This is a great point, and one that we're honestly in the process of trying to make progress on.
Ideally, I would love to have both:
- Writers and creators on Substack are in complete control of the brand and feel of their publication
And:
- All publications look & work well
- Readers get the benefit of already understanding some of what this thing is, which makes it easier to subscribe with confidence
- We can continue to ship rapid improvements across all of Substack
In practice, there are tradeoffs involved here and we're trying to figure out how to push both sides as far as possible, while maintaining a simple and powerful product.
First of all, huge respect to the Ghost team. Their open source contributions are valuable, and their approach to theming enables some great-looking things. That said, some important corrections:
Substack is not "powered by Ghost". Rather, we built our own theming API that’s compatible with themes built for Ghost, including those built by third parties.
The Free Press is using a modified Tripoli theme, built by Ahmad Ajmi, under a paid license. This is how this is supposed to work. It's good for the theme developer if we support this – you should check them out here.
https://aspirethemes.com/themes/tripoli
This was relatively quick to build for Substack devs, because the structure of Ghost sites matches Substack fairly closely.
With respect to the search library, this is an open source library that we are using in a fully compliant way. John's own screen shot shows that we don't load it "from Ghost’s own CDN", it comes from jsDelivr
https://www.jsdelivr.com
This is a standard way to use an open source library. It's pulling from the version that the sodo-search maintainers published to NPM (thank you!).
It is a good point that we should lock a version, so that if they accidentally published a minor version revision with breaking changes it doesn't cause problems for us. We’ve fixed that.
We’re grateful to the developer of the Tripoli theme and to Ghost for its contributors to open source work. We’re exploring ways to give writers more customization on Substack. This is one approach we’re considering but it’s too early to know if we’ll scale it up.
And @JohnONolan, thanks for the note at the end about potential collaboration. In our minds, we’re on the same side of an important battle for a better internet. We’re definitely up to chat.
I'm one of the Substack founders. My (obviously biased) view is that we are trying to do something that is genuinely important, and that we're willing to sweat the details in service of the people who use Substack. I'll try to give that quick barbell of lofty vision and practical details.
Big picture: we believe that what you read matters, and that therefore great writing (and great intellectual culture in general) is valuable. In our view the great promise of the internet is that it can unlock a flourishing of culture, but the first phase was dominated by a land-grab for attention which gave us the current landscape of ad-driven platforms which optimize for the wrong things and don't serve great writing. Substack is an attempt to create an alternate universe on the internet, with different laws of physics, that fulfills the original promise. Where writers can make real money by earning and keeping the trust of an audience who deeply values their work. Where readers can take back their mind, and decide for themselves who to trust and how to spend their attention.
Practical terms: we try to focus really hard on serving the people that use Substack. The writers, obviously, by giving them something that genuinely works. I think of this as "we do everything for you except the hard part". Which means if you can write something great, the job of the product is to handle the rest. This means not just the right features, but smart defaults everywhere that help you succeed. For readers, we try to make the experience very clean and frictionless, and communicate right from the get-go that this is a place that respects you and your attention, to the point that it might be worth paying for. The things others have noted - smart defaults, fast loading, clean design, etc. etc. are expressions of this. We're focused serving people, and we're not shy about using 'boring' technology (like email, for example) to make it happen.
Putting this together, the magic of the Substack model is not that it gives writers a way to get paid for doing the thing they might have done elsewhere. It creates a system where the kind of thing you do to succeed is qualitatively different and better than what succeeds elsewhere. For writers, that can mean getting paid -- sometimes very handsomely -- to do the work you actually believe in. It can unlock this for people who weren't professional writers before. And for readers, it means a lot of the best writing to be found anywhere is on Substack.
Yes we're keen on this. In fact, although you can't add RSS in the app (yet), you can add it on the web at https://reader.substack.com/inbox and it will show up in the app
However, I can't promise that we won't push Substack content. We will :)
Thank you this is an interesting perspective. Substack today is definitely targeted at people who want a full stack thing and don't want to worry about coding one's own template etc.
As someone who does want to self host, which Substack services would be valuable to you? What do you feel like we could offer you?
The model right now is really an unbundling. The direct relationship between writers and readers is what makes Substack work: as a writer, your incentive is to earn and keep the trust of the audience who deeply values your work. That's not just a good way to get paid for work you're already doing. It's a model that allows and rewards a fundamentally different and better kind of work that the work you would have to do if you were e.g. trying to please something like the Spotify algorithm.
That said, bundle economics are real. And so while we wouldn't and couldn't do some top down bundle, if there were a way to do bundling that maintained the direct connection, and put writers and readers in charge (e.g. writer self federation, or readers buying several subscriptions at once) that could be very interesting in the future.
That is incidentally a big part of the answer for (3). We are very public about how we think about this, and the first of those posts was written before there was any real pressure on this stuff. We talk about this with folks we are hiring, and it helps people choose for themselves if the approach we take is something they are excited to get behind.
Our goal with the app is to give a seamless upgrade to the email experience -- which is why the home page works just like an inbox -- while having writers retain ownership of their list (which therefore gives them exit rights.)
We don't want to be a walled garden. We want to make a great reading experience, with porous boundaries. If you publish on Substack, it goes everywhere - email, the web, other networks, but as the writer you can pull your most valuable audience to the place that you own and can get paid from. If you read on Substack, you can read things on Substack, and then maybe things from other places, like RSS etc. I like the idea of having emails that you can get stuff delivered to.
Reading articles may be an almost perfect fit for web, but we see a really clear pattern among the 70% of our users that read in mobile: they discover stuff to read on the web through links that get shared, but once they sign up they overwhelmingly read in their email client.
Email is actually pretty great for this, and email is especially powerful for giving writers direct connection to their readers. But there are limits to what you can do there, and stuff like community discussion, audio and video, and even 'not accidentally going to the promotions tab' can get a big upgrade.
Somewhat related: developing for email is still a pretty big pain. It's kinda like the bad old days when you had to support IE6 - lots of people still use old versions of outlook or whatever.
We'll get you a real answer about how the tech stack is today.
But for now I can tell a war story that might be fun. I'm one of the founders, and am in a very much non-technical role now as CEO, I did write a bunch of the very early code (some of which people curse my name for to this day.)
When we were starting, I was limited by how many new languages/frameworks I could learn at once. I started writing the backend in python, because I knew it a bit. But our first writer often needed to use Chinese characters, and in python 2.X I could never get unicode strings to work properly. I couldn't upgrade to python 3, because on google cloud I would have had to learn Docker and I was already learning too many things at once.
Eventually I got so frustrated I threw out several days work and started the whole backend over, with node + Postgres hosted in Heroku. This ended up defining much of the stack we use to this day, which might be good or bad depending who you ask. At least unicode works though :)
One of the good things Substack can do is put competitive pressure on traditional outlets. If people want to go independent, having a good way to do so helps them. But also that possibility creates pressure on existing institutions to give writers more freedom, pay them better, etc. etc. I don't think it's a coincidence that you see more legacy publishers starting "newsletter" divisions that give writers more leeway. All of this is good for writers in our minds and we're happy for it.