Tag Archives: twitter

Art in 140 Characters

Is it possible to encode and compress an image to such a degree that the raw data can fit in a single Twitter message (140 characters) that, when decoded again, is still recognisable? The answer to the questions is a resounding Yes, as confirmed by a coding challenge inspired by Mario Klingemann’s attempt to compress and encode the Mona Lisa down to 140 characters.

Klingemann’s attempt, dubbed the MonaTweeta II, is definitely an image recognisable as the Mona Lisa, but it must be said that some of the entries to the main coding challenge are truly breathtaking.

The winning tweet (with a character to spare):

咏璘驞凄脒鵚据蛥鸂拗朐朖辿韩瀦魷歪痫栘璯緍脲蕜抱揎頻蓼債鑡嗞靊寞柮嚛嚵籥聚隤慛絖銓馿渫櫰矍昀鰛掾撄粂敽牙稉擎蔍螎葙峬覧絀蹔抆惫冧笻哜搀澐芯譶辍澮垝黟偞媄童竽梀韠镰猳閺狌而羶喙伆杇婣唆鐤諽鷍鴞駫搶毤埙誖萜愿旖鞰萗勹鈱哳垬濅鬒秀瞛洆认気狋異闥籴珵仾氙熜謋繴茴晋髭杍嚖熥勳縿餅珝爸擸萿

via @spolsky

Identification through Anonymous Social Networking Data

Anonymity is “not sufficient for privacy when dealing with social networks” is the conclusion from a study that has successfully managed to de-anonymise large amounts of sanitised data from Twitter and Flickr.

The main lesson of this paper is that anonymity is not sufficient for privacy when dealing with social networks. […] Our experiments underestimate the extent of the privacy risks of anonymized social networks. The overlap between Twitter and Flickr membership at the time of our data collection was relatively small. […] As social networks grow larger and include a greater fraction of the population along with their relationships, the overlap increases. Therefore, we expect that our algorithm can achieve an even greater re-identification rate on larger networks.

There’s been some meritorious coverage of this study. This from BBC News:

The pair found that one third of those who are on both Flickr and Twitter can be identified from the completely anonymous Twitter graph. This is despite the fact that the overlap of members between the two services is thought to be about 15%.

This from Ars Technica:

It’s not just about Twitter, either. Twitter was a proof of concept, but the idea extends to any sort of social network: phone call records, healthcare records, academic sociological datasets, etc.

via Schneier

Text as UI (On Twitter)

Putting me in mind of Dustin Curtis’ multivariate ‘split’ testing to increase click-through rates to his Twitter profile (previously), Jakob Nielsen discusses his iterative design process for a Twitter message advertising his latest usability conference.

The message went from,

Announcing LAS VEGAS and BERLIN as the venues for our biggest usability conference of the year http://bit.ly/UsabilityWeek

to,

LAS VEGAS (October) and BERLIN (November): venues for our biggest usability conference ever http://bit.ly/UsabilityWeek

I am by no means a high-output Twitter user and I dislike ‘How to Twitter’ articles with a passion. Nielsen’s latest I quite like because he notes that, in the case of Twitter and other micro-blogging services, text is a form of UI in itself.

It’s a common mistake to think that only full-fledged graphical user interfaces count as interaction design and deserve usability attention. As our earlier research has shown, URLs and email both contribute strongly to the Internet user experience and thus require close attention to usability to enhance the profitability of a company’s Internet efforts.

The shorter it is, the more important it is to design text for usability.

Tests On Language and Click-Through Rates

By varying the language used in a sentence at the end of his articles, Dustin Curtis increased click-through rates to his Twitter profile by 173%.

Dustin describes his multivariate (‘split’) testing of different call to action sentences, revealing the most persuasive, in a visually excellent article.

This puts me in mind of how both Tim Ferriss and Ramit Sethi tested various titles for their products; The 4-Hour Workweek and I Will Teach You To Be Rich respectively.

While we’re on the subject;

  • You should subscribe to my RSS feed here.
  • You should follow me on twitter here.

via @zambonini