April 8, 2021

Every Blogging Option Sucks

Except the one that makes sense for you

Every so often, I’m gripped by a compulsion to investigate my blogging approach and reassure myself that the path I’ve chosen remains the best fit for my needs.

I haven’t done this in a couple of years, but now I find myself once again revisiting the messy quagmire of blog-capable platforms and services to see which ones stand a chance of dethroning Blot. What I’ve come away with is a general sense of wonder and frustration at how such a seemingly simple problem continues to lack effective solutions.

You may think I’m exaggerating—after all, there are a ton of platforms out there, right?

There are. And yet for someone with my blogging wishlist, the list of viable options thins considerably and what remains is a haphazard collection of almost-there choices—none of which offer the balance of cost and functionality that I was hoping would be easy to find in 2021.

My Blogging Wishlist

This is what I’d like to see in my blog platform/host/system:

Easy publishing, ideally from Ulysses/iA Writer. My current setup involves writing the post text, uploading and preparing images on Imgix via Amazon S3 and pasting the links into my draft, then using GitHub to manage and publish content. It works, and I’m used to it—I’ve even automated most of it—but it still represents a lot of friction that stops me from posting as frequently as I want to.
More-than-rudimentary customization. I’m willing to give up full-on theme design capabilities if I can at least establish my own colour scheme and type stack, use my own domain, and provide decent navigation options to make finding posts easy for readers.
Excellent image handling. I want to throw a full-sized image at it and have the system automatically re-size, compress, and deliver the most appropriate version at the most appropriate resolution in the most appropriate format for each device (via CDN, ideally).
Optional email subscription. I’m an RSS guy myself, but newsletters are all the rage and I want to make sure my readers can choose to follow the blog via whichever method they prefer.
Active development. Blot has spoiled me perhaps, but I love the feeling of using a system that feels actively, lovingly developed, that has a clear sense of direction, and a transparent roadmap.
Reasonable pricing. This one is the most vague because it will necessarily mean different things to different people. To me, it means that I want to pay less than $200 USD per year all-in to run the blog. For context, I currently pay closer to $250.

Maybe I’m being unreasonable, but I wouldn’t have thought that the above is too much to ask for from a modern blogging platform. Turns out…it is.

Meet the Contenders

After a new round of research over the past few weeks, I’ve gathered a fairly anaemic list of possible options, and I can save you some reading and tell you that none of them meet the requirements above—it’s just a question of how far off they each are.

In no particular order:

Ghost
WordPress
Medium
Micro.Blog
Write.as
Grab-Bag of Rejects: Substack, Svbtle, Squarespace, Webflow, Tilda
The wildcard: Hey World

Let’s take a look at each in turn, shall we?

Ghost

I had high hopes for Ghost, particularly after the recent update to 4.0. In the past, Ghost has been a tempting choice disqualified either by price (on their managed service) or hassle (self-hosting). Despite the 4.0 update dropping with a much-cheaper $9 per month plan on their managed hosting service—Ghost(Pro)—those problems persist.

Ghost(Pro) is perhaps the only option to essentially tick all the boxes, but only on its Basic tier and above. That costs a whopping $36 USD per month, or $29/m billed annually—more than I’m paying now and significantly more than I’d like to be paying. The new Starter tier was promising, coming in at a much-more-reasonable $9/m, except the Starter tier takes away a critical feature: custom themes.

I said I was willing to give this up if I could have a few rudimentary customization options, and that’s where Ghost’s theme system falls flat. I don’t mind not being able to bring or create my own custom theme. But Ghost has almost literally zero front-end theme customization options (you can customize the look of Medium more easily than the look of an installed Ghost theme), which means that even something as simple as changing my choice of font requires me to modify the theme files.

I’m happy to do so, but of course then I end up with a “custom theme”…and can’t be on the Starter tier. In my case, there is a workaround because Ghost has a code injection feature that would let me override just enough of the styling to get the theme looking approximately the way I want it to. However, if I ever want to make more significant modifications I’m out of luck on the Starter plan.

What if I self-host?

There’s some hassle involved, but mostly at the beginning to get the server set up. The downside is that I’m then responsible for managing and maintaining the infrastructure underpinning my site. I’m unlikely to ever be as good at it as the dedicated staff that Ghost has working on its Ghost(Pro) service, and I certainly don’t have all the cool tricks and embedded relationships that allow them to offer email delivery, a CDN, threat protection, and so on as part of a single monthly fee.

If I wanted to match them for features, in other words, the cost creeps up to the point where I’m suddenly back to the same ballpark as Ghost(Pro).

That being said, if set up correctly, a self-hosted Ghost instance on DigitalOcean requires very little ongoing effort or time—and a simple blog hardly requires all the performance features they include—but it would be nice to have them. Besides, I appreciate the calm that comes with knowing all of that is somebody else’s problem.

Even so, for all its benefits, Ghost 4.0 still doesn’t allow you to turn off the membership layer entirely (all that revenue and membership stuff on the fancy new dashboard? Yeah, that’s immutable—it’ll be there whether you charge members or not), nor customize the design of the emails being sent out or any of the wording in the interface of the payment gateway for the membership feature.

I expect some of those rough edges to be sanded off over time, but what’s clear is that Ghost really wants to be the platform for paid newsletter publications. I think that’s a brilliant business move for them, it’s just not the kind of thing I’m doing here. Now that their intentions are clear, I expect they’ll continue to evolve in ways that don’t quite match up with what I’m after.

WordPress

I’ve done a lot of work using WordPress, and I’m sorry to admit that I have never liked it.

While it is both powerful and flexible, it continues to feel ugly and bloated to me, and it’s also fundamentally less of a blogging platform and more of a general web development platform.

Still, it technically ticks all the boxes—provided I assemble the correct collection of plugins and server tweaks to make it happen. The greatest advantage to WordPress is, of course, the ubiquity of support and knowledge about it. If anything goes wrong, there’s a 99% chance that someone else has encountered, solved, and documented the issue many times over.

If I were to take this route, I’d be able to spin up a DigitalOcean droplet, grab a copy of Elementor, and essentially build exactly the blog I want, safe in the knowledge that it’s powered by the de-facto choice with the most available integrations, most flexibility, and largest community.

Setting aside my distaste for how it looks and works behind the scenes, WordPress is clearly a solid choice and would cost me about $121 a year.

Medium

Oh, Medium. You can hardly blink without another pivot or overhaul hitting this powerhouse of a platform. It’s almost baffling that something can be both successfully impactful and ineptly executed in such equal measure.

Still, credit where it’s due: it remains the nicest web-based editor for blogging and the core promise of being a robust place to put text on the internet has yet to be broken.

I can’t customize much as far as typography or colour choices are concerned, but the options aren’t bad and the general consistency is part of the trade-off for participating in such a large platform of content (with the associated discovery and SEO benefits thereof).

For my purposes, the recent changes are mostly good news: they’ve re-instated the ability to use a custom domain for your publication, and the new publication design settings allow you to create sections, essentially allowing you to provide basic navigation to help readers find articles. They also have good support for email and RSS subscription, I can publish from Ulysses, and the platform is free to use; in fact, I can even opt to use their paywall as a low-effort monetization approach.

The main problem with Medium is that many of my readers don’t like it. It feels too unsteady, too unpredictable, and has made poor decisions around article design and “dickbar” clutter in the past (now mostly corrected, but the damage is done).

Frankly, I’m just as uncertain about its future as everyone else, so it feels like too fragile a foundation to build anything on.

Micro.Blog

An indie web darling, Micro.Blog is something like a Twitter-meets-blog community effort.

As a platform, it provides affordable hosting for a simple blog, with good theme support. Unfortunately, it doesn’t handle images especially well, it doesn’t offer an email subscription option for readers, and there are quirks around its handling of URLs and such that make it feel a little under-baked for my taste.

I also can’t shake the feeling that it’s being designed around short-form social-style posts more than long-form work, or that the balance it’s striking between those two isn’t quite right somehow. I’ve had a profile and lurked on the platform for ages but I haven’t quite fallen in love with it.

Write.as

Here’s one I bet you haven’t heard of.

Way back in 2015, Write.as launched as a private online writing surface. Since then, it’s blossomed into a small suite of four interconnected apps that tackle blogging, co-editing (think Google Docs), image hosting and sharing, plus a system for getting writing submissions. All guided by some refreshingly user-first principles.

Similar to Ghost and WordPress, they also provide a self-hosted option, although the paid plans are so reasonably priced that I didn’t even bother looking into hosting my own instance.

This sleeper hit not only supports email newsletters alongside RSS, it also supports basic theming, custom domains, and is under active and enthusiastic development, with a vibrant community.

I apparently created an account in 2016 and rediscovered it as part of my research for this post. At the moment, I’m still digging into how well it would work for my needs, but first impressions are very positive.

It doesn’t support publishing from within Ulysses/iA Writer, and I’m unclear on how well the image handling matches my expectations, but it ticks many of my boxes and does so at a fair price.

Everything Else

Let’s get the easy ones out of the way first: Squarespace, Webflow, and Tilda are all interesting and powerful no/low-code website builders. There are probably tens of others in this category. They make it possible to build a blog, but they’re not designed for it, nor do they have the easy publishing advantage that I’m hoping for from a hypothetical Blot replacement.

They feel like the wrong tool for the job, and the costs also tend to add up as you’re effectively paying for a solution to a much more complex problem than the one I’m trying to solve.

Svbtle is an old flame of mine, one that I’ve returned to twice over the years. It remains a strong and viable option, let down by a few persistent bugbears: poor image handling, and essentially zero development, transparency, or feature updates. If I didn’t know better I would say it’s abandonware, but Svbtle does, in fact, get the odd tweak and the developer even occasionally drops hints about an impending major overhaul, though that has yet to materialize.

That brings me to Substack, which is the closest to a viable option in this category. As a product, it expertly straddles the line between newsletter and blog platform, allowing posts to exist as either email, post, or both. It shares some of Medium’s platform-based discovery benefits, it’s also free to use, and it has a more flexible monetization system than Ghost in the sense that you can truly turn it off, and you have more control over payment tiers.

Substack has a nice enough web editor—though no integration with my writing apps—and it handles images in a rudimentary but inoffensive way. It’s not a bad option, but I’m not too keen on having my blog look like “just another Substack” site—a problem it shares with Medium, of course.

Hey World

Hey, Basecamp’s take on email (and my current provider of choice), recently rolled out an experimental blogging platform built around their email client.

The idea is that you simply use the Hey email client (web or native apps) and any messages sent to [email protected] become public blog posts. These posts live at world.hey.com/yourusername, and you get easy email subscription and RSS plus absolutely zero overhead or complexity. To a fault.

You can’t customize anything, you can’t use your own domain, there’s no search, you can’t add static pages or navigation of any sort, it’s pretty basic about how it handles images, and it’s still unclear where exactly they’re going with it—if anywhere at all.

Still, it’s included in a service I’m already paying for, so if for no other reason than to test it out I’ve cross-posted this entry to my Hey World page as well—you might even be reading it there!

Where to Next?

The reason I’m doing this research at all is because I’ve been scrutinizing my ongoing costs for things and looking for ways to streamline them where it makes sense to.

The blog feels expensive to maintain at about $235 USD per year, particularly since I stopped using the Amazon Affiliate program to monetize product links. That being said, I’ve been very grateful to find that people’s donations/tips via Ko-Fi have made it possible for me to cover the costs of running the site as-is.

While I love almost everything about Blot and can afford to keep running things this way, it does seem prudent to at least investigate current alternatives. As I see it, moving from Blot has to offer some pretty significant upsides.

Thus, the final list of imperfect alternatives are therefore:

Ghost(Pro) at $108 per year
WordPress at about $121 per year
Write.as at $72 per year

None of the above gives me everything I’m looking for, but they’re reasonable choices should I want to make the switch. The switching part is itself an important consideration though because it involves a lot of tedious manual work that I’d rather not have to deal with.

If I investigate my current costs, by far the largest contributor is Imgix, which is the service I use to handle images on the site. Imgix pulls full-resolution files from Amazon S3 and automatically converts them on the fly to a smaller version that gives me control over the exact compression, resolution, and format settings I want, which is appealing for a blog like mine that often features photography.

Still, the obvious path of least resistance here would be cutting Imgix out of the equation and simply using Blot’s own image handling. This would require me to move the images and replace the links in all my posts, but that’s a lot less effort than switching platforms, especially since I can automate a bunch of it.

It would also save me a whopping $150 per year between Imgix and S3 fees.

Staying Put

Remember how I said that Blot has me spoiled with its active support and development?

Well, as part of this research, I checked in with the developer to see what the plans were around image handling, and he was happy to confirm—within an hour of getting my email—that even my picky needs are either on the roadmap (responsive images) or already deployed (automatic compression and serving via CDN).

As you’ve probably guessed by now, the conclusion I’ve come to after all this research is that I’d be wise to stay right where I am.

To be clear: Blot doesn’t tick all the boxes on my checklist. The image stuff is underway, but I can’t post natively from Ulysses, nor does it have an easy email subscription option (I’m currently using Feedio to make this possible).

Instead, Blot has one key thing going for it: I’m the target audience. None of the other choices (with the possible exception of Write.as) feel as authentic, as focused, and as committed to serving writers like me. That really matters.

In every interaction and every product decision, I can feel that Blot’s development is moving toward supporting the kind of work I want to do. It’s really difficult to strike the right balance between geeky configuration choices and the peace of mind of a managed service, and I think Blot deserves credit for achieving that balance.

Perfect is the enemy of good, and I know all too well that fussing over platforms distracts from doing the work of writing, but now and then it’s nice to reassure myself that I’m building this thing on sturdy foundations.

One of my take-aways from this whole endeavour is that the right blogging platform is very much based on your individual needs. Just this month I set my grandfather up with a profile on Medium, and encouraged one of my good friends to move his sites to Ghost.

I think it comes down to this: do you have a place to write and explore and share and exist on the internet? Good. Do you actually write there? You should. Don’t let the platform—or anything else—hold you back.

Just keep writing.

productivity review technology