I found the support docs fairly evasive in describing when large media is a good option. As far as suitability of numbers of files at different sizes accessed x times per y… the general outlook of the docs seems to be go for it, store as much as you want, but don’t use it for streaming.
But I suspect the reality is not that based on some other support requests (where the answer was basically “yeah… don’t do that”
Are there any guidelines for that?
(my particular use case is porting a wordpress podcast archive (just the old very infrequently accessed episodes; any new ones are hosted elsewhere) to a static eleventy site… which from the docs seems like it might work just fine, but my suspicion is that it really won’t )
The main reasons I recommend people not use it is:
that folks who are not git-savvy try to use it and get stuck in can’t-even-push-my-repo configurations. (the git config is a bit intense, but you’ve clearly done your reading and if you feel comfortable using git LFS anyway, this is probably not you :))
that the benefits for folks who don’t use the transformations are slim to none. You can use LFS on your own without the service, to speed up clones here and locally.
in case you don’t like it, we have to help you turn it off, and we may not see your request quickly to do so.
If you need the transformations and are comfortable writing code to use them, and can successfully set up LFS yourself, it makes a lot of sense.
If any of those conditions aren’t true, you might not get much value out of it.
haha… i’ve already enabled it… but am not sure that it will actually work once i push so i’ve been holding off… i was just imagining how long it would probably take to push over a gb of files on slow upload and what might happen if it fails partway through etc etc haha.
i’m also not doing any transformations… so ya your advice makes sense and i’m thinking about looking into alternate options
but i still feel like the docs are lacking… i was reading through them initially trying to understand what number/size of file would be feasible or not without really understanding the underlying architecture… and they mention streaming isn’t ideal but why not? i imagine it’s a bit up in the air because you’re live-scaling servers to deal with load behind the scenes so can’t really make claims but… i was pretty unsure what was being promised
my current use case is trying to pay far less to host a largely defunct hobby project that i’d still like to exist on the www, so i’m looking for cheap/free storage options for low access large files. someone recommended netlify-lm and it seemed perfect … since i’m already using netlify for a lot of static hosting, and from reading the marketing pages and even the initial support docs. but once i started going through setting it up i was starting to realize maybe it’s not the best choice and i don’t even know how it’s gonna behave in the git deploy/build ecosystem
)