DEEMOK: Firing up Online Discussion

IncoherentDEEMOK colleague recently posted an article from The Onion on our internal company Google+ page which, paradoxically, as a piece intended for humor, is entirely relevant to the DEEMOK story. To distill it, the article ‘reports’ about a  ‘passionate and incoherent discourse’ among friends, gathered around a table at a bar-restaurant, about hot topics of the day.  True to The Onion’s form, the exchanges are belly-laughing nonsensical gems:

“Just look at gun control—Bush vetoed the ban on assault rifles, and now shootings are happening all the time,” added a completely incorrect Wagner. “After Newtown, Congress tried to legalize background checks, but they couldn’t get the legislation through the courts. Same as Obamacare and same-sex marriage. It’s all about money.”

Now that’s  funny.  Unfortunately, however,  these types of exchanges aren’t so ‘fake’ in real life.  Today, discussions on hot-button issues, far more often than not, quickly degrade into either meaningless quibbling, or highly resonating echo chambers. We merely need to witness our own social media channels, and notice the innumerable and meaningless battles that ensue after every important event.

As discussed before in our previous blog entries, there are many reasons for this phenomenon.  For one, although the notion of online discussion is around 25 years old, a proper decorum that mimics face-to-face discussion, and its ensuing restraints, has yet to completely take form. As for another, the appropriate structures are still not available or are too immature to guide people on how to share and exchange information in a digestible, less disputable, and coherent way. Users aren’t entirely blameless either and mechanisms to incentivize them to be thoughtful are lacking as well. And finally, there is a dearth of ‘validated’ and ‘fact ranked’ information on topics of socio-political importance that can be relied upon consistently and subsequently referenced.  The sources are either abjectly incorrect or, for one reason or another, biased to an extent that the central facts either disappear or are distorted beyond recognition to fit the storyline.

We in DEEMOK, intend to address these problems by becoming THE central and unbiased online place for people to gather and discuss hot button social political topics of the day and earn credibility while armed with the best facts available.  How we intend to do that is described in our three part series.  Stay tuned.

Mo Moadeli, is the CEO and Founder of DEEMOK.  You can follow him on Twitter @mo_deemok.

To get access to our upcoming beta release sign up at

For DEEMOK updates send an email to or ‘like’ us at  You can also follow us on Twitter @DEEMOKCOM.



In Part II of DEEMOK’s introduction we described the various features of DEEMOK.  In Part III, we will bring it all together and describe how these features will impact credibility and facts on the internet.


User generated content (UGC) sites employ a various incentivization strategies (‘gamification’) to encourage users to contribute in a way that enhances credibility.  The examples are manifest and range from Quora and Reddit to Wiki Answers and Yahoo! Answers.  Stack Overflow is also worthy of mention although it is serves an important niche dedicated to highly technical computer language Q&A. Additionally, each of these sites target different types of users and their function is different as well.  For example, Quora, Wiki Answers, and Stack Overflow are all, to a large extent , ‘question and answer’ sites while Reddit is more a next generation free for all community forum with various sub-forums (‘sub-reddits’).  Regardless, all of these sites employ relatively well known incentivization mechanisms which credit the user for posting questions and answers, content, etc.  Users are then ranked by the total number of votes (‘reps’) received.  Even the level of user anonymity, that ranges from total anonymity to semi-anonymity (known to the website) to no anonymity, is incentivized.  These websites quite often also employ algorithms to measure the ‘quality’ of users and their posts. Quora’s ‘Answer Ranking Scheme‘ is one such example. These algorithms, in conjunction with the website’s incentivization protocols, assign additional credibility ranks to users.

However, the conventional incentivization mechanisms employed by these sites don’t entirely address the credibility-fact feedback loop, if at all.  The sites expend much effort on one side of the loop which is establishing user credibility.  However, not much is spent on establishing the content they present as ‘fact’ and no discernible mechanism is in place to accomplish this.  As we discussed in Part I, credibility and facts are mutually intertwined and both are required to establish the quality of information. With a missing leg (i.e.’ facts’) the feedback loop remains incomplete.

In DEEMOK, similar incentivization systems will be employed to measure source credibility.  However, DEEMOK, also makes an effort to not only measure the credibility of the sources but also the quality of the facts they provide.  In DEEMOK, users are able to rank the links others user provide to bolster their arguments.  DEEMOK  keeps a tally of the top voted links and presents them in the FACTBOX to users who can either use them as references for their own posts or as the go-to sources to answer the various DEEMOK POLL questions.

Additionally, since the FACTBOX top voted links are publicly available, the originators of the links such as media companies (CNN, Fox, etc)  or journalists will see how their content compares to other media  competitors or other fellow journalists as voted for by the DEEMOK user community.  In other words, now there is a competitive market for factual information and, as such, media companies and their journalists are incentivized to provide higher quality content.


Another weaknesses of user driven Q&A sites is that quite often the quality of user content is very low since users frequently post off-topic or post overly simplistically. After a while, as the number of users on the site grows, the quality of content diminishes significantly as well.  This dynamic has resulted in many users leaving these sites in droves.  Yahoo! Answers has experienced this for example. Google Answers was shut down entirely. An exception has been Quora.  Stack Overflow is another exception.  Research suggests that the reasons Quora has escaped this phenomenon is due to the overlay of a social network system on top of Quora content  in addition to the existence of real identities which, together, regulate content quality. Stack Overflow also has a strong content regulation mechanism where impertinent questions are regularly removed by a corps of senior (high credit) members and moderators.

DEEMOK, addresses a large part of the content quality regulation problem by being the source of the primary topics itself.  Since DEEMOK’s content focus is around sociopolitical trends this differentiation is a natural one.  DEEMOK identifies trending sociopolitical topics and subsequently generates the DEEMOK POLL which, as described in Part II, contains relevant questions and opinions and with their own discussions.  In other words, the content regulation is automatic in DEEMOK since topics are trend driven and not user driven.


As we develop DEEMOK, technologies both old and new will be implemented to tighten the credibility-fact feedback loop.  With these in place, we believe DEEMOK will be in an excellent position to be the go-to website to identify and interact with credible users and also high quality factual content in the sociopolitical arena.

Stay tuned–our limited release will be available in the near future.

Mo Moadeli, is the CEO and Founder of DEEMOK.  You can contact him on Twitter @mo_deemok.

For DEEMOK updates send an email to or ‘like’ us at  You can also follow us on Twitter @DEEMOKCOM.  



In Part I, we described the various scenarios where a solution to the dearth of internet credibility and facts was needed.  The solution we proposed  is DEEMOK (currently in development).  With DEEMOK, the goal is to  further facilitate identifying the ‘signal from the noise’ especially in the broad arena of sociopolitical topics.

To reiterate, the following are the key DEEMOK features:


What do these mean?  Let’s explore each in more detail:


First, allow us to emphasize the difference between an ‘opinion’ and a ‘fact’.  An ‘opinion’ is what it is: An opinion.  For example, one may have an opinion on the subject of gun control one way or another.  An opinion is neither right nor wrong.  It’s merely an opinion.  Same goes with one’s ‘opinion’ on the subject of climate change.  In one’s ‘opinion’, it exists or it doesn’t.  Again, it’s neither right nor wrong.

‘Facts’, on the other hand, are not up for dispute (normally). For example,  ’2+2=4′ is fact.  ’Neil Armstrong stepped on the Moon, July 21st, 1969′ is a fact.  Neither are normally open for debate.  In DEEMOK, this differentiation is important.  DEEMOK, ‘incentivizes’ users to express their opinions on a topic only after they have demonstrated they understand some basic facts about it.  This is done within the DEEMOK POLL.

The optional DEEMOK POLL, which closely follows trending national and global topics, is published on the DEEMOK website on a scheduled basis.  The DEEMOK POLLs are across various socio-political topics such as politics, society and culture, economy, religion, government, defense, environment, energy, national security, etc.

The ‘DEEMOK POLL’ is comprised of four components:

  1. DEEMOK OPINION: A true/false/yes/no ‘opinion’ question for user related to the trending topic.  This is the ‘opinion’ part mentioned previously. Users can’t answer the DEEMOK OPINION until they first exhibit basic knowledge of the subject via:
  2. DEEMOK QUESTIONS: –users exhibit a basic understanding of the topic by answering a series of optional easy/medium/hard questions related to the topic. This is the ‘fact’ part mentioned above.  By answering these questions users also earn DEEMOK CREDs that can be used later for discussion in the DEEMOK TALKBOX.
  3. DEEMOK FACTBOX:  The FACTBOX contains quick links to help users answer DEEMOK QUESTIONS.  Some are DEEMOK-provided links and others are the links in used in TALKBOX discussions that are highest voted by other users.
  4. DEEMOK TALKBOX:  Here is where users can engage in discussion with others.  Users will use a certain number of CREDs for each post. Additionally, users engage others with the same  DEEMOK CRED LEVEL or lower.  Users can’t initiate discussion with others of higher CRED LEVEL. Users also earn CREDs from their posts AND reference links by getting votes from other users.

We’ll explore the above DEEMOK POLL components further later in this article.

To summarize, once the DEEMOK POLL  is published ( the publishing frequency is related to frequency of  trending national and global topics), users can gain DEEMOK CREDs in various ways such as by answering questions, giving their opinion, and discussing topics with others in the TALKBOX.  The number of posts allowed by users in the TALKBOX is determined by the number of DEEMOK CREDs they have earned in the DEEMOK POLL.  As users attain more and more credibility points historically and across various DEEMOK POLLs, they attain higher CRED LEVELS and are enabled to do more things in DEEMOK such as creating a new DEEMOK MOVEMENT or becoming DEEMOK MODERATORS.


To assist with answering the DEEMOK POLL, a series of reference links (‘facts’) are provided by both DEEMOK and by users in the FACTBOX.  When users post in the TALKBOX, they often include reference links to bolster their argument.  In addition to the user posts, these attached links can be voted by other users as well.  DEEMOK keeps a tally of the highest voted links and includes them in the DEEMOK POLL FACTBOX.  The links can refer to any source whether CNN or FOX or Wikipedia or a blog.  In the FACTBOX it is easy to see which of these sources has attain a higher standard of credibility as determined by DEEMOK user votes.


As mentioned before, users can engage in discussion in the DEEMOK TALKBOX at any time.  The number of posts allowed depends on the number CREDs users earn answering the current DEEMOK POLL.  Although normally sufficient, it is possible for more prolific users to run out of CREDs. To earn more, users can request a new DEEMOK POLL and answer more questions to earn more CREDs.  Additionally, users can only initiate discussion with others at the same CRED LEVEL or lower.  All users can follow discussions by users with higher CRED LEVELs but they cannot initiate discussion with them.  This is what we call ‘stratified discussion’.  Additionally, since it is optional, users may decide to skip the DEEMOK POLL.  if they choose to do this, they can ONLY engage other users that similarly have not answered the DEEMOK POLL.  Needless to say, the quality of discussion at this level is naturally not as high as it would be otherwise, and the chance of trolling increases significantly as well.


It is evident by now, that a user incentivization ecosystem permeates DEEMOK.  One of the ways this incentivization manifests itself is through how a user gains credibility.  This is measured by how many DEEMOK CREDs have been earned and what the corresponding DEEMOK CRED LEVEL is.  CREDs can be earned in various ways such as answering DEEMOK POLLs, posting in the TALKBOX, and providing credible references in posts.   Currently, the DEEMOK POLL has an expiry date and earned CREDs after the expiry date is significantly diminished if not lost altogether. This encourages users to answers POLLs as soon as possible. Also, since in future phases of DEEMOK, interactions with local, state, and national elected officials will be included, additional CREDs can be earned when users optionally provide basic location information to enable DEEMOK to identify who their elected officials are.  Another example of the incentivization ecosystem is where users can earn additional CREDs by making their DEEMOK KREDO (see below) public.


DEEMOK has developed  a simple and powerful mechanism, called the DEEMOK KREDO, to enable users to set where they fall in the sociopolitical spectrum (i.e. liberal, progressive, conservative, libertarian, centrist, etc) and optionally publicize it.  Currently, the DEEMOK KREDO allows users to easily set their beliefs from 75 different options using an elegant ‘slider’ system.  Users can adjust their KREDO in three ways:

  1. personal/cultural range:  this ranges from moral conservatism to liberal freedom.
  2. government role range:  this ranges from government is best small to it is best large
  3. corporate role: this ranges from corporation are persons to corporations need regulation.

Users will also be incentivized to publicize their KREDO by earning CREDs.  DEEMOK NEW MOVEMENTS (see below) will also have their own KREDOs which can be set by movement founders.


In the next phase of DEEMOK development, which is Phase II  (all features above will be available in Phase I), once users attain a certain CRED LEVEL they will be able to launch a DEEMOK NEW MOVEMENT.  A DEEMOK NEW MOVEMENT can be anything, such a local gun enthusiasts club or a virtual national political party.  The ‘founders’ of each DEEMOK MOVEMENT can create a charter, logo and banner, set the MOVEMENT’s KREDO, invite  users of various KREDOs to join, launch their own POLLs and have their own TALKBOXs, and even have their own elections.  DEEMOK NEW MOVEMENTs will allow users to organize in a significant way. The DEEMOK NEW MOVEMENT will be a significant future feature for DEEMOK.


In another significant feature in Phase II, users can identify, monitor, and rate their elected officials at various local, state, and national levels.  Users can compare their KREDO with the voting record (and KREDO) of these officials and use that as a measure to publicly rate the performance of these officials.  Users can also communicate with their elected officials.  This DEEMOK feature will part and parcel complete the credibility-fact feedback loop in the context of the performance of politicians.

In Part III, We will bring all this together and and discuss how these feature will enhance the credibility and facts on the ‘internet’.

Mo Moadeli, is the CEO and Founder of DEEMOK.  You can contact him on Twitter @mo_deemok.

For DEEMOK updates send an email to or ‘like’ us at  You can also follow us on Twitter @DEEMOKCOM.  



A few years ago (about a couple of decades ago rather), many of us witnessed how the ‘internet’, was not only rapidly becoming a medium for dissemination of high quality and accurate information, but also a medium for the dissemination of poor quality content as well.  These inaccuracies, and their impact, were generally harmless as this new medium hadn’t quite made it’s way into all facets of our lives.  Needless to say, today, the internet is ubiquitous and a critical tool for people, government, media, and industry. As such, along with the ‘good content’, poor quality information also spreads rapidly to more important areas of discourse ranging from climate change and healthcare to national energy strategy, national security and electoral politics.  Today, not having access to credible sources and facts impacts us in highly crucial areas such as how we vote, where we decide government spends our taxes (or not), or how we balance our personal privacy vs our need to be safe from threats. The spread of poor quality information is even more amplified as people disseminate unverified or incorrect links and references through their social networks.  According to Pew Research Center,  67% of adults in the US actively use social media which contributes to the speed of the spread of information regardless of its veracity.  An innocuous example of how fast a meme spreads is the recently popular the ‘Harlem Shake’ meme and it’s various versions.   They went from zero to 7.4 million views in the span of a week.  There are many more examples.  ‘Truthy’, a research site run by The Center for Complex Networks and Systems Research at Indiana University, Bloomington provides powerful tools to visually analyze the spread of Twitter memes on the internet. Using Truthy, it is relatively simple to see how particular memes spread and what the impact is.

Although there are various ways to define how accurate information is, we have chosen to define it by:

1.  The credibility of the (human) source and:

2. How demonstrably factual the information from that source is.

The implication is that credibility and facts are intertwined and we need both to evaluate the quality and accuracy of information.  Let’s examine this further:


It is (or it should be) natural for us to evaluate the credibility of our sources when we are presented with information. A NASA rocket scientist has more credibility in conveying notions of rocket lift, payload, and aerodynamics than, say, a 5th grader or a lawyer.  However, what makes this interesting is that, as mentioned before, credibility is not sufficient in evaluating information in and of itself.  A rocket scientist can still err in providing information even if it is unlikely.  Which takes us to our second component of information: The ‘facts’.


We are fortunate that, even in this day and age of subjectivity, there are still vast bodies of discourse that can be practically classified into a simple binary fact/non-fact format.  These are usually bodies of science and mathematics.  Put simply, some things are demonstrably factual and will be forever.  Examples are’ 2+2=4′ or ‘the Moon orbits the Earth’.  A vast majority of us take these as a given and rightly so.

Unfortunately, a ‘fact’ in itself is insufficient to measure the viability of information. A medieval 7 year old child may randomly blurt out that the “Moon orbits the Earth”, but to an audience who is yet to be exposed to modern orbital mechanics, given the low credibility of the precocious child, this piece of information, however factual, won’t get too far. So again, one needs facts coupled with credibility of the source to measure the quality of information. In addition, not all bodies of discourse can be so easily broken down into a binary yes/no fact space as mathematics and physical sciences can.  Sociopolitical topics,  such as religion, economy, environment & energy policy, national security, etc, on the other hand, are ‘gray’ by nature and arriving to indisputable facts is inherently difficult.

Additionally, a strong credibility-fact feedback loop to incentivize higher quality content for sources of sociopolitical information does not exist.  Let’s take Yahoo! News or as examples of this disjointedness.  Note, both may publish separate articles on a given topic on their websites, and further below, in the respective comment areas, users initiate a storm of discussion on the topic.  Some might even question the validity of the article itself.  Unfortunately this mechanism does not provide sufficient incentivization for the sources of the two articles, namely the CNN or Yahoo! News writers, to make the effort to maximize the quality of their content.  Why?  Because the article and comments in each site is a closed system and  neither CNN nor Yahoo! News are able to see how users compare the quality of their respective articles with the other site.


Although user generated content sites (UGCs) (eg: Facebook, Quora, Yahoo! Answers, Reddit, etc) can be used as forces of good, they can as rapidly facilitate the spread misinformation whether intentionally or not.  A recent example is the community discussion website Reddit and the various sub-reddits that blossomed around finding the perpetrators of the Boston Marathon bombings.  This is a particularly acute case where a major NYC tabloid picked up on the results of these sub-reddit ‘investigations’ and erroneously splashed the pictures of the supposed perpetrators on its front page.  Facebook, is another good example. Due to the way Facebook facilitates dissemination of information across personal networks, its spread is particularly fast.  For example, we can see how often the hoax post “RIP Morgan Freeman” spread like wildfire across various news feeds, again and again.

Similarly, social network systems (SNSs)  like Facebook and Twitter are able to spread information on matters that are a lot more important than the supposed demise of an actor. We see this on our feeds and tweets regularly.  People posting links referencing certain ‘facts’ about a political candidate or cherry picked data on climate change.  All intended to sway a particular audience one way or another.  Even worse, arguments flare up on these subjects and people share and re-share the same unverified sources over and over again.  Sometimes, the spread halts because the critical mass of those who bought into the farce was never achieved.

Quite often, on the other hand, the bad information spreads to an extent that it becomes canonical ‘fact’.  So much so it even becomes a ‘given’ referenced by ‘credible’ others.  Fortunately, for better or worse, a sense of restraint seems to be prevailing on Facebook when it comes to serious sociopolitical topics. Less so because of the sudden rise of critical thinking of the public at large, but more so because the public at large does not look very well on posts and posters that are political and controversial in nature.  Either that, or they simply don’t see SNSs as a major channel for political activity, i.e. it’s simply not cool.  According to the Pew Research Center, a  majority (64%+) of social network users (Facebook, Twitter, etc) simply don’t post politically.  For those who are politically active in SNSs, 18% of ‘friends’ block the posts or posters with political bent and 16% of people friend someone based on political beliefs. Which means the net gain vs. loss is a wash. In other words, people do not cull or extend their SNS networks based on political beliefs.

So if traditional SNSs are not appropriate media for political discourse, then what is?


Our answer to the problem of tightening the credibility-fact feedback loop is DEEMOK.  DEEMOK will become the engine of credibility and fact in the realm of popular sociopolitical discourse.  To accomplish this, DEEMOK provides the following features to users:


What are these features and how do they impact the credibility and fact cycle we mentioned before?

Check out Part II of our DEEMOK introduction to find out more!

Mo Moadeli, is the CEO and Founder of DEEMOK.  You can contact him on Twitter @mo_deemok.

For DEEMOK updates send an email to or ‘like’ us at  You can also follow us on Twitter @DEEMOKCOM.