Journal tags: opinions

It always annoys me when a politician is accused of “flip-flopping” when they change their mind on something. Instead of admiring someone for being willing to re-examine previously-held beliefs, we lambast them. We admire conviction, even though that’s a trait that has been at the root of history’s worst attrocities.

When you look at the history of human progress, some of our greatest advances were made by people willing to question their beliefs. Prioritising data over opinion is what underpins the scientific method.

But I get it. It can be very uncomfortable to change your mind. There’s inevitably going to be some psychological resistance, a kind of inertia of opinion that favours the sunk cost of all the time you’ve spent believing something.

I was thinking back to times when I’ve changed my opinion on something after being confronted with new evidence.

In my younger days, I was staunchly anti-nuclear power. It didn’t help that in my younger days, nuclear power and nuclear weapons were conceptually linked in the public discourse. In the intervening years I’ve come to believe that nuclear power is far less destructive than fossil fuels. There are still a lot of issues—in terms of cost and time—which make nuclear less attractive than solar or wind, but I honestly can’t reconcile someone claiming to be an environmentalist while simultaneously opposing nuclear power. The data just doesn’t support that conclusion.

Similarly, I remember in the early 2000s being opposed to genetically-modified crops. But the more I looked into the facts, there was nothing—other than vibes—to bolster that opposition. And yet I know many people who’ve maintainted their opposition, often the same people who point to the scientific evidence when it comes to climate change. It’s a strange kind of cognitive dissonance that would allow for that kind of cherry-picking.

There are other situations where I’ve gone more in the other direction—initially positive, later negative. Google’s AMP project is one example. It sounded okay to me at first. But as I got into the details, its fundamental unfairness couldn’t be ignored.

I was fairly neutral on blockchains at first, at least from a technological perspective. There was even some initial promise of distributed data preservation. But over time my opinion went down, down, down.

Bitcoin, with its proof-of-work idiocy, is the poster-child of everything wrong with the reality of blockchains. The astoundingly wasteful energy consumption is just staggeringly pointless. Over time, any sufficiently wasteful project becomes indistinguishable from evil.

Speaking of energy usage…

My feelings about large language models have been dominated by two massive elephants in the room. One is the completely unethical way that the training data has been acquired (by ripping off the work of people who never gave their permission). The other is the profligate energy usage in not just training these models, but also running queries on the network.

My opinion on the provenance of the training data hasn’t changed. If anything, it’s hardened. I want us to fight back against this unethical harvesting by poisoning the well that the training data is drawing from.

But my opinion on the energy usage might just be swaying a little.

Michael Liebreich published an in-depth piece for Bloomberg last month called Generative AI – The Power and the Glory. He doesn’t sugar-coat the problems with current and future levels of power consumption for large language models, but he also doesn’t paint a completely bleak picture.

Effectively there’s a yet-to-decided battle between Koomey’s law and the Jevons paradox. Time will tell which way this will go.

The whole article is well worth a read. But what really gave me pause was a recent piece by Hannah Ritchie asking What’s the impact of artificial intelligence on energy demand?

When Hannah Ritchie speaks, I listen. And I’m well aware of the irony there. That’s classic argument from authority, when the whole point of Hannah Ritchie’s work is that it’s the data that matters.

In any case, she does an excellent job of putting my current worries into a historical context, as well as laying out some potential futures.

Don’t get me wrong, the energy demands of large language models are enormous and are only going to increase, but we may well see some compensatory efficiencies.

Personally, I’d just like to see these tools charge a fair price for their usage. Right now they’re being subsidised by venture capital. If people actually had to pay out of pocket for the energy used per query, we’d get a much better idea of how valuable these tools actually are to people.

Instead we’re seeing these tools being crammed into existing products regardless of whether anybody actually wants them (and in my anecdotal experience, most people resent this being forced on them).

Still, I thought it was worth making a note of how my opinion on the energy usage of large language models is open to change.

But I still won’t use one that’s been trained on other people’s work without their permission.

There’s a lot of emphasis put on decision-making: making sure you’re making the right decision; evaluating all the right factors before making a decision. But we rarely talk about revisiting decisions.

I think perhaps there’s a human tendency to treat past decisions as fixed. That’s certainly true when it comes to evaluating technology.

I’ve been guilty of this. I remember once chatting with Mark about something written in PHP—probably something I had written—and I made some remark to the effect of “I know PHP isn’t a great language…” Mark rightly called me on that. The language wasn’t great in the past but it has come on in leaps and bounds. My perception of the language, however, had not updated accordingly.

I try to keep that lesson in mind whenever I’m thinking about languages, tools and frameworks that I’ve investigated in the past but haven’t revisited in a while.

Andy talks about this as the tech tool carousel:

The carousel is like one of those on a game show that shows the prizes that can be won. The tool will sit on there until I think it’s gone through enough maturing to actually be a viable tool for me, the team I’m working with and the clients I’m working for.

Crucially a carousel is circular: tools and technologies come back around for re-evaluation. It’s all too easy to treat technologies as being on a one-way conveyer belt—once they’ve past in front of your eyes and you’ve weighed them up, that’s it; you never return to re-evaluate your decision.

This doesn’t need to be a never-ending process. At some point it becomes clear that some technologies really aren’t worth returning to:

It’s a really useful strategy because some tools stay on the carousel and then I take them off because they did in fact, turn out to be useless after all.

See, for example, anything related to cryptobollocks. It’s been well over a decade and blockchains remain a solution in search of problems. As Molly White put it, it’s not still the early days:

How long can it possibly be “early days”? How long do we need to wait before someone comes up with an actual application of blockchain technologies that isn’t a transparent attempt to retroactively justify a technology that is inefficient in every sense of the word? How much pollution must we justify pumping into our atmosphere while we wait to get out of the “early days” of proof-of-work blockchains?

Back to the web (the actual un-numbered World Wide Web)…

Nolan Lawson wrote an insightful article recently about how he senses that the balance has shifted away from single page apps. I’ve been sensing the same shift in the zeitgeist. That said, both Nolan and I keep an eye on how browsers are evolving and getting better all the time. If you weren’t aware of changes over the past few years, it would be easy to still think that single page apps offer some unique advantages that in fact no longer hold true. As Nolan wrote in a follow-up post:

My main point was: if the only reason you’re using an SPA is because “it makes navigations faster,” then maybe it’s time to re-evaluate that.

For another example, see this recent XKCD cartoon:

“You look around one day and realize the things you assumed were immutable constants of the universe have changed. The foundations of our reality are shifting beneath our feet. We live in a house built on sand.”

Perhaps the best example of a technology that warrants regular re-evaluation is the World Wide Web itself. Over the course of its existence it has been seemingly bettered by other more proprietary technologies.

Flash was better than the web. It had vector graphics, smooth animations, and streaming video when the web had nothing like it. But over time, the web caught up. Flash was the hare. The World Wide Web was the tortoise.

In more recent memory, the role of the hare has been played by native apps.

I remember talking to someone on the Twitter design team who was designing and building for multiple platforms. They were frustrated by the web. It just didn’t feel as fully-featured as iOS or Android. Their frustration was entirely justified …at the time. I wonder if they’ve revisited their judgement since then though.

In recent years in particular it feels like the web has come on in leaps and bounds: service workers, native JavaScript APIs, and an astonishing boost in what you can do with CSS. Most important of all, the interoperability between browsers is getting better and better. Universal support for new web standards arrives at a faster rate than ever before.

But developers remain suspicious, still prefering to trust third-party libraries over native browser features. They made a decision about those libraries in the past. They evaluated the state of browser support in the past. I wish they would re-evaluate those decisions.

Alas, inertia is a very powerful force. Sticking with a past decision—even if it’s no longer the best choice—is easier than putting in the effort to re-evaluate everything again.

What’s the phrase? “Strong opinions, weakly held.” We’re very good at the first part and pretty bad at the second.

Just the other day I was chatting with one of my colleagues about an online service that’s available on the web and also as a native app. He was showing me the native app on his phone and said it’s not a great app.

“Why don’t you add the website to your phone?” I asked.

“You know,” he said. “The website’s going to be slow.”

He hadn’t tested this. But years of dealing with crappy websites on his phone in the past had trained him to think of the web as being inherently worse than native apps (even though there was nothing this particular service was doing that required any native functionality).

It has become a truism now. Native apps are better than the web.

And you know what? Once upon a time, that would’ve been true. But it hasn’t been true for quite some time …at least from a technical perspective.

But even if the technologies in browsers have reached parity with native apps, that won’t matter unless we can convince people to revisit their previously-formed beliefs.

The technologies are the easy bit. Getting people to re-evaluate their opinions about technologies? That’s the hard part.