For the foreseeable future, this will be my last post on paragraph. But don't worry, I will continue writing and publishing content. But I will publish it directly on my X / Farcaster accounts until I found a better solution. Please follow me here and here.

The reason for this is the latest Paragraph update, which brought a number of changes. One thing that wasn't discussed much is that it eliminated the capability to use custom CSS to style your Paragraph blog. According to the Paragraph team, only a handful of content producers used this feature. Unfortunately, I was one of them. I spent hours carefully designing a cool brand identity for my Paragraph blog. The last update destroyed this by removing custom fonts and custom background colors, leaving me with broken designs.

Unfortunately, it looks really horrible now (nay!!)

While I did have the ambition to regularly publish content and continuously grow my audience, I was not a large Paragraph blog by any means. The platform will lose 0 users with my leaving - and that is fine. Every product team has limited resources, sometimes monetary, sometimes it's just developer attention, or maybe even real estate. By that metric, you cannot please everyone.

I also want to add that the Paragraph team has been nothing but nice to me. Not only have I been featured as an article of the week twice, but the team has also responded to my feedback in a professional manner and explained their reasoning.

Paragraph response to pushback on CSS customization

While I get where they are coming from, I think the mental framework that is driving this decision-making is essentially emulating what current Web2 publishing platforms are doing—standardization of content appearance to increase the distribution power of the central distributor. I think Paragraph can still be successful by offering a marginally better product than other blogging/newsletter platforms. I still wish the team all the best and hope they succeed. However, the reason I am leaving Paragraph is because, in its current form, I do not think it solves any critical problem worth solving anymore.

Paragraph: How it works and its value proposition

If you look at the value proposition on the Paragraph website, a lot of it comes down to the concept of ownership.

TLDR: Paragraph is essentially a blogging/newsletter platform that is partially run on crypto rails. Most notably, content is stored on Arweave, and articles can be sold as digital collectibles, allowing you to sell the content you have created. On paper, this gives you increased ownership of the content you create on your blog.

However, thinking about it more deeply, I think you fundamentally do not have true ownership of any content that is produced on Paragraph. Let me tell you why.

What does it mean to truly - own - something?

Ownership has always been a large part of the Web3 narrative. The birth fable of Ethereum was Vitalik thinking about the ownership of digital assets (what happens to your game items if the devs turn off the game servers?).

This resulted in the creation of Ethereum and subsequent smart contract platforms, including their primitives for ownership, such as NFTs. Over the years, blockchain technology has matured to massively expand the types of assets that can be owned and the quantity of assets. In simple terms: you can now own almost anything and as much of it as you want.

This makes sense because digital assets really thrive on abundance. The amount of digital assets and content is evolving at a growing rate and is essentially on a path to infinity.

However, ownership is conceptually more nuanced than that. First, it's important to understand that ownership is a societal, or at least a systemic, concept. It can only exist in a multiplayer world—for some entity to own something (owner), it requires at least one that does not own it (non-owner). If there were only ever one human, the concept of ownership would be entirely futile.With this established, we can get a better understanding of the different aspects of ownership.

Provenance: The necessary, but not sufficient aspect of ownership is the actual ownership of the asset. Is there a mechanism to distinguish the owner from the non-owner?
Monetization: The owner can sell the asset or access to the asset to non-owners.
Appearance: Appearance describes how the asset presents itself to non-owners. Is there a mach of what the owner perceives to own and how it is presented to non-owners?
Distribution: Lastly, distribution are the channels through which the provenance and appearance of the asset can be relayed to non-owners. Can the ownership be made widely known to non-owners?

There is no true ownership if one of these aspects fails

The technological emulsion of distributed ledgers and public/private key cryptography (aka blockchains) has made significant progress in these aspects. However, we can also understand that if some of these ownership aspects fail, the ownership of a specific asset can fall apart conceptually. For instance, do you truly own something if everyone else thinks you own something different? For instance, what if the only watch your videos is through a sepia-filter because either Youtube or your Internet Provider automatically applies it whenever you upload your content? This is essentially the situation with Paragraph as of now. The platform can decide to make my blog neon-pink and there is nothing I could do about it.

Failures can occur along all the characteristics of ownership. We can also get a better understanding of how Blockchain has helped prevent these ownership failures in general and what Paragraph contributes in particular.

As we see, Paragraph somewhat improves the Provenance and Monetization aspects of ownership, but it only provides limited value addition or even value deduction on Appearance and Distribution.

Which aspects of ownership should you compromise on?

Of course, perfect ownership is an ideal that can likely never be completely fulfilled. There are many attack vectors that can undermine ownership. However, we can measure if we are directionally moving towards that ideal or not. I believe in the past decade we have generally inched closer to the ideal. Blockchains have played a part in this, but there are also other technological progresses, such as decentralized energy production (mostly via renewable energy) or Starlink, which provides worldwide high-speed access.

Still, in the current state, everything is somewhat a compromise. The question arises: which aspects of ownership (Provenance, Monetization, Appearance, and Distribution) would you be willing to compromise on?

These compromises can be different for varying assets and platforms. For instance, we have seen memecoins and generally most crypto projects compromise on Appearance and Distribution by leveraging platforms like X and Discord for promoting their projects. For these kinds of projects, the reach traditional social media platforms provide outweighs their restrictions on content due to their sheer reach. In some ways, you can make the argument that due to the difference in scale, the ownership compromises are more beneficial than more ownership-aligned, but smaller-scale open platforms such as Farcaster or Lens. The fact that memecoins have become a multi-billion-dollar industry is an indicator of this.

However, things look different for media content. X penalizes distribution with links to external media platforms such as Substack. Writing about memecoins on their platform is TAM-expanding content for X, but linking to media content on external media platforms is TAM-restrictive. This is the issue that many Web3 media platforms face. Their value addition will only be net positive at a certain scale; until then, for many digital asset producers, it makes more economic sense to make other compromises regarding the ownership of their assets but have better distribution.

This is particularly true for Web3 media platforms that decide not to fully maximize on Provenance, Monetization, Appearance and Distribution control, such as Paragraph does. These are stuck in the middle by not providing enough additional ownership control to make it worthwhile to take a hit on distribution.

Which alternatives to Paragraph exist?

Where will my content live in the future? I see a number of pathways that would optimize for my mental framework of ownership.

Write on other publishing platforms, such as Medium, Mirror, Substack, or Ghost: All of these have their own pros and cons, but my feeling is that their compromises are not really meaningfully better than what Paragraph provides. It would feel more like a lateral movement.
Distribute via X and/or Farcaster and host content somewhere else: This would essentially mean splitting up the different aspects of ownership across several platforms. Likely, the best mix is to publish content on X/Farcaster first, and then have a blog with custom CSS to provide a quality-of-life destination where the content would look its best.
Keep posting on Paragraph and hope they refine the product mix: Potentially a good idea, but it will require a ton of additional work down the road for re-adjusting the appearance of content, etc. Still, for now, the blog will continue existing in case the product meaningfully changes again.

Right now, I am leaning towards the second option. Particularly, Farcaster provides a good balance regarding the different aspects of ownership. Potentially, there could be a way of publishing long-read content via Frames, having full control over the appearance and monetization this way.