Replying to @Heimdall_Bull
I imagine not every indexer is the same in terms of storage. That was the only number I had that was published, so it’s the one I used. It does sound like it’s on the high side.
My main point on index size is that you *can* design it to require less storage than all of block history, since block history already carries all the data. Ordinals goes the opposite direction, it’s always larger than all of block history.
1
30
i've spent too much time thinking about overlay protocols
Replying to @Heimdall_Bull
This deserves a proper response.
I believe a few things hurt it.
It tracks every koinu ever, so it has to index every UTXO ever not just the ones that end up as an inscription. There are more efficient ways to track ownership on-chain, like Dunes does with OP_RETURN. For example, you could still do the same commit-reveal mint, just output an OP_RETURN on the reveal with a 'register' operator. Then use OP_RETURN to trade like Dunes.
Opening up OP_RETURN to be able hold scriptSig size'd inscriptions would be even better, PR: github.com/dogecoin/dogecoin…)
Sybil cost is another.
This is more taste and application specific, but I generally believe inscribers should pay a premium to be picked up by indexers. Burning in the OP_RETURN output is a perfect place for it. This puts a hamper on spam and commits real value to inscriptions.
I also don't believe inscriptions *need* to be indexed forever. For example, you could require 'collections' to burn an indexing lease every once in a while. It can be cheap, and you could even set it up so anyone can pay for any collection. With that you have an index that drops off all the stuff nobody is interested in.
I do something similar with Pepecoin.surf (PepeNet Proof of Concept), the inscriptions live in OP_RETURN just a short name but require burn leases. Especially useful for TLDs because ownership needs to drop off if it's abandoned. Even has a on-chain marketplace with offers, listings, and trading, all powered by opcodes in OP_RETURN.
If you constrain index growth and don't rely on the UTXO set, you can make very very small nodes. Small enough to fit in a browser (dane.surf)
3
121
Replying to @Heimdall_Bull
This deserves a proper response.
I believe a few things hurt it.
It tracks every koinu ever, so it has to index every UTXO ever not just the ones that end up as an inscription. There are more efficient ways to track ownership on-chain, like Dunes does with OP_RETURN. For example, you could still do the same commit-reveal mint, just output an OP_RETURN on the reveal with a 'register' operator. Then use OP_RETURN to trade like Dunes.
Opening up OP_RETURN to be able hold scriptSig size'd inscriptions would be even better, PR: github.com/dogecoin/dogecoin…)
Sybil cost is another.
This is more taste and application specific, but I generally believe inscribers should pay a premium to be picked up by indexers. Burning in the OP_RETURN output is a perfect place for it. This puts a hamper on spam and commits real value to inscriptions.
I also don't believe inscriptions *need* to be indexed forever. For example, you could require 'collections' to burn an indexing lease every once in a while. It can be cheap, and you could even set it up so anyone can pay for any collection. With that you have an index that drops off all the stuff nobody is interested in.
I do something similar with Pepecoin.surf (PepeNet Proof of Concept), the inscriptions live in OP_RETURN just a short name but require burn leases. Especially useful for TLDs because ownership needs to drop off if it's abandoned. Even has a on-chain marketplace with offers, listings, and trading, all powered by opcodes in OP_RETURN.
If you constrain index growth and don't rely on the UTXO set, you can make very very small nodes. Small enough to fit in a browser (dane.surf)
Replying to @Heimdall_Bull
I got that number from github.com/bitcoinuniverseio…
Maestro claimed 'over 1 TB' in their docs and Ordinals Wallet specifically called out managing the data as being an issue in their discontinuation post.
1
38
Replying to @log4j @radioaktivetest
4
Basically, Doginals was poorly designed and everyone suffered for it. Doginals suffered the most.
1
6
170
Because Dogecoin reduced recommended fees in 2021, Doginals was a relatively "cheap" event.
It got a really good discount on Block/UTXO space.
Doginals is leaving behind a bloated UTXO set and a permanent tax on archival nodes.
83% of the UTXO set is only worth ~$10,000 (US), the remaining 17% is worth ~$9 Billion.
Next time we fork, we could do some cleanup.
Besides low activity, trying to index every Doginal also didn't scale well.
A few of those who dropped Doginals mentioned managing the data as a pain point.
So, Doginals is basically dead.
On-chain activity has been low and most of the major indexers/wallets have dropped it.
Noooooo, I don't want to play video games! 😭
Anyone want to fund a random dev $2k for Grok Ultra?
1
4
121
<<insert talking point here>>
Quad9's forced withdrawal from France highlights the dangers of mandating DNS-level blocking.
Read more: cfl.re/4f0cowv
3
175
Replying to @inevitable360 @chromatic_x
2
23










