"CERN Invented the Web" Isn't an Argument for Anything

I mentioned in passing in the Forbes post about science funding that I'm thoroughly sick of hearing about how the World Wide Web was invented at CERN. I got into an argument about this a while back on Twitter, too, but had to go do something else and couldn't go into much detail. It's probably worth explaining at greater-than-Twitter length, though, and a little too inside-baseball for Forbes, so I'll write something about it here.

At its core, the "CERN invented WWW" argument is a "Basic research pays off in unexpected ways" argument, and in that sense, it's fine. The problem is, it's not anything more than that-- its fine as an argument for funding basic research as a general matter, but it's not an argument for anything in particular.

What bugs me is now when it's used as a general "Basic research is good" argument, but that it's used as a catch-all argument for giving particle physicists whatever they want for whatever they decide they want to do next. It's used to steamroll past a number of other, perfectly valid, arguments about funding priorities within the general area of basic physics research, and that gets really tiresome.

Inventing WWW is great, but it's not an argument for particle physics in particular, precisely because it was a weird spin-off that nobody expected, or knew what to do with. In fact, you can argue that much of the impact of the Web was enabled precisely because CERN didn't really understand it, and Time Berners-Lee just went and did it, and gave the whole thing away. You can easily imagine a different arrangement where Web-like network technologies were developed by people who better understood the implications, and operated in a more proprietary way from the start.

As an argument for funding particle physics in particular, though, the argument undermines itself precisely due to the chance nature of the discovery. Past performance does not guarantee future results, and the fact that CERN stumbled into a transformative discovery once doesn't mean you can expect anything remotely similar to happen again.

The success of the Web is all too often invoked as a way around a very different funding argument, though, where it doesn't really apply, which is an argument about the relative importance of Big Science. That is, a side spin-off like the Web is a great argument for funding basic science in general, but it doesn't say anything about the relative merits of spending a billion dollars on building a next-generation particle collider, as opposed to funding a thousand million-dollar grants for smaller projects in less abstract areas of physics.

There are arguments that go both ways on that, and none of them have anything to do with the Web. On the Big Science side, you can argue that working at an extremely large scale necessarily involves pushing the limits of engineering and networking and working in those big limits might offer greater opportunities for discovery. On the small-science side, you can argue that a greater diversity of projects and researchers offers more chances for the unexpected to happen compared to the same investment in a single enormous project.

I'm not sure what the right answer to that question is-- given my background, I'm naturally inclined toward the "lots of small projects (in subfields like the one I work in)" model, but I can see some merit to the arguments about working at scale. I think it is a legitimate question, though, one that needs to be considered seriously, and not one that can be headed off by using WWW as a Get Funding Forever trump card for particle physics.

More like this

A number of people have commented on this LA Times op-ed by Steve Giddings about what physicists expect to come out of the Large Hadron Collider. It includes a nice list of possible particle physics discoveries plus a few things that will annoy Peter Woit, and also includes the obligatory note…
Back in late July, I got email from a writer for Physics World magazine (which is sort of the UK equivalent of Physics Today), asking my opinion on a few questions relating to particle physics funding. The basis for asking me (as opposed to, you know, a particle physicist) was presumably a post…
Michael Nielsen, who's so smart it's like he's posting from tomorrow, offers a couple of provocative questions about the perception of a crisis in funding for basic science: First, how much funding is enough for fundamental research? What criterion should be used to decide how much money is the…
A week or so ago, lots of people were linking to this New York Review of Books article by Steven Weinberg on "The Crisis of Big Science," looking back over the last few decades of, well, big science. It's somewhat dejected survey of whopping huge experiments, and the increasing difficulty of…

What Tim Berners-Lee did was invent HTTP, which went on to become the standard for what we call the World Wide Web. But there were competing protocols in the early 1990s: gopher and archie, and even FTP (an even older protocol which is still used today in specific situations). It was hardly inevitable that HTTP would become the standard.

The main thing about HTTP was that it was intended to be independent of client hardware and software. Some scientists in those days used VAX machines, and others used Unix boxes. Some even used MS-DOS (early versions of Windows were only recently available at the time and did not achieve significant market penetration until a few years later). HTTP didn't care what you had on your end, as long as you could connect to the server. Which made it a good solution for a problem experimental particle physicists had at the time: how to share their data with their collaborators.

So yes, "CERN invented the Web" is a valid argument for funding science in general, but it is not a valid argument for being more specific about what science is to be funded.

By Eric Lund (not verified) on 24 Mar 2017 #permalink

Wasn't the Internet itself a US Government-funded DARPA project? CERN only created the WWW as an easy way to connect and navigate it.

We could equally argue that GPS is a reason to fund differential geometry (no differential geometry => no general relativity => GPS doesn't work) and algebraic geometry (no algebraic geometry => no arithmetic of elliptic curves => what do you think internet security protocols are based on?). In general, I'd say what it means is fund *all* basic research, because you don't know which part of it will be applicable (and when).

Chad, your post is disappointing. It's a straw-man and it's parochial and leaves an inaccurate impression about my field. While I cannot speak for every single one of my particle physics (and yes, I work at CERN) colleagues, as a former chair of my department, former chair of the DPF, former chair of the Fermilab Users Executive Committee, former co-convener of the Energy Frontier working group for the (sort of decadal) "snowmass study" about the future physics in particle physics, and former chair of the US ATLAS Institutional board I have many times been called on to justify basic research in general and particle physics in particular.

I've never used the WWW as a reason to fund particle physics in the spirit that you suggest and certainly never seen anyone do so. "Steamroll" is just not fair, nor is the implication that particle physics gets what it wants because it "invented" the WWW. Of course it didn't. The infrastructure was in place in the US exactly as KAL notes, as a DARPA project and a matter of national preparedness in the cold war. And, as Eric points out, HTTP was Berners-Lee's contribution. That wasn't entirely trivial and it did break conceptual ground about how the large group communications problems that LEP experiments were facing might be facilitated.

We have often pointed at the WWW as an example of how problems get solved in science...maybe even problems that weren't necessarily recognized as problems at all. But we always do so in the broader context of increased funding for all of basic research. In fact, the GPS example from Barbara is perfect. Who would have thought that GR would be a practical need?

The best spinoff argument for particle physics (It's somewhat controversial as to whether "spinoff" should ever be used as an argument. I think it's fair.) is accelerator physics which enjoys the position of having two feet planted firmly in both the basic research landscape as well as an increasingly vibrant applied landscape. We don't do accelerator-based research anywhere in the world without them.

Particle physics doesn't get everything it wants. I don't know a single research group in the US that has not seen significant cuts in university research grants over at least the last 3 cycles. So again, you leave an inaccurate implication.

The "we'll do better with your money than you will" argument (again "steamroll" sticks in my craw) is not a winner for anyone. We've always argued on behalf of my physics - and yours - and everyone's basic research, especially at universities and I wish you'd do the same. I'm glad this wasn't in Forbes, frankly...which was a good article, apart from the short swipe at particle physics.

I enjoy your books, but this post, as I say, disappointed me a lot.

By Chip Brock (not verified) on 27 Mar 2017 #permalink

I'm sorry to disappoint (but then, I'll be happy to refund you the full cost of the post...), and glad to hear you don't use that argument in that way. If you've never heard it used as such, though, we're clearly not talking to the same set of people.

I've heard the WWW invoked innumerable times when the question of whether particle physics is a good investment relative to other areas of science. I've even had someone say to me "Return on investment? What, the Web isn't good enough for you?" (or words very close to those).

I agree that there's a good case to be made, particularly in the area of accelerator physics, that development of a next generation machine might offer good opportunities for technological advancements that will have impact well outside the fundamental physics community. I'd like to see that argument made more frequently, particularly in public outreach contexts. It's far more common, in my experience, for discussion of particle-physics spinoffs to begin and end with the Web, though, and I think that's a shame.

I find the accelerator argument to be totally bogus, because accelerators and accelerator-based treatments came out of nuclear physics back when the highest energy physics used cosmic rays to do experiments in balloons or on mountain tops rather than accelerators in a laboratory.

Eric, @ #1,misses the important detail that ftp offered most of what you need to share data. It wasn't neat and tidy if you didn't have a decent sense of how to put together directory tree, but it enabled the easy movement of data and was widely used for that purpose. It didn't have to be replaced if the users were all techno-geeks.

You all seem to miss the importance of Hypercard (a vast improvement over a directory tree and Readme files) to the idea of hypertext. That came out of Apple, not government research, but it was signficant that Apple machines were widely used on the non-computing side of physics research. HTTP was a very clever invention of a way to make the Hypercard idea work independently of hardware and o/s platforms. Maybe it wasn't inevitable, but I know people who argued ad nauseum that people with messy file systems to manage should use it, only to be met with "but I use a VAX", and never went further than to mutter under their breath. .

By CCPhysicist (not verified) on 01 Apr 2017 #permalink