The web is dying and its creators are watching

The inventor of the web and the founder of Mozilla on what we're losing, what's left, and whether it's too late.

Guest: Mitchell Baker, Tim Berners-Lee · ~42 min · generated by glm-5.2

~42 min · Guest: Mitchell Baker, Tim Berners-Lee

Title options

The web is dying and its creators are watching
Can the open web survive the platform era?
Two architects return to a house they built and barely recognize

Cold open (0:00)

30-second hook

There's a moment—I think it was maybe 2019—when I realized I hadn't typed a URL into a browser in weeks. Everything came from a feed. A search box owned by one company. A timeline owned by another. And the links, the actual links, the thing that was supposed to make the web a web, they'd become... decorative. Little anchors pointing to nowhere anyone clicked. So the question I can't stop turning over is: did the open web die, or did it just... stop being where people live?

Host intro (1:00)

Host

Thanks for being here. I'm talking today with two people who built the thing we're mourning, or maybe just diagnosing. Tim Berners-Lee invented the World Wide Web in 1989 at CERN — wrote the first browser, the first server, the protocols underneath everything you've touched online since you were a teenager. And Mitchell Baker co-founded the Mozilla Project, built it into a foundation that has been the closest thing the open web has to a standing army, and has led it for over twenty years. The topic is the death of the open web. Not whether it's happening — I think we can stipulate that — but how it happened, whether it was inevitable, and what, if anything, can still be reclaimed. Tim, I want to start with you. When you first imagined the web, what did you think the failure mode would be? Because I don't think you expected this one.

Main interview

Q1 · Host

Tim, I want to start with you. When you first imagined the web, what did you think the failure mode would be? Because I don't think you expected this one.

Why this question

This is the opening question of the episode and it frames the entire conversation — not as nostalgia but as diagnostic.

Follow-up plan

Listen for whether he says he expected corporate capture vs. government censorship vs. technical fragmentation. If he says he didn't anticipate platform consolidation, ask: 'What was the closest thing in your mind to what actually happened?' If he says he did see it coming in some form, ask: 'At what point did that expectation crystallize, and did you think you had time to prevent it?'

Q2 · Host

Mitchell, you came to this from intellectual property law, not engineering. When you look at what happened to the open web, do you see a legal failure first, or a technical one?

Why this question

Mitchell's legal background is underexplored in most interviews and reframes the open web question through property and rights rather than code.

Follow-up plan

If she says legal: follow with 'What specific legal infrastructure was missing — was it about not having the right licenses, or not having the right enforcement?' If she says technical: follow with 'Do you think the architecture of the browser itself was part of the problem — that it became a consumption device rather than a creation tool?'

Q3 · Host

Tim, Solid is your answer to the data sovereignty problem. But I keep wondering — is the reason people don't own their data because they can't, technically, or because they don't want to? Most people seem to prefer convenience over ownership. How do you design around that preference?

Why this question

This pushes on the most common critique of Solid without dismissing it — the adoption problem is real and Berners-Lee rarely addresses it head-on.

Follow-up plan

If he argues people would choose ownership if they understood the tradeoff: ask 'Isn't that what we said about privacy in 2012? And the needle hasn't moved much.' If he concedes convenience wins: ask 'So is Solid a tool for the minority who care, and is that enough to change the system?'

Q4 · Host

Mitchell, Mozilla laid off about a quarter of its staff in 2020 and pivoted toward AI. To someone watching from outside, that looked like the open web's biggest institutional defender deciding the open web fight was unwinnable on its own terms. Was it?

Why this question

The 2020 layoffs are the elephant in the room with Mitchell Baker and most interviewers either skip it or accept the press release answer.

Follow-up plan

If she frames it as evolution not retreat: ask 'What specifically can AI do for the open web that the browser couldn't?' If she acknowledges it was a retreat: ask 'Does the open web need a Mozilla-sized institution anymore, or has the model of institutional defense failed?'

Q5 · Host

Tim, you've said the original web was meant to be read-write — collaborative from the start. But the architecture you built, HTTP and HTML, didn't really enforce that. It could be read-write, but nothing in the protocol demanded it. Do you think the web was always going to drift toward read-only consumption because the protocol let it?

Why this question

This is a genuine architectural question that goes to whether centralization was latent in the original design, not imposed later.

Follow-up plan

If he says the protocol was intentionally permissive by design: ask 'Was that permissiveness the original sin — the thing that let platforms fill the vacuum?' If he says a different protocol would have been better: ask 'Is there a version of the web with identity and permissions baked in that you considered and rejected?'

Q6 · Host

Mitchell, you've spent decades arguing that 'open' is a value, not just a technical specification. But I want to press on that. Was the open web ever actually a values consensus, or was it just a brief window where nobody had figured out how to monetize it yet?

Why this question

This tests whether 'open' was ever a shared commitment or just an accident of timing, which is the deepest question about whether it can come back.

Follow-up plan

If she insists there was a genuine values moment: ask 'Who actually held those values — was it developers, users, institutions?' If she acknowledges the economic vacuum: ask 'Does that mean the open web was always contingent on the absence of a business model, and once one arrived, it was over?'

Q7 · Host

Tim, you've been remarkably public about your concerns, but you've also said you're an optimist. I want to understand what the optimism is actually anchored in. Not hope — what's the mechanism? What specifically reverses the trajectory we're on?

Why this question

Berners-Lee's optimism is often asserted but rarely interrogated — this asks for the actual causal chain, not the sentiment.

Follow-up plan

If he points to decentralized tech like Solid or Web3: ask 'Those have been around for years without broad adoption — what changes?' If he points to generational shift or public awareness: ask 'Do you think awareness has actually increased, or have people just accepted the trade?'

Q8 · Host

Mitchell, there's something I've been turning over. The open web movement has always argued that concentration of power is the problem. But the early web was also kind of chaotic and hostile and full of garbage, and platforms partly won because they made things work better. Is there a version of the open web that acknowledges people wanted the walled garden — that it wasn't just imposed on them?

Why this question

This is the question most open web advocates dodge — that platforms offered something people genuinely wanted, not just something they were tricked into.

Follow-up plan

If she concedes the user experience gap: ask 'Can open web tools ever match the polish of centralized platforms, or is that a category error — like asking a public park to have the amenities of a private club?' If she pushes back that the garden was a trap: ask 'Was it a trap from day one, or did it become one — and when?'

Q9 · Host

I want to ask you both this. When you talk to young developers now — people in their twenties who never knew the open web as anything but a concept — what do you tell them they're building for? Because 'open' doesn't seem to animate them the way it animated your generation. What replaces that word, if anything?

Why this question

This bridges to the generational question that determines whether the open web is a living project or a historical one.

Follow-up plan

If either says a new framing is needed: ask 'What word would you use — sovereignty, agency, ownership, something else?' If either says the framing still works: ask 'Why doesn't it land — is it the message or the audience?'

Q10 · Host

Tim, last question for you. If in fifty years someone writes the history of the web, and the chapter on the open web era is titled — what's the title? What was that period actually about?

Why this question

This gives Berners-Lee space to frame the whole arc — as history, not as advocacy — and often produces the most honest answer.

Follow-up plan

If the title is tragic or elegiac: ask 'Do you think that's the final chapter, or is there a sequel?' If the title is something like 'The First Attempt': ask 'What does the second attempt look like differently — what did the first one teach us that the next one has to get right?'

Pushback segment

Pushback 1

Setup: Tim, I want to push on something because I think you're too generous to the architectural argument. You tend to talk about the open web's decline as a problem of design — if we'd had better protocols, better data ownership, things would have gone differently.

Argument: But I think the counter is stronger than you usually grant. The web didn't centralize because the protocols were weak. It centralized because attention is a natural monopoly, because network effects reward the first platform that reaches scale, and because the economic logic of advertising demands aggregation. You could have built the perfect protocol in 1991 and Google and Facebook still would have emerged. The architecture was downstream of the economics, not the other way around.

Hand-back: If the economics were the real driver, can a protocol — even a brilliant one — actually reverse that?

Pushback 2

Setup: Mitchell, I want to challenge the framing of 'open' as inherently good, because I think that assumption is what made the open web movement vulnerable. You and Mozilla have always treated openness as axiomatic — the more open, the better.

Argument: But the open web was also the web of unregulated surveillance, of SEO spam, of misinformation without guardrails, of communities that became toxic because anyone could enter. The walled gardens didn't just offer convenience — they offered safety, moderation, trust. When the open web movement treats openness as an unqualified good, it's refusing to engage with why people left. They didn't leave because they were tricked. They left because the open web had real problems that the platforms, for all their flaws, actually addressed.

Hand-back: So my question is — what does the open web movement owe the people it failed before it asks them to come back?

Pushback 3

Setup: I want to push on the nostalgia that creeps into this conversation, and I'll put this to both of you because I think it affects the whole project. There's a version of this story where the 'open web era' was a golden age, and everything since has been decline.

Argument: But the early web was also exclusionary — you needed a computer, a connection, technical literacy. It was open in theory but accessible to very few. The platforms that 'killed' the open web also brought two billion people online who weren't there before. If we're honest, the open web was a small, privileged, mostly Western conversation that got outgrown by its own success. Framing the current moment as decline rather than transformation might be the thing that keeps the open web movement from actually reaching the people it claims to serve.

Hand-back: Is it possible the open web didn't die — it just became something neither of you recognizes, and that something isn't necessarily worse?

Mid-roll ad read (~18:00)

I want to sit with what Tim just said for a second, but first — a few things that make this show possible.

Sponsor: Quartermaine Hosting — Independent web hosting for people who actually care where their bytes live.

If you're building something on the open web — a blog, a wiki, a weird little project nobody understands yet — you need hosting that won't quietly change its terms on a Tuesday. Quartermaine gives you flat pricing, no upsell maze, and real humans who pick up the phone. Use code OPENSOURCE at quartermaine.co.

Sponsor: Vellum & Code — The reading app for people who think bookmarks are a form of procrastination.

Vellum & Code pulls long-form articles, research papers, and docs into one clean reading space — no ads, no pop-ups asking for your email before you've read a sentence. It's what the web feels like when nothing's trying to hijack your attention. Free for your first fifty articles at vellumandcode.com.

Sponsor: Halden Analytics — Privacy-first analytics that don't require a twenty-page consent banner.

Halden gives you traffic insights without cookies, without fingerprinting, without treating your visitors like data livestock. One line of script, GDPR-clean out of the box, and a dashboard that shows what you actually need. If Tim Berners-Lee wouldn't approve of your analytics stack, maybe fix that. halden.dev.

Okay — Mitchell, you were starting to say something about the moment you realized Mozilla's fight wasn't just about a browser anymore.

Listener questions

Listener Q1

We got a question that cuts right to the thing nobody wants to say out loud about the open web — which is whether we're mourning something that was already doomed by design.

Listener: If the open web was always going to consolidate because that's what markets do to platforms, was the open web ever really open, or were we just living in a brief window before the economics caught up with the architecture?

Listener Q2

Here's one from a listener who clearly spent some time staring at a ceiling thinking about this.

Listener: Tim, you've said the web was a permissionless space by accident — nobody had to ask CERN if they could build a website. What's the one architectural decision you'd change if you could go back to 1989, knowing what consolidation would look like thirty years later?

Listener Q3

And then there's this — a question I think Mitchell is uniquely positioned to answer, and honestly I don't know what she'll say.

Listener: Mitchell, Mozilla fought for the open web from inside a browser that most people use as a door to closed platforms. At what point do you look at the gap between what Mozilla stood for and what Firefox users actually do with it and ask whether the mission itself needs to be redefined?

Outro

Host

Tim Berners-Lee, Mitchell Baker — thank you. Both of you. The thing I'm going to carry out of this conversation is something Tim said almost in passing: that the open web wasn't a thing we built, it was a thing we failed to enclose. The walls came later, and they came because we let convenience outrun conviction. Mitchell, your framing of Mozilla as the canary, not the solution — that's the honest thing most people in this space won't say. So here's what I'd ask you to do. If this conversation made you uncomfortable in a useful way, share it with one person who'd argue with you about it. That's how the open web worked when it worked. Subscribe wherever you listen. Send us your questions — really, send them — to mail at theshow dot fm. I'm not going to tell you the internet is going to be okay. I'm going to tell you it's not decided yet. And that's the part that matters.

Show notes

Tim Berners-Lee and Mitchell Baker sit down together for the first time to interrogate the thing they each spent their lives building — and whether it's already gone. They argue about whether the open web was ever a values consensus or just a window before monetization arrived, whether people actually wanted the walled garden, and what you say to a 25-year-old developer who never knew the open web as anything but a nostalgia project.

Chapters

Timestamp	Section
0:00	Cold open
1:00	Host intro
2:00	Question 1
5:00	Question 2
8:00	Question 3
11:00	Question 4
14:00	Question 5
17:00	Question 6
20:00	Question 7
23:00	Question 8
26:00	Question 9
29:00	Question 10
32:00	Mid-roll: the ad read that earned its place
36:00	Listener question 1
37:30	Listener question 2
39:00	Listener question 3
40:30	Outro
42:30	End

SEO tags

open web Tim Berners-Lee Mitchell Baker data sovereignty platform consolidation

60-second social clip

Cut for X, LinkedIn, Shorts, TikTok