This project is needed to to provide news readers & news gatherers with a guide to the quality of information contained in news articles. I want to be able to color-code or otherwise identify (via dashboard indicators, perhaps) the following qualitative metrics:

    • number of unsourced statements (eg. "an administration official", "unnamed U.S. officials", "U.S. sources" etc.) - see also [this story] on sourcing for journalists (and links.)
    • uncorroborated assertions vs facts witnessed or otherwise validated
    • items taken out of quotes "an unnamed source said that "horseshit is burning......" "
    • temporal confusions - see Bill Clinton's famous Lewinsky speech abour the meaning of the word IS
    • headlines that do not match the content or contradict it
    • multi-level nesting of a news story (the AP reported that Reuters reported that the N.Y. times article contains the following information) if that occurs
    • undated, unnamed, unlocated news articles - reporters date & time etc. should be clear from the start.
    • incomplete contents - real events have who, what, where, when, why & how components.
    • absence of clear identification of the protagonists, actors & what they did
    • "a picture says a 1000 words" [see 10x10] - incongruency between pictures & headlines & texts. The pictures usually say the most.
    • From the other NLP ( NeuroLinguisticProgramming ) comes "hypnotic language" such as lost performatives - see [NLP Meta Model]
    • Things said "out of quotes" e.g. I met a man in the bathroom who said that you are a piece of shit! but, It was only him saying it not me(?).
      • One of the most insidious pseudo-sourcing phrases "critics say ....." & "critics of the president say...." when the reporter wants to get his own point of view into the article.
    • Saying something with a question like "Are we getting bogged down in Iraq?" - the idea may not have been there until we ask the question. Asking a question in a headline is also another example. Repeatedly asking a question seems to solidify the idea, etc..
    • Along with ReHeadlining there might also be a ReQuestioning - Asking the question that was actually answered in an interview, briefing or press conference - some people don't even pretend to answer the question asked & others spin in place until the questioner gives up.
    • Polls that make headlines different from the questions/results indicate - polls created to make a specific headline - DishonestPolling
    • Proportionality - is the news article out of proportion with other things in the news - is it as important as it seems from the placement & size of headlines - (front page above the fold or page 13 - either way - is it minimized?)
    • NewsDashboard
    • Google is getting closer - proof of life
    • There are some good definitions of political point of view in the Wikipedia such as [Libertarian] , [Liberalism], [Conservatism] & [Left-Right Politics] etc. - which may assist in determining an article's point of view. [Fifth Column] is also interesting in the light of terrorism, illegal immigration, the [ACLU] & [other] news of the day. [WikiNews] also has some interesting pointers on the reporting.

Blogs pose an interesting concern because the reporter is somewhere else. Blogs are mostly commentary. Blogs tend to be a megaphone for someone else's material.

So maybe we import the story into the wiki, click a button to apply the NL processor, then humans can go back and reedit the result. Cool i like it!



  1. information rating


Mark de LA says
See especially the news dashboard item.

Mark de LA says
Now I remember why I made these private.  They were from the original wiki. I wanted conversations to be in the fastblogit on new items not on the extracts of the old wiki.

Mark de LA says
seth 2012-12-27 11:36:49 1603
MR 2012-12-27 11:33:54 1603
Now I remember why I made these private.  They were from the original wiki. I wanted conversations to be in the fastblogit on new items not on the extracts of the old wiki.

i was just trying to find out where the "robotic polarity" started.
BTW, another reason is some of the old links don't work anymore.