There are no "rare" sats without the Ordinals system.
I collected things when I was younger too and can thus understand a bit the interest in collectibles. From my POV, If the Ordinals system was the "one and only" imaginable protocol to count Satoshis, or at least
by far the most straightforward/imaginable one, then perhaps I would be collecting "rare sats". At least those which are connected with some memorable event like the Pizza transaction, the starts and ends of the difficulty periods would not be that interesting for me.
Ordinals is a relatively straightforward protocol, but the problem is that there are another possible protocols to number satoshis. The FIFO aspect (inputs/outputs counted from first to last) is probably indeed the "most straightforward" solution. But the BIP already names several cases where a quite arbitrary decision was made:
- "underpaid" coinbase transactions - if someone mines only 49.999999 coins, then they will be treated as 50. IMO this is not very logical, although it probably makes calculations easier. But this decision in favour of "easy calculability" versus "logic" is problematic and would be a possible source for an "ordalt" protocol.
- duplicate transaction IDs could be handled in different ways, and as this occurs in blocks 91812/91842 and 91722/91880 a lot of sats would be moved. Here however the behavior follows Bitcoin Core, so this is a bit more "logic" than the previous decision.
And then there is the problem I already mentioned: which sats in a transaction go to the miner fees.
Example: two inputs with values of 2 and 3 sat -> one output of 4 sat + a fee of 1 sat.
If you number the satoshis of the inputs with [1-2] [3-4-5], then in the current protocol the 4 sat output will contain the satoshis [1-2-3-4] and the sat [5] goes as fee to the miner in the coinbase transaction. But we could actually also count the other way around: sat [1] could be the fee, and [2-3-4-5] those transferred to the output.
Inputs are ordered in the protocol and in blocks, so there is indeed only one straightforward way to count them. Fees however are outside of any transaction order, so this decision is
completely arbitrary.I am even tempted to start an "ordalt" software, rewriting some details which will change the ord number of some sats (e.g. the position of the fee), just to cause some confusion
.
Trading "rare" satoshis is not spam. You are literally sending and receiving bitcoins and nothing else, so I don't know how you could consider that to be spam.
The problem is that these transactions would not exist if Ordinals didn't exist. They do not help Bitcoin neither to establish itself as a currency (at least for "real products"), nor do they really contribute to the concept of a "store of value".
I consider them far less spam than BRC-20 transactions, but they're among the "more unnecessary" use cases. They could increase the presence of Bitcoin in collectionists groups, but I doubt these are really important to push Bitcoin adoption forward. Censorship resistance, Bitcoin's most important feature, is not really that necessary for the collectionist business.