20 years of signing up

I signed up for my first email account on Yahoo! just about 20 years ago. I didn’t know what to do with it. None of my friends had one, and the only person I knew with an email was a friend that signed up with me, and we saw each other almost everyday. Data limits back when this was a problem mean that my email archive starts 6 years later, in 2002. I had to delete all previous email to free up space. Looking back at that archive I see friends and acquaintances I’m no longer in touch with, comments and memories from a past self, as foreign to me now as if I was peeking into someone else’s email.

During these 20 years I’ve given up my data to hundreds of companies, websites and services. Along the years these companies have dissappeared, merged, crashed, or grown into multimillion dollar businesses. But what’s common to all of them is that once I signed up for an account, I lost whatever degree of control I had over my personal data. Most of them never offered a way to delete the account, and when they did, they were outright lying, as ownership of my data was a revenue stream and/or an asset they could use in a sale negotiation.

Services like Twitter, Google and Facebook offer the possibility to sign up on external services using their credentials. That is the closer we ever got to a personal data protocol we could use to easily manage the services we sign up for, and we paid a dear price by giving up even more of our privacy to the advertisement industry.

Privacy and the control over personal data is going to become a key civil and digital rights issue in the coming years. While I’m conscious of the danger of having centralized data providers that would become instant magnets for malicious hackers, I think it’s not inherently more dangerous than spreading our data over dozens of services who may dissapear, be sold, be attacked or transfer your data without your knowledge and consent. We need a new architecture that gives us back control over our own data.

Changing the metric of journalism

Slowly, led by the first group of digital natives like Upworthy and Medium, which measure traffic by time-on-site, and the Financial Times, which began selling ads based on attention last year, traditional journalists have been adjusting to what a change in Web metrics might mean.

via Can Tony Haile save journalism by changing the metric? – Columbia Journalism Review.

Top 5 essential skills for a data journalist

New York Times’ Aron Pilhofer answer to that question on the NICAR-L mailing list:

My top five (in order of importance):

  1. Know that the most important part of data journalism is… journalism. Reporting. In other words, you know how to report a story, you understand how to treat data as a source. You know how to pick up a phone, and not just assume that everything you get in data form (especially government data) is complete and accurate.
  2. You have at least basic data skills — meaning, you know your way around a spreadsheet. You can figure out for yourself how to import data, and do something with it. You also understand the basics of data analysis: rates, ratios, sums, averages, medians, and how to use them.
  3. You have command of more advanced data analysis skills, such as GIS, basic statistics, advanced SQL, etc. You also may know some basic programming techniques (using the language of your choice… Python, Perl, Ruby. ILENE.. shoot, even .NET) to scrape the web, get and clean data.
  4. You can apply your basic programming techniques to the creation of data-driven news applications using off-the-shelf tools like Google maps, MapBox, Fusion Tables, etc. At this point, you are not running servers, or serving database-driven apps. But you are creatively using what is available to you to add to your reporting online. This is probably where you need to get on the Javascript train.
  5. You have some skills with a web framework (Django, Rails, Grails) in order to enhance your reporting online through data-driven applications that you create from scratch and host.