Not a Good Riddance Post

by Leo Gaggl • WOL Notes

I’ve had a Flickr account since 2008. My own photostream sits past 4,800 photos now, and for a long stretch of that history it wasn’t just a place I put my own pictures. In February 2009 I wrote a HOWTO on this blog about posting straight from a Nokia S60 phone to Flickr, because, as I put it at the time, “a lot of my clients in the educational sector are starting to use Flickr as part of their online teaching.” I wasn’t a bystander watching open education discover Flickr. I was helping build the on-ramp for it, in a professional life spent almost entirely in EdTech, first at DETAFE, then building learning management systems at Brightcookie for over a decade.

Then, in 2011, I wrote a different kind of post on this blog, “Facebook - good riddance!”, deleting my account and declaring myself done with a company I’d never trusted. I illustrated it with a Creative Commons photo pulled straight off Flickr, credited to a photographer going by cogdogblog. That habit, reaching for a CC-licensed Flickr image whenever this blog needs one, has held for 15 years since, and it’s still the workflow described in this site’s own editorial instructions today.

This isn’t that kind of post. I’m not celebrating anything.

What Flickr Actually Was

The photographer behind that 2011 image was Alan Levine, known across the open education world as cogdogblog, or just CogDog. He’s been uploading to Flickr since early 2004, and by his own account has posted more than 73,000 photos there, every one of them released as CC0, public domain, since 2016. He didn’t do this quietly. He built tools for it: the Flickr CC Attribution Helper he shipped in 2014 is still in active use, a three-click way to generate proper credit in whatever format a blog needs. In a piece he wrote about Flickr as recently as July 2025, he put it better than I could: “almost more than anything on the internet, I love Flickr.” He’s spent two decades modelling exactly the practice this blog has quietly borrowed: attribute always, license generously, share by default.

He wasn’t alone, and this part is closer to home than a citation can show. Michael Coghlan, an eLearning facilitator who spent years at TAFE SA, was a colleague of mine in those days and collaborated closely with Levine himself, running sessions like “Teaching with Flickr” for the Images4Education community back in January 2009, the same month I was writing up Flickr HOWTOs of my own. Michael is a good friend, still. Between people like him and Levine, an entire generation of open educators treated Flickr as shared infrastructure: a place to put a photo once, license it properly, and let it get reused in a hundred classrooms and blogs you’d never hear about. I still open Flickr some mornings for one reason that has nothing to do with any of this: watching a handful of feeds travel, one captioned photo at a time, wherever each of them has got to next. Michael’s most of all, but not only his: a few people I got to know over the years through exactly this, scattered around the world, still posting. That’s the thing I know I’d miss most.

That credibility wasn’t only social, either. Flickr the company earned real open-source standing of its own, for years. code.flickr.net was where its engineers wrote openly about how the site actually worked, sharding, caching, scaling problems worked through in public rather than behind an NDA, and Flickr shipped genuine API client libraries on GitHub for anyone building on top of it. That’s the other half of what made Flickr worth trusting: not just a generous licence on the photos, but a company that behaved, for a good long stretch, like part of the same open culture it was hosting.

Stepping Back, Without the Relief

Flickr’s 2026 mobile app release is the reason I’m finally writing this down. In July, the company shipped a rebuild that dropped batch tagging and the mobile photo editor, and broke enough else that Flickr’s own Help Centre published an unusually direct admission: the app “hasn’t met the bar you expect from Flickr.” PetaPixel had already called the wider platform a thing that survives “on goodwill, habit, and a community that deserved more than it got,” months before that release shipped. It’s hard not to read some of that loyalty as something like Stockholm syndrome. Two decades of habit does that, and I don’t exempt myself from it, it’s a large part of why this post is 15 years overdue rather than five. It’s not only the app, either. code.flickr.net, the engineering blog I just credited, hasn’t posted since February 2022, and flickr-sdk, the company’s own official API client on GitHub, hasn’t had a commit since July 2025, over a year ago as I write this. None of that is a scandal by itself, most projects go quiet eventually, but it’s the same shape of drift as the app: things that used to be actively tended visibly aren’t anymore. It’s one data point in a longer pattern, not the story. The story is what happens next.

That’s a different feeling to 2011. Leaving Facebook was anger at where the company was taking things, and it came with real relief: I got out early, never looked back, and was glad I hadn’t let myself get pulled further into that cycle. Stepping back from Flickr wouldn’t carry that same relief, because Flickr, in the sense that mattered, the open community of people licensing their work generously and expecting the same back, was never the thing I needed relief from. It’s the thing I’d be sad to lose, and I haven’t lost it yet. I’ve cancelled my Pro subscription over the app disaster, a real step, not a symbolic one, but it’s still got a few weeks left to run. And in that window, credit where it’s due, Flickr’s actually walked the rebuild back rather than just promising to listen better next time. Their own help centre now says outright that they’re “rolling the app back to the previous version while we address the issues,” and on Android specifically, where the damage went further than that official notice lets on, they’ve reversed course too. So I’m still uploading, still opening the site most mornings to follow those same feeds, and genuinely undecided. The real decision lands when Pro lapses and I have to choose whether to renew, and I don’t yet know which way it goes: two decades of habit is exactly the kind of pull that talks you into one more year, and a couple of my Flickr friends are making that case. Whether that’s loyalty or the Stockholm syndrome I named a moment ago, it’s too early to tell, and I’m not going to write this post as if the door were already shut. What I do know, whichever way it goes, is that I’m not banking my archive’s future on any single company’s next roadmap decision, this one included.

The strongest case against leaving isn’t mine, and it predates the app trouble by five years. In 2021 Scot Hacker wrote “Flickr is no ghost town”, and most of it still holds: an interest group for every niche, full-resolution viewing, EXIF kept where Facebook and Instagram strip it out, an audience of other photographers who know what an image costs to make, and the line that sums up why I stayed as long as I did, “your images are decoupled from your social network, which gives you freedom”. He was right then and much of it is right now. What the piece leans on, almost in passing, is that “SmugMug have been great stewards” and that a better mobile app is surely on the way. That is the exact assumption the 2026 rebuild broke. The community Hacker described is still there. The stewardship the argument rested on is what wobbled.

What Coghlan and Levine built was never actually dependent on Flickr the company being well-run. It depended on an open licence, a norm of attribution, and enough people practising both in public that the norm became self-sustaining. Flickr was the venue, a very good one for a long time, but the venue and the values it hosted were always separable. This site has made the same argument from other directions: a community that feels decentralised can still be running entirely on one company's infrastructure. The commons that mattered can outlive the platform. It just needs somewhere else to live.

Your Keys, Your Photos

Norfolk Southern freight locomotives hauling mixed freight at sunset near Bement, Illinois
"Twilight of the District" by Leo Blackwelder (CC BY-NC-ND 2.0)

Here’s the part I actually want to spend this post on: where I think this goes, not what just broke.

There’s a phrase from the crypto self-custody world that I keep coming back to, stripped of the token speculation it usually travels with: not your keys, not your coins. Applied to a photo archive it becomes something simpler and more useful: your keys, your photos. Two real paths get you there today, and neither one depends on any single company continuing to behave.

The first is running your own backend. Immich, AGPL-licensed, past 100,000 stars on GitHub, is a genuinely complete Google Photos replacement you can run on a home server or a cheap VPS, automatic backup, face recognition, real search, all of it on hardware you actually control. No company sees the photos, no company can decide to rewrite the app over a weekend and break your access to them, because there’s no separate company standing between you and your own server in the first place.

The second is for people who’d rather pay for permanence than run a server: distributed storage networks like Arweave, built around paying once rather than a subscription that can lapse. I’ll come back to exactly why that distinction matters more than it sounds like it should, later.

FediFlick sits on top of either choice rather than picking one for you. It’s a posting layer, not a storage decision: what actually holds your photos is a separate question from where they get shared. Right now that’s five destinations from one composer: Mastodon, Bluesky, Pixelfed, Flickr, and Nostr (the most obscure of the five, mostly of interest to the hard-core distributed-network people I count myself among). Flickr sitting inside that list rather than outside it is deliberate. Posting to the Fediverse and still uploading to Flickr aren’t opposing choices here, and nothing about using the tool asks you to pick. That separation is the whole point, the archive and the posting tool shouldn’t be welded together the way Flickr welds them today, because that weld is precisely what makes an entire archive hostage to one company’s next roadmap decision.

What’s Actually Shipped, and What Isn’t Yet

FediFlick, the posting tool itself, is real today, released under the AGPL, and it already treats Flickr as one of those five equal destinations rather than something to abandon. Exactly how, and exactly who I actually built it for, is worth its own post rather than a paragraph here.

That licence choice isn’t incidental, and the reason deserves precision rather than treating “open source” as one interchangeable label. The GPL, and Stallman’s whole free-software project behind it, was a legal hack from the start: using copyright’s own mechanism to protect the right to share rather than restrict it. AGPL adds the one clause plain GPL doesn’t have, closing the exact loophole cloud computing opened up decades later: run modified code as a hosted service instead of shipping a copy, and GPL’s sharing obligation never technically triggers. AGPL’s network-use clause closes that gap. I’ve written before about who’s built policy specifically to avoid that clause, and why https://gaggl.com/blogs/2026-07-14-what-the-commons-built/, this is that same choice, made in a Flutter app instead of a manifesto.

The Immich-or-Arweave dream above is bigger than that, and I want to be honest about the gap. Running Immich well still means running a server, or trusting someone else’s hosted instance, which is its own smaller version of the same problem. The Arweave side of it is still design, not shipped code, a multi-year, separately-funded effort. None of this is about hauling 15 years of my own images off Flickr, though. The originals have always been backed up on my own NAS, so Flickr was never the archive, just one of the places it gets shown, and FediFlick posts to Flickr as one destination among five rather than pulling anything out of it. What’s changed already is smaller but real: new work doesn’t have to go into a system I can’t influence, and I’m no longer betting an archive’s future on one company’s next roadmap decision.

Thank You, Not Good Riddance

I don’t think Flickr owes anyone an apology it hasn’t already tried, badly, to give. What I do think is that Coghlan, Levine, and thousands of people like them built something real on top of a platform that, for a long stretch, was worthy of it. That deserves saying outright, not folded into a complaint about an app update. The photo I used in 2011 to say good riddance to Facebook came from someone still, 15 years on, actively practising the thing that made Flickr worth using in the first place. That practice, not the app, is what I’m carrying forward, onto whatever combination of Flickr, a home server, and paid distributed storage ends up holding it for the long term.

Part one of five in the FediFlick & Digital Commons series. Next: .

Sources

Related reading: The Decentralisation Paradox on why a community that feels decentralised can still be running entirely on one company’s infrastructure.

←Enclosure by Good Intentions

Comments

Be the first to comment! Reply to this post from your Mastodon/Fediverse or Bluesky account, or mention this post's URL in your reply. Your comment will appear here automatically via webmention.

Follow this blog on Mastodon at @gaggl.com@web.brid.gy or on Bluesky at @gaggl.com