semanticpool

Mensch, ich sollte öfters mal den Heise-Newsticker lesen. Da war vor einigen Tagen ein Artikel über Soziale Netze. Es ging wieder einmal um den Missbrauch der Benutzerdaten — dieses Mal von Seiten der Geheimdienste.

Zudem gabs auch einen Verweis auf ein interessantes Paper was ich auch noch nicht kannte:

Semantic Analytics on Social Networks: Experiences in Addressing the Problem of Conflict of Interest Detection (PDF-Dokument)

Like TechWeb said, “Digital Dirt” can be a problem for people who apply for a new job:

[...] 35 percent said they dropped a job candidate because of information uncovered online.

Uuh, so watch out which information you leave in the internet and use for example my survival guide for social networking. ;)

From Wil Harris’ “Why Web 2.0 will end your privacy“:

[...]
So Murdoch knows everything about MySpace. The financial gurus at Yahoo know all about your personal thoughts, pictures and bookmarks. The guys at Google know everything about your search habits, and you can bet they want to link ‘em up to your email and calendar and whatever else you end up using online. How much is that data worth? With marketing spends online going ever upwards, as more and more of the world ‘logs on’, you can bet that it’s only going to get more and more valuable.

And where it’s valuable, it will be bought and sold. Our social networks, searching habits, visual identifiers and personal preferences will be mercilessly sold to anyone who wants to get their hands on our particular demographic.
[...]

My last posting dealt with privacy in social networks. In one of the comments Tristan Louis made the following statement:

The balance between privacy and openness is an interesting one. I’m not sure of how to address it… yet :)

I thought about it and my first idea was: Disguise your data.

Its not bad if social networks share their data. The bad thing is, if your profile-data from network A can by easily connected to your profile-data from network B. With “easily” i mean especially in an automated way.

So where are the possible junctures between profiles?

• First of all your Email-Address. Its an unique identifier.
• Your chosen username.
• Your name. Not unique but a possible connection.
• Your address.
• A picture of you. Modern image recognition algorithms are available and can answer the question if two pictures are equal or not.
• The contacts you invite.
• The URL of your homepage.
• …

So what can you do to avoid the connection of profiles?

1. One social network = one new email-address. Use different email-addresses in every network.
2. Use different usernames.
3. Yes, also use different names. Ok, you don’t need to rename yourself into ‘Peter’ if your name is ‘Daniel’. But make little changes in your first- and especially in your lastname (the firstname is often used if the social platform gets in contact with you. The lastname is not so important.)
4. Always use a fake address. No social platform needs your real address.
5. If you want to use a personal picture in the social network always use a new one for each network you take part in.
6. Do not always invite the same people in every network. Your personal circle of friends is also like a huge identifier.
7. Avoid to reveal the URL of your personal homepage.
8. Change your day of birth. Minor “modifications” are sufficient (e.g. change your day +/- 5)
9. …

If you follow these tips, every network you use will get a unique profile. So its nearly impossible to find a connection of profiles between different networks — in an automated way.
Ok, if a real human wants to get the connection he will get it. But you can complicate it. ;)

Why Scott Stevenson is not willing to switch his OS. There is so much truth in his words.

… – i hope so.
Just read about the helpful “see if the new Dashboard widget is the same that we have on our servers here at apple.com“-feature since the update on Mac OS X 10.4.7.

So do we have to worry?

Whow, i just tried to register at digg-service. It was the longest time a online-registration has taken in my whole life. :)

What has happened:

1. The first 10 usernames i have chosen were not available (took approx. 3 minutes)
2. My password contains special characters – they are not allowed (“only the characters a-z, A-Z and 0-9 are allowed in passwords“). (took another minute)
3. Then the hard part: The captcha. It costs me nearly 4 minutes to enter the right letters. I think i am not human. ;)
4. And some time elapses because my passwords didn’t match.

I think its time for a cool indentity-management on the internet. All the current registration-processes are really antiquated.

PS: Finally i got my digg-account. Hooray!

Who still doesn’t know Wil Shipley’s “Pimp My Code” postings in his blog should take a look at them. Very informative for Cocoa developers. Start with Part I.

Just found a future-scenario by Apple from the year <= 1988 (i dont know it exactly): the Knowledge Navigator (mov video). It shows really nice collaborative functions, a perfect voice recognition and a lot of other stuff which every computer-user is dreaming about. ;)

Ok, some things are available right now, but i think that it will once more take 20 years to create such a all-powerful system.

Tristan Louis lists some reasons why social networks may fail. One reason is that people pay attention to their privacy and are not willing to share all their private data. Tristan also says that such networks have to get more integrated with other applications and have to stop their existence as “walled gardens”.

But i mean, if you would follow these proposals and increase the integration and share more data with the outside world, the lack of privacy would become much more bigger. And then we would return to the first reason “privacy concerns”.

In my opinion the most of the social networks fail, because there are too much of them. There are some well-known which run excellent and most of their users doesn’t care about privacy. The rest of the networks – all the copyists – will die in the next 15 months. And between the players there will be a selection.
Its because the users don’t feel like maintain their private data in X networks. It consumes to much time and results in nothing. But we will see. ;)

Vorhin entdeckt, scheint es aber schon etwas länger zu geben. Ein Java-basierter Browser für RDF-Graphen: Welkin. RDF-Datei laden und *zack* erscheint der passende visualisierte Graph.