If someone asked me to name a modern-day equivalent to the Library of Alexandria then I would probably pick YouTube. Not Wikipedia because a library is not a place where anyone can go and edit the books. I have personally learned and benefitted so much from having access to it. How to fix (almost) anything I have in the home. How to cook. How to improve my drumming ergonomics. About boxing. Philosophy. Movies. The news and all those countless time wasters that we hold so dear. My YouTube export's watch-history.html is 28MB long. I don't know how much that it is hours and I'm not sure I want to know. But here's my question. When's the last time you've found a video that's not on YouTube? Where would you even go to look for them (outside of porn)? Archive.org? Which brings me to the next question. Do we really want a single company controlling access to all that information? To be the sole proprietor of basically all human endeavors? And if the answer is "no" then what are we going to do about it? Because continuing on this path will just keep making the situation worse. There was a time when Google was just some new search engine that worked faster. Indexing content that it did not host, and more importantly only having access to data the publisher explicitly chose to share. Not analyzing your entire presence on the Internet. Not reading your emails. Not knowing where you are or where you're going. Or what you're saying at home. Or powering most of the internet-enabled mobile devices. And not selling all that data in some pre-digested form to the highest bidder. The web was supposed to be a mesh network. It works best when it's a collection of things working together. An open marketplace for ideas not just in the content, but also in the implementation. Google has benefited enormously from the open web and the innovation it enabled. We somehow managed to take this incredibly flexible thing and turn it into these massive silos filled with who knows what. Controlled by who knows who. Incredible amounts of labor, energy and natural resources. Abstractions upon abstractions. Just to store, process and analyze all this stuff at this *planetary* scale. By a handful of companies.

160 lifetimes per second

I don't know exactly how much time is spent on YouTube, but according to this thing it's roughly 12 000 hours per second. How many lifetimes is that? About 160? Every second. The financial sector likes to talk about how this "competition for attention" is affecting the Video-On-Demand (VOD) business, but what is it all doing to our lives? All those hours spent watching this stuff. Could it be that one of the reasons productivity is declining is simply because there's so much out there distracting us from what we should really be doing? Could it be that the increasing rates of depression and anxiety among our young is because they spend too much time staring at never-ending content on these highly-addictive platforms and disassociating them from reality? And then there are ads and the neverending cat-and-mouse game between YouTube and adblock developers. Not to mention the Kafka-esque nightmare that is the YouTube copyright system. So, while I really respect what all these brilliant people have accomplished, I must also say that I don't like this trend of putting all our data into these gated silos. I say "gated" because they're not completely closed. You and I just don't control the gates. And some of those rules for controlling the gates are stated in the EULA. Which nobody reads. And then there are bugs. Lots and lots of bugs. Which nobody takes responsibility for. Data is the first completely human-made product that has some value sometimes even tremendous value. Monetary value, political value, cultural value. Human value. All that music we've enjoyed over our lifetimes, all those photos of your family and your life, are just data. It's high time we start treating it as something valuable. Which includes being deliberate about how and where we share it. Yes, we need means to share multimedia content, but we don't need for all of it to be in the same place. Sometimes I think - as do probably many other people - could there ever be another "Google"? And the honest answer is no. But I think the more interesting answer is "no and there shouldn't be" Because websites still work and browsers are actually in many ways better than ever. And streaming something to anyone is still just copying a file to a web server. Links still work despite the silos' best efforts trying to break them. People still don't waste 12 000 hours every single second watching content that *Google* created. YouTube has only changed their conception of "where" because of all the great content that millions of people have shared. The thing is, I only had 60 subscriptions on YouTube and I've been logged into that thing for quite a long time. I would say less than 10 of them had more than 2mil subscribers. I would say that at that level it's probably best to stay on YouTube because you're now running a media business. For the rest of us. Let's build websites. If we want to share something, let's just do that and not worry about how many people see it or like it. Where they're from or how old they are. Because then we can be on the best platform of them all - the open web. For anyone and anything to see. They just have to find you.

Getting back to my point 😊

This post was originally supposed to be about how to download all your subsriptions down from YouTube. I don't know how it morphed into this epic rant about the Library of Alexandria, Big Tech and the future of humanity. My mind just seems to wonder there whenever I think about these things. I almost deleted most of this text leaving just the bit that follows, but honestly believe I'm on to something here so I'm posting this as is. I will probably revisit many of the topics touched upon here in the future. If you're like me who wants to spend less time on YouTube while still seeing value in it (not blocking it completely), here's one way to do that. The method I found involves using RSS. Gotta give YouTube credit where it's due - Google Takeout works and they still publish channel metadata in RSS. And the export really does contain your entire search history while you were logged in meaning of course that Google still has it too. So, after downloading the subscriptions.csv file, we can run something like this: awk 'BEGIN { FS = ","; print "<?xml version=\"1.0\" encoding=\"UTF-8\"?><opml version=\"1.1\"><head><title>YouTube Subscriptions</title></head><body>" }; /http/ { print "<outline text=\"" $3 "\" title=\"" $3 "\" xmlUrl=\"https://www.youtube.com/feeds/videos.xml?channel_id=" $1 "\" htmlUrl=\"" $2 "\" type=\"rss\" version=\"RSS\"/>"; } END { print "</body></opml>"; }' < subscriptions/subscriptions.csv > /tmp/youtube_subscriptions.opml ... resulting in /tmp/youtube_subscriptions.opml which you can import into your favourite newsreader. Why do this? Personally, I found that using RSS just lets me spend a lot less time in front of YouTube (or any other site for that matter). It changes my behaviour to just skim the headlines and check out the links that seem either truly relevant to me at that moment (not that many) or stuff that's extra interesting as opposed to being distracted by all those shiny things the YT algorithm dangles in front of me. Part of it is also the lack of images. On that note, there's a neat little Firefox extension that makes YouTube a lot less addiction-inducing called Unhook. Older posts