How I’d like to consume interact with my news

(Prompted by a discussion today with Doc Searls, newly minted editor-in-chief of the reborn Linux Journal, and full of plans to redefine the relationship between publication, reader and advertiser. With some notes from over a year ago after prodding by Dave Winer)

  • My personal news portal
    • I follow about 30-40 sources of news (from CNN to Breitbart, from Spiegel to the Tehran Times, from CNET to Hacker News). Currently I have bookmarks for each of them, but it would be nice if they were integrated into one portal.
    • I also follow hundreds of people on Twitter, Facebook and the like. There is no reason why what they post appears in a different place, so I want those feeds to be part of my personal news portal.
  • River-of-subjects
    • Let’s group articles from different sources by the same subject (like Google News do, and Memorandum). That’s a great way to actually understand what is going on: different view points on the same event.
    • Those subjects flow, like Dave’s River-of-News, in chronological order but of when the event occurred (not when the article was written).
    • But I should be able to keep a subject around if I’m not done exploring it, even if I temporarily go to articles in the next cluster.
  • Don’t show me articles and subjects I have read before.
    • If an article is updated, I don’t want to see the whole article again, just the diff.
    • If I read a piece on news X from one publication, and I feel that’s about as much time I want to spend on that subject, don’t show me a piece on the same news item from another publication.
  • Auto-bookmarking and indexing of what I have read
    • I want to be able to easily find what I read, by keyword, by when I read it, etc. and even if the original article is since gone. So auto-bookmark everything I’ve read, with local copies, and a full-text database.
  • Visual prioritization
    • Things that are more important (on the score, see below) should be visually standing out and be in the middle of the flow. Newspaper front pages figured that one out. Less important items should be less prominent and off-center.
    • Breaking news (“at the speed of Twitter”) needs its only visual language so I can follow along in real-time without getting lots of redundancy.
  • Sources:
    • Obviously, I should be able to choose my sources, and unchoose them, without constraints.
    • A personal news portal, of course, should start with good defaults, and I don’t mind additional recommendations.
    • My extended social network would be a good starting point (extended meaning: aggregate of all the relationships I have on social networking sites, people I interact with in e-mail / messaging etc., news sites where I have bookmarked articles before etc)
  • Importance scoring
    • My local repository of bookmarks and full-text index is a good base from where to decide whether or not I am going to be interested in another piece of incoming news.
    • The amount of time I spent, and how many articles I read from different sources, would be a good indicator how important I consider this subject.
    • Some machine learning could probably be applied to get this right, with the parameters just for me.
  • Tagging, sharing and commenting
    • When bookmarking a piece of news, I should be able to tag the item, and comment on it
    • The tags and comments will be stored together with the bookmark in my personal data store
    • Optionally, I can share the bookmark with tags and comments by “publishing it” to my own website
    • Anybody who is subscribed to my website, via their own personal news portal, will receive my posts / bookmarks / comments / tags as part of their news feed. If they comment as well, we get a distributed comment thread that spans our respective websites, #IndieWeb style.
  • Trustworthiness / reputation
    • One particular type of tag (with an associated score) indicates my confidence level that this story is true or false
    • As I receive incoming from multiple sources with those tags, my personal news portal can automatically derive a confidence score not just on each article, but on the event itself.
    • By correlating confidence scores with the sources that provided the news items, the news sources themselves acquire a confidence-for-truth rating.
    • (This could obviously be much more sophisticated)
  • Privacy
    • What news I read is nobody’s business. I don’t want my personal news portal to leak what I read and what I didn’t to anybody
    • I want private news sources (access-controlled RSS feeds, news from member-only websites, etc) to be a seamless part of this experience. The personal news portal needs to manage access and credentials for me.
  • Payment:
    • I have not paid for news in many years since I canceled one paper after the other, the last one being the Economist. However, I would pay for this on monthly basis: flat-fee, regardless of how much news I read.
    • Revenue from my subscription would be shared among the people / sites / organizations whose news I read.
  • Advertising:
    • I don’t want to be tracked. If there are to be ads, the technology serving them must prove to me that they cannot track me.
    • However, I am willing to indicate to advertisers if I’m in the market for something. So if my personal news portal said “I see you are reading a lot of articles on electric cars, would you be interested in seeing some electric car ads”, I might very well say “Yes” (as long as there is no tracking)

Ok, so when can I have it? :-)

Leave a Reply