At Atomic Form, we’ve been strong advocates for improving NFT metadata standards & usage. Why? We see a pattern emerging from two previous generations of the web. In both cases, content producers that embraced/applied metadata benefited big time. Key benefits:

1. Their content ranked higher against other content.

2. Their content received more real estate on the page.

3. They controlled the exact title, image, and description that would appear.

These were transformative changes. So how did each generation do this?

The metadata within the HEAD tag of an HTML document was read by search engines. This allowed Google (and others) to extract the exact information to be used on their search engine results page (SERP). Before that, Google would have to scrape the page and make an educated guess. Many times it was right, but other times it might pull a random image from the sidebar or pull the wrong summary of text.

By giving Google metadata that included the explicit, optimized information, SERPs became more accurate for end-users, resulting in better experience for end-users. As a result, Google would boost a results ranking AND more prominently feature thumbnails and other information.

Social media platforms faced challenges similar to the early search engines. If you posted a link on your feed, Facebook would scrape the page and determine what to show. In the early days, Facebook would just pull 10-20 thumbnails (effectively every single image on the web page) and make the user decide. It would also require the user to edit the description because it was often wrong.

Facebook created the Open Graph Protocol to fix this problem. These new metadata tags allowed web pages to tell Facebook the exact title, image, description, etc. The result? Facebook gave an accurate rendering each time (no work required by the person sharing!). It also expanded the space given to the media and description, allowing the content to stand out against other shares.

See the pattern yet?

The better you feed your indexers, the better your data indexes, ranks, and displays.

Here’s the challenge in web3. There is so much history and backstory to NFTs (both as a collection and as an individual token), but it’s disconnected from the NFT collection and individual tokens. Some of it exists within Twitter threads and conversations. Some of it exists as an OpenSea collection page. Regardless, it’s hit or miss and generally lackluster.

If you go to OpenSea and look at Crypto Punks (arguably one of the most important historical collections), this is what you see when looking at token ID #1.

It’s important to note that the About CryptoPunks section is not a part of the NFT metadata, but rather information added on OpenSea by the account that controls this collection.

If you look at raw metadata, say for another popular project like Board Ape Yacht Club NFT, you’ll see a limited set of information.

Imagine trying to create a compelling search engine result or social media share with just this information. Where’s the name? Description? What would be an optimal image size to use for Twitter vs Facebook?

Fortunately, many NFT APIs attempt to fill in some gaps. However, this is done without the content creators input, scraping offchain information (such as OpenSea description). We’re back to the reliance on guessing and scraping to determine how to best represent a web3 NFT any place we want to experience it.

We know this challenge first hand at Atomic Form. We build gallery-quality NFT displays, and only have access to a limited amount of information about an NFT (like the title and mint date). However, we would love to expose additional information (the history behind and the story beyond the mint).

This is both the challenge and opportunity we see ahead of us in web3. If we start providing a richer set of NFT native metadata, then all consumer apps and dapps using this information will be able to offer up the same benefits as previous generations of the web. Better metadata = better indexing, rank, and display. It just makes sense.

This is why metadata matters!

So, how do we get there (standards and adoption)? We have some ideas (check out Let’s Talk About a Schema.org for NFTs). But if you’re interested in joining the conversation, please connect with us on Twitter at @atomicform and DM us!

At Atomic Form, we’ve been strong advocates for improving NFT metadata standards & usage. Why? We see a pattern emerging from two previous generations of the web. In both cases, content producers that embraced/applied metadata benefited big time. Key benefits:

1. Their content ranked higher against other content.

2. Their content received more real estate on the page.

3. They controlled the exact title, image, and description that would appear.

These were transformative changes. So how did each generation do this?

The metadata within the HEAD tag of an HTML document was read by search engines. This allowed Google (and others) to extract the exact information to be used on their search engine results page (SERP). Before that, Google would have to scrape the page and make an educated guess. Many times it was right, but other times it might pull a random image from the sidebar or pull the wrong summary of text.

By giving Google metadata that included the explicit, optimized information, SERPs became more accurate for end-users, resulting in better experience for end-users. As a result, Google would boost a results ranking AND more prominently feature thumbnails and other information.

Social media platforms faced challenges similar to the early search engines. If you posted a link on your feed, Facebook would scrape the page and determine what to show. In the early days, Facebook would just pull 10-20 thumbnails (effectively every single image on the web page) and make the user decide. It would also require the user to edit the description because it was often wrong.

Facebook created the Open Graph Protocol to fix this problem. These new metadata tags allowed web pages to tell Facebook the exact title, image, description, etc. The result? Facebook gave an accurate rendering each time (no work required by the person sharing!). It also expanded the space given to the media and description, allowing the content to stand out against other shares.

See the pattern yet?

The better you feed your indexers, the better your data indexes, ranks, and displays.

Here’s the challenge in web3. There is so much history and backstory to NFTs (both as a collection and as an individual token), but it’s disconnected from the NFT collection and individual tokens. Some of it exists within Twitter threads and conversations. Some of it exists as an OpenSea collection page. Regardless, it’s hit or miss and generally lackluster.

If you go to OpenSea and look at Crypto Punks (arguably one of the most important historical collections), this is what you see when looking at token ID #1.

It’s important to note that the About CryptoPunks section is not a part of the NFT metadata, but rather information added on OpenSea by the account that controls this collection.

If you look at raw metadata, say for another popular project like Board Ape Yacht Club NFT, you’ll see a limited set of information.

Imagine trying to create a compelling search engine result or social media share with just this information. Where’s the name? Description? What would be an optimal image size to use for Twitter vs Facebook?

Fortunately, many NFT APIs attempt to fill in some gaps. However, this is done without the content creators input, scraping offchain information (such as OpenSea description). We’re back to the reliance on guessing and scraping to determine how to best represent a web3 NFT any place we want to experience it.

We know this challenge first hand at Atomic Form. We build gallery-quality NFT displays, and only have access to a limited amount of information about an NFT (like the title and mint date). However, we would love to expose additional information (the history behind and the story beyond the mint).

This is both the challenge and opportunity we see ahead of us in web3. If we start providing a richer set of NFT native metadata, then all consumer apps and dapps using this information will be able to offer up the same benefits as previous generations of the web. Better metadata = better indexing, rank, and display. It just makes sense.

This is why metadata matters!

So, how do we get there (standards and adoption)? We have some ideas (check out Let’s Talk About a Schema.org for NFTs). But if you’re interested in joining the conversation, please connect with us on Twitter at @atomicform and DM us!