dConstruct

9 Sept 2022, Brighton, UK

George was one of eight presenters at the last dConstruct conference, opening the day with “How to write a 100-year plan,” a provocation about the future of Flickr.

Here’s the presentation, and the full script is available below.

CC-BY 2.0 The Flickr Foundation.

George Oates at dConstruct 2022 from Clearleft on Vimeo.

SLIDE 1

Hello. I’m George Oates, and I’m honoured to be opening proceedings today. Thank you for the invitation, Jeremy. 

You’re looking at a picture taken by Garrett Coakley at dConstruct in 2007. That’s me on the right, with Denise Wilton. We talked about what it was like designing for huge online community and I think that’s one of the first conferences I’d attended where people had a chat on a couch. It was good! And I think it’s on the conference website if you’d like to listen to it.

That was 15 years ago! About three years after a photo sharing website called Flickr had launched. I was the first designer at Flickr, which is – I think – why I got this gig.

I worked at Flickr from its beginnings to late in 2008, and now I’m back. I’ve returned to lead the new Flickr Foundation, which exists to figure out how to make Flickr last for 100 years. It turned 18 years old this year.

It’s certainly one of the older web platforms online today. We can probably all list 3 or 4 web things that are about the same age. It’s grown into one of the largest image collections humans have ever assembled. I think roughly only Facebook is bigger. And, well, Meta owns Instagram too, so that’s bigger still. But nonetheless, Flickr’s huge.

But, it’s not just size that matters, guys. To me at least, it’s how you use it. No… to me, the most interesting part is how many of us have contributed it. To me that’s what makes it valuable. It can hold millions of different histories.

SLIDE 2

A thinker and writer I enjoy, John Berger, explains photographs as “observable moments.” This is a definition I like, especially in the Flickr context, because it gives you a sense of what Flickr has become. It contains billions of “observable moments” that have happened all over the world since photography was invented to just a moment ago.

This collection isn’t just user-generated content, it’s user-generated histories.

SLIDE 3

So, let me take you on a trip down memory lane to 2004, and to this picture, taken at a conference called eTech on the 10th February 2004 by a nice chap called Boris, who was in the audience when Stewart Butterfield, Ben Cerveny, and Eric Costello launched Flickr.

https://www.flickr.com/photos/bopuc/2220/

SLIDE 4

And here’s a photo taken in the same room of the guys getting settled just a few minutes earlier, by another nice chap called Chris Heathcote, who I’m proud to say is now a good friend of mine. I found this one because Chris left his photo as a comment on Boris’s. 

https://www.flickr.com/photos/antimega/2201/ 

SLIDE 5

If you look closely, you can see a crunchy opening slide for their talk, which was called Transcendent Interactions. You can find a copy of that on Slideshare, and we’ll be grabbing one for the Foundation archives – it’s great. Talks about designing context for interactions, not just designing applications.

Let me quickly repeat that – design a context for interaction, not just an application.

https://www.flickr.com/photos/antimega/2201/ 

SLIDE 6

Here’s a tiny record of what the site looked like in that first year or so.

It was really weird when it first started, but people stuck around and began to share their photos with their friends and it took off. 

Here’s a description written by Dan Tynan in an article from June 2004 called Photo Sharing Gone Wild, in PC World:

Moblogs aren’t the only way to embarrass yourself online. For example, Flickr.com is a chat-centered photo-sharing service. Click the site’s FlickrLive button to open a Flash-based chat window in your browser; upload images from your PC and drag them into a chat room for all to see, or send pix directly to your Flickr pals.

Following a clever interactive demo, the site logged me in to a chat where people were swapping pix of their cats. I have a recurring nightmare about being trapped in a chat room where people talk about their cats, so I logged out and entered another chat. Here, a heated match of rock-paper-scissors was under way: One player would post a picture of a hand in scissors mode, another would follow with a shot of a fist (“rock”), and so on.

I surmised that Flickr users must be young, artistic, and chronically idle. But the emphasis is more on community than celebrity; sharing photos is just another way of saying “I’m strange. Are you strange? Let’s be strange together.”

SLIDE 7

https://www.pcworld.com/article/116576/article.html

SLIDE 8

Here’s a graph of the uploads over time. 

Please note: it’s approximate. I know the start point and the end point, and I’ve just filled in the gaps…

That’s me, rewritin’ history. It’s that easy.

SLIDE 9

The millionth picture was uploaded in October of 2004. That’s eight months after it launched.

SLIDE 10

And here it is. It’s called Aaron S. vs. the sail.

I had to resist the temptation to scale it up in Keynote because I wanted to make the point that this is the biggest size (on Flickr). Those other pics we saw of Stewart and the crew were 500 by 375 pixels.

SLIDE 11

We’re at billions today. Yes, billions. From millions of people.

There aren’t that many, if any, traditional cultural organisations in the world that have holdings made by millions of different people. And none that I know of that have collections in the billions.

And as the great philosopher Susan Sontag said, “to collect photographs is to collect the world.”

We’re putting our cultural heritage into content platforms now. YouTube, Reddit, Flickr or Facebook, and people don’t especially expect these things to survive. It’s weird. We’ve been conditioned to expect they’re probably ephemeral (because they mostly have been). 

As I’ve been telling friends and people I meet that I’m back at Flickr working on the Foundation, they usually say something like 

“OMG I’ve got 10,000 family photos there, and that’s the only place I’ve got them.” or
“Oh, yeah. I used to use that. I loved it! Who owns it now?” or
“Ugh it got so much worse after Yahoo bought them.”

SLIDE 12

Fun fact – Yahoo bought Flickr in March 2005.

Most of us moved from Vancouver to San Francisco that year.

SLIDE 13

Yahoo Photos closed down in 2007. All Yahoo Photos account holders got an email suggesting they migrate to Flickr. That was an interesting year for the community on Flickr, welcoming hundreds of thousands of new friends all at once, and, I don’t know how much got deleted from Yahoo servers.

SLIDE 14

Then in 2017, Yahoo, and therefore Flickr, was acquired by a bigger fish called Verizon Media. And here’s the thing most people don’t know, and that’s that Flickr was nearly destroyed back in 2018, about a year later.

SLIDE 15

The story goes something like this… 

This is a Verizon executive, Mr. Gowrappan, mentioned in a January 2019 Wall Street Journal article called “Verizon To Lay Off 7% of Media Group Staff.”

He was one of the people doing a bunch of reorganisation at Verizon that Flickr was caught up in.

The article says he “has halted some efforts to sell brands within the media business, opting instead to shutter flagging brands or services.”

https://www.wsj.com/articles/verizon-to-lay-off-7-of-media-group-staff-11548270277 

SLIDE 16

“He viewed the process of selling photo-sharing service Flickr last year, for example, as too pricey and time-consuming, people familiar with the matter said.” 

Wouldn’t it be interesting to see the material he used to reach that decision? Was it a spreadsheet? Who were the other people involved? The real story is obviously way more complex than I’ve explained.

I mean, their revenue had dropped to 1.8 billion dollars for Q3 2018, so of course they needed to trim some dead weight. Of course. [shrug]

https://www.wsj.com/articles/verizon-to-lay-off-7-of-media-group-staff-11548270277 

SLIDE 17

[shrug] That’s the economic pressure of divesting. 

That’s what nearly killed Flickr.

It was “too pricey” for them to sell it. That’s what they decided. Cheaper for them to destroy it. Luckily, there were folks on the inside who helped arrange the sale, and the transfer of all the stuff to the team at SmugMug. I hope I find out who those folks were one day so I can take them to a nice dinner.

Flickr isn’t just a brand. It’s an archive.

SLIDE 18

Have a look at this headline from USA Today… that’s the article that announced SmugMug had acquired Flickr in 2018. 

Exclusive: Flickr bought by SmugMug, which vows to revitalise the photo service.

You don’t hear a company “vow” to do something very often, do you?

SLIDE 19

But, I’ve appreciated the language the SmugMug folks have used since they rescued Flickr from the jaws of Verizon execs.

“Rest easy, [our products] aren’t going anywhere.”

Interesting, isn’t it. This new and reassuring tone.

In fact, SmugMug are a member of a US-based collective called the Tugboat Institute, whose purpose is to “build and scale private, profitable, enduring, and market-leading businesses that make a dent in the universe.” 

So the MacAskills are already working on building a long company. SmugMug is actually older than Flickr. It was founded in 2002.

SLIDE 20

As I mentioned, I’ve been back at Flickr for about a year now, so we’ve been talking about the Foundation a bit here and there. I wanted to direct you to this tweet from April-ish that stopped me in my tracks and made me go hmm. It’s from Jason Scott, who has also presented at dConstruct – he’s a boisterous digital archivist who runs Archive Team, and works on the software archiving at Internet Archive. 

He tweeted:

“But as Flickr tries to refashion itself with a new goal of sustainability and continued access, please, MacAskill, remember that whether you intended to or not, you bought a World Heritage Site that comes with all the same headaches, but all the same weight.

https://twitter.com/textfiles/status/1516864234876985347 

SLIDE 21

Huh.

A World Heritage Site.
A digital World Heritage Site.

If you believe that’s what we might be dealing with, shouldn’t we be treating these huge platforms more carefully? More sustainably? More as a commons?

To be fair, 50 billion images is the Wild West. It’s so huge we would just say enjoy, and best of luck. We’ll do our very best to keep it looking nice. 

But, it’s full of networks and close looking and photos you probably remember and friends you’ve made. It’s more than the sum of its parts.

https://twitter.com/textfiles/status/1516864234876985347 

SLIDE 22

Here’s one of the MacAskills – that’s Ben, the COO and President.

He’s the person who reached out to me in the pandemic to see if I’d like to come back and help. He’s also the person, along with his brother, Don, and the SmugMug team who forked over cash and time and big effort to extract it from Yahoo, or was it Verizon?

SLIDE 23

And he’s also the person who supported creating flickr.org.

[pause]

Deleting Flickr now would be like burning 100 big libraries all at once.

Sure, it’s durable, all on Amazon Web Services and all that. But, now that we’ve had the web for – what – about 30 years(?) we’ve all witnessed how many durable things have disappeared without a trace. 

Actually, I think we’re still really young in our terms of robust and durable digital things, especially things that are more than a bucket of files. Flickr is full of connective tissue as well. Social and descriptive connective tissue. So an archive in the form of a folder of JPGs just doesn’t cut it.

SLIDE 24

How can we align Flickr to a future public domain?

Flickr can do things in ways traditional archives can’t. 

The social cataloguing that happens there is unique and deeply interconnected. It’s an archive made of millions of voices, and there’s room for millions more.

This digital world heritage site must be treated differently. And the thing about a commons is that it’s not only about resources managed in the commons like photos and software and hardware and stuff, but it’s also about the social norms that surround and shape that management.

  • We have to investigate the general morality of corporate deletion and recenter to what a 21st century commons could be like. 
  • What does owning a 21st century commons even mean? 
  • How does it need to work when there are millions of commoners? 
  • How could we do it differently?

Just thinking back… as a Flickr member (and long-time account holder) I don’t recall being asked either for my opinion or my permission in any of the corporate takeover deals that happened… We were all just told it had happened.

It’s time for us to make the conscious shift from cool applications and technologies to meaningful persistent infrastructures.

SLIDE 25

So, How to write a 100 year plan.

The short answer is – I don’t know – but, I think we need to start figuring it out, otherwise we’re going to keep loosing all the cultural heritage we’re pumping into these online services.

Writing a plan to last a century should be a group project. 

And it’s going to take at least that long to dissolve the patriarchy.

https://www.flickr.com/photos/107895189@N03/48016326542/ 

SLIDE 26

So, while we’ve been getting the foundation set up, I’ve started doing research about the plan. We’ve done 7 or 8 “How to write a 100-year plan” workshops now, and we’re hoping to do more in the coming years. I’m keen to gather people’s ideas and responses to thinking in this kind of timeframe. I’ve done a couple of workshops with the Flickr team itself too, though this idea of a 100-year plan could be put into all sorts of other contexts. 

SLIDE 27

Some very early patterns are emerging about our desire to protect the natural world. There’s also a theme about the sorts of things we’ll need to design to create an organisation that can move in the same direction across generations. Things like mentorship, documentation, and translation. The workshops are fun and informative – we don’t often think in centuries.

I don’t think there will end up being much mention of technology in the plan… We don’t even know if we’ll have JPGs in 10 years. What we can do is plan for the inevitable technology questions to be thoroughly addressed at regular intervals, by the team of the day. There doesn’t seem any point today in trying to anticipate how many servers we’ll need to buy in 2050.

SLIDE 28

I’ve also been wondering about possible futures for Flickr, and I’ve come up with five main ones to share with you today. 

SLIDE 29

Let’s start with the worst case scenario: Future 1 – ANNIHILATION.
Perhaps it will be destroyed. Perhaps it will be deleted.

There’s the giant bin of destruction.
And there’s Flickr going into it.

GONE. 

SLIDE 30

Future 2? Perhaps Flickr will continue to grow. Perhaps it will become truly gargantuan. It’s already one of the biggest collections we’ve ever assembled. Imagine if that rate continued for 100 years.

Is that we want?
Is it good? [shrug]

SLIDE 31

There’s a possible future where we don’t even make pictures ourselves anymore… I think that’s unlikely – that’d we’ll completely stop, but, there are already new picture makers, although they’re still directed by us, more or less.

This picture is from a New York Times article called We Need to Talk About How Good A.I. Is Getting. You can see a machine-generated image of a “black and white vintage photograph of a 1920s mobster taking a selfie” generated by OpenAI’s DALL-E.

Interesting to watch how copyright is developing around machine-generated pictures. 

https://www.nytimes.com/2022/08/24/technology/ai-technology-progress.html

SLIDE 32

And then you zoom in and you see how creepy and monstrous it is at the edges and you’re like UGH or maybe that’s just me?

But that’s a whole other conversation about the already centuries-long research project called “what is a photograph?” and well beyond my 30 minutes.

SLIDE 33

Perhaps it could become read only. Perhaps the service could stop accepting new members or new uploads. Even if no more photos were added to Flickr, I believe it’s still worth preserving. Most traditional institutions are read-only to the public.

We put our cultural heritage directly into corporate spaces now, at astonishing rates. But we don’t treat those platforms like our great libraries and archives. We don’t expect the same stability. I think we need to change that.

SLIDE 34

Perhaps it will be scattered to the winds. This sometimes happens with archives. This is actually not a bad idea for digital archives – make lots of copies.

So decentralisation and diffusion are definitely approaches to explore.

Dandelion illustration by hati.royani.

SLIDE 35

But, dandelions can suck because then you’re dealing with proliferation of metadata as well.

Two copies of the same photograph in two digital places, how do you know which is the “golden copy” and in particular if the metadata diverges, which version is right? It’s quite fiddly stuff.

Sometimes a photograph can even have two different licenses associated with it on two different platforms. This is not good.

Some archivists will tell you that the context of a thing is almost more important than the thing itself.

SLIDE 36

Perhaps it will slowly fade or degrade. This is quite likely.

SLIDE 37

And actually, the process of deaccession is normal in traditional cultural organisation, although it can be controversial at times. It’s usually linked to either physical space or the need for cash.

These orgs will have collection development policies, which outline how and when and under what conditions things may be deaccessioned.

It’s really a huge question: “What do we keep?”

Not all things are meant for the ages, and not all of us expect or want our stuff to be kept.

SLIDE 38

On the challenge of degrading and what to keep, I’ve been really inspired by a tiny bit of this book by Adrian Hon.

It’s called A New History of the Future in 100 Objects, and I wanted to tell you a story about one of the objects, which is Object 43, called Rituals for the Secular, set on Earth in 2033.

The Forgetting by Adrian Hon

The Society of Lethe encourages individuals to purge all their online data and social media every seven years in a ritual called The Forgetting.

In the month before the purge, people celebrate their most outstanding online contributions with their friends and family, and select just seven pieces to keep for the future.

In the week before, they acknowledge the hurt and confusion they might have caused through social media, and compose seven contributions asking for forgiveness.

In the seven hours before, they stay offline completely to experience the world as it was before everything was remembered forever.

At the moment of the Forgetting, they give the Society’s server permission to access all of their online accounts and permanently delete their data and erase all backups. Some adherents use a ceremonial hammer to smash a replica of an ancient spinning hard disk to commemorate the loss and the opportunity for growth.

SLIDE 39

I mean, it’s going to be bloody hard to keep 50 billion digital things alive, and that’s if it doesn’t grow any bigger.

So, I’m wondering whether we could we make a particular way to gather specially selected things for “deep keeping” in some kind of capsule?

Maybe everyone with a Flickr account could choose one of their pictures to deep keep and maybe even one from someone else. These would be guaranteed to survive in some way. Printing is an obvious solution here. We’re really good at keeping paper. But! Printing doesn’t capture the true nature of a Flickr picture. 

The V&A – “Street Walk” (or something like it) – Annual event where staff are unleashed into London to collect ephemera.

SLIDE 40

So, the future of Flickr is likely to be some of blend of these possible futures.

SLIDE 41

And that’s the territory for the Foundation to explore.

To finish up, I want to give you a quick glimpse of two projects we’re developing. 

(There will be lots more information on flickr.org up soon too, in case you want more info on projects, or programs, or the team, and jobs and all that.)

SLIDE 42

I’m calling the first project Data Lifeboat. It’s about planning and designing for a possible ending.

The idea is that we can just look each other in the eye and say yes, there’s a chance Flickr might go down, or be destroyed. So, let’s plan for that and design for it, so it’s a resolved end, and not scattered or partial one.

Naturally, I visited the Maritime Museum in Greenwich to learn a bit about the first lifeboats… Mr. Lionel Lukin’s design was made in 1785, which he informed the Prince of Wales in 1806 with this letter.

His design had features like:

  1. watertight bulkheads or compartments
  2. buoyant gunwales
  3. cork and other lightweight materials
  4. false iron keel to keep it upright

What might the analogues to those features be in a digital archive? What’s digital buoyancy?

SLIDE 43

Could we design a lifeboat for Flickr photos? 

How many lifeboats do we need?

How do we make them float? 

How can we make sure it stays usable on whichever beach it appears? 

Importantly, and perhaps most difficult is, how do we maintain the “original order” in a Data Lifeboat? That’s a concept central to most archival practice, that you keep the stuff you have to archive in the order in which its creator placed it. 

How do we represent a networked object in its network if we can only archive a bit of the network? How do we describe that in a way that it could be reconstituted, or at least understood?

Data lifeboats are about designing for that possible end. Since 1996, the Internet Archive has been archiving the web, and doing an amazing job. They’ve developed a practice and technology to listen out for reports of dying services, then doing their best to archive as much of that service as they can before the lights go out. There are lots of reasons this isn’t comprehensive, and I think it’s time for web platforms to help by making things easier. (There’s a whole other talk in this point too.)

SLIDE 44

There’s another idea I’d like to explore in the context of the Flickr Foundation and I thought some of you may be interested to consider in your work too, and that’s the idea of a Digital Day Book. 

I was lucky enough to do a tour of some fantastic London-ish archives this year, where I got to look into the archives at places like Windsor Castle and Waddesdon Manor (which was built by Baron Ferdinand de Rothschild, who was scorchingly wealthy). It was there I discovered that in the 18th and 19th centuries, businesses would often employ a clerk whose responsibility it was to maintain a Day Book. This was a log of everything that happened in that business. “Mr. Roberts arrived to purchase 100 brazil nuts.” “Mrs. Khan signed a lease on the Wapping property.” If you look at books like this today, you can reconstruct the actions and life and characters in a business. 

https://www.flickr.com/photos/public-domain-archief-alkmaar/49940973713/ 

SLIDE 45

We don’t have that sort of thing today. Our work is diffused across lots of different digital tools and services like Google Drive, Slack, Salesforce, WordPress, Email, Github, and thousands more.

The Digital Day Book is an idea that I hope will underpin the 100-year plan. It’s going to be important for future folks at the Foundation to be able to see the old stories of the foundation. To learn and know the purpose of the work; to see and read histories of how decisions were made, when, how, and by who. In fact, We’re going to hire an archivist in the first crew to create the first digital day book so we can see the actions and life and characters at the Foundation.

The history of Verizon’s near-fatal decision to delete Flickr is hidden in email, on long-erased whiteboards, and probably Google docs, and maybe some kind of project tracker but probably not because corporate executives know what it can mean to record decisions that will bite them on the ass later. 

Picture is an Ekphrastic Poem Network (poems about works of art)
https://www.lisarhody.com/ekphrasis-as-an-lda-network-in-nodexl/

SLIDE 46

So, to finish up, this story isn’t a new story. You can look back in human history to lots of moments where we’ve said “oof that was too close we should probably archive this properly”. I do think the way we choose to do that now could be really different, because we can choose to redirect and realign how we look after these huge digital things currently maintained by corporations towards the commons.

I’ve hope I’ve given you a quick taste of the forces at play against the future of Flickr in its current configuration. We’re in a bright positive phase at the moment, but who knows where we’ll be in 2060 when some young folks get wind of it and want to use it.

My hope is that the Flickr Foundation will become a context for interaction around figuring out how to make Flickr and perhaps even other things like Flickr survive destructive economics and corporate whimsy to last well beyond our lifetimes.

SLIDE 47

If this talk has piqued your interest, or, in particular, if there’s a challenge I’ve mentioned that you have an answer to, come and find me. We need you.

Thank you.