N and I discovered that both of our fields have developed ridiculous amounts of hype over very similar and equally non-existent professions. N falls asleep of sheer boredom when he hears about “data scientists” and I do the same when I hear about “specialists in digital Humanities.” Both terms are heavily used by unemployable folks who are trying to make themselves sound relevant.
18 responses to “Fake Professions”
-
I have a real profession not a fake one. But, for a long time it didn’t offer me any real pay. Now all that is changing in September. 😉
http://jpohl.blogspot.com/2016/07/on-move-again-kurdistan.html
LikeLike
-
Hummmm, what your sister thinks about data scientists?
LikeLike
-
And what do you think about econometrics and epidemiology?
LikeLike
-
I can’t speak to econometrics, but epidemiology is an actual thing. It’s the study of diseases and the spread of those diseases, and I believe there is a lot of overlap with parasitology, virology, and immunology. Still, it’s a huge area of study, and overlaps with a lot of other areas of biology and chemistry. It’s a specialization that often couples with the medical and medical research fields.
LikeLike
-
The main overlap in epidemiology is with statistics.
LikeLike
-
But Klara (with the Mars vowel) doesn’t work very well with an American r (and I don’t think a Russian r will fly in English)
A very good mnemonic device for Americans tends to be rhymes.
“It rhymes with Sarah” might work if Clarissa wants an American pronunciation (unless Sarah is now also too exotic).
LikeLike
-
My comment above was not meant for this thread, sorry.
LikeLike
-
-
-
-
-
Well I know a few former astronomers who left academia and who are quite gainfully employed as data scientists, for whatever that’s worth. Although my impression as of late is that making that jump is much harder now since you actually have to prove you know something useful in order to get hired these days instead of training on the job. I don’t recall what you’ve said N’s job is, although I know it’s something largely numerical. A statistician maybe? That’s really all data scientists are, statisticians who try to pull useful information out of large datasets to help their company in some way. The folks I know either work for consulting firms, so they tackle a variety of problems, work for the national weather service to improve weather or climate models, or work in silicon valley on various things like improving targeted advertisements or spam blocking algorithms. In any case, I’m pretty convinced “data scientist” is a real job (but with an ambiguous name) that basically means a statistician who is also good at coding and developing new models to explain the stuff they find (more or less like scientists do).
-
Here’s an abstract of a paper in digital humanities entitled, “Critical Behavior from Deep Dynamics: A Hidden Dimension in Natural Language. Can you translate?
“We show that in many data sequences – from texts in different languages to melodies and genomes – the mutual information between two symbols decays roughly like a power law with the number of symbols in between the two. In contrast, we prove that Markov/hidden Markov processes generically exhibit exponential decay in their mutual information, which explains why natural languages are poorly approximated by Markov processes. We present a broad class of models that naturally reproduce this critical behavior. They all involve deep dynamics of a recursive nature, as can be approximately implemented by tree-like or recurrent deep neural networks. This model class captures the essence of probabilistic context-free grammars as well as recursive self-reproduction in physical phenomena such as turbulence and cosmological inflation. We derive an analytic formula for the asymptotic power law and elucidate our results in a statistical physics context: 1-dimensional “shallow” models (such as Markov models or regular grammars) will fail to model natural language, because they cannot exhibit criticality, whereas “deep” models with one or more “hidden” dimensions representing levels of abstraction or scale can potentially succeed.”
http://arxiv.org/abs/1606.06737v2
Is this a serious field?
LikeLike
-
The language certainly reads like gobbelty gook to an outsider, but I know enough stats to recognize some nuggets of familiarity in there. Take for example “mutual information” – it actually means something to a statistician but it sounds like random words being put together to someone who doesn’t know stats (it’s the probability of given event A you know event B will follow; so in language processing, given the letter “q” in a word how likely do you know what the next letter will be “u”, or something along those line). Similar thing are going on for their usage of “shallow” vs. “deep” and “critical” in this context. There are definitely aspects of this abstract that makes me want to roll my eyes about theorists being overblown about the importance of their work… but I do not recommend dismissing the value of the results just because you don’t understand the jargon. You just aren’t the target audience for this abstract. Now if they can’t explain it to you lay terms… well then it’s totally reasonable to be skeptical and concerned they’re making shit up.
LikeLike
-
It’s serious but it’s not for the humanities.
LikeLike
-
It basically says that one of the most popular methods of generating strings of information automatically produces strings of information with different statistical properties than the ones of the strings they’re trying to mimic, compares it with some newer, better methods that are quite poorly understood, and says that the newer methods are better because they generate strings of information that are similar, from the point of view of this one statistical property they care about here, to the strings they’re mimicking. In the process of doing this, they come upon a new method of evaluating machine learning performance for string-of-information-generating tasks.
It’s an ok computer science paper, but I cannot for the life of me understand why anyone would link this with the humanities, considering the fact that it doesn’t say anything new in that direction.
LikeLike
-
-
My favorite fake profession is product owner… of course data scientist is quite a amusing I suppose they must distinguish from those scientists who can figure things without any data, observation or experimentation 🙂
LikeLike
A conservative literary critic and college professor
Clarissa's Blog 
Leave a comment