I'm particularly excited about FUTO funding Immich [0], a photo management app similar to Google Photos. The money allowed the FUTO devs to go full-time [1]
Yes, immich is incredible! I have my entire family's photos (not all technical users mind you) synced and backed up thanks to immich server and its phone app. The web UI is easy, fast and they've even started adding basic editing capabilities recently.
I see a decent amount of skepticism and negativity in this thread, but I'm glad to see that someone is funding projects like this.
> From its founding, FUTO has been funded entirely by investment from its sole owner, Eron Wolf.
This sort of thing concerns me. So if Wolf has a change of heart, suffers some sort of financial catastrophe, etc., then... the project just dies, and any project depending on them for a significant portion of their funding has a crisis?
> a few talented engineers can produce delightful software that rivals anything big, established firms can produce.
Sort of, but... not really? In my experience, more resources means more polish and more scale. It's the usual 80/20 problem: with a small team you can get to 80%, but the last 20% bogs you down forever. Certainly large teams come with their own problems, but that last 20% is hard.
In general, though, I love the idea of this. One thing I really want to try out is the FUTO keyboard: I currently use the stock GBoard on my Pixel, but would prefer something that just doesn't connect to the internet at all. I'm curious to know how well their swipe-based keyboard works (GBoard has become so hit-or-miss lately with that), as well as their voice recognition (Google's is of course famously good, but I don't love the privacy implications of sending all the things I might say to my phone to Google's servers).
Still, though, privacy-wise: why should I trust this guy? I like what he says on the website, but I've never heard of him, and don't really know anything about him.
> One thing I really want to try out is the FUTO keyboard
I've been using it for a few weeks and it works well. It seems to require some learning period because it started out making unhelpful suggestions for the predictive typing but is now pretty good.
> Sort of, but... not really? In my experience, more resources means more polish and more scale. It's the usual 80/20 problem: with a small team you can get to 80%, but the last 20% bogs you down forever. Certainly large teams come with their own problems, but that last 20% is hard.
Consider this: AAPL is valued at 3.4 trillion USD. Both the polish and breadth of their software took a nose dive compared to early 2000s. There’s no more Aperture, Logic, Final Cut, Shake…
I think it's GP probably intended something like "more resources spent".
That aside, valuation doesn't seem like a meaningful metric in this context. It doesn't tell you how many resources are being spent on a given thing and it's not necessarily obvious whether that given thing is connected to the value either.
Exactly. My point is: a small team can have huge impact writing software. The biggest public company right now could be investing a rounding error on software, but chose not to.
> This sort of thing concerns me. So if Wolf has a change of heart, suffers some sort of financial catastrophe, etc., then... the project just dies, and any project depending on them for a significant portion of their funding has a crisis?
> Still, though, privacy-wise: why should I trust this guy? I like what he says on the website, but I've never heard of him, and don't really know anything about him.
Do you prefer to use tech presented by opaque committees with institutional funding, but only if they also consist of people you're acquainted with?
That's quite the straw man you've constructed there. Please don't put words in my mouth. That's not what I said, or intended.
Try the HN guidelines on for size: "Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."
This is indeed a pain point, a sole owner, one source of money, essentially a king. Or dictator. Or BDFL. Who knows.
The idea is great, but like most (FOSS) ideas, it needs money, and money needs to come from somewhere. Existing commercial software doesn't track people for funsies, but as a way to make money. Granted, paid software didn't start out this way, but free software and services got in this realm pretty quickly as it's one of the very few viable ways we get mass market usable products (where usable doesn't mean it's any good, just that the mass market can actually do things with it).
Even if everything is above board and ethical and perfect, at best it will just last a few decades and then the money is gone, the owner is dead, or the amount of projects has grown to a point where the money pool is too diluted to make the difference it set out to do. As-is, it is not sustainable. But it's great while it lasts.
While I don't disagree, what's the solution in today's world? Seems they're doing the best with the existing legal framework we have here, for better or worse.
But immich was kept agpl and no cla which is a good thing.
We are about to package a few of our pieces of software across several repositories. Not sure what issue you think might come up with that? If we are not allowed because of some restriction you can just manually add our repos to your /etc/apt/sources.list which is perfectly normal.
> A license is allowed if Fedora determines that the license is a free software / open source license. At a high level, the inquiry involves determining whether the license provides software freedom, and (equivalently) making sure that the license does not place burdens on users' exercise of its permissions that are inconsistent with evolving community norms and traditions around what is acceptable in a free software / open source license.
The BUSL is on their list of not allowed licenses[2], so I find it highly unlikely your similar "non-commercial use only" license will be permitted.
what FUTO and others (eg bruce perens) are trying to figure out is how to make sure that profits made from the software go back to the original developers.
in the past linux distributions were needed because distributing software was difficult an the distributions helped get the software to the end users.
today distributing software is easy, and if i want my application usable on debian all i need is to create a debian package that i can distribute myself. which means i can use a more restrictive license and prevent big companies from taking advantage of my work for free.
it is not clear yet what the right approach here is, but getting my application into debian proper is unlikely to help me make any money, so why bother?
i therefore appreciate FUTO experimenting with the licensing model to see if there is a way to get software to the world without allowing profitable companies to freeload on our work.
I love the sixth of their "Five Pillars of FUTOey Software":
> 0. Don’t Suck This applies to all software, FUTOey or not. We have accomplished nothing if our software is sluggish, unreliable, or lacks key features. Our clients need to be delightful. Our servers need to help our clients be delightful.
And then I click through the polycentric sign-up flow, and it breaks after 3 steps, and has very page-like things for which the back button doesn't function, but it's okay, they provide their own back button in a non-standard location that disappears if you resize the window.
I've switched from SwiftKey to the FUTO keyboard and it's worked well enough that I haven't switched back. After many, many years on SwiftKey that's a huge accomplishment. There's some more advanced features I miss and one broad issue I have with the autocorrection feature, but overall it's quite usable.
The feature I'm still missing is having 2 dictionaries active at the same time. I text a lot in both English and German and this not being available prevents me from making the switch
44 comments
[ 3.1 ms ] story [ 78.9 ms ] thread[0] https://immich.app/
[1] https://immich.app/blog/2024/immich-core-team-goes-fulltime/
I see a decent amount of skepticism and negativity in this thread, but I'm glad to see that someone is funding projects like this.
This sort of thing concerns me. So if Wolf has a change of heart, suffers some sort of financial catastrophe, etc., then... the project just dies, and any project depending on them for a significant portion of their funding has a crisis?
> a few talented engineers can produce delightful software that rivals anything big, established firms can produce.
Sort of, but... not really? In my experience, more resources means more polish and more scale. It's the usual 80/20 problem: with a small team you can get to 80%, but the last 20% bogs you down forever. Certainly large teams come with their own problems, but that last 20% is hard.
In general, though, I love the idea of this. One thing I really want to try out is the FUTO keyboard: I currently use the stock GBoard on my Pixel, but would prefer something that just doesn't connect to the internet at all. I'm curious to know how well their swipe-based keyboard works (GBoard has become so hit-or-miss lately with that), as well as their voice recognition (Google's is of course famously good, but I don't love the privacy implications of sending all the things I might say to my phone to Google's servers).
Still, though, privacy-wise: why should I trust this guy? I like what he says on the website, but I've never heard of him, and don't really know anything about him.
I've been using it for a few weeks and it works well. It seems to require some learning period because it started out making unhelpful suggestions for the predictive typing but is now pretty good.
Consider this: AAPL is valued at 3.4 trillion USD. Both the polish and breadth of their software took a nose dive compared to early 2000s. There’s no more Aperture, Logic, Final Cut, Shake…
That aside, valuation doesn't seem like a meaningful metric in this context. It doesn't tell you how many resources are being spent on a given thing and it's not necessarily obvious whether that given thing is connected to the value either.
> Still, though, privacy-wise: why should I trust this guy? I like what he says on the website, but I've never heard of him, and don't really know anything about him.
Do you prefer to use tech presented by opaque committees with institutional funding, but only if they also consist of people you're acquainted with?
Try the HN guidelines on for size: "Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."
The idea is great, but like most (FOSS) ideas, it needs money, and money needs to come from somewhere. Existing commercial software doesn't track people for funsies, but as a way to make money. Granted, paid software didn't start out this way, but free software and services got in this realm pretty quickly as it's one of the very few viable ways we get mass market usable products (where usable doesn't mean it's any good, just that the mass market can actually do things with it).
Even if everything is above board and ethical and perfect, at best it will just last a few decades and then the money is gone, the owner is dead, or the amount of projects has grown to a point where the money pool is too diluted to make the difference it set out to do. As-is, it is not sustainable. But it's great while it lasts.
No thanks
That's exactly the right thing to do and is what they should do for everything.
Where is that not allowed?
> A license is allowed if Fedora determines that the license is a free software / open source license. At a high level, the inquiry involves determining whether the license provides software freedom, and (equivalently) making sure that the license does not place burdens on users' exercise of its permissions that are inconsistent with evolving community norms and traditions around what is acceptable in a free software / open source license.
The BUSL is on their list of not allowed licenses[2], so I find it highly unlikely your similar "non-commercial use only" license will be permitted.
[1]: https://docs.fedoraproject.org/en-US/legal/license-approval [2]: https://docs.fedoraproject.org/en-US/legal/not-allowed-licen...
in the past linux distributions were needed because distributing software was difficult an the distributions helped get the software to the end users.
today distributing software is easy, and if i want my application usable on debian all i need is to create a debian package that i can distribute myself. which means i can use a more restrictive license and prevent big companies from taking advantage of my work for free.
it is not clear yet what the right approach here is, but getting my application into debian proper is unlikely to help me make any money, so why bother?
i therefore appreciate FUTO experimenting with the licensing model to see if there is a way to get software to the world without allowing profitable companies to freeload on our work.
They use the term "open source" because the sources are available, but because of their disagreement with the OSI, they prefer "source first".
Their license is a proprietary, source-available, "donation-ware" type licence.
It is not my thing, but it is their work, they get to set the rules.
> 0. Don’t Suck This applies to all software, FUTOey or not. We have accomplished nothing if our software is sluggish, unreliable, or lacks key features. Our clients need to be delightful. Our servers need to help our clients be delightful.
https://futo.org/about/what-is-futo/