a measurement of power, I see many of them doing all they can to get more followers.
I’ll not discussed here why it’s a non-sense to rate reputation based on number of followers (ratio between following and followers is a lot more important).
Yesterday, I received an “@lmau” notification.
First thinking is “Nice, someone is including me in his “follow Friday” recommendations !”
But who is @HoldemTalkRadio ? 
A look as his bio : 
@HoldemTalkRadio is a guy named “Jamie L. Simon”.
Jamie lives in Los Angeles, California.
Welcome Jamie, Nice to meet you.
More interesting, Jamie is listed in 1,174 lists !

According to the title of the lists, Jamie seems to be really power user : “top-im-rockstars”, “klout-influencers-coachforyou”, “awesome-engaging-people”, …
A quick look on Mrtweet also give me good feedback on Jamie : 39 people have recommended him (while 0 have for me …)

But something is strange… Jamie recommended me in his #ff, but we’re not connected. He doesn’t follow me, I don’t follow him…
Another strange thing is that Jamie follow 19k+ persons, which is more than the 17k+ that are following him…
Next strange thing is that Jamie has already send more than 100 000 tweets.
A quick look on twittercounter.com show me that he sends few hundreds tweets every day (when big users like @loic “just” send around 50 / day….).
When I look closer to yesterday tweets from Jamie, I see that he sends hundreds of ramdom #followFriday tweets.
Conclusions : 
  • Jamie is just a robot
  • Jamie is just generating noise
  • I’ll not follow Jamie.
  • People who have honored Jamie by adding him in their list… are certainly also generating noise. I’ll not follow them
  • people who have recommended Jamie in MrTweet are also generating noise. I’ll not follow them.

What should twitter do to improve that ? Here’re a few ideas :
  • automatically detect accounts that are sending more tweets than a human can, and automatically tag them as “robots”
  • propose special functions to filter, manage these “robots” users (in my case, I’ll just filter them, because I do not what to see their noise)
  • measure the “noise” in the total number of tweets generated by these robots.

Have a nice day Jamie !
Reblog this post [with Zemanta]