28 Oct 2023
The Co-op Digital newsletter occasionally had stories in it, tiny fictions to explore what retail and the world might look like as software and big tech tried to remake it. They looked at the bigger picture of what technology change might mean for humans and communities, and the second order effects. The earlier stories are here, and these are the 3 that I signed off with:
Supermarkets and many other types of retail businesses are often high volume/tight margin machines, looking to increase volume and find small gains and complementary products and services with higher margins. Will this always be the case?
As climate change bites into the world, making it less predictable, will the cost and availability of materials change the dynamic for retail? Do supply chains fracture? Will managing the entire lifecycle of products and packaging add cost that makes retail unviable, or will it unlock new business? Can shopping fit into circular economies if recycling and reuse becomes central? What if shops bought things from you, instead of selling them to you?
2040
Aya looks the length of the warehouse. There’s a haze in the distance where the unfactory starts and above the superstructure is barely visible behind the ambient dayglow. A city inside a building.
She pulls the door shut on her housing unit and walks to the bus stop. There’s only a couple of minutes to wait before a driverless bus pulls up. These carbon grey boxes crawl the building, patiently looping between the different work, housing and leisure zones.
Aya gets on the bus and sits. Her glasses immediately pop up an interactive about hitting this month’s KPIs. With a glance she flicks it away, and looks the other way to access shopping. A carousel of plaid shirts flickering across the field of view, partly obscuring the view out of the window of apartment blocks, terraced 2-up 2-downs, community centres, and behind her in the distance parkland and the executive mcmansions. Aya considers buying a shirt that says 2005 and has all the confidence of that era. The checkout dances and says: “Hey you know you could put that 229.50 into something more long term? Say Pension to put it to work for you.” Aya sighs, shuts her eyes and whispers “Cancel”, the UI an afterimage in reversed colours.
Aya steps off the bus at the warehousing zone and passes through two layers of perimeter security: ID badge and then a tongue print which performs ID verification and a recreational drugs test. Now she is matter-side. She steps through the air wall into a cavernous space: kilometres of racking, ducting, assembly areas, disassembly areas, recycling, storage, cabling, transport, uncountable shelves of racking and gigantic bays of raw material.
10 years ago, at peak shopping, this building was Amazon’s largest fulfilment centre, the entire building performing inventory, pick and pack - all of it heading into cardboard boxes, out the loading bays and to everyone’s home via the Amazon-only lanes on the motorways.
Since then, the building has grown and evolved in response to 15 years of carbon budgets and circular economy targets. The shadow of the smile logo is still faintly visible on the wall. The work is now two-way: they’re doing as much inbound for recycling as they do outbound retail: the increasing value of material feedstocks made disassembly and recycling higher margin businesses. Across the way, building crews are pulling out the pick and pack racking, and hauling in more recyc stations.
Conveyor belts of toys, consumer goods, electronics, plastic items, everything. Some things are automatically sorted, separated by size or material type into smaller flows on smaller conveyors, a river delta. The metal goes off along a conveyor to be remelted in a distant area. The plastic is ripped apart into uncountable white-edged grains of feed stock.
Further away, and behind more layers of security, are the goop vats where the synthetic bio voodoo happens. There’s a smell of ammonia and long chain polymers everywhere, which the air ducting cannot ever remove entirely.
Here though, the river belt meanders through rows of stations, where human-robot pairs work next to each other, pulling objects off the main belt and moving them to several smaller tributary belts for onward sorting.
Aya walks up to the only unstaffed station. A fixed robot is at the wide conveyor belt. Its body is a 7-axis spine of rotatable stainless steel segments that look more like muscles than components. Its arm ends in an array of cameras, lidar sensors and rubber tipped and suction manipulators. Around the robot is a 2 metre no-go zone hashed out in hazard orange. It has placed several objects in that space around its base. A Casio watch. A christmas tree toy. A clown painting on black velvet. Lego pieces.
The supervisor approaches and shrugs at Aya: “You make it work, if you can. I in’t going near it though…” Aya swipes her badge into the work station. The robot’s arm monitors everything that passes on the belt under it, occasionally twitching as if indecisive. But it is not moving and not sorting items.
Aya watches the belt for a while and reaches out to pick up an old lunch box in blue and red plastic. She shows it to the robot, and then with great care Aya places it on the floor, inside the no-go zone, and steps back. The robot’s arm bends, the array inspecting the lunchbox for several seconds.
Then it straightens, moves back over the belt, and starts grabbing objects and sorting them.
What if shops did more? What if shops were slower? What if they were more meaningfully embedded into the communities they serve? Typically technology makes products repeatable, processes automatable, data insight-able (etc), and reduces costs. Can it do more than that? Can it free up all of colleague time, to spend with customers?
Anyway, here is the newsletter and new conversational AI ChatGPT trying to write a story about a supermarket together. Assuming that tools like ChatGPT continue to improve, it’s also worth thinking about what the world looks like when machines can initiate human conversation. And when anyone can generate plausible text (or imagery or…), using tools whose models may have opaque biases and insufficient consent. What happens to trust?
9 Dec 2022 (the pronoun change halfway through seems to be my fault, not ChatGPT’s):
2035
Let’s write a story together. I’ll go first and you respond. OK? Here’s the opening: Xavi parks their ebike at the rack outside the Freshco supermarket. It bloops to let Xavi know it’s secured and charging up.
[ChatGPT] Xavi smiles to themself as they walk into the Freshco. They love their ebike, it’s been such a game changer for them. As they enter the supermarket, they grab a cart and start to make their way down the aisles, picking up all the items on their shopping list. As they move from one aisle to the next, they can’t help but feel grateful for their ebike, which has made it so much easier for them to get around town and live a more sustainable lifestyle.
Inside, Xavi shivers after the cold draft of anti-viral air at the entrance.
[ChatGPT] They quickly make their way to the produce section, where they select a few fresh fruits and vegetables for their meals this week. As they move through the aisles, they notice that the Freshco is not as busy as usual. Xavi wonders if it’s because of the cold weather outside or if it’s due to the ongoing pandemic.
Down the right side of the store, shoppers are using the kiosks, and at the counter is a screen and a worker, both with a smile.
[ChatGPT] Xavi continues to make their way through the store, grabbing everything they need for the week. As they approach the checkout, they see that there are several kiosks available for self-checkout, but they decide to go with a human cashier instead. They approach the counter and greet the cashier, who smiles and greets them back. Xavi notices that the cashier is wearing a mask and gloves, and there is a screen between them to help protect against the spread of the virus.
Xavi checks out, but then remembers that he needs to talk to the community service team, and heads over to an open area with armchairs. Xavi puts the bag on the floor and drops into a chair. Evie, a staffer, joins him, says Hi, and the green padlock on the chair’s arm lights up, indicating that their conversation will be private. Evie: “How can I help?” Xavi: “Well, …”
[ChatGPT] Xavi pauses for a moment, gathering his thoughts. He looks at Evie, who is patiently waiting for him to continue. Xavi then says, “Well, I wanted to talk to you about the community service programs that Freshco offers. I’ve been thinking about getting involved and volunteering my time, but I’m not sure where to start. Can you tell me more about what’s available and how I can get involved?”
[ChatGPT] Evie nods and smiles. She begins to explain the different community service programs that Freshco offers, and how Xavi can sign up to volunteer. She tells him about the different causes they support, such as food banks, environmental conservation, and community outreach programs. She also gives him information about the schedule and requirements for volunteering, and how he can sign up online or in-store. Xavi listens intently, taking notes on his phone as Evie speaks. He is grateful for her help and is excited to get involved and make a difference in his community. He thanks Evie for her time and assistance, and they say their goodbyes. Xavi picks up his bag and heads out of the Freshco, feeling motivated and inspired to make a positive impact in the world.
He turns back and says “Yeah, but where to start?” Evie says: “Look, it’s more… dynamic than looking at what’s available. I can point you at a project right now, or you can look for a team online. Either way, they’ll need to decide if you can help make a difference to their mission happen.”
[ChatGPT] An error occurred. If this issue persists please contact us through our help center at help.openai.com.
Evie sighs at the error message, and uses her team badge to restart the openai service.
Xavi leaves with the shopping, considering what “making a difference” means. Xavi forgets to collect the ebike.
[ChatGPT] Too many requests, please slow down
Are you ok ChatGPT?
Local stores offer a convenient geographic proximity - they’re just round the corner. Could they offer a second kind of proximity, and be a critical part of communities? Could they help communities thrive with services, support, infrastructure as well as just the shop and the local causes bits? Can a shop turn into a true member organisation? Can it help repair and maintain community?
These questions matter because the world is changing ever faster, and everything (products, services, institutions, infrastructure, jobs, cities etc) will need looking after, improving or remaking.
2030
“Can you fix it? Because this bike and I have been through a lot.” says Reenie at the bike shop.
Jim, behind the counter, nods. “Your battery has done a lot of kilometres and it is very tired now. I can swap it out for a new one for 499 which is the battery and the labour. Or I can sign you up for this Gobattery service which is a new battery every year for 199 annually and the labour’s free.”
“Which would you do?”, asks Reenie.
“Um, probably not Gobattery”, Jim murmurs quietly.
“OK. What happens to the old battery?”
“And your old battery will go have a relaxing retirement doing grid backup for a shopping centre or somewhere like that.” Then more brightly, Jim adds: “We’re a member of The Co-op, are you? I can add you in the app, it only costs 20. You get discounts in shops, vote on decisions, and help each other out. I can give you a 50 discount right now, so the card will cost you nothing. But it’s like you’re part of” - Jim’s hands wheeling - “you know, a… Thing.”
“Yeah OK, sounds interesting? Thanks.”
Jim taps at the terminal and there’s a quick transactional handshake of lights and bings. An app downloads to Reenie’s mobile, which briefly pops a green tick saying that the app is properly signed and isn’t hoovering up too much personal data. Reenie opens the app - it has already onboarded her - which says both of their Co-op memberships have earned 200 points for the transaction, and a green leaf grows on the screen: a small but positive update to the carbon budgets of Reenie, of Jim, and of the Co-op itself.
“Nice. It all adds up, you’ll see”, says Jim.
Several weeks later, and late in the evening, Reenie’s Co-op app bleeps out a red alert. A fellow member needs help: their premises is flooding. Reenie shrugs, grabs a coat and heads for the bike. The map takes her back to the bike shop.
Before she can make a joke about the emergency services arriving, Jim shouts: “The river burst its banks! Move things up, or put sandbags here, please?”.
And Reenie and a couple of others get to work, placing sandbags, lifting everything onto higher shelves, bailing water out of the shop. The culvert is cleared and the flood water starts to recede. It takes two hours. The bottom layer of the shop is now a landscape of grey mud over unidentifiable product shapes. The same mud that coats them.
Jim says: “Thank you so much. This stuff needs to dry out but some of it’ll be ok. Hey, take anything you want too, seriously, thanks. And I’ll update the app and send you points tomorrow.”
“OK cool. But it’s not about the points.”