Overview Link to heading

Academia is in a crisis of communication. Academics do a lot of communicating, but we rarely reflect on how and why we communicate. There are very active debates on the reform of academic publishing, but they rarely make it into the mainstream of academia.

Both “the paper” and “the book” have become the standard units of currency in the academic world. So entrenched are they that in the modern day it is often required to publish a set number of articles in journals with a specified amount of prestige (or books with a prestigious publisher) to have a chance at an academic career. The research really is secondary, it is the venue of publication that matters to career advancement. Academics do the research, write the articles, review the articles and read the articles for free, but for some reason it is the commercial publishers that make fantastic profits off all this activity (see this post by Björn Brembs for more background).

I’ve been thinking about academic communication a lot and I decided to write up some thoughts as an essay, perhaps make a series out of it. By “academic communication” I mean communication as an academic, whether this is within academia among academics or if it is between academia and business, academia and civil society or academia and the general public. In this first essay I’ve chosen to focus on some modern means of academic communication: websites, social media and e-mail. Each of them is not enough to make an academic career, but they represent some of the best means we have to evade traditional academic gatekeeping.

Table of Contents Link to heading

Website and Blogging Link to heading

I’ve experimented with and considered a lot of publishing channels and venues for my academic and personal work, but by far the one I enjoy the most is my website and blog. It is my personal corner of the internet where I can write whatever I want to, make it look like whatever I want and can try adding code, diagrams and citations wherever I please. The website is in dark mode by default, because I like it. Some people don’t and that is fine, but of all readers I probably spend the most time having to look at it, so dark mode it is.

I wouldn’t go quite so far as to recommend that every academic have their personal website, but creating websites has become so easy that I believe everyone should at least consider it. Most people think this means having a full-fledged blog and tons of content, but in reality a basic CV, a list of publications/data/code and some contact information (e-mail, social media profiles) is perfectly fine as a first website. Why? See below on gatekeeping, but the short answer is that a personal website has far fewer gatekeepers than social media or even an institutional page and is portable across corporate failures and academic transitions.

Essays Link to heading

Essays allow me to experiment with my writing, try new things, be informal, be formal, be cranky, be professional and everything in between. Some essays have citations, some have diagrams, some have code, some have none of these. My essays are certainly too long by any traditional standards of web publishing, but on my own website there isn’t really anyone who can stop me. I don’t use Large Language Models (LLM) for any of my writing, because the writing is the point of it, not the product. Personally I think this comes close to how scholarship is the most fun: chaotic, experimental, curiosity-driven and unique.

RSS Feed Link to heading

I have an RSS feed to push new posts to people using feed readers, free of charge, no account required, no tracking involved. Some people consider this ancient technology, but it is in fact a modern, important, open, standardized and decentralized means of publishing content that many ignore because we have forgotten that there are good things that are not expensive Software-as-a-Service (SaaS) “platforms” with aggressive vendor lock-in. The entire podcast industry runs on RSS and I hope that RSS will make a comeback among the general population with an easy-to-use feed reader. I use RSS a lot to get my regular news intake, check academic journals for new articles and read my favorite blogs.

Gatekeeping Link to heading

Almost every communication channel these days has some gatekeeper that must be convinced to publish your work. Some gatekeepers are more hands-off, some have very strong opinions on what can be published and how it should be written. You’d think that traditional academic publishing has the strongest gatekeepers (seeing as there is serious science to be done) but personally I think it is social media.

Social media algorithms are just brutal in their selection of which content gets attention and which doesn’t. If an academic journal rejects you, you just try the next one. It might not be as prestigious as your first choice, but there are usually at least a dozen alternatives to choose from. If your work gets published elsewhere, it’s usually just as well as long as the article itself is good, except if you really need a publication in a Nature-equivalent in your field to get a job.

With social media on the other hand, there is no easy way to switch platforms (except the Fediverse, which allows communication between all the different ActivityPub-compatible clients like Mastodon, Friendica, PeerTube, Pixelfed etc.).¹ People cross-post, of course, but most social media algorithms tend to prioritize the same type of content (catchy, short to match modern attention spans, emotional, divisive, simple), resulting in heavy pressure to conform to a particular archetype of content. If your content doesn’t fit the type, it ends up in algo-purgatory, never to be seen by more than the three or four die-hard friends who check the chronological feed on your profile every week. The most worrying development in social media is the quasi-suppression (or down-ranking) of external links, turning platforms into strictly walled gardens and cutting off means to escape the platform. This hostility to linking is really a threat to the open web and together with AI plagiarism it damages the entire information ecosystem.

By contrast there is no real gatekeeper to your website’s content. Yes, discovery can be an issue if Google doesn’t send generic traffic your way. With “AI search” this is going to become much worse, because the AI algo-washes your content and presents it as its own — without attribution, of course.² However, there are many other ways of discovery, some which can be just as relevant to academics as Google. Academics don’t need millions of visits per day, they need a dozen relevant people in their field to deeply engage with their ideas.

People learn about you when you meet them for coffee, when you meet them in person, when you give a presentation at a conference, every once in a while through social media posts, through personal recommendations and so on. As soon as people hear about you as an individual and understand that you have a website, you have potentially broken Google’s hold and freed yourself from all gatekeepers forever. Usually this takes years and sustained work, but it is an option.

Asynchronous Publishing Link to heading

A website stores and presents everything you put on it, whenever you choose, in whatever structure and design you choose. You might merrily go your way publishing for months or years before a potential reader learns about you, but when they do, they have access to all the content you published before. This gives them a ton of material to choose from, to establish your legitimacy as an author and to forward to other people, if they think it’s good. One of my happiest internet experiences is discovering a mature blog I had overlooked before and going through the archives to see what else the person has written. I enjoy forwarding links to old articles, because some ideas are still valuable years later.

The contrast is, of course, social media feeds, where anything you write pretty much vanishes after a day or two, never to be seen again, except be intrepid explorers of your profile who can stomach scrolling through your unstructured flow of thoughts at length on a platform that makes it as painful as possible. I wonder how much wonderful ideas we lose forever, simply because social was the only place they were ever posted and they never received the audience they should have because the computer said no.

Speed Link to heading

This may be a pet gloat of mine, but my website loads fast. It’s not because I added some clever technology, but because of the absence of technology. There is (almost) no Javascript that must be loaded to display the website,³ there are no tracking scripts that must be loaded, there is no overloaded Javascript framework involved, just some HTML, CSS and occasionally some down-scaled image files, if the page or post warrants it. Contrast this with bloated sites like GitHub or Linkedin that take several seconds to load, even on a 100 MBit connection and a modern computer. Even regular blogs these days are so full of unnecessary frameworks that it makes me wonder why I am loading all this Javascript to read some simple articles. And I do have to load it explicitly, because I have NoScript on by default. Since I have to check and re-check my website often to revise it and upload new content, its speed makes this a surprisingly enjoyable experience.

Analytics Link to heading

My website doesn’t have analytics. While I have some strong philosophical opinions on tracking and generally oppose it, the best part about being tracking-free is that I save a ton of time not looking at analytics. Tracking never mattered to the purpose of my website in the first place. Checking analytics all the time can consume you.

In the past I spent much too much time looking at analytics on Linkedin. What did I learn from it? Nothing useful, really. Some posts do really well, some don’t. I often can’t say why, except it being down to luck that a few people liked the post in some crucial early window during which the algo decided whether to promote it and then it spread fast. A lot of people distribute likes primarily to those whom they like and respect as a person to help their content get noticed. This is great, but again, I can’t tell if people like the content or not unless they leave a substantive comment. Thoughtful comments are rare and again, you’ll usually get them from people who enjoy your work in general, not so much from a cross-section of the population who honestly respond to your work.

In the end, analytics do not measure what matters to me. It matters to me that people closely engage with my writing, learn from it, perhaps change their behavior and become better legal and technical professionals. I usually discover if this worked when people write me thoughtful e-mails, have a good conversation at a conference or recommend my work to other people. None of these feedback signals can be captured by a regular analytics platform, so I don’t bother. In the years I’ve been running this website, I never felt like I made a mistake.

Social Media Link to heading

Social media is kind of the new kid on the block in terms of digital distribution channels, even though it has been around for years now. Social media quickly become critically important for academics to disseminate their research, in a way that it is hard not to have a social media presence as a young academic these days. I’m not sure if Facebook was ever important for academics, but Twitter quickly became the go-to place for “serious” social media accounts, including academics, journalists and politicians. Of course Twitter later fell off a cliff after the Musk takeover and turned into a Nazi hellhole now called X, but it was useful and valuable to academics while it lasted.

In most disciplines in the academic world there is a sliding scale of importance in academic journals. Papers published in “good” journals traditionally receive more attention than papers in “mediocre journals”. Journal importance is both blessing and curse. On the one hand, getting into the right journal provides much-needed attention for good work, but on the other hand, the more important the journal, the more conservative it is with regards to methods and content. Social media levels the playing field a little. Even if you don’t make it into Nature, social media is still a chance for your research to be noticed. Perhaps only be your peers in your circle of PhD students, but you’ve got to start somewhere. This is especially important for people like me who work with very much non-traditional methods in a very conservative discipline (law). Social media often allows me to make an end-run around the traditional gatekeepers.

Mastodon Link to heading

Mastodon is my absolute social media favorite. I created my account in November 2022 during the big Twitter migration, have been active on the site for three and a half years and haven’t regretted it at any point. Mastodon does a lot of things right in a way similar to early commercial platforms, before they deteriorated in quality to wring out every last bit of attention from their users and advertisers.⁴

These are my personal Mastodon highlights:

• Unexpected and unique perspectives that would be canned by any standard social media algorithm
• Plenty of academics from all disciplines, expert tech people and non-traditional individuals
• Mastodon software is open source
• Powered by the ActivityPub protocol, an open standard for social networking
• Developed by a non-profit
• No suspicious algorithm
• No ads in the software itself (however, some accounts, esp. companies, post ads as part of their regular output, so Mastodon is technically not ad-free)
• Limited analytics (counts of boosts/favorites/followers, but that’s it)

While there are many good sides to Mastodon, it admittedly is a relatively small network compared to the major social media behemoths. The Twitter migration brought a massive influx of new users, but many didn’t stay with Mastodon. The website mastodon-analytics.com indicates that Mastodon had around 400,000 monthly active users (MAU) in mid-2022, peaking at around 2.5 million MAU after the Twitter meltdown later in 2022. The Fediverse Observer shows we are currently at around 750,000 MAU in February 2026. So Mastodon almost doubled its long-term active user base over a few years. Total monthly users are around 9.5 million, which is small compared to Linkedin’s claimed 1 billion users. However, there is still the wider Fediverse which is interoperable with Mastodon, and that adds a good number of users.

In the long-term I am confident that Mastodon will win in social media market share. Either Mastodon itself or other open social media platforms will slowly gain dominance. This may take decades, but I’m in this for the long run. UI/UX is improving slowly but consistently. A lot of people come to stay. However, the most important feature is its technical foundation on open standards and open source. Linux took a long time to take off, but it has gained in reputation and legitimacy every decade and is now threatening even traditional Windows strongholds such as gaming and government use.

Linkedin Link to heading

Yes, I have a Linkedin account and I maintain it semi-actively. I am not terribly excited about Linkedin (nor Microsoft) and have never been, but I must admit Linkedin has been useful at times. The German legal community is quite active on Linkedin, same as the business community in general. I believe this is deteriorating due to AI overload, but more on that later.

Working conditions in academia are quite poor in Germany (low pay, mostly part-time and short-term contracts, very few permanent positions, strict hierarchies), so it is helpful for academics at all stages pre-tenure to be able to earn money in industry. I’ve done so for a number of years now and the mix between low-paid/unpaid academic work and well-paid industry work has allowed me to succeed in paying the bills and keeping me intellectually curious.

In terms of maintaining one’s sanity it would certainly be better to read fewer posts from salespeople (including most CEOs, by the way) trying to sell inferior products with outrageous lies about productivity gains to scared customers traumatized by the AI hype machine. Unfortunately, not everything in life pleasant. Sometimes we must endure. I certainly have won jobs, contracts, connections and a reputation from being active on Linkedin. For, now, I think trading some of my sanity to be able to feed my family is worth it, but I do wonder if this will be the case for much longer. I find the value of Linkedin has been steadily decreasing since the release of ChatGPT.

Despite the huge difference in the size of the user base between Linkedin and Mastodon, I see more high-value posting activity on my Mastodon feed than on Linkedin. I follow about 3,000 to 4,000 people on each site, so I should roughly be seeing a similar amount of useful content on each site. However, since the advent of AI the number of valuable posts from people I find interesting has dropped off a cliff on Linkedin and hasn’t recovered. I assume it is some combination of the algorithm having become bad at detecting human posts, AI slop completely overwhelming the system and people deciding to opt out of posting because of poor reach. I see this at work in my own posting habits. The reach of my Linkedin posts and reactions to them are down to about 10% of what they were before 2025. At some point you start to wonder whether the effort is worth it.

Linkedin knows this. Some time in 2025 they changed the creator dashboard to not show daily impressions/reactions by default, but instead only show the cumulative impressions/reactions per interval (7 days by default). What this means, of course, is that the numbers and the line always point upwards. You don’t do this if you are confident that many active people read and comment on timely and valuable posts. This change, more than anything else, tells me that Linkedin is failing, possibly irreversibly.

My personal approach to Linkedin at the moment is to avoid reading the feed, check for messages and comments on published posts every couple of days and occasionally post something about my work. Since 2025 I have rarely written posts targeted at a Linkedin audience, but instead write them for Mastodon and cross-post them if it feels right. I still announce larger items (papers, open source projects, blog posts) on Linkedin, but again with the text that I write for Mastodon.

E-Mail Link to heading

E-Mail is old. It primarily used to be a means of direct communication among individual scholars and groups of scholars. There were newsletters, of course, but this usually indicated some kind of institutional setting, making it necessary to regularly communicate information top-down and one-to-many.

This has changed in recent years with the rise of professional newsletter writers who produce a regular newsletter for a group of paying subscribers and, as far as I can tell, for some academics and journalists this is a viable means of making an income or at least some nice side-money. This is has been made easier by SaaS platforms targeting this market segment, which I prefer not to name because you probably have heard of them and don’t hold them in high regard. Nevertheless, technological developments have made publishing professional paid newsletters much easier than before.

While e-mail is still my medium of choice for professional communication among peers, I’ve opted against publishing a regular newsletter or even implementing some kind of RSS-to-E-Mail mechanism that mails out updates from my website to subscribers.

There are a number of reasons why I do this:

• E-mail channels are overloaded. Most people I know are drowning in e-mails. No one knows what inbox zero looks like. Add to that the fact that very few people know how basic keyword-style e-mail filtering works, inboxes around the world are exploding with e-mail. Why write a newsletters that just gets lost in a mountain of text?⁵
• There are too many newsletters. There are all the traditional newsletters related to work and civic life one has to watch. There are many more new newsletters being created to look for a viable source of income as a writer. There are all the unwanted newsletters people add me to whether I like it or not because I needed an account somewhere and some e-commerice business development executive thinks that is an invitation to bombard me with e-mails every week. I have no idea how people who deal with this without e-mail filtering.
• Newsletter writing feel like a chore. I enjoy writing when the urge hits me and I don’t enjoy writing when it doesn’t. Turning occasional blogging into a regular newsletter would require me to publish new articles on a regular basis to not lose the attention of my audience and I just don’t feel like turning something fun and intellectuall freeing into a chore.
• Managing a subscriber database is work. Building an e-mail subscriber base requires time, persistence and a lot of marketing. Then you have to make sure the e-mail database stays safe and never leaks. All in all even with technological assistance it is a lot of work that I don’t need to do if I stick to blogging.
• Long-term storage of witing. A blog keeps everything as long as I want it to. People delete e-mails. People should delete e-mails, particularly newsletters, to free up space and retain their sanity. My own e-mail program is set to auto-delete e-mails from certain folders and my own newsletter folder is set to auto-delete everything older than 30 days. On my blog I can keep my articles around for as long as I like, in whichever order I like.
• Editing and Updates. Sometimes I make mistakes. Sometimes I want to update some post with new information. When you hit “send” on an e-mail it is gone and issuing corrections or updates requires sending new e-mails, irritating everyone. Blogging can do that, e-mail cannot.

There are certainly valid reasons for pursuing a newsletter writing strategy as an academic or journalist. Human attention from search is becoming less and so-called AI produces a ton of traffic only to plagiarize content and keep all the valuable human attention for itself. Controlling distribution via an individual subscriber database sounds nice, but in practice this is only viable for the most popular writers. All others end in social media recommendation hell on one of the typical newsletter platforms, trying to please the recommendation algorithm while never building a subscriber base large enough to leave said platform.

I’m fine with blogging.