Blair Fix

![](https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=150%2C150&ssl=1){.attachment-thumbnail
.size-thumbnail .wp-post-image width="150" height="150" decoding="async"
srcset="https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=150%2C150&ssl=1 150w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=1200%2C1200&ssl=1 1200w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=800%2C800&ssl=1 800w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=600%2C600&ssl=1 600w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=400%2C400&ssl=1 400w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=200%2C200&ssl=1 200w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=450%2C450&ssl=1 450w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=60%2C60&ssl=1 60w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?resize=550%2C550&ssl=1 550w, https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?zoom=2&resize=150%2C150&ssl=1 300w"
sizes="(max-width: 150px) 100vw, 150px" attachment-id="8469"
permalink="https://economicsfromthetopdown.com/2024/04/27/five-years-of-economics-from-the-top-down/five_cover/"
orig-file="https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?fit=2000%2C1600&ssl=1"
orig-size="2000,1600" comments-opened="1" image-meta="{"
aperture\":\"0\",\"credit\":\"\",\"camera\":\"\",\"caption\":\"\",\"created_timestamp\":\"0\",\"copyright\":\"\",\"focal_length\":\"0\",\"iso\":\"0\",\"shutter_speed\":\"0\",\"title\":\"\",\"orientation\":\"0\"}\"=""
image-title="five_cover" image-description="" image-caption=""
medium-file="https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?fit=300%2C240&ssl=1"
large-file="https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/five_cover.png?fit=723%2C578&ssl=1"}

My how time flies. As of April 11th, 2024, I've been blogging for five
years. To celebrate, I thought I'd engage in some obligatory naval
gazing.

### Why blog?

I
[started](https://economicsfromthetopdown.com/2019/04/11/economics-from-the-top-down/){target="_blank"
rel="noopener"} this blog on a whim. In the spring of 2019, I was one
year post PhD and busy publishing pieces of my
[dissertation](https://osf.io/preprints/thesiscommons/7uqw9){target="_blank"
rel="noopener"}. It was about as much fun as licking sandpaper.

The problem, I now realize, is that I hate academic writing. Although
academic prose may be the orthodox dialect of science, it's also the
surest way to kill the intrigue of scientific inquiry. Science is full
of wrong turns, heated debates, huge breakthroughs, and foolish
mistakes. In short, science is like a courtroom drama. But the academic
presentation of the drama is the equivalent of legalese ... dry and
unreadable.

What's odd, though, is that in 2019, I didn't know that I hated academic
writing. Instead, I thought I disliked *writing* itself. You see, for
seven years of grad school, I'd leapt into new research, delighted by
the tendrils of data that led to unexpected places. But when it came
time to write up the results, I felt mostly dread.

In hindsight, the problem was that I hated writing like an academic.
Grad-school training had me convinced that writing should be self
important, needlessly complicated, and utterly humourless. So when I
started blogging, I had an immediate [sense of
relief](https://economicsfromthetopdown.com/2019/05/28/resources-for-recovering-academic-writers/){target="_blank"
rel="noopener"}. Instead of writing tortured prose, I could write
honestly about what I thought. It was hugely gratifying.

To be sure, I still find writing difficult. Often, I struggle for hours
(sometimes days) to express an idea clearly. But while this churn is
frustrating, the reward of clear communication is worth it. When the
idea is finally on paper, crystallized in a form that (I hope) the
average person can understand, I feel like I've crafted something worth
sharing.

Alternatively, if I've done something scientifically foolish (which I'm
sure I have), I'm proud to do it boldly and clearly.

### The joy of writing research notes

Besides enjoying the tone of blog prose, I like that blogging is a way
to resurrect the 'research note'.

A century ago, scientific papers were concise affairs. A scientist would
discover something interesting. Then they would write a short research
note that said 'here is something interesting'. The published note would
have minimal references, and a only a cursory nod to other literature.
In short, the paper was the equivalent of a modern blog post.

Over time, though, the requirements for a formal scientific paper grew
more onerous, especially the need to conduct a literature review. The
result, as Figure [1](#fig-lit-review) shows, was the persistent rise of
the phrase 'literature review' in written English.

[]{#fig-lit-review}

<figure class="wp-block-image size-large">
<img
src="https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2024/04/lit_review.png?w=723&amp;ssl=1"
decoding="async" data-recalc-dims="1" />
<figcaption><strong>Figure 1: Scientists, thou shalt review thy
literature.</strong></figcaption>
</figure>

To be sure, reviewing what other people have said about a topic is good
scholarship. And there is certainly a place for well-crafted reviews of
past thinking. (In a dissertation, for example.) That said, requiring
every scientific paper to have an exhaustive 'literature review' is
frustratingly redundant.

What's also redundant is the need to funnel science through academic
publishers. These organizations are essentially vampires that suck 30%
profit margins out of publicly funded research. [We don't need these
monsters](https://economicsfromthetopdown.com/2021/03/17/why-are-most-textbooks-still-proprietary/){target="_blank"
rel="noopener"}. Science is a public good, and it can be conveyed
swiftly and easily through the public internet.

In short, blogging is an easy way for scientists to document their work
without the burden of formal publishing. So if you're a young scientist
(or even an old one), I'd encourage you to think of blogging as the best
way to convey your research.

### Blogging as note taking

Now to some obligatory life lessons.

During my masters degree, I once had a committee member bombard me with
criticism of my thesis. After several minutes of onslaught, he noticed
that I wasn't taking notes. The meeting did not end well.

Now the truth was that I thought the committee member's points were
fairly silly, and hence, not worth writing down. But the flip side is
that I've always been bad at taking notes. I had no system then, and I
have no system now. I just read stuff and remember whatever my brain
happens to remember.

That said, I now realize that blogging is how I take notes. It was Cory
Doctorow who tipped me off to how this works. (It's what he calls the
[memex
method](https://pluralistic.net/2021/05/09/the-memex-method/){target="_blank"
rel="noopener"}.) The idea is that no matter how great your system, note
taking on your own computer tends to devolve into entropic decay. You
take notes. Then you forget about them. Repeat.

In a sense, the problem is that when you write notes solely for
yourself, there's no pressure to be coherent. Unfortunately, my
experience is that it's the pressure to be coherent that makes notes
memorable. Because blogging is public, it forces a moment of reflection
before you write something down. It's this reflection --- driven by the
implicit threat of public scrutiny --- that staves off the entropic
decay of personal note taking. Or at least, it does for me.

The other nice feature of blog-based note taking is that it makes your
ideas easily searchable. All modern website builders come with full-text
search that is generally better than what you get on your computer. Add
the fact that your public blog will be crawled by search engines, and
you have a built-in recipe for indexing your ideas for future reference.

### Don't write SEO spam

Now to some obligatory advice about blogging. First up, don't write SEO
spam.

Some backstory. When I started blogging in 2019, I quickly learned that
business bloggers were obsessed with 'SEO' --- search engine
optimization. The idea was that you should write content that catered to
certain key words --- words that would get you ranked highly on search
engine results. I thought SEO targeting was stupid then, and I continue
to think it's stupid now.

The difference is that today, we know that SEO bullshit is ruining the
internet. SEO is a way to [game search
engines](https://en.wikipedia.org/wiki/Spamdexing){target="_blank"
rel="noopener"} --- a way to place vacuous, ad-driven clickbait at the
top of search results. The consequence is that if you search for
something on the modern web, you have to sift through a mountain of
useless spam. Unfortunately, the popularity of large language models has
only made things worse, vastly increasing the ability of internet
enshittifiers to churn out SEO swill.

If you're starting a blog today, take no part in this enshittification
industry. Write for humans, not search engines.[^1^](#fn1){#fnref1
.footnote-ref target="_blank" role="doc-noteref" rel="noopener"}

### Search-engine traffic is over-rated

Speaking of search engines, there's a common assumption that
search-engine traffic is the route to internet success. But for the most
part, it's not.

Think of search engines like the tourist industry. Some websites are the
equivalent of Hawaii --- a place that depends on tourist traffic for its
survival. Such sites are almost always selling something, and their
content caters to the lowest common denominator. But other corners of
the internet are like a great local restaurant --- places that thrive
mostly by word of mouth.

A science blog (like this one) will invariable land in the second
category. Sure, some people will stumble upon it by chance. But most of
these tourists will eat one meal and never come back. The real audience
for your science blog are the readers who become 'locals' --- the ones
who read your stuff regularly.

Now the truth is that building a community of 'locals' is a tough slog.
There's no shortcut other than to keep at it. Sure, the odd post may go
viral. But my experience is that this short-term traffic has almost no
relation to long-term readership.

For example, when captain Elon Musk
[tweeted](https://economicsfromthetopdown.com/wp-content/uploads/2024/01/musk_dk.png){target="_blank"
rel="noopener"} a link to my post on the [Dunning-Kruger
effect](https://economicsfromthetopdown.com/2022/04/08/the-dunning-kruger-effect-is-autocorrelation/){target="_blank"
rel="noopener"}, it created a boom of tourist traffic. But almost all of
the visitors ate a single meal and never returned. (That's probably a
good thing. If a Musk-bro liked my research, it would suggest that I'm
doing something wrong.)

To summarize, while you're ignoring SEO, forget about targeting search
engines as well. Write for the group of engaged readers who like what
you do.

### Blogging is dead. Long live blogs {#blogging-is-dead.-long-live-blogs}

Now to some obligatory reflection.

The funny thing about 'blogging' is that the term itself has long been
an anachronism. Named after a contraction of the phrase 'web log', the
'blog'
[arose](https://en.wikipedia.org/wiki/History_of_blogging){target="_blank"
rel="noopener"} in the late 1990s as a kind of online diary, devoted to
the minutia of everyday life. But by the mid 2010s, this way of blogging
had been mostly killed off by social media. What survived were 'blogs'
that focused on issues of public importance. Science communication
flourished, as did web writing about all kinds of niche topics.

By the 2020s, people continued to declare that 'blogs' were dead.
Meanwhile social media had so thoroughly killed off traditional
journalism that reporters were fleeing to the blogosphere in
droves.[^2^](#fn2){#fnref2 .footnote-ref target="_blank"
role="doc-noteref" rel="noopener"} Of course, these journalists
rebranded their blogs as 'newsletters'. But pot*a*to, pot*ah*to. Call it
whatever you want ... my point is that self-published, internet-hosted
writing is alive and well.

### Thoughts on the future

Now to some obligatory speculation about the future of this blog. What's
in store? The short answer is *more of the same*. I'm going to continue
to do research that interests me, and I'll continue to write about it on
the internet.

That said, lately I've been thinking about the massive rise in audio and
video content on the web. Part of me dislikes this content, as it's
prone to gimmickry and meaningless pyrotechnics. But another part of me
realizes that it fills a huge void. The truth is that often, it's easier
to get knowledge from a talking person than from a word-laden page. For
example, I'm an avid consumer of YouTube talks about astrophysics.
Usually, the talks are accessible in a way that scientific papers are
not.

Turning to my own work, I try hard to make my research accessible. But
my guess is that by virtue of it being written only, I'm excluding a
(potentially large) group of people who'd prefer to have audio-visual
versions of my posts. The problem is that pontificating on video is
*not* my thing. I think slowly. So if I appear eloquent in print, it's
because I've agonized over my choice of language. Give me an open mic,
and I'll mostly sound like an idiot.

That said, I've noticed a trend in which bloggers record themselves
reading their own posts. This approach *does* interest me. Still, my
posts are typically data heavy, and charts don't translate well into
audio. So what to do?

Looking to the future, I'm toying with making audio versions of my posts
that come with embedded slides. (So basically, a YouTube slideshow.)
This approach appeals to me because I could make some whizzbang code
that automates the slideshow from the figures that are already in my
posts. Then I'd just have to record myself reading the text that I've
already written.

Anyway, in the near future, you may find audio/video versions of my work
that accompany the normal text version. Perhaps you will find it useful.

### Thanks for reading

This ends my anniversary trip gazing at my own naval. Regular
programming will resume soon, with an upcoming post about the [Jevons
paradox](https://en.wikipedia.org/wiki/Jevons_paradox){target="_blank"
rel="noopener"}.

To everyone who reads this blog, thank-you for joining me on this
scientific journey. If you have the means, the best way to support my
work is by [becoming a
patron](https://economicsfromthetopdown.com/membership/){target="_blank"
rel="noopener"}. And the second best way is to recommend the blog menu
to passing tourists. Some of these folks may become regulars!

------------------------------------------------------------------------

#### Support this blog

Hi folks. I'm a crowdfunded scientist who shares all of his
(painstaking) research for free. If you think my work has value,
consider becoming a supporter.

[![member_button](https://i0.wp.com/economicsfromthetopdown.com/wp-content/uploads/2023/05/supporter_button-1.png?w=220&ssl=1){.aligncenter
decoding="async"
recalc-dims="1"}](https://economicsfromthetopdown.com/membership/){target="_blank"
rel="noopener"}

------------------------------------------------------------------------

#### Stay updated

Sign up to get email updates from this blog.

:::: {.jetpack_subscription_widget}
##  {#section .widgettitle}

::: {.wp-block-jetpack-subscriptions__container}
Email Address

Keep me up to date
:::
::::

------------------------------------------------------------------------

[![](https://economicsfromthetopdown.files.wordpress.com/2021/03/by.png?w=723){.aligncenter
decoding="async" width="150"
recalc-dims="1"}](http://creativecommons.org/licenses/by/4.0/){rel="license"}  
This work is licensed under a [Creative Commons Attribution 4.0
License](http://creativecommons.org/licenses/by/4.0/){rel="license"}.
You can use/share it anyway you want, provided you attribute it to me
(Blair Fix) and link to [Economics from the Top
Down](https://economicsfromthetopdown.com/).

------------------------------------------------------------------------

### Notes

::: {#footnotes .footnotes .footnotes-end-of-document role="doc-endnotes"}
1.  [As Cory Doctorow [puts
    it](https://pluralistic.net/2024/01/15/passive-income-brainworms/){target="_blank"
    rel="noopener"}, "No one reads 2,000 words of algorithm-pleasing LLM
    garbage over an omelette recipe and then subscribes to that site's
    feed."[![↩](https://s.w.org/images/core/emoji/15.0.3/72x72/21a9.png){.wp-smiley
    style="height: 1em; max-height: 1em;"}︎](#fnref1){.footnote-back
    target="_blank" role="doc-backlink" rel="noopener"}]{#fn1}
2.  [More precisely, it was socially media *ad-tech* that killed off
    traditional journalism. As Facebook became a content aggregator, it
    sucked up the ad revenue that would have gone to newspapers. Google
    did the same thing, translating its search monopoly into an ad-tech
    monopoly. So while people may still *read* newspapers, for the most
    part, businesses no longer *advertise* in
    them.[![↩](https://s.w.org/images/core/emoji/15.0.3/72x72/21a9.png){.wp-smiley
    style="height: 1em; max-height: 1em;"}︎](#fnref2){.footnote-back
    target="_blank" role="doc-backlink" rel="noopener"}]{#fn2}
:::

The post [Five Years of Economics from the Top
Down](https://economicsfromthetopdown.com/2024/04/27/five-years-of-economics-from-the-top-down/)
appeared first on [Economics from the Top
Down](https://economicsfromthetopdown.com).


My how time flies. As of April 11th, 2024, I’ve been blogging for five years. To celebrate, I thought I’d engage in some obligatory naval gazing. Why blog? I started this blog on a whim. In the spring of 2019, I was one year post PhD and busy publishing pieces of my dissertation. It was about as much fun as licking sandpaper. The problem, I now realize, is that I hate academic writing.

Five Years of Economics from the Top Down

Jeroen Ooms

The latest version of the rOpenSci [av
package](https://docs.ropensci.org/av) includes some useful new tools
for working with audio data. We have added functions for reading,
cutting, converting, transforming, and plotting audio data in any
popular audio / video format (mp3, mkv, aac, etc).

The functionality can either be used by itself, or to prepare audio data
for further analysis in R using other packages. We hope this clears an
important hurdle to use R for research on speech, music, and [whale
mating calls](https://youtu.be/F8Zt3mYlOqU).

## Audio analysis in R

CRAN has serveral cool packages for audio analysis, most notably the
[tuneR](https://CRAN.R-project.org/package=tuneR) package, maintained by
Uwe Ligges. However, getting your audio data into R can be tricky,
because real-world audio come in all sorts of formats and codecs, with
varying sampling rates, channel layouts, and so on.

I got stuck on this myself when trying to use sound from youtube videos,
or extract a short fragment from a long recording. The existing packages
mostly assume small wav files, but it is unclear where to get started to
analyze the complete oeuvre of Taylor Swift or a 3 hour recording of
bird sounds.

The av package builds on the
[FFmpeg](http://ffmpeg.org/documentation.html) libav libraries, which
provide a extensive, high performance implementations for streamable
reading, editing, and writing of media in any format. The av package is
already used in R for working with video, for example to create
[animated
graphics](https://ropensci.org/technotes/2018/10/06/av-release/) or
[sampling pictures from a camera
stream](https://ropensci.org/technotes/2019/09/27/ropensci-docs/) for
image analysis. By adding audio functionality, we hope to make it more
feasible to analyze real-world sound data directly in R.

## Converting between audio formats

The simplest use of `av` is converting an audio file into another
format. The
[av_audio_convert](https://docs.ropensci.org/av/reference/encoding.html)
function will convert any sound input (even from a video file) into
another output format, optionally adjusting the number of channels,
sampling rate, and start/end time.

Let's run some examples. First make sure you install the latest version
of the package from CRAN:

::: {.highlight}
``` {.chroma tabindex="0"}
install.packages("av")library(av)
```
:::

Suppose we want to read a certain piece of audio from a youtube
recording. We use the `youtube-dl` utility to download a full video from
youtube:

::: {.highlight}
``` {.chroma tabindex="0"}
# You need youtube-dl: https://ytdl-org.github.io/youtube-dl/download.html# MacOS: brew install youtube-dlsystem("youtube-dl https://youtu.be/F8Zt3mYlOqU -o whale.mp4")# Show some infoav_media_info('whale.mp4')#> $duration#> [1] 138.507#> #> $video#>   width height codec frames framerate  format#> 1  1280    720  h264   3460        25 yuv420p#> #> $audio#>   channels sample_rate codec frames bitrate layout#> 1        2       44100   aac   5965  125588 stereo
```
:::

This gives us a 75mb HD video with an AAC audio stream, which is not
something other packages can read. Suppose that for your analysis you
only need the first 10 seconds of sound, in mono `wav` format. Use
[av_audio_convert](https://docs.ropensci.org/av/reference/encoding.html)
to convert the input video into a shorter wav file:

::: {.highlight}
``` {.chroma tabindex="0"}
av_audio_convert('whale.mp4', 'whale10.wav', channels = 1, total_time = 10)#> Output #0, wav, to 'short.wav':#>   Metadata:#>     ISFT            : Lavf58.29.100#>     Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 48000 Hz, mono, s16, 768 kb/s#>  Adding audio frame 503 at timestamp 10.01sec - audio stream completed!
```
:::

Now we have a nice small wav file that we can read with any of the
standard R packages. Note that this entire conversion above took 0.02
seconds and uses very little memory because it streams directly from the
input to the output file.

## Reading raw audio samples

For this example we use an MP3 demo that is included with the package.
The [av_media_info](https://docs.ropensci.org/av/reference/info.html)
function shows some information about this file (without reading it in
memory).

::: {.highlight}
``` {.chroma tabindex="0"}
# Our example datawonderland <- system.file('samples/Synapsis-Wonderland.mp3', package='av')av_media_info(wonderland)#> $duration#> [1] 30.04082#> #> $video#> NULL#> #> $audio#>   channels sample_rate    codec frames bitrate layout#> 1        2       44100 mp3float     NA  192000 stereo
```
:::

As we can see, this file has one audio stream and no video streams,
which is what we would expect. The function
[read_audio_bin](https://docs.ropensci.org/av/reference/read_audio.html)
reads audio files (in any format) into binary PCM samples. By setting
`end_time = 2` we only read the first two seconds (approximately),
keeping the data relatively small.

::: {.highlight}
``` {.chroma tabindex="0"}
pcm_data <- read_audio_bin(wonderland, channels = 1, end_time = 2.0)plot(pcm_data, type = 'l')
```
:::

![pcmplot](Ve6HraL.png)

The docs page for
[read_audio_bin](https://docs.ropensci.org/av/reference/read_audio.html)
shows parameters for reading audio, in particular to reduce the number
of channels (i.e. convert to mono) or downsample to make data a bit more
more managable.

## Frequency data and spectrograms

Raw data is not very informative, most analyses require that the signal
is transformed into frequency data. The
[read_audio_fft](https://docs.ropensci.org/av/reference/read_audio.html)
function converts audio directly into frequency data using FFmpeg's
built-in FFT. Because the FFT is done on-the-fly when reading the audio,
it requires relatively little memory. It returns a matrix with the
time-frequency data, from which we can plot the spectrogram.

::: {.highlight}
``` {.chroma tabindex="0"}
# Read 5 sec of data and directly transform to frequencyfft_data <- read_audio_fft(wonderland, end_time = 5.0)dim(fft_data)#> [1] 512 860# Plot the spectrogramplot(fft_data)
```
:::

![fftplot](https://docs.ropensci.org/av/reference/read_audio-1.png)

The docs page of the av package has an entire article dedicated to
[spectograms](https://docs.ropensci.org/av/articles/articles/spectrograms.html),
comparing the various R packages. This vignette also shows the
[av_spectrogram_video](https://docs.ropensci.org/av/reference/capturing.html)
function which generates a video that overlays the audio on an animated
spectrogram with moving status bar:

::: {.highlight}
``` {.chroma tabindex="0"}
# Create new audio file with first 5 secav_audio_convert(wonderland, 'short.mp3', total_time = 5)av_spectrogram_video('short.mp3', output = 'spectrogram.mp4', width = 1280, height = 720, res = 144)
```
:::

Your browser does not support the video tag.


The latest version of the rOpenSci av package includes some useful new tools for working with audio data. We have added functions for reading, cutting, converting, transforming, and plotting audio data in any popular audio / video format (mp3, mkv, aac, etc). The functionality can either be used by itself, or to prepare audio data for further analysis in R using other packages.

Publicaciones de Rogue Scholar

Five Years of Economics from the Top Down

Working with audio in R using av