On 3-20-17, I sent an e-mail to Lina Rodriguez of the media-redesign group that is editing and formatting the colloquially known “Eli Pariser” document, or at least it’s colloquial in the constant re-telling of the story when relatives ask where I’m working. I haven’t really talked to any professionals about it. Anyway, this e-mail is an extension to the previous e-mail I’d sent, which you’ll find below this e-mail. Within it some new ideas and refined ways of organizing everything. It is still not a finished product – but getting closer.
I sent out a rambling e-mail to some people, outlining the steps I believe would really change the environment that makes disinformation or fake or partisan or bad news not financially worthwhile. I tried to send a copy to you, but it wasn’t a working address. This is a rewritten proposal with a few new ideas. Any notes would be appreciated. I know how little time you must have, so I mean those words.
Originally I caught your attention with the six-step process. The six different elements, in my mind, feed off of each other and should be organized into one non-partisan-but-activism-oriented entity. Essentially, I would like to see a major, organized governing body of the fourth-estate, because I feel like that is the only way a fourth-estate could harness any power. I am working strictly from the perspective of the journalist and truth-seeker – I am not versed in statistics, big data or programming, so the idea might seem too rudimentary to be effective – hence no responses – and it very well might be that a far better solution exists. I just have not found one all-encompassing, driven solution.
I see the problem always emerging from the corrupting force of money. The main theme is transparency-as-an-antagonist, and a lot of inspiration I took from Tom Rosentiel’s and Bill Kovach’s book “Elements of Journalism, 3rd Edition.” I’m not sure if it’s listed on the “Eli document” but if not, I recommend.
I’ve been researching the feasibility of automating things, and what organizations exist, and what jobs they already would or could provide to the process. Credibility is key, and I view it developing in two different ways…the first is organic and independent and slow, and the other is with an existing, trusted organization, which could speed up the process by a number of years, but be subject to their motivations. Wikipedia probably could and should take this on. I don’t think anything else could incorporate it, but there probably are many that could build it from scratch.
Either way, for this to work, I would need cooperation from the publishers, eventually, because I propose that the metadata of the content distributed through RSS feeds be made to include a byline and a defined subject (to which debate question does this article provide substance?). This is, of course, after the title, date, and publisher, which is generally already present. The only way I could do this is to establish a relationship with the publishers. In effect, their sharing of the data could be used as a verification process, which is a better alternative to rooting out the pop-up fake news outlets than direct silencing. They must stand up to scrutiny – something akin to the process of applying for a patent.
So first, my entity would provide a database of journalists, publishers, and their articles/content. The ISSN
could be close to accomplishing this, because they have relationships with publishers in 89 countries so far, but they don’t seem to be very visible with the publishers, or collect data that is very relevant for researching or scrutinizing. It’s also not free, and of no use to non-academics and professional researchers. Ideally, I would propose a free database, with funding from the major journalism/media associated endowments/philanthropists, that presents a database of all the articles written up to that day, browsable by author, by publication, and by debate topic, in chronological order. Each article contains links to the original source, and then all the aforementioned information, at minimum, which could easily be automated. After that, any rebuttals of an article should also be presented with the original article. I know of one company providing a backbone for this. Frankly, any linkbacks
should be provided, and there is no perfect tool that provides this yet. I would propose professional, original article analyses for the most popular or divisive articles to also be included, commenting on the strength of journalism, transparency, logic, and sourcing.
For a long time, years really, I was concerned with generating a “likelihood-of-truth” score, but I think the better solution now is to incorporate all of the other similar scoring projects into the article’s data page. The formula Tim McCormick used for his Diffr
project was exactly what I had in mind and should work for relative likelihood-of-truth, although his formula was intended for how partisan an article is, something I detest, since it doesn’t address confirmation bias and forces topics to have an equivalent political measure from an arbitrarily dualist conception. I think it is better to publish all scores and researchers can study the differences in score. There is a difference between Snopes and MediaWindow
, and I intend to use them all (and there a lot of these projects). Perhaps, if the publishers are able to break down their content data a little more, all of the services can use a universal encoding language.
The problem all AI-generated scoring will have is that money and might force stories on people, so a lot of it is public relations and propaganda, and there is nothing inherent about either that falsify or support claims. I can’t imagine a language-parsing program having the ability to determine the degree of the likelihood of an entity having a vested influence on an article, but even if one can and is successful, no doubt another AI will counter it. Unless explicitly stated – a desired ideal in journalism – there is still no way of determining the engineering of the story, which is what I’m striving for, or at least progressing towards.
The last part of the vision that I suspect will emerge out of this new collection of information gathered, with all the AI-incorporated projects in the field in mind, is that this database could direct the scope of what journalists cover. Perhaps, with hope, it could be big enough to lobby the publishers to counteract the corporate or governments interests. Some ideas related to this are to set up a direct blockchain-type donation system for rewarding journalists, so that they have more working freedom. What I would like to see driving coverage is a user-based voting system displaying the least researched questions in debate, with a Reddit-styled algorithm determining the length of visibility.
In my mind, this data can open up new branches of research in different genres of journalism. I mean, what the hell is everyone going to do once everything is automated? My hope is that they become citizens. What this all comes down to is transparency, and I would be asking a lot from journalists…that they stick their necks out first. With it would come celebrity and danger, presumably, but, such is the balance of life.
The major problems I see, in the short-term and small-scale, are publishers with paywalls, and how to deal with them. For my project to work, I would have to drive the price down to nothing, which might happen eventually, but in the interim, I would probably need a major negotiating checkbook balance or develop a new form of payment. This idea rings a bell
(The Klopping fella reformatting pricing through paywalls). In the long-term and abstract, this is also (arguably) the first step in willed-deanonymization, and I just don’t know how I feel about that. How necessary are secrets? Is lying the sweeping problem I think it is, or is it just as fundamental to nature as truth? Who becomes weakened? What will the unintended consequences be?
I have wavered on the importance of different selections, and there are other content/media project ideas I have that I would build out of the database, but that is essentially how I had foreseen implementing what I’d summarized in my first remarks to you. Your impression might be different. Your organization might have gone in different directions. I’m interested in hearing if you think this is still worthwhile, original, implementable, etc.
Once again, I’m writing this because I’m having the feeling of throwing a bucket of water at a forest fire. I feel like my particular conception here is elegant, but I’m also unemployed and can’t afford to be so. Thus, my priorities and anxieties are elsewhere, and my time for this may have already passed. Since you know the landscape, please let me know if there are any paying opportunities anywhere remotely close to the field. I am very close to New York City. Understandably, being a non-coder results in a lot of silence from the others I’ve reached out to, but I can learn quickly, and have some good contacts in programming.
Feel free to use any part of this in the document, if there’s anything new.
Thanks again for the time, and continue doing great work,
As you can tell, from reading this, that there have been some past conversations. I have yet to hear a response, although, I am expecting one, as this person has reached out in the past and extolled the sense that this project is worthwhile. I will post if anything warrants posting.
On 2-8-17, I sent an e-mail to different parties that might be interested in funding or hosting a project like this. Surprisingly, there is not a working Press Council in the U.S., only organizations that fund, direct, host and teach journalism and associated enterprises. But there is no major organizational body, and that is part of what my vision is.
The recipients of the e-mail were the following:
Tom Rosentiel: Head of the American Press Institute and major inspiration vis a vis his book with Bill Kovach, “The Elements of Journalism”
Jane Elizabeth: American Press Institute, Head of Accountability Journalism Program
Herbert Sandler: Finance billionaire who also funds many of the major left-leaning organizations that I consider having very positive affects on society
Reid Hoffman: LinkedIn Founder, Billionaire, and head of many different philanthropic organizations, who, like Sandler, funds things I consider very helpful and positive. Oddly, he funds them both furtively and publicly, through different organizations
Connie Moon Sehat: Director of Newsframes at Global Voices, an organization working on several projects similar to my “system.” We had shared correspondence before, and I wanted to let her read this
Craig Newmark: Founder of Craigslist, Billionaire, and very philanthropic. Same as Hoffman and Sandler
Sam Gill: A senior advisor at the Knight Foundation, who provides money for many different journalistic organizations. I admired his background and was looking more for advice or networking than funding
Jeff Brown: President and Founder of Fourth Estate. Seemed like an organization that most corresponded with my goals, though I was unsure what their method to attaining it was
Claire Wardle: Head of FirstDraftNews. Very active and intelligent thinker in the journalism community
Daniel Sieberg: Head of the Google News Lab
Jeanne Brooks: Genius working on several projects associated in journalism
Sally Lehrman: Head of the Trust Project at UC Santa Clara
Not Surpisingly, to date, I have received one serious reply and one wise-ass reply. I sent out what I am publishing here, which, in my estimation, needs some editing, for length, quality, and updating. Unfortunately, my brain has always been the type to want (it almost feels like a need) to finish a project more than to ensure the quality of the product. A volume guy. The only addition I am adding is a rough sketch, so that the idea hopefully makes more sense.
In 2010, I was reacquainted with my love for quality journalism having been assigned “Trial by Fire” by David Grann for a journalism class. Since then, I’ve subscribed to 30 print publications for the duration; I’ve tinkered with building websites that advocate these types of articles, much like longform.org
, and having become something of cynic, I simultaneously tried to answer the question, “How does bad news make it to print, and is it related to our (American) general divisiveness in politics?” As fake news became a narrative during the election, my thinking culminated into what I believe is the best solution for the state of journalism, and the events allowed me to survey the entire (again, probably American) landscape as to what others were doing. I have reassessed and will lay out the entire process for the thirteen of you that I’m emailing. I hope, if you may grant me the time, you can look this over and offer criticism, advice, or help of any kind, because I am at a critical juncture financially, and I believe the information I possess, the work I’ve done, is very valuable.
There are hundreds of projects dedicated to combating fake or bad or unwanted news that are too narrow in scope to be efficient on their own, subject to being squandered in the crashing wave of competition or global attention-span, and don’t target the people that need them most, while not, and this of utmost importance to me, infringing on the rights of citizens to be able to question, report, analyze and share freely online. So my proposal is not to attempt and force out the bad stuff, which sets a dangerous precedent and will likely happen in the interim anyways (see Facebook, Twitter, Reddit), but, instead, my proposal is to eliminate the conditions that enable bad news to exist in the first place. That would be done through a laborious, collaborative and integrated process similar to how Sydnor and Powell and the Koch’s created the anti-government/pro-business paradigm that prevails in typical water-cooler conversations now.
My proposal is a six-stage process, as far as I can see now, whose fundamental concept is transparency, and coalesces into one independent, international organization. Because my reach is limited, I have begun building this process myself, though I imagine that an existing organization with the resources and reach I don’t have would be able to carry this out much quicker. Unfortunately, I am not a programmer or coder, however, I’m intuitive enough to where I believe I’m capable of eventually building the requisite processes myself. At this point, nothing on my website, ThingsRelevant
is currently in its ideal form, but one can gauge the idea.
The first and second parts are very closely related, in that an index/wiki of every publishing entity and every journalist be made freely and publicly available online. The publishing index should contain profiles of every owner and staffer, operating costs, editorial history, and potential conflicts of interest. It would be optimal to establish a relationship to verify the publisher, and then automate much of this information, with the argument that a transparent publisher is a trusted publisher. Once a major publisher agrees, then I hope and imagine the rest will feel pressured to do so.
Concerning the journalists index/wiki, I would push for more information. I would propose including their education history, think-tank associations, every article authored along with editor’s name, links to all professional and publicly available personal sites, compensation, and potential conflicts of interest. This would be something of a loss of privacy, to which I have some suggestions to compensate. First, I recommend including in this index a method to provide a way for anyone moved to send micro-(or macro)donations to the authors of the pieces that they feel are deserving. When my curious nature is satiated by a great piece, I certainly have that feeling. I believe that the utilities already exist to simply send the money, via PayPal or bitcoin or some of the new cashless tipping apps, but I would like that to be transparent as well. Perhaps that is a new industry idea? (Paypal with blockchain technology?) My other suggestions are pieces of my other processes, which I will address later.
These two databases, to my knowledge, do not exist. Currently, what Wikipedia offers is general, secondary-sourced information, and as far as journalists go, only the ones lucky enough to score a publishing deal are represented, in line with nature of their operations. Customarily, the publishing companies set up the websites, and they go untouched after a book’s release. Some do exist with independent organizations, but not nearly as comprehensive as I am suggesting. Given the nature of the business, no journalist works their lifetime for one publication only, so it wouldn’t make sense for the publisher to provide a journalist index, and an aggregated, easy-to-navigate index allows for both academic and casual media study research. It’s also worth mentioning that publications do not use strong search engines, and as articles age, they become tougher to locate, either for the newspaper’s case – to save space they’re deleted, or they vanish behind firewalls or page-rank algorithms.
The third step is to employ article analysis. Over the years, my hobby has been to read one magazine fully through in the morning, and provide an analysis of one select article. I saw this as an educational process that doesn’t require a lot of time for the consumer – sensitively directing it towards those that need it most. These incremental steps were the small things I could do to make the world a better place. Improve the information diet of those I’m surrounded by. I would love to continue doing that, and probably will if I don’t hear from any of you. But I do understand that this is not a large-scale solution. It is the start of one though.
There are quite a few fact-checking sites, and even fact-checkers of the fact-checking sites, and this should be considered article analysis. They are useful, but they are subject to bias or mistake, as nature shows, and the larger problem is that people aren’t necessarily swayed by facts as much as what the experts recommend be done with those facts. I suggest a simpler, data-driven approach, in which information is collected for URLs with content from the verified publishers. Information such as page views, page shares, linking, word count, bot-text-analysis, use of facts (logical analysis) and sourcing, etc. would be useful, and once a system is running one can measure timing and relationships to other articles and content producers. One company, Rbutr
has identified a critical measure of an article’s integrity by focusing on rebuttals. Their system, like many others, seems to see their solution as recommendation, which I prefer to stay away from, but nonetheless, it provides useful data.
These first three steps are primarily objective-oriented roles that intend to produce data, to which I suggest keeping public for outside analysis. Browser extensions that relay this data are the common direction, but I believe the architecture exists to provide a better option, which are annotations
, if I understand the concept correctly. I suppose, theoretically, that they function quite similarly, as it would be up to the user to employ the tools, unless the annotation crowd has a method to develop a new browser and limit moneyed interests from their visible layers. Regardless, this is merely a matter of external presentation – all of the data will be stored and propagated from a primary source.
Because of the capitalist and controlled nature or model in which content is produced, propagated, and shared, we are not just handed facts. Even if it were the case that all objective material were factually accurate, our content is embedded with the ways we (should) use these facts. We are manipulated, not necessarily intentionally, into reacting on a desired or predictable spectrum. So, my solution, as the fourth step, is to create a debate database, open to all. I have been working with someone whose project is to clearly define different debate topics, hold a live debate on a given topic, and have a small group of chosen people reformulate the best arguments and publish them as “living encyclopedias,” and the process would work in a feedback loop, debating the book and publishing new issues after each debate. It is called the Truth Engine
. The other is a “Wikipedia for Debates
,” which does not yet have a solid form, but is web-based and crowd-sourced, and much more accessible. The creator, Timothy High writes about it at Medium,
and his ensuing articles to the one I linked argue the flaws of existing debate content.
All the ideas have merit, but I propose, in line with a desire to stay non-partisan, developing a simple debate concept much like debate.org
, or incorporating debate.org
, and linking each article in the database from section 3 to specific debates. As the databases for the journalists, publications, and articles grow, with links to the questions people are trying to answer and the problems are trying to solve, so too might the user, or citizen-generated debates and their topics. Having established relationships with the journalists and publishers, the hope here is that editorial decisions be directed from the questions and arguments arising from the people, and not so much from editors, lobbyists, or think-tanks. It can create a demand for certain journalists to do certain stories, or compete for stories, driving the value of the work up, so long as cheating doesn’t occur, which should be covered by my call for transparency. My hope is that, with credibility and funding, not only does it guide what the journalists cover, but it also guides, because of the openness and the knowledge that it will be investigated, what politicians and their funders can get away with. This is an attempt to create a major voice of and for the oppressed, which is also the majority; a true fourth estate.
With each article, linked to a debate or problem, as well as a publisher and writer, you can begin to mine information and draw editorial-content maps. The next step is to use the data from all the relevant sources to produce a rating or score for each article, which represents the percent chance that the article is likely to be truthful. I must admit, at this juncture, I am equally unsavvy mathematically as I am with programming, so I may be extolling impossibilities. The idea I had conceived of is very close to the formula Tim McCormick uses for this project, Diffr
(Begins on page 18), and the means of producing the score are very similar. My idea is to have a small number of experts rate articles, while introducing the concept to college professors, so that they might assign it to their students, and weigh the scores appropriately with a random sample of a mass audience – perhaps using an extension introduced by my theoretical organization. The difference in McCormick’s goal that he aims to measure the political bias of a given article, which doesn’t seem to have any value. Producing a political score will likely not only perpetuate the “filter bubble,” but the two-party system in general, which I see as a developmental issue to the nation.
The last step in the process, I propose, is to be coordinated simultaneously as the other steps are unfolding. It is journalism-oriented think-tank and political action. First, is the effort to push for journalists and their employers to be more transparent, to make the other processes easier, and to improve credibility with the citizenry. As the steam begins to build, then there should be efforts to not only promote journalism, but to promote the integrity of the profession and the vulnerabilities that journalists willfully take on. Journalists should lead very public lives, and should be compensated for doing so. They should take an oath, like a doctor or an FBI agent – to protect the truth. Judging by the early actions of the current administration, there will need to be a lot of money spent to preserve or repossess the rights of the press and the flow of information. If momentum can be built to this point, I want to address the issues surrounding journalism, much like what all of you do. I would like to see concerted efforts – journalism coalitions perhaps, in fighting for government transparency, whistleblower protection, limits on political spending, and whatever else may be a pressing need…I think there will be many.
My only cause for concern, which would only manifest in the latter stages of implementation, is that my perspective is too isolated. Obviously, there are things that I don’t know. A part of me wonders if secrets are necessary, and if the feeling I have about “the truth” is necessarily “right,” “good,” and worth pursuing. These are major issues to contend with; I do not know what humanity is and should ultimately be striving for. If I am to believe there are operators and they have agendas and they contradict mine – what makes theirs wrong and mine right? I only know that I have a strong conviction to tell the truth. But, at the very least, I would like to have the conversation about it, and judge for myself. And I want others to have that chance too.
I am writing in a state of emergency. I do not expect all of you to read this, and even less to respond. I am aware how grandiose this might all sound. My primary goal is to find a way to make a living as a fighter for others. I believe I have an idea – one that I’ve wanted to establish as a non-profit corporation and build on my own. But, I am also unemployed, and need to live. I do not intend to make money on this project, although there are some areas that could conceivably generate revenue. If this does resonate, there are numerous ways that I am willing to make this work. I may try a Kickstarter or GoFundMe campaign. However, If someone is building something similar, or would like this to become part of their organization, I would like to contribute. If you might know someone not on this list that I didn’t consider, please pass it along. And at the very least, if you believe that I could add value to an organization in a semi- or unrelated fashion, keep me in mind.
All criticism is welcomed and appreciated.