Written by Gal
on 

Making the most of the Internet

As we all know, the internet is an incredible body of knowledge. To put things in perspective: today, anyone with internet access (be it in the US or in Malaysia) has – at their fingertips – access to a body of knowledge that even the most well-positioned scholar of a mere century ago couldn’t reach in a year. Yet, simultaneously, as we also know well, it is also home to the world’s largest collection of cat video compilations. I am by all means not taking the perspective that using the internet to educate yourself is good and for watching cat videos is bad. Therefore, a better title for this post would come with the following disclaimer: how to best “unlock” the amazing potential of the internet, if – and when – you want to.

I have organized my thoughts on this issue in the form of a 4-part “roadmap” towards achieving this goal. This isn’t to say that this is a recipe that anyone else should follow; it is simply my way of creating structure in the insights I’ve gathered on this topic.


#1: Exposure.

The first step towards unlocking this potential is of becoming aware of things that are actually “worth knowing”. If you think about it, this isn’t a simple task. This object which we refer to as “the internet” is in fact nothing more than a huge graph (the nodes being the webpages and the edges being the links between them). It is safe to assume that what counts as interesting to you is a not only a very small fraction of this graph, but also a very scattered fraction.

As such, navigating this graph and finding interesting content is a formidable task. Indeed, for all intents and purposes, we need an interface - a “gateway” - for the internet. For a long time, search engines served as this gateway. And indeed, this is a useful way of exposure: using the appropriate search term, Google will reliably take care of the task of finding which web-pages contain relevant results or information. I call this active exposure: if you know exactly what you want to know, you’ll find it. But what if you don’t know?

Recent years have seen the rise of a new alternative: social media. The idea of a “news-feed” has essentially become the new gateway. This is appealing in two ways. First, it promises a form of passive exposure: all you have to do is keep scrolling: the feed is the gift that keeps on giving. Second, and even more appealing, is the promise that this feed is personalized: it suits your taste (and in fact, bears the promise that the more you interact with it the better it will become at knowing which content you want to consume). This sounds very promising, and indeed if people will be polled for the first thing they do online, I would be surprised if the first answer would not be some form of social media website.

When I started thinking about these issues, this was precisely the state I was in. I was scrolling through my never-ending Facebook feed, hoping to discover interesting content tailored for me: new music, book recommendations, insightful news articles. The more time I spent there, however, the more I realized that Facebook was hardly serving this purpose. An overwhelming portion of my time there was effectively spent scrolling through images from the (probably fictitious) lives of the people who were once a part of my life.

What to do? Psychologist Kurt Lewin famously said that there are two ways for changing people’s behaviour. The first is the famous carrots and sticks approach: convince, pressure or persuade them, using either positive or negative incentives. But there is a different approach. Ask the question: why aren’t people behaving in the way I want them to? What are the obstacles they face? and work to remove these obstacles. The quintessential example of this lesson applied in practice is in organ donation: we could try to pressure people in different ways to donate their organs after their death; but if we recognize the fact that people are merely lazy and don’t like filling forms and making choices, then a much simpler solutions presents itself: make donation to be default.

Applying these ideas in my case, yes, I could – in theory – get rid of social media all-together. But it’s hard, and perhaps more importantly, it’s not clear what the alternative is. RSS feeds? They are static and require too much maintenance. Many years ago I remember using the enjoyable website StumbleUpon to randomly find interesting things on the internet, but it is now deceased.

The solution in my case had a “make lemonade out of lemons” flair to it. Rather than attempt to get rid of social media altogether (throwing away the good stuff with the bad stuff), find a way to use an appropriate form of social media to fit your needs. In my case, this amounted to using Twitter, and actively maintaining a list of people whose thoughts, links and musings I really want to follow.


#2: Curation.

At this point, I found myself with a feed full interesting content. Is that it? As I quickly realized, the answer was no. I had access to the content I wanted, but I wasn’t actually consuming it. Confusing, yes, but to be expected, if you give it some thought. One thing social media platforms all have in common is that they end up promoting superficial interaction. You are rewarded for “Likes” and comments, maybe viewing a video, but not for anything that ends up navigating you outside social media (such as actually opening the link and reading the article). The platforms aim to maximize user’s engagement, and as ex-Googler Tristan Harris cleverly argues, they are often alarmingly good at it. This leads to a conflict of interest, with users often spending significantly more time than they would have desired scrolling through endless feeds. (In fact, this is a can of worms, with numerous other problematic effects. One that I find especially alarming are Youtube’s terrible recommendations).

The solution that I found was practical and simple: keep using Twitter to find content, but once I found something worthwhile - put it aside. “Aside” is the important point here, and means outside social media. I keep a Google doc with things I want to read, and sometimes send them to my Kindle to maximize the probability that I would read them undistracted.


#3: Knowledge Acquisition.

Having a “reading list” distant from social media proved to be insufficient for me. The incentives promoting superficial interaction (just get through “as many” of the items as quickly as possible) were still there. Now, when I sit down to read something I actively practice going beyond a superficial interaction: acquiring knowledge, insights. This can take many different shapes and forms. One thing that works for me is to practice distillation. Can I tell another person about it? Can I write a few lines about what I understood?


#4: Knowledge Persistence.

The problem with steps 1-3 is that with very high probability, whatever you truly know today you will forget in a year from now. This seems discouraging, and to potentially undermine the very thing we’re doing here: what’s the point of knowing something, if this knowledge will likely not last?

The resolution, perhaps, is that we need a new definition of knowledge. What does it mean to know something in the 21st century? I like the definition of prosthetic knowledge suggested in the title of this blog:


n. Information that a person does not know, but can access as needed using technology.


If you think about it, this is nothing new. In fact, software developers practice this type of knowledge literally all the time. For example, I never remember the exact syntax for saving and loading a Python pickle object, even though it’s something I do very often. However, what I do remember is that if I type in Google the query “Python load pickle”, then the fourth result is a Stack Overflow whose top answer has exactly the code-snippet that I need. Then, do I know how to load a Python pickle? The answer is of course (albeit under a slightly extended definition of knowledge). In fact, this is exactly what made Stack Overflow such a hugely popular website.

A natural question, then, is why stop at code-snippets? Under the “prosthetic knowledge” approach, forgetting something is perfectly fine - so long that you remember how to look that piece of information up when you need it. Applying this perspective to the context discussed in this post yields – at the extreme – a type of “personal knowledge base” (or digital brain). A huge hash-table of everything you know!

Honestly, this requires a level of organization that for me has (so far) proved difficult, but I find the idea itself incredibly compelling. Perhaps one day I will be there too.

(Comment: This was a writeup of talk of the same name I gave at the Weizmann Math & CS Women’s Forum on March 26th, 2019.)