So having saved that file offline on my harddrive, I can later re-add it to IPFS.
ipfs add Downloads/1337.png
added QmWjXFUVNjavM4hDzXeQYGXGCXjV86mP7kPEaTeP5bohae 1337.png
140.68 KiB / 140.68 KiB [=============================================] 100.00%
Notice that because I added the same file to my local ipfs, it got the same id as it had when I retrieved it.QmWjXFUVNjavM4hDzXeQYGXGCXjV86mP7kPEaTeP5bohae
So, IPFS doesn't even require that the file is indefinitely pinned.
And if I have a public node, others can now once again retrieve the file using the same known id.
The person that re-adds it doesn't have to (nor can) specify the id of the file; that's calculated from the hash of it.
(Although, IPFS does also have a way to serve mutable data, so in those cases the id is not calculated directly from just a hash. But in the case of the image we looked at here we see that it was.)
The difference between needing to pin forever and merely needing to have a copy of the same file somewhere outside of IPFS which is then re-added to IPFS and gets the same ID is a fine distinction perhaps, but still an important one IMO.
And I think that even if all of the IPFS nodes that currently have a copy of the Bored Apes would go offline, there is probably someone out there who would notice that and for whatever reason be motivated to make all of the apes available again on IPFS with their same original id's that they had by re-adding the original files.
I guess the best thing to do if you own an Ape is to store the image on your HDD and put a copy in a safe-deposit box, and then pin it yourself whenever you can.