View profile

Etcetera #17: Content, Recommendations, Greebles

Matthew Culnane
Matthew Culnane
@coldbrain
Hello. Nice to be back after a week at the seaside with mediocre weather followed by a week stuck inside at work with great weather. Sigh.
The main links this week are about content and its quality, discoverability and value. Funny how ‘content’ has gained pejorative connotations in the last decade or so—essentially coming to mean entertainment that’s been created in service of some other, grander, profit-driven aim, such as ad revenue or data collection—yet people are keener than ever to label themselves ‘content creators’. Maybe I’m just old.
All of this and more is discussed below. People seemed to like the revised format of more in-depth stuff at the top and the assorted links below so let’s continue with that.
Reply with your thoughts and suggestions. See you next Friday.

Paid in Full
Paid in Full
Drew Austin on the emerging dream of an internet where every interaction is a financial transaction. In “Web 2.0"—the participatory internet—there was a currency of attention, usually counted in terms of followers, likes and retweets and supported by platforms who made the overwhelming majority of the available money from the content that we as participants created. Quantity was often rewarded with increased visibility, ensuring popularity begot popularity.
These platforms still dominate but they have been joined in a pragmatic "Web3” by a swathe of supporting services that provide the infrastructure (distribution, payment portals, marketing tools) to divert some of that money to the creators—while keeping a slice of it to themselves, naturally.
Turning more of this legacy attention into revenue is the next challenge for lots of people and, while the mechanisms for doing so are now within the reach of many more, those with the greatest success tend to be those that benefited the most from the participatory internet, whether that’s because they simply put more time and effort into it, and/or more likely because they belong to a class of people (typically white, male, middle-class, tech-adjacent) that have traditionally benefited in all eras of the web to date.
The article goes on to discuss crypto pyramid scheme excrescences such as NFTs as a mechanism for some of this value and exchange stuff, but this is of no interest to me; I’m more confident than ever that we’ll look back on such dreadful things as a minor footnote.
reallifemag.com
See also: Back in Etcetera #4 we saw an article about the imperfection of the Indieweb and this is important in understanding a broad nostalgia for a gratis and libre web. In an interview with Ezra Klein, Tressie McMillan Cottom offered her thoughts:
There was a class of thinkers, a class of writers who came up in that web 2.0 that does feel like, yeah, we lost something there.
There was a humanity there for good or for bad. Humanity is messy, but there was a sense that those ideas were attached to people, and there were things driving those people, there’s a reason they had chosen to be in that space before it all became about chasing an audience in a platform and turning that into influencer and translating that into that — before all that happened, the professionalization of it all. And that’s what I think we’re missing when we become nostalgic for that web 2.0. I think it’s the people in the machine.
Having said that, I am very resistant to nostalgia as a thing because usually what we are nostalgic for is a time that just was not that great for a lot of people. And so what we were usually really nostalgic for is a time when we didn’t have to think so much about who was missing in the room, who wasn’t at the table. So when I talk to friends, and especially younger people coming up behind us either in the internet or in writing spaces, we’re like, that time was horrible for young queer people.
They talk about looking for little safe pockets of space in web 2.0 world where it was still very OK to be homophobic, for example, in those spaces and our casual language and how we structured that kind of thing. And they love being able to leave that part behind in this new world of whatever the web is now, both a consolidated and a disaggregated new web.
The discussion of internet counterculture that we saw all the way back in Etcetera #1 touches on some of this with regard to identity and authenticity:
This is very different from, say, 1990s ideations of IRL counterculture, where there was a premium on unmediated authenticity and “being real” (think MTV Unplugged). Now “selling out” is tying your online identity to your IRL life and real name.
In 2020 Jason Parham wrote about related issues in Everything Is Becoming Paywalled Content—Even You and described the changes as a move to an internet “built around more purposeful connections"—only to the things, people, and experiences we want. In this sense some of our subscription fees are there merely to bypass algorithmic choices.
Aside: It would be remiss of me not to mention that I have activated the paid component of this newsletter service. Please note that I am not Geldofing you here, I do not currently intend to paywall this weekly newsletter, and I won’t be plugging membership on anything like a regular basis—my views most closely align with Tim Carmody’s ideas of giving people the opportunity to unlock the commons for others. If I started a site or newsletter based around, I don’t know, advice and how-tos for digital publishing—a sort of written version of some of the consultancy I do—then I would consider that to be a ‘product’ that I might charge for. Until then, I’ll content (lol) myself with this as a general interest newsletter with optional membership and will merely hope for more of those delicious attention metrics.
What the ephemerality of the Web means for your hyperlinks
What the ephemerality of the Web means for your hyperlinks
A different aspect of a shifting internet is linkrot. This is where content (or the site hosting it) moves or is removed and the link to it becomes useless. This article by John Bowers, Clare Stanton and Jonathan Zittrain discusses their research into links from nytimes.com to external sites to discover how many have been subject to linkrot.
This is something I have worked closely on with almost every professional job or contract I have had. There are a bunch of imperfect, siloed ways to mitigate it—from adding more intelligent link-checking functionality into a CMS to using the Internet Archive to keep content available after it has been removed. Linkrot is tough to keep on top of, and without a more standardised solution it will only get harder and more resource-intensive in the short-to-medium term.
www.cjr.org
The History of Pitchfork’s Reviews Section in 38 Important Reviews
The History of Pitchfork’s Reviews Section in 38 Important Reviews
A more celebratory take on content. I don’t visit Pitchfork as much as I did in the 2000s and 2010s—their Best New Music feed is still in my feed reader, so I just look at those reviews and the occasional feature—but I am pleasantly nostalgic for the way it deepened and broadened its output over the years from its scrappy zine beginnings into a wider variety of genres.
This article is essentially an oral history of some of the site’s most notable reviews. I had a somewhat strange experience reading it, as I kept hoping to see particular albums from this era featured (You Forgot It In People, Funeral, Body Talk, Visions, Black Messiah, The Epic) and, well, the hits just kept on coming. (Except Clap Your Hands Say Yeah, wtf? That review seemed a big deal at the time.)
pitchfork.com
Aside: The article accompanies a new explorer tool where users can quickly find albums from a given artist to see related reviews and a small piece of datavis on score distribution. It’s good, but it could be even better. I cannot understand why this is not done more by any website that publishes lots of the same content type—reviews, recipes, articles, podcasts, whatever. Let people find things more easily!
I once had a conversation with a very famous publication about adding a content explorer tool where you could say, for example, I have 5 minutes and I want to read an explainer piece on Middle East conflict that was written in the past 2 years. Nothing came of it, as is often the case—most publishers are either happy with a very basic filtering system, or not having one at all, and you’d be surprised at the number of them that say things like “we’re going to do it with AI!” despite them not really having any idea what that means, and the project never ships. Alas.
A Brief History of Netflix Personalization
A Brief History of Netflix Personalization
More on surfacing content for users: this is a history of Netflix personalisation and recommendation systems from a former employee. It’s interesting but fails to explicitly mention two key things, likely because they demonstrate the imperfections of the algorithm.
The Napoleon Dynamite Problem is a longstanding issue where a film (or other piece of platform content) is very polarising and the recommendation algorithms find it difficult to predict whether a given person will love or hate it, regardless of their other preferences.
The other, related issue is that Netflix never really tries to recommend you the perfect film. It’s too risky. As Hannah Fry says, platform recommendation algorithms merely offer you songs and films that are safely good enough to insure you against disappointment. Sure, giving you something that you’d rate with 5 stars is great, but they only need to do this very occasionally to generate that feeling of great discovery. The primary strategy is one that could be considered safe, perhaps defensive; it aims to keep you using the service by steering you away from the 1-star material, and ‘good enough’ is, well, good enough. As CEO Reed Hastings famously said, Netflix’s main competitor is sleep.
gibsonbiddle.medium.com
Etc.
The Millennium Falcon, covered in greebles—bits of model kits.
The Millennium Falcon, covered in greebles—bits of model kits.
Brian Moore
Told off by my 4 yr old twins for being unkind to Bing. All I said was I hope he drowns
  • Most weeks I include an album recommendation and some people have been kind enough to reply and say that they like them. Here’s a new playlist that I intend to update each week. To start, I’ve included some things that have appeared in the newsletter along with some old favourites that I listened to this week.
  • If you’re looking to find interesting newsletters, I’ve found The Sample to be useful. You tell it what topics you like, it sends you a new newsletter each day, then you rate them so that it learns more about your preferences. Thinking again about the Netflix recommendations, I’ve subscribed to The Sample for a couple of weeks and I now rarely get a dud. Lots of 3- and 4-star stuff, which is good enough. Hopefully like this newsletter.
Don’t miss out on the other issues by Matthew Culnane
Did you enjoy this issue? Yes No
Matthew Culnane
Matthew Culnane @coldbrain

Interesting & in-depth articles on lots of topics with added context & further reading from my archives.

You can manage your subscription here.
In order to unsubscribe, click here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Created with Revue by Twitter.