The internet archive is a universal archive of the internet rather than just what is exposed trough their wayback machine. You are comparing a quantifiable well indexed archive of any online with one which is not quantifiable due to being lost to history. There are similarities between the efforts of the Library at Alexandria and the internet archive which make them comparable, so lets.
>'size' is really not that important. [...] each one a historical artifact.
The uniqueness of content in the internet archive is significantly greater by virtue of there simply being more unique knowledge at this point. Knowledge which would be lost (often intentionally) if not for efforts such as the internet archive, or that of the Library at Alexandria.
The library made copies of the knowledge for the original owners rather than having the only copy. making every copy most likely non-unique. The value in its indexing is immense especially when you realize that useful information is only useful within a context where it can be applied and cannot be valued outside of it.
>Losing the Library at Alexandria would be like ...
This did not happen.
>If the Library continued to exist in its fullest form, we may very well have much more of our history [...] orbital physics.
Imagine if a similar effort to archive exists today and in a few thousand years we would be able to use it, this is how I see the internet archive. The archive and efforts like it are used as a first source of knowledge by wikipedia, students, developers, researchers, anyone needing to verify a business existing prior to the investor call. If there was not an archive of the Feynman lectures I would know nothing about orbital physics now, if the current hosts are lost I am very very glad that the internet archive has a copy.
>If the contents of IA [...] important).
Personally I would like to know if homedepot existed 14 years ago if I wanted to do business with them and they claimed so, or if they tried to censor anything (that case of the US government having honeypot sites archives with code comes to mind). Things which would be lost if not for the specific functionality you are mentioning. Also there are many many other efforts, unlike you seem to state (e.g. archive.is, ipfs, the-eye.eu).
Perhaps the most powerful argument: https://web.archive.org/web/*/news.ycombinator.com
edit: I have archived our comments for historical context for anyone studying internet discussion in the year 3020
https://web.archive.org/web/20200611105240/https://news.ycom...
https://web.archive.org/web/20200611105414/https://news.ycom...