data vizualisation

Reading stats from Pocket

I’ve been a fan and heavy user of the Pocket app for a long time, but one good-to-have feature that is missing is Reading stats. Pocket doesn’t allow users to look at their reading stats directly so I looked up some resources on how to extract data using Pocket’s API. 

It took me an embarrassingly long time to do this  so I’m surprised no one has come up with a plug-and-play solution yet. I did find something but it wasn’t up to the mark. You can check that out here –

Let’s get to the stats. 

Number of articles added20142436
Number of articles read8701080
Number of articles favorited 251118
Word count (average, read)12331362

The trickiest bit of analysis here was the one on tags associated with each article. The JSON from pocket was nested 5 levels deep and a bit difficult to manipulate for someone who’s never used Python before. 

Top tags from 2017 

Top tags from 2018

I also made word clouds using this free website. Super easy to work with. 

Word cloud for tags from 2017

Word cloud for tags from 2018

That’s all folks! 

If you want to learn how to extract your data using the Pocket API, here’s a good primer:

For flattening your JSON and creating a pandas dataframe out of it, you can refer to this:

reading list

Highlights from my Reading List – Week 17


  1. Why monopoly is the future and we love it – IdeaFaktory
    Who controls the tools of disruption and what that means for our future.

    An interesting take on the shift towards monopolization.

  2. Is The Cult Of Self-Esteem Ruining Our Kids? – The Last Psychiatrist
    I won’t summarize this one. Just read it. It’s brilliant. 

  3. My best Extreme Contract lately – Jacopo Romei
    A real life story of how to negotiate an extreme contract – negotiating under uncertainty and complex, dynamic environments. 

  4. Academic Activists Send a Published Paper Down the Memory Hole – Quilette
    We’ve entered dangerous territory in academia where political correctness and lack of external checks have bred an unscientific environment. 
  5. Raised by YouTube – Atlantic
    An in-depth look at Kids’ channels on youtube and the rise of youtube parenting. 

  6. The New Atlanta Billionaires Behind An Unlikely Tech Unicorn – Forbes
    The story of Mailchimp – now a $4B company. 
reading list

Highlights from my Reading List – Week 16


  1. Populism – Kevin Kwok
    Why is populism growing?

    So in combination we have
    1) a re-centralization of financial capital and personal capital,
    2) reasons to believe that even if economics were net doing well there are certain countries (US) and demographics (white male) we might suspect are doing relatively worse than they were historically,
    3) social capital becoming stronger and more distributed, and
    4) people no longer believing in their personal growth–which is the source of all alignment.

  2. Empty Realm – Evan James (Jacobite)
    An in-depth discussion on the NPC meme and what it signifies.

    NPCs are characters with scripted dialogues and no original thinking. In modern parlance, the word has a pejorative connotation and describes people who are products of mass culture.

    How much of it is true and why are people offended when being referred to as NPCs?

  3. The Dunbar Number as a Limit to Group Sizes – Christopher Allen
    The Dunbar number is often associated with the number 150 being thrown around. That is a disservice to the ideas and implications of Dunbar’s theory.

    Dunbar is an anthropologist at the University College of London, who wrote a paper on Co-Evolution Of Neocortex Size, Group Size And Language In Humans where he hypothesizes:
    “… there is a cognitive limit to the number of individuals with whom any one person can maintain stable relationships, that this limit is a direct function of relative neocortex size, and that this in turn limits group size … the limit imposed by neocortical processing capacity is simply on the number of individuals with whom a stable inter-personal relationship can be maintained.”
    However, ideal group size varies depending upon the mission/purpose of the group (survival or leisure). Group satisfaction, trust and thus efficiency are functions of the Dunbar number and have multiple discontinuities.

  4. Analyzing Experiment Outcomes: Beyond Average Treatment Effects – Matthias Lux (Engineering @ Uber)
    At Uber, we test most new features and products with the help of experiments in order to understand and quantify their impact on our marketplace. The analysis of experimental results traditionally focuses on calculating average treatment effects (ATEs).
    Since averages reduce an entire distribution to a single number, however, any heterogeneity in treatment effects will go unnoticed. Instead, we have found that calculating quantile treatment effects (QTEs) allows us to effectively and efficiently characterize the full distribution of treatment effects and thus capture the inherent heterogeneity in treatment effects when thousands of riders and drivers interact within Uber’s marketplace.
  5. Bold Takes — Conversation with Sar Haribhakti
    For those who have never met or spoken to Sar Haribhakti, he has a vibrant personality and a very unique background. This interview covers a wide diversity of topics, ranging from Sar’s upbringing to how he thinks about startup studios to much much more.

    PS – Sar is one of my favorite follows on Twitter. You should probably be following him if you’re reading this list. His twitter is @sarthakgh

  6. – Francis Pedraza
    An exploration of the idea – What if you could take an equity stake in an individual?

    Super long read but worth the time. 

  7. World’s First Ceramic Knife Shave – Tom Blodgett
    I was chasing down a rabbit hole on whether razors can be made from ceramics given their hardness and long shelf life and came across this blogpost.

    This guy knows his shit – the video is him using a ceramic knife to shave. :O

    The idea is probably worth exploring commercially. 

  8. Institutional Memory and Reverse Smuggling – Quanticle
    Institutional memory comes in two forms: people and documentation. People remember how things work and why. Sometimes they write it down and store that information somewhere. Institutional amnesia works similarly. The people leave and the documents disappear, rot, or just become forgotten (as it were).

    This is a lovely article on institutional archaeology (figuring out why stuff works because the institution is so old that reasons and instructions are forgotten, only processes remain). 

  9. Thoughts on India’s Prospects – Prashant Abhishek 
    Prashant is the co-founder of AltCampus – a code bootcamp that is free until you get a job.

    I like his optimism and bullish view on India’s future. 

  10. Kinky Labor Supply and the Attention Tax
    Over the past few decades, labor force participation has sharply dropped for men ages 20-34.
    Theories about the root cause range from indolence, to a lack of skills and training, to offshoring, to (perhaps most interestingly) the increasing attractiveness and availability of leisure and media entertainment.
    In this essay, we propose that the drop in labor participation rate of young men is a result of a combination of factors:
    (i) a decrease in cost of access to media entertainment leisure,
    (ii) increases in both the availability and
    (iii) quality media entertainment leisure, and
    (iv) a decrease in the marginal signalling utility of (conspicuous) consumption goods for all but the highest earners. 

    Super long but worth the read. Ties in nicely with the NPC article and the one shared last week on Sex robots. 


Social Media, systems thinking, technology

Violence on Social Media: Feature or bug?

Ever receive a hate comment or insult because you expressed an opinion on social media? Does ‘Send bob’ sound familiar? How about libtard or bhakt? 

Violence, hate and harassment on social media platforms have been a fixture since their inception a decade ago. However, articles like “Instagram has a massive harassment problem” are doing the rounds with increasing fervor and frequency. 

Like most critiques of technology, the prevalent narrative is that of platforms being incompetent and irresponsible in preventing this degeneracy. The truth however remains buried, within the structure of these systems and in the human psyche. 

I’ll try to back the assertion that violence on social media is a feature not a bug using systems thinking and Rene Girard’s theory of mimetic desire.

Note: for a thorough and nuanced treatment of this assertion please read [1] in references which is the source of this idea in the first place; not breaking new ground here.

Also, here’s the original twitter thread that got me started. 


Girard’s theory of mimetic desire and systems thinking in the context of social media violence –   

  • Mimetic desire – the concept that we desire things not because there is an innate need or instinct but because others desire them. 


  • The model – our desires have little to do with the objects themselves and more to do with the subjects that possess or desire them. In a way, our desires are mediated by an external subject who functions as a ‘mediator’ or ‘model’.


  • Mimetic rivalry and violence – the fixation of two or more subjects on similar objects of desire leads to rivalry and violence. Girard termed this as mimetic violence. 


  • The scapegoat mechanism – society developed a mechanism to purge itself through periodic acts of bloody violence on victims who were made scapegoats. These scapegoats were sacrificed as a symbol for all that was evil and peace was restored. 


  • Violence and modernity – the modern world abhors violence. The march towards a peaceful society has not quelled our innate desires for violence but has taken away the old ways of sacrifice and purging which were meant to be channels for collective violence. “The modern era is characterized by a discrediting of the scapegoat mechanism, and therefore of sacrificial ritual, which creates a perennial problem of how to contain violence. [1]”


  • Social media as a channel for mimetic violence – Girard postulated that the intensity of the rivalry between subjects is a function of the distance between the object of desire and the subject. Social media has reduced that to virtually zero.


  • “Social media doesn’t alter the structure of social relations it merely makes some of them salient. [1]” Social media amplifies mimetic desire; the richest, prettiest, smartest, XYZ-est people are known to everyone.


  • Who wields power? “For Girard, to wield power is to control the mechanisms by which the mimetic violence that threatens the social order is contained, channeled or expelled. [1]”


  • Social media is the newest conduit for channeling mimetic violence. “It suggests that abuse, harassment, and bullying – the various forms of scapegoating that have become depressing constants of online behavior – are features, not bugs: the platforms’ basic social architecture, by concentrating mimetic behavior, also stokes the tendencies towards envy, rivalry, and hatred of the Other that feed online violence. [1]”


  • Trying to control hate speech on social media by having a zillion external moderators is naivete. No amount of reviewing will prevent this behavior because the cost of producing such content is essentially zero while the cost to identify, remove and moderate is finite. Given such lopsided dynamics, moderation (manual) can never catch up with production. 


  • Violence is an emergent property of the system – meaning it is systemic to the platform and its innate design. For any change to have material impact on this behavior it has to affect the core design and functionality of the product. Twitter CEO, Jack Dorsey even suggested they’re thinking of removing the ‘like button’ from Twitter. Will that be enough?


I’ll end this essay on the following note from Sundar Pichai’s interview with NYT, and whose situation can be aptly described by the following meme given the mess Google finds itself in:

Image result for this is fine meme

But there’s a deeper thing here, which is: Technology doesn’t solve humanity’s problems. It was always naïve to think so. Technology is an enabler, but humanity has to deal with humanity’s problems. I think we’re both over-reliant on technology as a way to solve things and probably, at this moment, over-indexing on technology as a source of all problems, too.

This brings us to question the following about narratives around tech-hate – are platforms *really* responsible for the collective behavior of people? Are they complicit if the downside of being intimately connected are violence, hate and trolling? 

I suspect that new social media platforms will learn, and focus extensively on these rather subtle aspects of features that result in harmful emergent behaviors. In the meantime, let’s just use mute and block liberally – it works fine.  



[2] Rene Girard’s theory of mimetic desire – Wolfgang Palaver