A Speed Limit for Computers(caolan.uk)
caolan.uk
A Speed Limit for Computers
https://caolan.uk/notes/2026-07-02_a_speed_limit_for_computers.cm
30 comments
> Though no one cares, I feel compelled to put down my 2 cents.
I am not sure if you are insulting everyone who reads your point, or your own ability to communicate.
But maybe interpret the ideas less as a dichotomy you can't untangle and so must reject, which is a fragile place to reason from, and instead identify what are good points, what are not, and improve upon either.
Because there is certainly a great deal of truth to the problems being addressed.
I am not sure if you are insulting everyone who reads your point, or your own ability to communicate.
But maybe interpret the ideas less as a dichotomy you can't untangle and so must reject, which is a fragile place to reason from, and instead identify what are good points, what are not, and improve upon either.
Because there is certainly a great deal of truth to the problems being addressed.
What truth? The whole article is absurd. Speed limits exist to reduce fatalities, not reduce inequality. Road speed limits in my area have been increasing, not decreasing, as road designs have improved. Distance didn't scale with speed either - the West was settled well before the automobile.
> Speed limits exist to reduce fatalities
In the USA the national speed limit was implemented by Richard Nixon in response to the oil crisis of 1973. The motive was to reduce energy consumption, not to reduce fatalities.
https://en.wikipedia.org/wiki/National_Maximum_Speed_Law
In the USA the national speed limit was implemented by Richard Nixon in response to the oil crisis of 1973. The motive was to reduce energy consumption, not to reduce fatalities.
https://en.wikipedia.org/wiki/National_Maximum_Speed_Law
As you might note in that link, the states already had their own speed limit laws in place, and that the national limit was later repealed. Thus, the speed limit laws are not in place to reduce gas usage.
Violating fundamental human rights sounds like a very serious charge. I kind of understood human rights to be self-evident and universal. In a transportation context, I’m hard pressed to think of much transit infrastructure without speed limits. That leads me to believe they’re not obviously abhorrent to everyone.
I have felt endangered when people careen around me at speeds that feel good to them but dangerous to me. In many cases that’s because we’re both moving at speed appropriate to our level of risk, but I’m vulnerable in ways that they are not (for example, they’re driving an SUV and I’m driving a bicycle).
With respect to TFA’s more general argument, it seems intuitive to me that more of most anything is good to a point, but there’s also such a thing as too much. How does that impinge on human rights?
I have felt endangered when people careen around me at speeds that feel good to them but dangerous to me. In many cases that’s because we’re both moving at speed appropriate to our level of risk, but I’m vulnerable in ways that they are not (for example, they’re driving an SUV and I’m driving a bicycle).
With respect to TFA’s more general argument, it seems intuitive to me that more of most anything is good to a point, but there’s also such a thing as too much. How does that impinge on human rights?
I don't know if it's a bad idea or not, but I'm struggling to understand how the idea as presented in the post would be a violation of fundamental human rights. Do you care to elaborate?
> I don't know if it's a bad idea or not, but I'm struggling to understand how the idea as presented in the post would be a violation of fundamental human rights. Do you care to elaborate?
Nowadays calling something "a violation of fundamental human rights" often really just means "I really don't like it."
Nowadays calling something "a violation of fundamental human rights" often really just means "I really don't like it."
So I suppose I should have added a bit of a disclaimer: I genuinely believe free access to computation and information to be a human right; up there with the press 100 years ago.
Restricting computers, morally, is like restricting the printing press. Many people look back on the latter with disdain, and today want to censor the Internet.
Restricting computers, morally, is like restricting the printing press. Many people look back on the latter with disdain, and today want to censor the Internet.
I'm still thinking through this and you seem to have strong opinions. What do you think of speed limits for cars and e-bikes?
Well, to start with, If you drive your car too fast, you might kill people. If you drive your computer too fast, it might get data corruption and/or reboot unexpectedly.
Or you might build a data center that poisons a community's water and drives up the cost of energy for your neighbors. We can't pretend there are zero negative externalities that accompany unconstrained compute.
To be clear I'm not necessarily agreeing with the idea, but to be fair, there's more to it than you're suggesting.
To be clear I'm not necessarily agreeing with the idea, but to be fair, there's more to it than you're suggesting.
There's a great deal of demand for compute. Data centers are a very efficient way of providing that compute in a single place with limited resources. If computers were restricted to being slower and less efficient than they are now, people would build even more, and larger, data centers.
How does a person running a computer too fast cause them to build a building? Maybe make that part illegal, not the indirect cause. Otherwise we may wanna outlaw breathing, it may cause people to murder.
Like I said, I don't necessarily agree with the idea and I don't feel strongly enough about it to really argue in its favor, but to answer the question: the same reason why OpenAI doesn't operate out of Sam Altman's garage.
At a certain level of compute you need specialized infrastructure -- such as a purpose-built datacenter -- for the energy needs (and really, I think the stronger argument to be made here is about energy, not raw speed, and where the argument might fall apart is the historical fact that compute tends to become more energy-efficient over time).
Not sure whether the breathing/murder analogy is apt, but I get where you're coming from and I would probably agree that a blanket restriction on computer speed wouldn't be appropriate.
At a certain level of compute you need specialized infrastructure -- such as a purpose-built datacenter -- for the energy needs (and really, I think the stronger argument to be made here is about energy, not raw speed, and where the argument might fall apart is the historical fact that compute tends to become more energy-efficient over time).
Not sure whether the breathing/murder analogy is apt, but I get where you're coming from and I would probably agree that a blanket restriction on computer speed wouldn't be appropriate.
Data centers don't poison a community's water and the people claiming that they do are lying because they dislike generative AI and want to weaponize populist environmentalist sentiment and state regulations against the institutions doing AI inference and training.
They also don't particularly drive up the cost of energy any more than any other phenomenon in human society that uses energy, and the right solution to this is to build more power generation capacity, which is something we should be doing anyway because abundant energy is the foundation of all human prosperity.
They also don't particularly drive up the cost of energy any more than any other phenomenon in human society that uses energy, and the right solution to this is to build more power generation capacity, which is something we should be doing anyway because abundant energy is the foundation of all human prosperity.
I'm not anti-AI or anti-data center, in fact I lean more towards "pro-AI" overall, but to say that everyone who claims data centers can contaminate water is lying is a strong claim that doesn't really hold up to scrutiny. If they're saying "every data center without exception poisons water" then sure, they're lying, but that's certainly not what I'm saying. A couple examples from this year are linked below. I understand if you're skeptical of politicization in the highly-publicized Georgia case [1], but it's harder to dismiss what happened in Wyoming [2].
[1] https://news.bloomberglaw.com/environment-and-energy/epa-to-... [2] https://www.wyomingnews.com/news/local_news/cheyenne-bopu-tr...
As for energy: honestly I agree with you that the solution is to build more power generation capacity, but that doesn't change the fact that in the meantime energy prices are already increasing substantially in many areas because of data centers [3].
[3] https://www.eenews.net/articles/data-centers-drive-76-surge-...
[1] https://news.bloomberglaw.com/environment-and-energy/epa-to-... [2] https://www.wyomingnews.com/news/local_news/cheyenne-bopu-tr...
As for energy: honestly I agree with you that the solution is to build more power generation capacity, but that doesn't change the fact that in the meantime energy prices are already increasing substantially in many areas because of data centers [3].
[3] https://www.eenews.net/articles/data-centers-drive-76-surge-...
I'm actually more skeptical of the politicization of the Wyoming case than of the Georgia case, after reading the linked article.
The thing that Wyoming News describes happening was:
- a construction contractor working on building a data center for Meta was using normal utility water for some construction purpose that might have been related to preparing the building's coolant system for use (the article is actually not clear about whether that was the specific source of the water), and then discharging it into the normal sanitary sewer system
- the local utility district was doing routine testing of their wastewater for fecal bacteria and accidentally discovered "Cupriavidus gilardii is a rare type of bacteria found naturally in the environment, such as in soil and water", which they tracked down to the construction discharges
- because the city of Cheyenne uses reclaimed and treated wastewater for some irrigation purposes, and because this bacteria has a risk of being harmful to immunocompromised people, out of an abundance of caution the utility district banned the construction project from discharging wastewater into the sanitary sewer
- it's still unclear exactly where the bacteria was coming from
These facts strike me as basically having nothing at all to do with data centers being harmful in a way that literally any other aspect of the built human environment isn't. Every construction site for anything uses water industrially and discharges it somehow. It's basically normal for wastewater to have bacteria in it, because it's wastewater - and ignoring the fact that it's not clear the bacteria have anything to do with the construction site being a data center as opposed to any other type of building - it's not clear that this type of bacteria being in the wastewater is actually harmful to anything. Every time I've seen reclaimed wastewater being used for irrigation somewhere I've seen signs posted saying so, in order to prevent people from thinking that the water is potable or treated.
The article says "Instead, BOPU now requires industrial companies using closed-loop cooling systems to construct separate collection systems so any water from cooling equipment or associated floor drains is directed into storage tanks, rather than the city’s sanitary sewer." and I'm frankly extremely skeptical that this is a good policy. It makes constructing a building that uses a closed-loop cooling system more expensive, which includes (some) data centers but also lots of other types of building, and I don't think there's any reason to assume that the health risk from this bacteria being in wastewater justifies the cost, nor that the policy even prevents this bacteria from getting into the wastewater.
I'm sure that despite Meta's public words about "being a good neighbor in Cheyenne", the people actually working on this project are pretty pissed off at the local utility district officials for making their job harder and more expensive, and I don't think they're wrong to feel this way. And I strongly suspect that the reason this story is being spread, and possibly the reason the journalists at the Wyoming Tribune Eagle chose to write about it to begin with, is because it involves the keywords "data center" and "water" and superficially seems bad. Certainly if this same incident happened during the construction of, say, a grocery store, I'd be unlikely to hear about it - unless a lot of journalists had an ideological problem with that grocery store, in which they'd report about it in exactly the same terms they're reporting on this incident.
The thing that Wyoming News describes happening was:
- a construction contractor working on building a data center for Meta was using normal utility water for some construction purpose that might have been related to preparing the building's coolant system for use (the article is actually not clear about whether that was the specific source of the water), and then discharging it into the normal sanitary sewer system
- the local utility district was doing routine testing of their wastewater for fecal bacteria and accidentally discovered "Cupriavidus gilardii is a rare type of bacteria found naturally in the environment, such as in soil and water", which they tracked down to the construction discharges
- because the city of Cheyenne uses reclaimed and treated wastewater for some irrigation purposes, and because this bacteria has a risk of being harmful to immunocompromised people, out of an abundance of caution the utility district banned the construction project from discharging wastewater into the sanitary sewer
- it's still unclear exactly where the bacteria was coming from
These facts strike me as basically having nothing at all to do with data centers being harmful in a way that literally any other aspect of the built human environment isn't. Every construction site for anything uses water industrially and discharges it somehow. It's basically normal for wastewater to have bacteria in it, because it's wastewater - and ignoring the fact that it's not clear the bacteria have anything to do with the construction site being a data center as opposed to any other type of building - it's not clear that this type of bacteria being in the wastewater is actually harmful to anything. Every time I've seen reclaimed wastewater being used for irrigation somewhere I've seen signs posted saying so, in order to prevent people from thinking that the water is potable or treated.
The article says "Instead, BOPU now requires industrial companies using closed-loop cooling systems to construct separate collection systems so any water from cooling equipment or associated floor drains is directed into storage tanks, rather than the city’s sanitary sewer." and I'm frankly extremely skeptical that this is a good policy. It makes constructing a building that uses a closed-loop cooling system more expensive, which includes (some) data centers but also lots of other types of building, and I don't think there's any reason to assume that the health risk from this bacteria being in wastewater justifies the cost, nor that the policy even prevents this bacteria from getting into the wastewater.
I'm sure that despite Meta's public words about "being a good neighbor in Cheyenne", the people actually working on this project are pretty pissed off at the local utility district officials for making their job harder and more expensive, and I don't think they're wrong to feel this way. And I strongly suspect that the reason this story is being spread, and possibly the reason the journalists at the Wyoming Tribune Eagle chose to write about it to begin with, is because it involves the keywords "data center" and "water" and superficially seems bad. Certainly if this same incident happened during the construction of, say, a grocery store, I'd be unlikely to hear about it - unless a lot of journalists had an ideological problem with that grocery store, in which they'd report about it in exactly the same terms they're reporting on this incident.
The article seems to be saying that lack of equity is worse than lack of innovation. Obviously we can strive for having both equity and innovation, but it really seems like the points made in the article are arguing that, for example, if there exists a discrepancy in what speeds are available to different people, this is worse than only the lowest speed being available to everyone.
The linked article from 1973 is very strange to read in 2026 and honestly it's hard to take seriously. It seems to actually argue that China and India should stop developing because development encourages dependence on energy. It says that machines are slaves that modern people are required to master?
The speed equity discussed in the linked paper has actually increased significantly as well. More people have cars, and more people can afford plane trips, and bicycle infrastructure is better all over the nation (the world, probably) compared to 1973. I agree with the general principle that cities should encourage diverse development so more people can choose not to use cars if they want, or if they don't have access to a car. In the paper, this would be achieved by somehow state limiting the amount of energy people would be able to consume per capita. I'm glad that world didn't come to pass.
Basically it sounds like the point of view of the article is that at some point the state should tell people they're not allowed to try to make a certain process more efficient in terms of time or resources, because that might drive demand for better productivity, which would be bad because it drives us further from nature and community. Some of the points intersect with beliefs I hold, but I strongly oppose this way of going about it.
The linked article from 1973 is very strange to read in 2026 and honestly it's hard to take seriously. It seems to actually argue that China and India should stop developing because development encourages dependence on energy. It says that machines are slaves that modern people are required to master?
The speed equity discussed in the linked paper has actually increased significantly as well. More people have cars, and more people can afford plane trips, and bicycle infrastructure is better all over the nation (the world, probably) compared to 1973. I agree with the general principle that cities should encourage diverse development so more people can choose not to use cars if they want, or if they don't have access to a car. In the paper, this would be achieved by somehow state limiting the amount of energy people would be able to consume per capita. I'm glad that world didn't come to pass.
Basically it sounds like the point of view of the article is that at some point the state should tell people they're not allowed to try to make a certain process more efficient in terms of time or resources, because that might drive demand for better productivity, which would be bad because it drives us further from nature and community. Some of the points intersect with beliefs I hold, but I strongly oppose this way of going about it.
The author is a socialist, quoting a socialist from 1973 whose ideas if implemented would've made everyone's lives worse. The idea that a lack of equity is worse than a lack of innovation is a core principle of socialism and it's the core reason why socialism is bad.
We have speed limits for vehicles because speed kills.
In computing, waiting kills (indirectly, by wasting time). Speed is life.
Some roads have minimum speed limits. If we're talking about limits, that's the kind of limit we want.
In computing, waiting kills (indirectly, by wasting time). Speed is life.
Some roads have minimum speed limits. If we're talking about limits, that's the kind of limit we want.
I am partial to this speed-is-life sentiment, and it made me think of this beautiful and wonderfully poignant demoscene production by Farbrausch and Haujobb: "Time Index". It is made in the memory of a friend of theirs that died.
The softsynth soundtrack includes lyrics and one of them is "we slow down" which I always interpreted as a kind of demoscener's lament since making things go fast is sorta the whole point!
https://youtu.be/fngv1dCFrdo?si=tR-1uQ4vKIPLHZu3
I am not arguing for either side of the fastness debate, while I certainly adore fast computers I don't like the mental image of our computers all blazing away doing stuff mainly humans care about, while the the world 'outside' is steadily getting hotter and more polluted.
The softsynth soundtrack includes lyrics and one of them is "we slow down" which I always interpreted as a kind of demoscener's lament since making things go fast is sorta the whole point!
https://youtu.be/fngv1dCFrdo?si=tR-1uQ4vKIPLHZu3
I am not arguing for either side of the fastness debate, while I certainly adore fast computers I don't like the mental image of our computers all blazing away doing stuff mainly humans care about, while the the world 'outside' is steadily getting hotter and more polluted.
People are too hooked on their feeds to accept an imposition that doesn't decrease their risk of physical harm.
Self-imposable speed limits like screen time limits and blocking services and proxies do exist already and certainly could reduce overall non-physical harm if imposed on everyone - but we struggle as a species to agree on the impact of non-bodily harms.
Self-imposable speed limits like screen time limits and blocking services and proxies do exist already and certainly could reduce overall non-physical harm if imposed on everyone - but we struggle as a species to agree on the impact of non-bodily harms.
As a developer, I first thought about a different point when I saw the title. We usually use high-end devices that mask the performance issues in the software we develop. There is some effort and tooling, but I don't think it's enough.
Artificial limits or more regulation might not be the answer, yet post-AI scarcity can lead us to interesting places.
Artificial limits or more regulation might not be the answer, yet post-AI scarcity can lead us to interesting places.
I like the idea, and it reminded me that the „Vim book“ had the subtitle: „Edit text at the speed of thought“.
How much faster should an ‚automated thinking‘ than a ‚manual thinking‘ be (allowed to be)?
How much faster should an ‚automated thinking‘ than a ‚manual thinking‘ be (allowed to be)?
This is a blog post in search of a solution in search of a problem. IE nothing.
Why are you comparing apples to oranges? Implementing a speed limit, it size limit on a computer is just idiotic. It works be again to saying, no you can't use a calculator to do the arithmetic, you must use pen and paper, because I said so. Which means who ever doesn't implement this arbitrary speed limit is going to eat you for lunch. This is a Luddite argument.
The device to point at in this context is the Remarkable e-ink tablet. It isn't an iPad. It's a digital sheet of paper. What it makes up for in lack of features is what sells it. It doesn't have a webbrowser or Instagram or TikTok. That would be missing the point. A "simple" digital pen and paper. Optionally a keyboard. Has wifi and cloud connectivity.
Such a "speed limit" is an unfathomably bad idea, in every way, and moreover is a violation of fundamental human rights. I deeply oppose it. That is all.