Comment by imiric | Hacker News Reader

Comment by imiric | Hacker News Reader

Kudos to the team! It looks like a really polished product.

I've been wanting to set it up at home for a while now, but I have some reservations...

This type of fully featured open source software that competes with a service by a giant corporation with infinite resources is ripe for the typical OSS rugpull, and/or enshittification, and/or being acquired, followed by rugpull/enshittification. I've been burned many times by this before, and I'm reluctant to go all in on a product that demands so much practical investment: importing all my data, categorizing/labeling, using mobile apps, relying on "AI", etc. I'm tired of the modern software churn.

Immich does have an optional commercial license, which is good, and it's part of the FUTO brand of products, which seems to have a noble vision. But it's an unproven model that doesn't guarantee none of the above will happen.

In a weird way, I would like this type of software to be less polished and have less features. Both to make it less attractive for hostile practices, and to make development/maintenance easier and more cost-effective. Give me something that I can easily sync my photos to, view them in a web browser, and securely share them with close contacts. I don't need a Google Photos replacement. I need something simple that does a few things well, and that I can trust is not going to disappear on me. Is there such a product?

I'm aware of apps like Nextcloud which probably does this, but that's even more complex than Immich. Currently I'm just using `rsync`, and a static file web server, but that's a pretty hacky solution. I would need just a bit more polish. I could probably do this myself, but already have a long list of unfinished projects I would like to get to first.

> ripe for the typical OSS rugpull

Immich is AGPL-licensed and doesn't have a CLA, which makes it functionally impossible for us to do any kind of rugpull. We're already "aqcuired" by FUTO since a solid year ago[0] and it's only been up since then.

> I don't need a Google Photos replacement.

Then there are many other options in this space that might be better suited to you. [1] has a really good overview.

[0]: https://immich.app/blog/immich-joins-futo

[1]: https://meichthys.github.io/foss_photo_libraries/

> Immich is AGPL-licensed and doesn't have a CLA, which makes it functionally impossible for us to do any kind of rugpull.

More difficult, perhaps, but certainly not impossible. And there are many other ways the software can be functionally crippled without involving a license change.

> We're already "aqcuired" by FUTO since a solid year ago[0] and it's only been up since then.

Like I said, FUTO seems like a fair steward of OSS so far, but I have no reason to trust them long-term. It is funded by a single investor, which doesn't inspire confidence. That relationship may end at any point.

To be fair, I am being overly cautious of what seems like an excellent project by any measure, especially compared to the alternatives. In my defense, I've been conditioned over the years to be wary of software that looks too good to be true, because it inevitably disappoints me, and to prefer simpler and even jankier software that doesn't promise the world, because it rarely does.

Commercializing free software is a difficult and unsolved problem. I'm not against mandatory licenses to unlock features that only a few large/"enterprise" users would generally use. Or to have some sort of a tiered licensing system for more advanced features. This would go against your current philosophy of not locking any features behind a paywall, which is noble, but it also has the drawback of making payments entirely optional. It is essentially donationware, which is not a stable source of income.

Also, subscriptions, as much as users don't like them, are really a requirement for this type of software. It pays for continual maintenance which wouldn't be covered by a single lifetime purchase. There are many ways this could be implemented that makes it fair for both the user and the developer.

In any case, thanks for your work, and good luck with the project. And thanks for the link with alternatives.

Reply on news.ycombinator.com