All Highlights

• ∎ The Serendipity of Streams - Reading Session 202306081350

  Last updated Jun 8, 2023 | Originally published Jun 8, 2023

  The Serendipity of Streams

  A neat article about the structure of (digital) streams of information and their propensity for serendipity and innovation.

  A stream is simply a life context formed by all the information flowing towards you via a set of trusted connections — to free people, ideas and resources — from multiple networks.

  What makes streams ideal contexts for open-ended innovation through tinkering is that they constantly present unrelated people, ideas and resources in unexpected juxtapositions. This happens because streams emerge as the intersection of multiple networks.

  This means each new piece of information in a stream is viewed against a backdrop of overlapping, non-exclusive contexts, and a plurality of unrelated goals. At the same time, your own actions are being viewed by others in multiple unrelated ways.

  As a result of such unexpected juxtapositions, you might “solve” problems you didn’t realize existed and do things that nobody realized were worth doing. For example, seeing a particular college friend and a particular coworker in the same stream might suggest a possibility for a high-value introduction: a small act of social bricolage. Because you are seen by many others from different perspectives, you might find people solving problems for you without any effort on your part. A common experience on Twitter, for example, is a Twitter-only friend tweeting an obscure but important news item, which you might otherwise have missed, just for your benefit.

  ---

  [In a stream, t]he most interesting place to be is usually the very edge, rather than the innermost sanctums.

  Not sure I agree with this. The author is binding a bunch of factors into “interesting,” but the truth is that there are different kinds of power here, and whether you want to be in the center or at the edge depends on what you’re trying to do.

• ∎ The Last Answer - Isaac Asimov - Reading Session 202301282203

  Published Jan 28, 2023

  If you were an amoeba who could consider individuality only in connection with single cells and if you were to ask a sperm whale, made up of thirty quadrillion cells, whether it was one or many, how could the sperm whale answer in a way that would be comprehensible to the amoeba?

  𖠫 The Last Answer - Isaac Asimov

• ∎ Data Science and Prediction - Dhar - 2013 - Reading Session 202301271312

  Last updated Jan 27, 2023 | Originally published Jan 27, 2023

  In situations where we are able to design randomized trials, big data makes it feasible to uncover the causal models generating the data.

  The author doesn’t make this point very big but it is actually quite important. There are many cases in which it is unethical to undertake a conventional scientific study of certain phenomena… Even though we know that those phenomena are happening anyway. If the data is collected “naturally,” however, we may be able to trace and track patterns in these data without putting ourselves in precarious ethical situations

  • Data
  • Research-Methods
  • Research

• ∎ Mexico bans solar geoengineering experiments after startup’s field tests - Reading Session 202301191446

  Last updated Jan 19, 2023 | Originally published Jan 19, 2023

  The company, called Make Sunsets, conducted the field tests without prior notice or consent from the Mexican government.

  This is one of the scary consequences of democratizing technology: volatility. It is getting easier for small teams to take big actions without oversight.

  And this is a well-intended initiative. The opposite of this would be ecological or environmental terrorism against businesses or governments perceived to be direct contributors to climate change, which surely will happen as climate change advances and people get desperate.

  At least this test was small:

  Iseman says he launched two balloons in Baja California last year, each carrying less than 10 grams of sulfur dioxide. That’s a tiny amount of the compound that’s typically released into the air by fossil fuel power plants and volcanoes in much larger quantities — so the release isn’t likely to have had much impact.

  The business model is interesting:

  Founded in October 2022, Make Sunsets started with the grandiose vision of releasing enough sulfur dioxide to offset global warming from all the world’s CO2 emissions annually. It’s already selling “cooling credits” for the service at $10 per gram of sulfur dioxide — even though it has yet to achieve any measurable impact and can’t guarantee that releasing sulfur dioxide at a bigger scale wouldn’t trigger any unintended problems.

  This has obvious parallels with Climeworks, who was recently paid by a few big tech companies to pull carbon from the atmosphere. It is hard to imagine this business model working at scale, though… surely there is a kind of prisoner’s dilemma at play that will keep every company from chipping in. Perhaps we need regulators to require businesses to purchase credits like these to properly recognize the environmental costs of business.

  • Signal
  • Climate-Change
  • Systemic-Change

• ∎ Design Science Dysfunctions - Listening Notes 20220525

  Published Oct 26, 2022

  # Notes from listening to “Design Science Dysfunctions,” an episode of This IS Research

  • Recker, Berente, and Gregor identify three problems with design principles:
    1. Design principles rarely build on existing work (They should be developed over extensive amounts of research (eg review articles), not one-off papers)
    2. Design principles are rarely very surprising or useful (“the UI has to be easy to use”)
    3. They often depend too much on context → indeterminacy
  • Interesting: even though (to her chagrin) she is the “[Design] Theory” person, Gregor doesn’t always include or look for a design theory in design science research. She said “A theory is just a body of work” in defense of why she doesn’t sometimes include a full Design Theory.

• Note titles as API calls

  Published Oct 13, 2021

  i saw someone say that note titles should be like API names ( from @davestridsr on the Obsidian Discord)

  It was Andy Matuschak.

  An app adds an API when its developers want other developers to be able to extend and work with the app’s functions, e.g., to develop other features or to let services interact with one another.

  Some “good” features of an API in software development might be that the API is: (1) deep (you can get at as many functions of the underlying app as are useful), (2) expressive (you can fine-tune interactions between your extension and the app), and (3) intuitive (by looking at an API function’s title and parameters, it’s pretty easy to guess what it allows you to do).

  Andy’s saying that a good note title fits those parameters too. The metaphor goes: you “extend” an atomic evergreen note by invoking its title in a new note. By doing so (and if done right), you can grasp the details of that note easily while, at the same time, you can reinterpret it in its new context.

  • Highlights
  • PKM

• Unknown

  - > Cognitive load theory is the idea, first published by Sweller in 1988, that instructional design should focus on not overloading a learner’s mental effort when designing instruction. “Learning is hampered when working memory capacity is exceeded in a learning task” (de Jong, 2009).

  Page 2 - > Germane cognitive load – This is the on-task mental ‘load’ or activity during learning. This can and should be influenced by the instructional designer. If one defines learning as schema acquisition and building, the ‘germane’ cognitive load is that which contributes to such schema acquisition, and the ‘extraneous’ cognitive load is that which does not.

  Page 2 - > As soon as I first read about germane cognitive load (good) in 1998 vs. extraneous cognitive load (bad), cognitive load theory became unfalsifiable in my opinion. You can justify any experimental result after the fact by labeling stuff that hurts performance as extraneous and the stuff that didn’t as germane.

  Page 2 - > Can’t distinguish between germane and extraneous cognitive load. Related to the above – one can’t objectively and before the fact tell whether something will be germane or not. Sometimes something that induces extraneous load may also induce germane load and vice versa. The type of load is highly dependent on learner characteristics and learning objectives (Moreno, 2009).

  Page 3 - > So is cognitive load theory a failure or wrong? Is that important? Like I said, the question is a joke. From one perspective Newton’s laws are wrong and were superseded by Einstein’s theories, but of course Newton’s laws are still quite useful and correct enough for everyday scenarios. The more important question is whether a theory is useful, or is there a better, more useful theory.

  Page 4 - > I believe an area of future interest should be in exploring how post-cognitive theories may provide more useful explanations for some of the phenomena uncovered by cognitive load theory research. Kaptelinin and Nardi describe some post-cognitive theories in chapter 9 of their book Acting with Technology. Unfortunately, chapter 9 was taken down from the First Monday site for some reason. But the four theories they compare include activity theory, phenomenology, distributed cognition, and actor network theory.

• Unknown

  - > Applicants should have completed or be nearing completion of a Ph.D. in information systems or related field and have a demonstrated potential for excellence in research and teaching.

• Unknown

  - > Efforts to advance knowledge that is based on ill-conceived classifications can prove futile, and even harmful. At best, they might result in wasted time spent arguing over terminology. More seriously, they can misdirect research efforts and funding. And at worst, in cases such as the misclassification of medical conditions, the result can be serious harm, including misdiagnosis, improper treat- ment and even death.

  Page 1 - > Phenomena can be categorized in many ways on the basis of the properties they share. However, categories are useful only if they make it possible to infer further informa- tion, and only if they do so consistently and over a reasonable time period. To distinguish a general category from a more useful one with inferences, we call the latter a ‘class’. Whereas a category simply reflects a repeat- ing pattern of properties, a class additionally indicates that relationships exist between these properties, even if the mechanisms behind the relationships are unknown.

  Page 1 - > classifica- tion is a purposeful human activity that reflects observations about relationships among prop- erties of phenomena.

  Page 1 - > In structuring knowledge, scientists should aim to identify classes rather than categories, and controver- sies over classification should be understood and resolved in that context.

  Page 2 - > an optimal level

  Page 2 - > between predictive power and stability needs

  Page 2 - > to be found.

  Page 2 - > the main problem was

  Page 2 - > the misattribution of the

  Page 2 - > property ‘cannot grow in

  Page 2 - > the acidity of the stomach’

  Page 2 - > to the class of bacteria.

  The persistent misattribution of properties two instances that are not actually in the class is actually a kind of “source forgetting” problem. People learn to believe that an instance — erroneously labeled as a member of a given class — has a certain property. Then, later, even though they get evidence that the original classification was incorrect, they continue to associate the instance with that property, because they forget why they originally learned that instance–property association.

  Page 2 - > Taking a classification perspective on scien-

  Page 2 - > tific discourse suggests a sequence of questions

  Page 2 - > to ask when studying a domain of phenomena.

  Page 2 - > What are the properties of interest of these

  Page 2 - > phenomena? Are there stable sets of properties

  Page 2 - > common to these phenomena? Are there stable

  Page 2 - > relationships in some of these sets? And finally,

  Page 2 - > and most importantly, what is the evidence or

  Page 2 - > rationale that these relationships reflect the

  Page 2 - > true nature of the phenomena?

  Page 2 - > The important thing is

  Page 2 - > to avoid the trap of imposing a categorization

  Page 2 - > solution that fits some organisms onto others

  Page 2 - > that it does not fit.

  Page 2 - > researchers should recognize that several correct and useful schemes can coexist.

  Page 2 - > The more shared proper- ties per category, the greater the potential predictive power, making highly specialized categories attractive.

  Page 2 - > very specialized categories are not stable over time.

• Unknown

  - > a concern. our ratings. A simple analysis of mean ratings across raters could be used to supplement a correlation coefficient if this is differences in ratings do matter, and we want to know how much of an impact they have on

  Page 1 - > The intraclass correlation coefficient (ICC), also known as the G-coefficient from generalizability (G) theory, is the most flexible consistency index because it can be calculated in different ways to either account for or ignore systematic score differences, and it also works with ordered score categories. Thus, ICC addresses the limitations of both the agreement indices and the correlation coefficient.

  Page 1 - > And fixed raters is labeled ICC(3) or two-way mixed.

• ««
• «
• 1
• 2
• 3
• 4
• »
• »»