# Ordinals

Ordinals is a scheme for assigning names to satoshis. These names can be linked to digital assets to make them transferrable.

The Global Share Memory protocol is closely related to the ordinals protocol. It is based on a scheme for assigning names to outputs instead of to satoshis. The purpose is the same, an asset can be linked to a name, thereby making the asset transferrable.

There is an important special case where the two protocol coincide: Recall that an output is a "memory update" if its index is smaller than the number of inputs of the transaction. We call a memory update conservative if it stores the same amount of satoshis as the output that it spends. Conservative updates preserve ordinal ranges.

A Bitcoin Computer smart contract that does not use the _amount keyword only produces conservative updates. Therefore a smart objects defined by such smart contracts preserve ordinal ranges.

For example consider an NFT contract:

class NFT extends Contract {
  constructor() {
    super({ data: img })
  }

  send(to) {
    this._owners = [to]
  }
}

If you create an NFT it will be stored in an output and that output is identified by the revision of the smart object. Say that output contains ordinals 5-10 (we are assuming some OrdinalsApi that can return the ordinals stores in a specified output).

const nft = await computer.new(NFT)
const oldOrdinals = await OrdinalsApi.getOrdsFromOutput(nft._rev)
expect(oldOrdinals).to.deep.eq([5, 6, 7, 8, 9, 10])

If we send the NFT it will get a new revision assigned. The important thing is that the new revision will have exactly the same ordinal range.

await nft.send(...)
const newOrdinals = await OrdinalsApi.getOrdsFromOutput(nft._rev)
expect(newOrdinals).to.deep.eq([5, 6, 7, 8, 9, 10])

Therefore you can build smart contracts like swaps that cannot be built with ordinals alone. You can store ordinals inside a smart object, pass them around in smart contracts and safely recover them, for example after they have been exchanged.