Comments on: Personas Online Data Portraits

By: pete

pete — Thu, 10 Jun 2010 00:10:34 +0000

nice one but sadly i have a sportsperson with the same name as me skewing my persona results

By: GypsyMyKitten

GypsyMyKitten — Mon, 24 Aug 2009 07:20:38 +0000

Erm, every time I enter the same name, I always got the same exact results. I did it for each of my immediate family members in which we have a really rare last name. For me, my sister, and my mother, “no digital traces found”, yet for my father, it was weird with mixed results of the truth and some crap than doesn’t apply to him at all.
As for my fiance’s name several times whom has a common last name? Only two different results – one shows only “news” and nothing else, and the other says “education” and “news”. I don’t believe he’s known for anything news-worthy, but he too well educated (though no common sense, lol).
Sorry to say, this “art piece” is useless to me. But, hey, at least it caught my attention for about 20 mins.

By: keysa

keysa — Sun, 23 Aug 2009 16:34:35 +0000

People… get a clue…. THIS SITE IS MEANT AS A JOKE!!!!!!

By: Ken

Ken — Fri, 21 Aug 2009 04:03:34 +0000

Thanks Aaron. I do dig that video. thanks.

By: Aaron Zinman

Aaron Zinman — Thu, 20 Aug 2009 22:13:15 +0000

@ken: Take a look at the larger context.

http://techtv.mit.edu/genres/25-humanities-arts-and-social-sciences/videos/3315-metropathologies

By: Ken

Ken — Thu, 20 Aug 2009 22:09:53 +0000

Sheesh, Mr. Zinman. “Questioning the modern world overflowing with information and it’s subsequent consequences” Well at least we know bullshit isn’t owned solely by the ad community :)

By: Aaron Zinman

Aaron Zinman — Thu, 20 Aug 2009 17:46:38 +0000

Hi internet goers, creator of Personas here.

You guys should REALLY read the text on personas.media.mit.edu, I didn’t put it there by accident.

You see, this is an art piece critiquing data mining. It’s not suppose to be “useful”. Inaccuracies are a part of the intended design, and non-repeatability is part Yahoo returning different results, part me taking a sub-set of a larger set of results.

@SwissRant glad you think that my work is worthless. Before you make such claims, you should know that this is not my thesis work, but a side art project I did as a part of a larger installation questioning the modern world overflowing with information and its subsequent consequences. It has much more conceptual depth than you give it credit for, and the natural language process is using nearly-state-of-the-art algorithms. I didn’t do it simply to ‘look pretty’.

By: cempson

cempson — Thu, 20 Aug 2009 13:43:36 +0000

would have been a lot cooler if it didn’t give drastically different results every single time. still, it is an interesting idea.

By: David B

David B — Thu, 20 Aug 2009 12:51:50 +0000

This looks similar to http://www.spezify.com/ , but not quite as useful.

By: SwissRant

SwissRant — Thu, 20 Aug 2009 06:47:41 +0000

It is kind of depressing what some MIT students spend their time on. The design is nice enough, and the pretty boxes all very well, but the relevance is very weak.

The fact that it didn’t want to work in IE8 regardless of Flash Player 10 being installed is hopeless.
I can see why designers would like this, but from a functionality and usability standpoint, it is rubbish.

By: Mmmatze

Mmmatze — Thu, 20 Aug 2009 05:57:56 +0000

Yeah, if you have somewhat of an Allerweltsname, nothing good will come up. Sadly I have one. But I like the idea, though.

By: bryan

bryan — Thu, 20 Aug 2009 04:45:25 +0000

I apparently am into sports… I guess out I’ll check out a baseball match tonight…

By: Ken

Ken — Wed, 19 Aug 2009 21:47:58 +0000

This looks great, design-wise. But is style only. There is no substance. The final boxes have nothing in common with anything that might have been searched for. Kind of like electronic palm reading.

By: nerdski

nerdski — Wed, 19 Aug 2009 19:03:14 +0000

Fun stuff but I was wondering why each time I did it the outcome was different.

By: Amanda

Amanda — Wed, 19 Aug 2009 16:16:52 +0000

Ha! I did this, but sadly, I share my name with a b*tchtastic soap opera character, so nothing good came up!

By: David Airey

David Airey — Wed, 19 Aug 2009 16:12:12 +0000

I tried it too, and pulled some big chunks for “illegal”, “aggression”, “medical”, and “religious”.

Thankfully, my girlfriend need not fear, for there are also chunks for “domestic”, “family”, “fashion”, and “fame”.

By: websnap

websnap — Wed, 19 Aug 2009 15:53:21 +0000

Well, there is no way to narrow down if you have a common name. Location, or a way to distinguish you from other’s who share your name would help I guess… Luckily my name is quite rare so it was a bit more accurate…

By: wl

wl — Wed, 19 Aug 2009 15:14:41 +0000

it’s amazing and wonderfully designed though yep I still don’t quite understand why the identity varies everytime.

By: jérémy

jérémy — Wed, 19 Aug 2009 14:47:24 +0000

hmm.. i’m a bit surprised that it is absolutely not deterministic.. have you tried it several times?
or maybe it’s because the corpus of information available is not big enough for me (compared to you..)

plus, it won’t let me type the accents of my first name. ;)
and what’s that big ‘aggression’ chunk?

but an interesting project nevertheless, thanks.

By: devolute

devolute — Wed, 19 Aug 2009 14:37:41 +0000

This is excellent. My largest ‘chunk’ is “online”. Guess I need to get out more, although the 2nd largest part is “illegal” so maybe not!

By: Kimberley

Kimberley — Wed, 19 Aug 2009 14:24:24 +0000

I have no online digital data apparently. Nice idea though.