Andrew Mitchell

Photo of 12 soundscape researchers in front of Lake Zurich.

Having a great three days at my favourite sound conference, the Urban Sound Symposium in Zurich!

Screenshot of article titled Soundscape perception indices (SPIs): Developing context-dependent single value scores of multidimensional soundscape perceptual quality. Available at: https://doi.org/10.1121/10.0034417

Paper Fig 1. Steps for calculating the SPI. (1) Define and parameterise the target circumplex distribution; (2) sample the target distribution and prepare the test distribution; (3) compare test and target distributions using the distance metric [two-dimensional Kolmogorov-Smirnov (KS) distance, DBKS]; and (4) calculate SPI 100(1 - DBKS)

Our latest article and the capstone paper from the SSID Project!

We've spent 6 years working towards creating useful single-value indices of soundscape quality. We present SPI, our method for defining perception-focussed, context dependent Soundscape Perception Indices!

doi.org/10.1121/10.0...

'Generative AI is not the solution to every problem': Survey reveals officials' 'misconceptions' about AI GDS research finds many civil servants continue to associate artificial intelligence solely with generative tools

"The way they conceptualise AI has been shaped by the prominence of gen AI tools. The misconception that AI means LLMs and chatbots can make managing stakeholders’ expectations complicated. Generative AI tools are not the solution to every problem."

www.civilserviceworld.com/professions/...

See also:
- creating a folder
- the concept of a filepath
- savings anywhere other than the desktop
- what their downloads folder is
- googling for information
- adapting code examples in the course notes for literally the smallest change

The amount of time I've spent in my programming support office hours just talking students through how to copy and paste a file is, frankly, concerning.

Hi Elon,

1. None of your business pt. 1
2. None of your business pt. 2
3. You wouldn't understand it if I told you, so again none of your business (pt. 3)
4. Revised none of your business pt. 2
5. Wrote this dumb email.

Regards

Elsevier/ScienceDirect?

www.nature.com/articles/d41...

Veronica Coan: “Daddy, do scientists just spend most of their time saying, ‘this is bullshit, this is bullshit, and this is also bullshit?’”

Jim Coan: “Yes, my love, that is what scientists spend most of their time doing.”

Effective, efficient, and fair peer reviewing 2023 Effective, efficient, and fair peer reviewing Ben Bond-Lamberty Based on a presentation at the AGU Early Career Scientist Workshop, 2019 1

Scientists are expected to do peer reviewing but no one tells you HOW. After talking with several folks about this today at #AGU24 , it seems a good time to link to this presentation I made several years ago -- feel free to use/share! docs.google.com/presentation...

No problem! I started out studying Music Education at VCU when I took a Physics of Music class. Ended up transferring to get a BSc in Physics & Music and then a PhD in Acoustics and ML.

I'm always disappointed we don't teach acoustics to musicians as standard. Nice to see this!

This looks like a very nice first introduction! I started my now decade-long journey in acoustics through music and I still often come back to that for how I think about acoustical principles.

Any musician or student interested in exploring acoustics should enjoy starting here!

Is it commonly recommended to actually cite a genAI? The guidelines we've developed are to disclose it in the acknowledgements.

I'd only cite it if I were doing AI research on the model to give credit to its authors and document the version. The same way I'd cite software (R Core Team, 2024).

To offer some perspective as a researcher whose been in the field longer than it was trendy:

It comes from the name of the research field: the study of how to make intelligent machines. It’s not a declaration of success.

But if you want to refer to specific technologies use their names, Eg LLMs.

Agreed, we don't need to make it harder to access, we need robust regulation and enforcement on how that data is used. This is a societal issue, not a technological one.

I really hope that doesn't happen! I'm hopeful that the infrastructure setup of BlueSky and ATProtocol can help prevent that restriction, but it's definitely not certain.

Just to clarify though, we're not being scraped by AI, but for AI training, which is arguably worse...

bsky.app/profile/mitc...

Haha I guess BlueSky is the place for acousticians to come together for some amateur political theorizing

I think there's space and justifications for both capitalist and socialist viewpoints and there's no need to vilify either. But then, yeah, I sit somewhere within that fuzzy Keynesian/mixed economy/market socialism space, so of course I view them both as valid to some degree.

Yeah, I think this is where the main disagreement lies. And why I wouldn't call myself a Socialist, because I disagree with Marx that capitalism inherently and inevitably leads to feudal capitalist collapse.

That move in the US was intentional and took work from people who idolize Ayn Rand.

I agree! But I think Americans do care as far as they are terrified of being called a Socialist. And then they've been convinced the only form of capitalism is laissez faire, so they just take those policies as default and any step away needs to be justified in terms of how it won't disrupt profits.

I would hope even socialists would see the utility in allowing a middle ground. As a progressive who is much more opposed to neoliberal capitalism than I am of socialism, I'd love to see our overton window shift so that Keynesian capitalism is the political right and socialism is the left.

Just from a strategic POV I think there's value in allowing space for people to take progressive positions and still consider themselves a (Keynesian) capitalist

It shouldn't be true but there are plenty of people who want something else but are just terrified of being called a socialist

What neoliberals and Republicans did was hoover up all other definitions of Capitalism and also demonize the socialist label.

So, anyone who opposes LF gets labelled a socialist which is *scary* so people run away and think the only non-socialist option is neoliberalism.

From my point of view, neoliberalism ate all other forms of capitalism and shat out neofeudalism. Socialism is still there as its own ideologies, but there's no longer any space for non LF capitalist ideas.

What do you think the distinction is between market socialism and Keynesian capitalism?

As an American progressive, I very rarely talk to progressives who would define themselves as a market socialist.

As you said, how I or Brits define Liberal (and I agree is closest to people who call themselves progressives in the US) is closer to Keynesians, which is definitely still capitalism.

Progressives are still generally capitalists though, just not laissez faire capitalists. I agree they broke away and redefined themselves away from neoliberals. But being opposed to LF Capitalism doesn't automatically make someone a socialist or not a capitalist.

Some people use it to mean just 'the Left' or Democrat (including neoliberals, progressives, social democrats and [some] socialists) which I don't find useful. But we already have a label for neoliberals, so I'm not sure I see the point in just calling them liberal.