Update EIP-721: delete broken openzeppelin links

[konojunya]

In the NFT Implementation and Other Projects column, the 17th link is to openzeppelin, which is already an invalid link and also a link to SafeERC20. openzeppelin has an ERC721 implementation.
The link should be to ERC721.sol

ref: https://eips.ethereum.org/EIPS/eip-721#references

[eth-bot]

Hi! I'm a bot, and I wanted to automerge your PR, but couldn't because of the following issue(s):

---

(fail) eip-721.md

classification
updateEIP

• eip-721.md requires approval from one of (@fulldecent, @dekz, @nastassiasachs)

[Pandapip1]

+0. I would prefer the links to be removed altogether, but I'd rather the link be fixed than a broken link stay.

[xinbenlv]

Thank you for the contribution!
This is a historical doc, to fix it

1. Editors are working on a link policy per EIP-5757 (Add EIP-5757: Propose process for allowing external links #5757 ), let's wait for that to be finalized
2. Links are generally suggested to use permlink to reduce chance of such broken links in the future

But still, thank you for your contribution! Greatly appreciated

[konojunya]

@xinbenlv Yes. Exactly. This EIP is referenced by many people and we should wait for the Editors reply as it is a change to a part of that document.
I too recommend using permlink, but since the source of the change points to the default branch, I made the change using the same policy.

[gcolvin]

Links are generally suggested to use permlink to reduce chance of such broken links in the future

So, in this case, would the better fix be to replace the broken link with a permlink to the last SafeERC20, rather than a link to ERC721?

And, given these links are no normative, and this one is not discussed in the proposal, I'd be happy to see it removed.

[konojunya]

@SamWilsn @axic @gcolvin @lightclient

How are you?
Let us know what you think!

[SamWilsn]

I think my preference for this would be to remove the links entirely. I don't want to establish a precedent of updating broken links in old EIPs.

Should probably be discussed on EIPIP...

[konojunya]

@SamWilsn In my opinion, I leave the update or deletion to the outcome of the discussion, but the current situation with the wrong link should be changed.

Even if it were removed, the openzeppelin implementation of EIP-721 would have little impact because it is widely known, but the EIP-721 page does not have a reference implementation like EIP-4907, so newcomers could lose the link to a valuable source of information.

[Pandapip1]

I think my preference for this would be to remove the links entirely. I don't want to establish a precedent of updating broken links in old EIPs.

This is my preference too. However, I would like to establish a precedent for updating broken links in old EIPs. I would just like there to also be a precedent of deleting those old links too.

[konojunya]

I understand your basic policy. I will add a commit in the form of a deletion rather than an update to the link.

[fulldecent]

Please close this PR in favor of #6012

---

The link to ERC-20 implemented by OpenZeppelin is correct and intentional.

This citation is an important part of the justification of how ERC-721 (and frankly, all) token standards currently work. The text explains the risks of the old ERC-20 behavior and advocates that when transactions fail they should throw, something many now take for granted.