But in their early days, their praise was well earned. I was a designer with a focus on typography at the time, and medium's defaults were an order of magnitude better than any other non-custom publishing platform at the time.
Big, readable serifs, intentionally constrained column widths (and corresponding line length), and lots of attention to the details and best practices that ease the pain of reading things on the web. They're cluttered now, but they were intentionally uncluttered then, which isn't exactly typographic but is worth a lot in terms of cognitive burden.
To the extent that this is table-stakes now, and no longer a differentiating factor, medium is mostly to blame IMO. They showed that doing web type the way typographers told you to got slam dunk results. Everyone copied their style and the web is vastly more readable for it.
So while they lived long enough to become the villain, there's some gratefulness due to their former selves as well.
----
I personally do things the way you've described (markdown > jekyll > GH pages), but I think medium's value prop now is ease of distribution. You get bolted onto the homepage and other people's articles as part of a graph of other people's work, and writers find that valuable.
But zero ways to get people to read it. I only read his post because it's on HN. And I read Medium stuff regularly on Medium.
The chances of most people discovering some random blog? Not so much. Whereas Medium has concentrated eyeballs and knows their preferences to serve them stuff that's close to their tastes...
I totally get where you're coming from, but I'm not convinced that Medium helps at all. I mean, this article was posted to Medium. Medium didn't put it in front of you. So how is Medium helping?
I used to cross-post everything I wrote to Medium, because I thought it was a good reader acquisition strategy. But posting on Medium doesn't automatically get you readership -- you still have to go find readers and rely on them sharing stuff on traditional sites like HN/Reddit/Twitter/Facebook. And none of that sharing process requires your content to be on Medium.
For all of the effort I put into formatting articles well, and making sure that I matched popular writing styles, and adapting things, a post I put on Medium will get <20 readers over the course of a week.
In contrast, I had one good post get traction on HN, and on average my self-hosted blog now gets more visitors than that per day. So now I don't even bother cross-posting, because I don't see how Medium helps me get content picked up on other social sites.
Medium has a ton of recirc features that drive page views within the ecosystem. That is the entire reason why Ev started Medium, it wasn't about giving people better typography or whatever.
The reason the product is kind of messed up is because they made some financing decisions that may have not been correct in retrospect, so now they've had to go back and try to retroactively design the product to match their capital structure. But it's not like they wouldn't have had tons of other problems if they were bootstrapped, the reality is there just aren't any good financing options for consumer web startups right now.
Maybe this is the problem. I'm sure as a professional writer you need to keep up with reader interests (rather than setting up to do the blog equivalent of Finnegan's Wake), but you don't want to be too interchangeable either.
It's like you had to decide each month from each cake you want to eat (there's an infinite row of cakes) and you vie for the huge ones -- together with countless other people, so you get really small crumbs. Go for a smaller cake and you may get actual slices with frosting and all.
Medium put it in fronts of tons of other people though -- someone of which posted it here.
And I discover posts on Medium alone every day.
Plus, Medium posts don't feel like a wasteland -- they have a healthy number of comments.
About 80-90% are from users with a large following on Medium.com, and many of those are also 'online influencers' with decent sized Twitter/YouTube/LinkedIn/whatever fanbases as well.
Oh, and forget it if your story isn't a paid story. About 90+% of Medium's promoted stories are members only stories now, with the home page I got just now literally having every single story outside of the 'popular on Medium' column being a paid one.
If you're not charging for the story and don't have a pre existing fanbase, at best you've got the end of article suggestions and latest stories link to rely on and the latter has about twenty pages of mediocre stories for every good one.
Like on many platforms now, it seems the popular just get more popular, and the rest battle for the scraps or the off chance they'll break out.
It does through emails and their homepage, which is one major reason people publish on Medium despite the downsides
But in practice, is Medium currently helping people build audiences in a significant way? My experience suggests no. It was easier for me to do that with a self-hosted site.
I'm not sure exactly why this is -- maybe it's that Medium's discovery algorithms take into account correlations between existing readers, so if you don't already have a bunch of readers you'll never get picked up. Maybe Medium is just super crowded, so the odds of you getting recommended over someone who's already popular are low.
All I know is that my readers are return viewers (which self hosting helps a lot with) and people coming from Hackernews, Reddit, Facebook, or curated link aggregators. I don't personally see any data that suggests Medium is driving any traffic to my content. And at this point, universally across the board, my self hosted blog performs better than my Medium blog.
I'm sure that varies from publisher to publisher, but I think it's simplistic to say that Medium by default drives traffic. I suspect that for small sites it's a lot more complicated.
I actually don't think Medium ever helped people build audiences in a significant way, with relatively few exceptions: Medium has always been structured to build an audience first for Medium and second for "publications" hosted on Medium. But it's never been particularly good at building audiences for individual authors.
While I'm aware this is anecdotal, there appears to be very little correlation between the number of followers one has and the number of views any given article gets: you can have several hundred followers and have some articles that get thousands of views (usually because they were included in someone else's publication or linked from somewhere else on the web), and others that get maybe a couple dozen views.
> I don't personally see any data that suggests Medium is driving any traffic to my content.
It is not. Readers that start on Medium by and large stay on Medium.
Yes, Medium takes into account who you follow and what subjects you've liked, which is one reason Medium article collections like HackerNoon exist.
I feel Medium is for writers who don't take the time and effort necessary to self market their own content, which is not easy for many people. The platform is just like everything else in life. Are you willing to make concessions for its benefits? I don't see anything wrong with it as long as people are aware of the price of the Faustian deals they make in exchange for a little magic. The same applies to Alexa, Facebook, Waze, etc...
Well, yeah, but a post on a self-hosted blog can be posted to HN just as easily as a post on Medium can. And it's not like having "(medium.com)" next to the submission is going to drive upvotes any more than having "(yourdomain.xyz)" would. (For me it actually has the opposite effect; there's so much empty "thought-leader" drivel posted at Medium that seeing that domain next to an HN submission makes me less likely to click through.)
So I guess the question becomes: does Medium function well enough as an aggregator of its own content to make posting there worth the tradeoffs? Does it drive enough traffic from other Medium readers to your own Medium posts to justify giving up everything you have to give up to use the platform? (That's the theory behind Medium, of course, but what I'm wondering is how well it actually works in practice.)
I observe, Medium posts are now the equivalent of most TED talks.
Most are useless but the rare good ones make it hard to write the platform off as a self-promotional junk delivery mechanism.
Exactly, the platform isn't an indicator of quality. In the long term, how could such a platform be?
I stopped doing both, though: you can't read too much for free, but more than half of the content was not worth reading (for me), except maybe for quick skimming. There definitely are articles on Medium that are gems (again, from my POV), but I hope them to be brought to me by social networks anyway.
For authors, though, Medium must be pretty valuable, bringing relevant readers, and maybe some ad revenue. It does have a discovery feature which may be not brilliantly great, but is still quite OK, especially if you have time to spend on lighter reading.
Not very high if you browse, but adequate if you get a link or two from other sources.
I wish they adopted the bandcamp model: quite a number of tracks for free per day, but you can instantly pay for the one you liked, the amount you like (above a certain minimum). This way, the articles I personally find valuable would be both easier for me to find, and paid by me higher than the slice of the common subscription funds they receive now.
I personally rarely click on them, but I can imagine some titles can be hard to resist. That make interaction-driven recommendations, which are becoming the dominant way of content exploration, a bit of a dark pattern.
Perhaps we're only an inch away from 'article autoplay'.
Have you ever seen news sites that let you seamlessly scroll to the next article, by dynamically appending it at the bottom of the one you're reading, and updating your URL as you enter it? The ones where it's ridiculously easy to get lost, because scroll space keeps expanding (sometimes both ways) and the URL keeps changing?
Someone's gonna figure out soon that the code for loading those articles could use a recommendation engine.
Otherwise I just ignore Medium articles. Maybe it is me but most of the Medium articles I stumble upon are pure garbage.
It's not just you, a lot of them are.
Look at how many are just "python deep learning tutorial" recycled for the millionth time, but with some fundamental misunderstanding or shortcut, or they're just lifted almost verbatim from the starter example or the documentation with nothing actually meaningful added. Often there's lots of "here's how to do x in y" that somehow manages to not understand good practices at all.
Most of our traffic comes from hackernews, reddit and similar. However because our blog is too niche?
Edit: just because a few people (myself included) have a negative reaction to a Medium link doesn’t change the fact that many Medium links are upvoted on HN.
EDIT: adressing your edit, do you have any evidence they wouldn't be if they weren't on medium.com?
I only remember a handful of them, one in particular that is now a regular bookmark of mine (it’s https://www.valdemarsro.dk and in Danish so likely uninteresting to most), and the one thing that the ones I remember have in common is that they are their own things.
I don’t think there is anything beneficial to being a medium blog for this exact reason. I may read your article, I may even really like it, but it takes more than one article to get me to stick around, and because every medium article looks the same I can easily end up reading more than one without noticing it.
Having a unique domain helps as much as having a unique look.
Perhaps someone should develop a platform that creates the same discoverability, but while allowing people to retain control of the content?
I have somewhere between seven and nine[0] articles out and my work has been translated into French, Portuguese (twice), Russian, and Uzbek.
Medium isn't a source of traffic unless you're one of the very few writers they intentionally push. People share high quality work or new ideas, and audience engagement helps too, so link to your Twitter and share your email.
[0] Since starting the site I've started writing for other publications and the audience flows back and forth between the sites, so it depends on what you think counts as an article.
Maybe what we need for blogs isn’t a platform, where authors come on and content is hosted centrally; but an aggregator, where each author self hosts but content is still discoverable by a wider audience.
A good example of this would be Apple news.
You could always post on medium and on your own blog though, and post your own blog here, for example.
Getting other people to read your post is as difficult as getting other people to read other people's content - you post it where you frequent, and if people like it, they'll reshare it. Building a larger audience requires work, and any attempt at cheating that is unsustainable at scale.
I actively avoid links to Medium, even on HN.
[0] https://casual-effects.com/markdeep/
[1] https://github.com/susam/texme
PS: For more ready-to-fork (open source) themes for GitHub pages, see the Dr Jekyll Themes directory / listing (200+ and counting) @ http://drjekyllthemes.github.io
Instead I recommend people buy a domain, create a DigitalOcean account, have DigitalOcean manage it for them, and then create a single $10/month Droplet with their one-click deployment of Ghost. Now you have your markdown files and a self-hosted, beautiful blog.
That being said I am a "power user", so I use Hugo and Netlify, but there are deeper requirements to make that work. I would love to see a one-click deployment of a Netlify based blog.
For an official reply, you can send an email to the mods [email protected] and they can take a look at the history of comments of the user.
[For example, this is an example of a comment that is not relevant at all https://news.ycombinator.com/item?id=16961492 ]
[1] It's fine to reply to the technical questions that appear about your page/app/product. It's not fine to add a link to your product in unconnected threads.
https://news.ycombinator.com/item?id=16961492
Dan answers if he likes you. It's as simple as that. If he doesn't, he either filters your emails and never sees them, or refuses to answer them.
You are one data point. Please don't extrapolate it to those of us who have tried for months to get answers to specific questions.
Since you won't stop posting abusive and misleading comments to HN, we're not going to unban you.
https://news.ycombinator.com/item?id=18412225
https://news.ycombinator.com/item?id=18294434
https://news.ycombinator.com/item?id=18224895
Your memory is very, very different from mine. I remember Medium being heavily criticized by (sorry, lack of a better term) "tech nerds" for stuff like the article brings up when it first came out. People complained about it not working when Javascript was disabled and bloat for a text website, load times, etc., all sorts of stuff. People said they refused to visit Medium. In fact, it was so negative I'm a bit surprised it got as popular as it did with tech blogs.
>Now, we need all kinds of tricks, add-ons and blocks to just read five-minute posts during our commute.
Only "tech nerds" care about those sorts of things. Joe User reads Medium just fine without any tricks, add-ons and blocks.
It still needs some more slimming down, and small aesthetic changes would be welcome, but here it goes.
Currently trawling your archive
I'm a terrible writer, but, there's only one way to get better! I really, really like your own blog, too :)
Is this only a thing for phones? I never use my phone for web browsing, but I use my web browser. All I have on that is uBlock Origin and I have always found Medium pages completely clean and pleasant to look at with no clutter or ads.
It's 2018 and there's still no standardized format for a blog's complete content (text, images, comments/credentials for comment provider) that we could use to move quickly from one site/setup to another. It would be nice to just get all Tweets even into Hyde or whatever.
I just need RSS and a decent data parser for the blog entries list and it'll be all set. Not recommended for you to use it, just build one, it's fun! And easy, and you'll learn a lot.
Or maybe objecting had the same effect it had back then as it does now: absolutely nothing.
It's like me saying "the government is corrupt!" and, any time something happens or some time has passed, "shoulda listened! ;)" -- Seems a bit masturbatory at best.
It's the same with OSX, often praised but at some point the UX got so bad that you'd have to install system updates through their music player iTunes.
It's 2018, and there are a gazillion ways to get yourself in control of your stuff. The next free framework that comes, which promises a clean, non-distracting view, will also (most probably) eventually add advertisements, because, well, corporations need money.
In my humble opinion, just write your Markdown files, and use one of the bajillion Static Sites Generators to create/host them, or fork a framework that already does this such as [1], copy-paste them in, and you're set.
[1] https://github.com/barryclark/jekyll-now