4 ms·
All of those “features” come with with a heavy price—responsibility. Running your own instance is increadibly time consuming and difficult to understand how to
by johnklos 4y ago
All of those “features” come with with a heavy price—responsibility. Running your own instance is increadibly time consuming and difficult to understand how to secure and scale. It also can be quite expensive if it grows to any meaningful amount of use.
I think it's important to note, which I think the article fails to do, that this isn't a problem with other Fediverse software, but something unique to Mastodon:
1) Ruby on Rails is, well, bad. It was in the past, but I hoped things had changed. I tried running Mastodon, and in the two weeks I was playing with it late last year there were two dependency updates which broke Mastodon, until Mastodon was updated. It's untenable, and it certainly adds to the idea that running your own is "increadibly time consuming", if you have to worry about breaking things any time any dependency gets updated.
2) The failure of Ruby on Rails leads to insecurity. Most people don't update because of #1 and just let things run. This is all fine and dandy until there are real security issues, at which time people sometimes panic, update and break things, or there's a compromise.
3) Mastodon (and Ruby on Rails) isn't performant. A decent modern server, particularly one that could be bought for $5,000, should easily be able to handle 30,000 users, if one were to extrapolate from some of the articles written about Fediverse hosting... But the articles showing very good performance aren't about Mastodon. Some specifically talk about migrating to other software because of Mastodon performance.
So is it time consuming? Yes, because it's slow, breaks easily and doesn't scale well. I think it's demonstrably true that other software (Pleroma, for example, or Akkoma for those of us who aren't fans of the Pleroma admins) is faster (and therefore scales much better), isn't remotely as fragile (and therefore can be kept up to date and secured with ease), and less time consuming (most updates take literally a minute or two).
Perhaps the author should consider the idea that the failure of Mastodon is not the same thing as the failure of the Fediverse precisely because other options exist.
- JustinGarrison 4y ago[OP] Did I say the fediverse would fail in the article? I tried to not imply that in any way. But Mastodon is the largest fediverse network and I don't see it scaling or accommodating the general public.
- johnklos 4y agoPerhaps it was a bad choice of words in this specific place where you lumped in Mastodon with the Fediverse, but this, as it reads, suggests that with the failure of Mastodon we'll also see the loss of the ideals of the Fediverse: > It has nothing to do with data ownership, identity portability, or federated networks. The ideals of Mastodon and the fediverse will be lost once established products use it. Also, you made some broad generalizations which I don't see evidence to support: > And you will be blocked from the broader public discourse just the same Why would large instances pull a Google on smaller ones? I've never seen that, nor do I see anything indicating that somehow it's inevitable. Do you have any evidence showing this will happen? As someone who has run email servers for a quarter of a century and as someone who is more than happy running my own Akkoma instance, I haven't seen even a mention of people who are marginalized because their server is somehow not big enough. I'm curious if you have, though.