199 Matching Annotations
  1. May 2023
    1. For example, if the only covariates collected in a study on people are gender and age category, the data can be stored in grouped form, with groups defined by the combinations of gender and age category.

      I wonder if my model of success of startups could be refactored in a similar way? HAS_FEMALE_COFOUNDER could be a group.

  2. Jan 2022
    1. bridge the subsymbolic methods

      Not wholly convinced

    2. Neural methods for NER offer a key advance, because they can distinguish between different senses of words based on context

      Distinguishing the topics is good: something else we've talked about with Michael is pulling together hierarchical models of documents and domains, so that we know which topics relate to which; recommendations could then be made on that basis. BTW, Zoe Corneli has done some recent work with recommendations in the shopping domain (Amazon Shopping).

    3. we expect to be able to reliably link symbolic identifiers to their definitions, an entirely novel advance

      This is still part of our current thinking: however I'd also say that in addition to "novelty" we should talk about why this matters. What is it about linking symbolic identifiers that improves our understanding of documents? I think a core concept here might be that of "fingerprint databases". Understanding symbolic text (or other text in the formal register) should help us build better fingerprints. https://arxiv.org/abs/1304.3866

    4. six month project

      This was very helpful for formulating something with a concrete focus; however, 18 or 36 months would allow us to do quite a bit more.

    5. Can AI be used to help make this technical material easier to access, understand, and apply?

      In subsequent thinking about this project I've been looking a bit more broadly. Maybe it's better to come back to this core focus on Arxiv.

      The vigorous growth and development of Science, Technology, Engineering and Mathematics (STEM) begins to suggest a radical idea: to treat STEM itself as a source of big data...

  3. Jun 2021
    1. Architecture Roles and Analysis People and Process Domain Specific Techniques OO Techniques Non-OO Techniques

      Look at evolution of PLoP itself (this would be a bonus-round)

    2. This paper makes a persuasive case about how we would want to use design patterns, this will have to be replicated.

      Zettlekastens, program specifications, etc., are in a family of related computational objects.

    3. he problem with a lot of futures studies is that it’s short on philosophical depth. Let’s understand it by looking at layers litany, system, worldview, and myth. If we look at the events, then what are the causes, the deeper causes, and the deepest causes.

      Probably a thesis statement: what if you make a pattern language and you don't think about the assumptions underlying it.

    4. generative patterns

      we had quite a bit last time, and there seems to maybe be more to say here about organic growth — maybe this could be related to the issue of speed, since we somehow want generative things to make rapid progress.

    5. Leitner’s name for what Alexander left nameless is ‘perception of quality of life’. Even have things like the car in relationship to the horse! Not just the Mustang… now cars and motorcycles have inherited that

      In our modern day we have to do things quickly — this seems implicit.

    6. So, OK, if we want to understand how to move to a post-carbon world, look at how they moved to the current carbon-friendly world.

      Maybe better for the discussion but a well-developed thought.

    7. The City is not a Tree — the designer’s view It wasn’t design this way — this is the organic lifelike quality

      Taking a deeper dive into Christopher Alexander. How much of his work has been brought in here? If we reduce it down to the schema you left it out.

    8. There’s something curious going on here, though, relative to another key concept for pattern theory, namely, the concept of wholeness.

      Wholeness is definitely in here at this level.

    9. ellipse

      We have at least a constant and plane to chose after we nominate the context and community — what might these represent.

    10. “sacred spaces” and the “sacralisation of space”

      Here we might want to talk about provisions about who are stakeholders, who are not, what roles they occupy (e.g., 'noble savage'), who becomes a stakeholder later (e.g., inheritors).

    11. Community was implicitly there Stakeholders, that was a community.

      The different ways they experience the conflict may be interesting.

    12. Context problem solution… Concept of problem and solution comes from mathematics or physics

      A scientific worldview exists within a larger context that will involve other views.

    13. critique of KAKU — which was a critique of HIS VISION OF THE FUTURE

      Let's move some discussion of this into the Discussion section.

  4. May 2021
    1. Patterns are shared as complete methodic descriptions intended for practical use by experts and non-experts

      This becomes a pretty interesting or central idea.

    1. By contrast, in our work we have centred circumstances of cultural diversity and human relationships.

      I'm glad to see that we didn't retreat from the critical stance that we initially brought forward. Our perspective was that Kostakis et al. tended towards a 'provisionist' and paternalistic sort of development trajectory, a kind of neo-colonialism, even though they are pitching their work from within peer-to-peer and free software settings. The same issue could easily come up with patterns, i.e., if they are offered as the solution. Realistically, it's possible to read a lot of the patterns literature this way (including Alexander), regardless of what they say about adapting their patterns for local circumstances. The medium of print holds the more egalitarian process of pattern development hostage, and makes pattern authors seem much more arrogant than they ever intended to be! This is why we've focused on participatory processes of pattern development here. Furthermore we're not saying that this is an exhaustive treatment of that topic! Hopefully it is clear (though, if not, this will take more work) that what we're trying to develop is the opposite of 'provisionist' thinking. (BTW, for this term, Cf. Boud and Lee "‘Peer learning’ as pedagogic discourse for research education"

    2. Pattern language development can give us another way to think and talk about the “process of language development” that already goes on inside institutions

      This part could be relevant for thinking about 'pattern mining' and how to contextualise it. Consider the various situations where we run into a problem of language, i.e., where we don't have any language to talk about something that actually urgently matters to the parties involved. Things can either end badly there, or the parties can engage in a process of language development.

  5. Apr 2021
  6. Mar 2021
    1. This simplistic approach leads to false positive hits, for instance, in words that have multiple part-of-speech uses or words that have both technical and everyday meanings.

      Seems related to "wordpiece tokenization", at the concept level?

    1. L. Zia, I. Johnson, B. Mansurov, J. Morgan, M. Redi, D. Saez-Trumper, and D. Taraborelli

      BTW, we might make a goal of shrinking the number of references somewhat. Maybe "at least 50%" would be a good estimate as we go into the next round of revisions.

    2. Poli

      JC: I found that reference to Kull that I sent around; it could go here, too.

    3. We thank

      R2: In conclusion, I do encourage the author(s) to continue their work and possibly resubmit this work in the future once more progress has been made.

    4. Which brings us to you, dear reader. We assume you wanted to learn about the subjects in the title: or perhaps this paper was assigned to you in school and you are reading it for a grade?

      R2: For this paper to be publishable requires building an argument that would, for instance, connect various observations, premises or ideas about the context of collective anticipation and its key dimensions with ways (i.e. strategies, approaches or solutions) to instigate collective thinking within communities about the future.

    5. Our main results outlined several ways in which design patterns can be used to model future-directed activities.

      R2: I would encourage the author(s) to continue this work but with more focus on delineating the key characteristics of road-mapping, scenario planning and play; together with an analysis that justifies how these notions contribute to different situations (contexts and problems) where collective anticipation is required.

    6. Our presentation at the Anticipation 2019 conference took the form of a scripted dialogue that we invited audience members to read aloud.

      JC: BTW, this is a third cold open!

      R2: As a result, the paper lacks the focus that is needed in order to develop a strong argument for collective anticipation through the use of design patterns and the resulted design pattern are essentially reduced to a general instruction to use road mapping, scenario planning and play.

    7. This work can be guided by further patterns, e.g., to develop - a language for projects → Roadmap - a language of future scenarios → Participatory Scenario Planning - a language of roles → Play to Anticipate the Future

      R2: More specifically, the paper discusses a number of design patterns for anticipating futures around three key approaches: road-mapping, scenario planning and play.

      These discussions (in section 2) could be potentially interesting and original, but the different sections tend to cover broad and diverse ideas/observations about participation rather than an analysis of these key concepts and their contribution and meaning in the formation of a design pattern for collective anticipation.

    8. After defining and giving examples of how design patterns have been used in other fields, we then examine how they can help create peer learning experiences to explore and prepare for the future.

      R2: However, the current version of the paper appears to lack focus, in particular in term of argument development, and the proposed design patterns for collective anticipation are very general.

    9. To begin our exploration of how design patterns relate to futures studies, we refer to Schwartz (1996, Appendix, pp. 241-248), viz., his “Steps to Developing Scenarios”.

      R2 finds this, and related aspects of the paper, to be valuable and interesting content.

      R2: In particular, I found interesting (in the next section 2) the comparative analysis of the work of Christopher Alexander and Peter Schwartz. The author(s) made an observation about the difference between proposing a schematic solution (as in the original patterns) and that of working with diverse scenarios. These short of observations can be key directions in the development and proposition of design pattern that would facilitate people to engage with anticipatory thinking.

    10. we get limited insights about how to engage together

      If we are going to criticise existing work on anticipation we should be a bit more careful to identify the gap. For example, of course participatory scenario planning is a thing, we even talk about it in our paper! So what is the gap?

      R2: The paper appears to identify a 'gap' in the literature: that is on page 5 with the claim that are limited studies on 'how to engage together' to anticipate the future. Strictly speaking this is not true but it is true, to my knowledge, that design patterns have not being (explicitly) formulated as a vehicle for anticipatory thinking or anticipatory scenario building.

    11. Of course, people have been finding new ways to think about the future together for a long time.

      R2: Overall, I think the subject matter of this article could be relevant for the journal and anticipatory studies. Some of the ideas presented in the paper may also have the potential to offer an original contribution to the vast amount of studies and applications of design patterns.

    12. In the midst of the Covid-19 pandemic

      Reviewer #2: The paper proposes a design pattern methodology for collective anticipation. The study aims to demonstrate how design patterns can be used to develop a mindset for collective anticipation. The paper analyses three approaches of thinking about the future - roadmaps, scenario planning, and play - and proposes these approaches as key design pattern methods for articulating future scenarios.

    13. Reviewer #1: The aim of this article seems to be to present the methodology of design patterns to structure anticipatory futures work. I think this aim is highly relevant and interesting for the discussion around anticipation and that this focus does have the potential for an interesting contribution. I also want to acknowledge the hard work that the authors have put into this. They have demonstrated that they are aware of a lot of related work at the intersection of design and futures. In order to support the development of the thinking, I will outline some of the main issues that I found when reading the manuscript below.

      At present the manuscript does not adequately explain what design patterns are and how design patterns can actually support anticipatory futures work in a way that is distinct from existing design and futures methodologies. The broad definition of design patterns as "solutions to problems arising in the context of repeatable activity" doesn't seem to provide clarity on what is NOT a design pattern. Are all solutions design patterns if there is some aspect that is repeated? Are all existing design and futures methods patterns? What is helpful about thinking about these activities as patterns if they are already happening? To make a contribution with this work, the conceptualization of patterns needs to be unpacked much more and the examples can't just be described but there is a need to show how the pattern itself is particularly useful in the example. In what way are thinking about these as patterns generative? How it is different from existing alternative approaches?

      In addition, the structure of the manuscript could be improved. Starting with COVID-19 as the hook for this work and then coming back to it at the end to show how patterns could have helped the pandemic, feels somewhat disconnected and unconvincing. Also, in Fig. 5 much of what you included bags questions about if such activities really could have helped in preparing for the pandemic. For example, on what basis can you argue that planning the game Pandemic beforehand would have helped leaders contain the virus. Furthermore, throughout the text, there is the integration of many loosely connected ideas but you often hop quickly from one to the next without fully explaining yourself. For example, the use of Heidegger on p. 11 felt thrown in and not well explained. I would try to ask yourself if the idea or reference is critical to your overall argument and if not leave it out to make space to better explain and connect what you do include.

      There is also a need to find a more convincing way to motivate why this methodology of design patterns is needed - what are existing methodologies at risk of? What is the issue if this type of approach isn't developed? What urgent needs does it meet that others do not? At the end of p.5 you say that this fills a gap around participatory anticipatory learning for complex problems requiring a group response, but are there not already many pre-existing methodologies that focus on that, including the many examples that you refer to in the paper?

      In addition, there seem to be a central tension between patterns and anticipation of change that goes unaddressed by the authors. If patterns are conditioned by repeated activities from the past, might they actually be inappropriate for thinking about evolving futures or conditioning us to think about the future by projecting the past? Or why is it that we need patterns to help us make sense of and prepare for the complexity and uncertainty of the future? I would love to see this paradox better grappled with in this work.

      The approach of the paper needs further explanation and rational and could perhaps be re-worked to better show the value of patterns as an anticipatory methodology. First you introduce Open Futures Design, but it is not clear to the reader what this really is. Is it a pattern that using patterns is useful for social cohesion? This overall umbrella needs to be made clearer. In addition, it is not understood why you have selected the three patterns of the roadmap, participatory scenario planning, and play. Why those three and not others? In what way do they in particular reflect the intentions of open future design? What makes them great examples of relevant patterns?

      In addition, throughout your patterns you draw on many examples of existing methods/methodologies or activities that relate to these examples. Their inclusion feels somewhat random. There is likely too many examples and they are not well justified as to why they are included or how they inform the usefulness of patterns as a methodology. In particular, if you are pointing to design futures methods that already exist and then put the label of pattern on it, how is this methodology new or what value is it adding to existing work to call it a design pattern? If you could develop much more how thinking about these as a pattern could be helpful, you would have stronger support for the development of your main claims. In addition, if you are pointing to other examples that were not necessarily thought about as patterns at the time of use, it is not convincing as to how than it helps to think about them as patterns. Please elaborate on this.

      In most of your figures you use metaphors and also throughout some of the text. However, these metaphors are not well explained and connected to your points in the text so they become somewhat distracting and leave questions for the reader. For example, when reading about Open Future Design, I was left wondering about how exactly it was like a water cycle and how that helped me in understanding your points. I love metaphors and think they can be powerful guides, but I couldn't quite grasp this. Also, you seem to break your own pattern of visually showing metaphors for the patterns on Fig 4, which directly seem to reflect play. I also felt like the language patterns you articulated under each image "Context… If… Then…" felt overly simplistic and left me wondering about how they could really be helpful and how they could be used by myself or others moving forward. You didn't really explain how these patterns you are creating could be used in practice, but that seemed to be your intention. This too needs further development.

      In general, I think the idea of working with patterns as an anticipatory methodology is an interesting one. However, more work needs to be done to explain how this is novel and what value it brings in relation to existing methods and methodologies in this same space. Best of luck with your future developments of this and related work.

    14. Reviewer #2: The paper proposes a design pattern methodology for collective anticipation. The study aims to demonstrate how design patterns can be used to develop a mindset for collective anticipation. The paper analyses three approaches of thinking about the future - roadmaps, scenario planning, and play - and proposes these approaches as key design pattern methods for articulating future scenarios.

      Overall, I think the subject matter of this article could be relevant for the journal and anticipatory studies. Some of the ideas presented in the paper may also have the potential to offer an original contribution to the vast amount of studies and applications of design patterns. The paper appears to identify a 'gap' in the literature: that is on page 5 with the claim that are limited studies on 'how to engage together' to anticipate the future. Strictly speaking this is not true but it is true, to my knowledge, that design patterns have not being (explicitly) formulated as a vehicle for anticipatory thinking or anticipatory scenario building. In particular, I found interesting (in the next section 2) the comparative analysis of the work of Christopher Alexander and Peter Schwartz. The author(s) made an observation about the difference between proposing a schematic solution (as in the original patterns) and that of working with diverse scenarios. These short of observations can be key directions in the development and proposition of design pattern that would facilitate people to engage with anticipatory thinking.

      However, the current version of the paper appears to lack focus, in particular in term of argument development, and the propose design patterns for collective anticipation are very general. More specifically, the paper discusses a number of design patterns for anticipating futures around three key approaches: road-mapping, scenario planning and play. These discussions (in section 2) could be potentially interesting and original, but the different sections tend to cover broad and diverse ideas/observations about participation rather than an analysis of these key concepts and their contribution and meaning in the formation of a design pattern for collective anticipation. As a result, the paper lacks the focus that is needed in order to develop a strong argument for collective anticipation through the use of design patterns and the resulted design pattern are essentially reduced to a general instruction to use road mapping, scenario planning and play. I would encourage the author(s) to continue this work but with more focus on delineating the key characteristics of road-mapping, scenario planning and play; together with an analysis that justifies how these notions contribute to different situations (contexts and problems) where collective anticipation is required. This paper to be publishable requires building an argument that would, for instance, connect various observations, premises or ideas about the context of collective anticipation and its key dimensions with ways (i.e. strategies, approaches or solutions) to instigate collecting thinking within communities about the future.

      In conclusion, I do encourage the author(s) to continue their work and possibly resubmit this work in the future once more progress has been made.

  7. Feb 2021
    1. of tensors {Xk}, then there is a built-in function that will return the tensors {dC/dXk}

      This is some math

  8. Jan 2021
    1. Note 5.41 (Roundtripping). The easiest test is: can we import a document into the system and then export it again, and find it unchanged?

      This was a major objective and would be a super useful feature, but it has yet to be completed!

      One contemporary (2021) approach might be to try to build this feature into org-transclusion, perhaps with the additional feature of search to define the list of nodes that are pulled together in one buffer to be edited.

    2. get-text-property

      OK, one way we can tell that Arxana 2010 doesn't have all the same features is by looking at how text properties are used here. In Arxana 2005, they were used intensively to mark regions (so that when things moved around, the linked annotations could be preserved). Here, they're barely used at all.

    3. The current system is an almost-completely rewritten variant, bringing in a shared database and various other enhancements to support multi-user interaction.

      So, the database is certainly an enhancement, but I wonder if the annotation features from the previous version were carried over at all? Does Arxana 2010 perhaps mainly deliver a "graphical model" of a document? I suspect that the markup of scholia is NOT present in this version, but want to dig a bit more to confirm. If I'm correct then we would still need to build a merged version to include all the features from all the versions...!

    4. Elephant2525 25 http://common-lisp.net/project/elephant/

      This is the Lisp Elephant backend, not to be confused with McCarthy's Elephant 2000.

      I seem to remember having a nice chat with the author at the Hungarian Pastry shop the first time I went there.

    1. Howm is an already existing Emacs package for both forward (‘‘goto’’) and backward (‘‘come-from’’) linking

      So it's been around for a while!

  9. Dec 2020
  10. Nov 2020
    1. other agents, very much like ourselves

      Maybe robot rights is the wrong question, or even in the short term dangerous, but it may be heading in the right direction. If robots can do these things, then why don't they get rights...

      What about the biosphere, or communities?

      Why are we even having this debate about robot rights? If that's the wrong question, then maybe it's still trying to address a serious issue that needs to be addressed sooner rather than later. Cog Sci, computer science, and so on, affect the foundations of what humans can do. Literally speaking we don't have it but as a thought experiment it's valid b/c we have things that are close enough.

      As a philosophical issue, if we respond to this as trolling, then the whole thing becomes a flame war rather than thinking about how Cog Sci and so on has helped us think about what it means to be human.

      What then does it mean to have human dignity in a situation where we have current technology? Where do things like rights go w/ embeddedness?

      Maybe there's some continuous thread from rights of kings to rights of man and beyond — what if we take the fundamental unit to be the community, in which case, then we get different conclusions. Maybe even Burke has something like this — the thing about the isolated individual was the community. Communities can have have special communities this is why you get such things as rights of kings: this is why the king gets special rights. It wasn't just that it was exploiting kings rampaging around with other responsibilities.

    2. A deep appreciation of this embeddedness does not entail artifacts should be seen as ‘agents like ourselves’

      What if they are part of us, or we are part of them; and this then isn't reducible to one select piece. If you're dealing with simple tools like hammers then it makes sense to reduce it to the human — but only if you're keen on reducing things b/c there's a whole theory that says the rise of tools is what helped us become human...

    3. . Because instead of steadily progressing towards a happy community of humans and ‘sentient AIs’,

      Within this system we localise all of the intelligence within the human! A deeper post-cartesian perspective, the humans wouldn't have the same special role — we'd have such things as "sociality with tools" or holistic properties of the whole group which you wouldn't reduce to the individuals in the group. A problem w/ the oppressive systems is when you pull out things like king: isn't at a deeper level this what's causing the problems?

    4. Based on the post-Cartesian embodied perspective we hold that while human being may incorporate

      This still centres the human: we could take a post-Cartesian view that said "We exist, therefore intelligent machines are possible."

    5. (neo-)Cartesian distinction

      Aristotelian logic wouldn't have led to machines in just the same way as the A-B-C-D-F logic of Descartes.

    6. But what is the self-image we use as a model? AI from its early days attempted to engineer a cognitivist interpretation of human thinking in the machine,

      Emergence and stuff still come in. It's not that we started with a blueprint for building a copy of a human being! There are certainly some things engineered but now we still have problems explaining what a neural network does. And, if you leave out the Global, Endo-, levels that are discussed e.g. by Sloterdijk... with all of the engineering & science this calls into question the basis for "dignity of humans". What would be a more robust basis for "dignity of humans"? Protagoras & al. presumed that humans would have a unique dignity, but now machines are doing things that we assumed that only humans can do — so, just like Descartes had to rethink things... maybe now we have to rethink human dignity now.

    7. We discuss how AI, rather than being the potentially oppressed, is used as a tool by humans (with power) to oppress other humans

      This could be viewed in a different way, e.g., tools, communities, media, knowledge bases, human labour under the hood that's organized in different ways — not necessarily via big companies. Things like PlanetMath, Project Gutenberg, and so on did exist, but we didn't get support at a higher level. A cynical view would be that maybe there was even opposition from higher levels. At the time of Hardt starting with Project Gutenberg and so on, things looked almost like performance art. In 1984 then we started to actually have personal computers and the science fiction was starting to come to reality. In principle in 1970 they could have been thinking about "what about author's rights when books go difficulty." People would have dismissed Hardt and Nelson as visionaries but we are now living in the future that they predicted. But we do have the scenario of Kurzweil.

    8. post-Cartesian, phenomenologically inspired position, human being is a lived

      Thinking about rights corresponds to e.g. Locke, you'd have atomistic individuals to even bring in a community. If you have individuals and they all have human rights then you could get to a community. This is different from other theories where the community starts and ultimately you derive the individual. There are many ways of conceptualising things.

    9. With the rise of machine learning, there is an increased appetite to hand much of our social, political and economical problems over to machines bringing with it corporate greed at the expense of human welfare and integrity

      This is the sort of thing that brings us into a "post-Cartesian" way of thinking. Before Galileo & Copernicus we had heavens made out of quintessence, things in heaven went in circles while things down on earth fell down, and so on. People speculated about humans being machines but it wasn't concrete; now we have computers that can do a lot of things that humans can do. This questions the foundations of things like ethics! The basis for "human rights" might have been: humans are unique and precious.

    10. In comparison, iRobot’s Roomba, while portrayed as a harmless household machine

      Cf. Stephen from Django returns...?

    11. Meanwhile popular culture talks about actual AI and robots as if the intelligent machine is already there, while in fact, it is not.

      What if the problems of the present are because of ineffective preparation in the past — or, even worse, a cynical preparation (like the tobacco industry claiming that cigarettes aren't unhealthy).

  11. Oct 2020
    1. AI for Good

      Related to Ethical AI

    2. while there are many ways to apply AI within CI [3], the present project moves in the direction of building computational models of CI itself

      There are a number of interesting feedback loops on this: for example how do computational models of social behavior but also increasing literacy with modelling, helping people learn how to access information. Keeping in mind that most people will have gone down path-specific learning & development trajectories that make grappling with complex issues difficult!

    1. Can autonomous computational agents build an explicit, functional, model of the knowledge and epistemic processes that underlie a large technical corpus written by many authors?

      Build an Agent-based "simulation" of Stack Overflow.

    2. The LEAPQA project will use artificial intelligence to create an online learning support tool for technical training in computer science and mathematics.

      Turn online open source materials into interactive tutorial.

    3. “If you can’t solve a problem, then there is an easier problem you can solve: find it”

      There's a social variant: If you can't solve a problem, there's someone out there who has solved it. Find them.

    1. Agents writing institutions

      Maybe this also relates to 'AI planning' (or, anyway, writing plans)

    2. Publication: IJCAI

      One nice thing about this plan is a "divide and conquer" strategy towards having a reasonably specific output after 1 year.

    3. Time off and plan Year 3

      A "towards" way of thinking that's sufficiently expansive is motivating: this needs to be balanced against doable tasks & achievements.

    4. Questions are similarly rated.

      We can also consider a new repository oriented around questions.

      • E.g., "what is the evidence for climate change?" (Org)
      • Link out to other files that perform analyses of existing data
      • "What is the best measurement we have over time of carbon in the atmosphere?" (this isn't the usual starting point b/c people wouldn't think of this)
      • When do I encounter something that's worth doing some computations about (an Org Roam network of computable objects/processes)

      Potentially deploy with a reboot of Dewey's "laboratory school" (they did cooking...).

    5. Agents will gain points for asking and answering questions about Stack Overflow content.

      If we're interested in understanding social & economic dimensions of things then we should also be thinking about what agent based models can tell us... and what about other things?

      • Seeing ourselves as a potential competitor to fivethirtyeight.
    6. Active Inference bootcamp

      This is the second upskilling task, thinking about how to design agents that learn. Beren Millidge has been working on this stuff: could any of that work be taken over? Do we want to explore other possible.

    7. ML/NLP bootcamp

      This would be the first key bit of upskilling for me, though in principle I could work with people who are already working on this sort of thing, and that are interested in teaming up on a piece of work here.

    8. Milestones and deliverables for first 24 months of the project.

      This was something I proposed for "just me, for 2 years", so this is something that I should consider again now, as I pick up a 2 year contract with Oxford Brookes.

    1. Our long-term vision is computational intelligence based on collective intelligence.

      Though, given the 1.5° C buffer, it may not be very suitable to have a "long term view" that doesn't take account of climate change & human adaptation. So, maybe this needs to be re-jigged around a "why" that is much more pressing than just the hedonistic interest in learning stuff. Furthermore part of an answer to this concern is going to be through "social networks" not just in "scientific computing."

    1. Specifically, they leave out some of the eight core software features listed above.

      Yikes!

    2. Beginning about 2013, the Wikimedia Foundation began offering two new interfaces specially designed for mobile devices: the Wikipedia mobile web interface and the Wikipedia mobile app.

      I didn't know about the app! Does it make editing significantly easier...? Not that I would prefer to edit on my phone overall but maybe I should try it.

    3. observe their surroundings and orient themselves prior to deciding and acting.

      Github also uses similar stuff, but it has a higher barrier of entry.

    4. Wikipedia users have ready access to information, such as what changes have been made and who made them, that allows them to meet their own needs.

      This is another interesting claim. What kinds of hands-on knowledges might they need to support this kind of process? (Relevant if we were thinking about a bot-based re-design for Wikipedia for example.)

    5. In my view, the challenges inherent in contributing to the ODP did not bode well for its long-term survival, especially when compared to sites like Wikipedia, where working together is easier.

      This is a nice parsimonious theory, and the 8 dimensions that back it up seem like good ones. What about applying them to other sites of activity, or looking at how they are threatened in cases where Wikipedia doesn't work well (e.g., the Knowledge Engine example)

    6. the ability to perceive or act without editorial intervention

      has this been used as a lens to analyse B. Mako Hill's "Almost Wikipedia" case studies?

    7. “observe–orient–decide–act” (OODA) loop, which originated in military strategic theory. The OODA loop theory emphasizes that easy access to relevant data is a crucial component of effective decision-making.

      maybe we could combine this with the PAR

    8. I explained to Cunningham that I believed Wikipedia’s software supported collaboration by giving editors ready access to a fairly complete set of relevant data—the eight software features listed above.

      so this is ultimately the thesis of this essay maybe?

    9. Cunningham preferred to trust in the good intentions of his site’s user community, so he initially resisted writing software code to capture old versions.

      I wonder how this would relate to the ambition to use wikis together with pattern methods. A pattern would evolve in its context, but wouldn't necessarily have a version history separate from that context.

    10. content and discussion were segregated

      an interesting design feature for sure!

    11. which conveys exactly what changed between any two revisions, and indicates who’s responsible

      So, that's an interesting level of analysis but I wonder about taking it further to consider that author's previous editing history — looking at 'why' they made the changes they made?

    12. Those same data can help Wikipedians find others interested in the same topics

      I remember talking about a concept of 'paradata' that would serve related purposes

    13. The ability to find this kind of information can feed a Wikipedian’s sense of confidence.

      Yes, it builds confidence about the commits, but this meta data language is content-free, so does not directly help with understanding the content itself...?

    14. Wikipedia users all have access to the site’s underlying data, analogous to the company’s accounting data

      Do they actually use this data in significant ways? What data literacies might be needed?

    15. If and when a robust theory of Wikipedia’s “magic” emerges, I believe it will give prominent attention to a collection of about eight mutually supporting software features.

      I'll be curious to know what they are and what other dimensions the theory might encompass! This reminds me of Ray Puzio's CCC theory for PlanetMath.

    16. The half-serious notion that no theory can capture the magic of Wikipedia caught on

      maybe this was due to an anti-theory sentiment more than anything? Tim Gowers, in math, writes about 2 cultures, one towards theory building and the other towards problem solving. Maybe there was something similar here.

    17. emphasized the importance of using judgment, rather than adopting “hard-and-fast rules.”

      This has come up again... w/ conservatives on the supreme court, apparently no agency can do things that aren't made explicit by law. But agencies are designed to interpret things, so this hamstrings them.

    18. “fun for the contributors” as the most important guiding principle

      That could relate to our recent discussion of Play to Anticipate the Future .

    19. help resolve conflict between Israelis and Palestinians

      That doesn't seem to have happened?

    20. seemingly on the wane in the broad public sphere

      You might think that a better tooling system would now be in place in the US government? Is it?

    21. exposed data about who was making changes to what

      per I.P. address?

    22. The opacity of the entire situation was baffling.

      This seems to be a significant concern for the author b/c it is applied to Wikipedia's processes too!

    23. Wikipedia dared to defy conventional wisdom, in order to magnanimously welcome the good faith of anyone and everyone.

      It seems more like it defied management practice than wisdom? Similar things are discussed here about programming: https://www.dreamsongs.com/MobSoftware.html

    24. have those efforts been driven by a fulsome understanding of why Wikipedia succeeded?

      if there are no other examples it will be hard to make a theory

    25. it enabled them to accomplish incredible things at an inflection point in the Internet’s evolution

      If they were sloppy in their management practices that's one thing. If they were flexible that's another. Anyway, since I don't know the history this sounds a bit mythical...

    26. a complete picture of the activities involved in producing the site’s content

      So, is this a Peeragogy Handbook like HOW TO guide?

    27. architects, both staff and volunteer

      Is there a public history of Wikimedia's funding receipts anywhere? For VC funded companies often you can get a chart of their raise amounts over time.

    28. . The email list discussions are rife with aspirations and idealism

      Are there learnable patterns to any of this? Should we want to "replicate" Wikipedia's success (in this regard) or are they a special case?

    29. many of us have come to expect greater transparency from more traditional publications as well

      I guess PubPub is meant to be an example? Are there other examples?

    30. facilitating reader insight into writers’ actions and motivations

      That sounds like a good idea for us to adopt in the peeragogy project too!

    31. But the original wiki software they had adopted in 2001 wasn’t fully up to the task. Key policies and software features had to be developed

      That's interesting, is there a guide to the design choices that were made in building MediaWiki?

    1. The site has relentlessly kept its focus on its main goal of providing information — even to the exclusion of chasing money from advertisers or by reselling user data.

      Avoiding advertisers was pretty much there from the start. I think the perennial question in the WIkipedia space is: is Wikipedia dying?

      BTW there was a brief question about this same sort of thing in the peeragogy project. Do we "revile" anything to do with money? NO. But we do have an orientation that brings us closer to the "gift culture" world.

    2. it doesn’t use algorithms to predict and guide what you encounter online, and it doesn’t capture and analyze user data

      Do you suggest here that algorithms inherently problematic? Or is it more the business model behind the familiar social media algorithms that are controversial?

    3. all the other WB_wombat_top web sites

      Wombat?

    4. The plan was hatched with little regard for the values that drive the Wikipedia community, and was ultimately scuttled, following a full-blown revolt by Wikipedia’s users and the Foundation’s staff.

      I guess if we think about "ethical AI", it's good to engage the entry-level threshold question: should we be using AI for this in the first place?

      Even if the code for the AI system was open source, it might not be easily understandable; or it might be a "black box" due to the very nature of the algorithm. That said, maybe these issues aren't entirely deadly to open source community-oriented AI projects...

    5. Wikipedia’s front page is the same for all users. Wikipedia’s volunteer editors openly discuss what content to feature.

      OK, the front page is the same for all users, but the individual user experience is rather different for each different user. Consider for instance the argument (I believe this is made by Lessig) that in the old days there were only a small set of news channels. Now there are many. This leads to "bubbles". While there is only one major English Wikipedia, it's so large that it may feel very different to different people.

    6. Each social media company closely guards its algorithm as valuable intellectual property, even as they tinker and test new versions.

      I listened to a BBC segment on how this works in the world of online dating. In that case, they want to maximise engagement with the site — which means actually helping people find long-term satisfying partnerships isn't in the site's business interest!

    7. different from social media platforms, which have a more complex project

      Agree that it sounds like Wikipedia is less complex, but I'm sure that Wikipedia could also be describes as a kind of multisided market, possibly just as complex as social media, but without money changing hands? Actually this brings to mind some research of mine that was used to theorise Wikipedia as a "creative space". It's in here, I'd have to look around for an open access version: https://link.springer.com/chapter/10.1007/978-3-030-10889-2_3

    8. At the core, Wikipedia’s design and governance are rooted in carefully articulated values and policies, which underlie all decisions.

      Where did these come from? And why are they robust to time & its vagaries? (Are they?)

    9. Everybody has an opinion about how to govern social media platforms.

      Which brings to mind work like "Modular Politics" (https://arxiv.org/abs/2005.13701)...

      Also this:

      « I’m especially interested in alternative governance models for social media: nonprofit organizations, government-run, worker and user cooperatives. I hope to explore these in a later essay. » - https://medium.com/@jasminewsun/jane-jacobs-social-media-83b4265a1d12

  12. Aug 2020
    1. Roles for participants in real-time meetings

      We tend to refer to these as "Rheingoldian roles" since Howard Rheingold introduced them in his HRU courses (and maybe in earlier courses as well).

    1. based on its expected utility

      ... Now, don't we also have considerable uncertainty about the actual utilities of actions?

    2. calibrate

      Is this ever defined?

    3. how to interact with the latent phenomenon encompassed within the true data generating process

      This looks a lot like the set-up in scattering theory.

    4. Regardless of its interpretation we presume that this variation is sufficiently regular that, while an observational process cannot be quantified by any single observation, it can be quantified with a distribution of observations.

      A key assumption, but this does constrain the setting quite a bit. For example, imagine a delicate ancient manuscript that could be read 3 times only before it crumbles to dust.

    5. aleatoric randomness

      Just for the record: aleatory processes are related to games (like dice games). That's different from 'stochastic' processes.

    6. probe

      Cameron suggests to think about this in relation to "projection"

      In category theory, the above notion of cartesian product of sets can be generalized to arbitrary categories. The product of some objects has a canonical projection morphism to each factor. This projection will take many forms in different categories. The projection from the Cartesian product of sets, the product topology of topological spaces (which is always surjective and open), or from the direct product of groups, etc. Although these morphisms are often epimorphisms and even surjective, they do not have to be.

    7. formalize learning and decision making

      This is the sort of thing that we're interested in studying, both in research on collective intelligence and artificial intelligence.

    8. a familiarity of probability theory at the level introduced in my probability theory and conditional probability theory case studies

      Foundational topics.

    9. presumes the existence of a true data generating process

      I guess this relates to philosophical realism?

    10. however, even the most rigorous utility assignments will depend on circumstances we do not know with perfect certainty

      Interesting to wonder why this is! I am reminded of the idea (which I think is incorrect) that generating the maximum number of possible answers will lead to more creative outcomes.

    11. utility

      Related to reinforcement learning, this can create ordering on the actions.

  13. May 2020
    1. To double down on this agenda our aim will be to build the company around open source software.

      From the RM Unger whitepaper:

      "The more one knows and discovers, the easier it is to make the next discovery. If the process of production can be organised on a model of scientific inquiry and experimentalism, innovation can stop being episodic and become permanent. Continuous innovation undermines the basis for the constraint of diminishing marginal returns."

    1. knowledge economy

      Worth having a look at the whitepaper produced by R.M. Unger for Nesta, "Imagination unleashed: Democratising the knowledge economy"

  14. Apr 2020
    1. Dame Wendy Hall and Jérôme Pesenti conclude that “AI could positively affect every area of STEM education”

      I worked briefly in the SOCIAM consortium with Dame Wendy Hall. I wonder if some of the stakeholders in the AI report mentioned here could be approached to ask how they are sourcing technical talent. Benevolent AI would be one interesting possibility.

      A first step could be to follow up with the Southampton folk about the Innovate UK proposal.

    2. [33] “Uncovering the Dynamics of Crowdlearning and the Value of Knowledge”. Proc. Tenth ACM International Conference on Web Search and Data Mining. 2017 [34] “Emergent Complexity via Multi-Agent Competition”. arXiv preprint arXiv:1710.03748, 2017

      These look like particularly worthwhile references.

    3. Universities UK

      Universities are a primary incumbent for any kind of skill sourcing! https://www.universitiesuk.ac.uk/

    4. Wolfram Research

      Another possible speculative angle. They will have a broad interest in sourcing technical talent, not just programmers but also mathematicians.

    5. developing new business models for education using open source software

      London-based experts in this aspect of things include Canonical.

    6. Elsevier

      It's a bit of a strange angle but at the level of a brainstorm we could look at how publishers source technical talent. They are probably not the biggest employer of programmers out there, but it would be a non-trivial number.

    7. Building our Industrial Strategy

      "The aim of the Industrial Strategy is to boost productivity by backing businesses to create good jobs and increase the earning power of people throughout the UK with investment in skills, industries and infrastructure."

    8. IBM

      I have a couple of rather remote links to people who work at IBM, I wonder how they go about sourcing talent there.

    1. Simulating Developer Communities

      I found some earlier work, though it's not very fine grained.

      • Modeling the Free/Open Source Software Community: A Quantitative Investigation

      Also:

      • Agent-based Simulation of Open Source Software Evolution
    1. A short bio and link to LinkedIn would be perfect.

      A link to Github would perhaps be more relevant for many developers. Indeed, wouldn't a good starting place be to crawl Github and try to recruit people whose profiles look suitable?

    1. Practice meditation. It occupies only ten minutes of my day

      Some text relating that back to the kinds of reflection that you do when you process your task list could be thought-provoking. I'd say that each of the little actions with your task list is a kind of mini-meditation.

      Since you're interested in meditation I will take a moment to advertise a book I absolutely love, called After Buddhism. It was probably the most clarifying book I read on the topic after nibbling around the periphery for 10 years or more.

    2. 5. Matthew Walker. 2018. Why we sleep. Scribner.

      An excellent book!

    3. try squeeze

      try to squeeze

    4. coffee

      Well, if I maximized my intake of coffee beyond a couple of cups a day, I might not get any sleep!

    5. It is important to understand that your seemingly perfect colleagues are anything but, and often struggle with the very same issues.

      I think this is a good insight, and that attention to how people think about things, how they get things done, can further bring home the notion that we're all just human. Some people may have more effective "tools" of one kind or another but mostly these different toolsets are learnable.

    6. I take notes in the simple, yet expressive markdown format, which is both expressive and very simple.

      The text here is redundant and yet also repetitive.

    7. Take notes of meetings

      Completely technical point here: if you have a look in the sources of sub-pages on hyperreal.enterprises you'll see the Javascript plugin that I use to create anchors for linking to sections. Maybe useful.

    8. polices

      Just etymologically, the notion of a "policing" function is linked to the notion of a 'polis' or communal inhabitation. It seems like with these tools you are in a way multiplexing your mindset so that you can explicitly look at things from different perspectives. Again, the notion of a shared language is important: there is a difference between the 'polis' and a simple conurbation.

    9. Taking a look at your work log makes it easier to realise when you’re off track

      I've been thinking about the degree to which this sort of feedback could be automated -- but reflecting on some of the earlier comments, I think it's also (and maybe even more) beneficial to keep track of the things that are going well, and understand those better so that you can play to your strengths.

      For example, the time spent exploring infeasible ideas might not be "wasted", e.g., maybe they are only infeasible now but could become feasible later. Thinking of this as "wasted" time ignores the long-term learning and the possibility of recontextualisation later on. By the same token, if you could detect early on when you were potentially about to go down a rabbit hole, maybe there is some information that you could bring in to the current context that could afford a speedy resolution (if that is what was truly needed). This seems to be less about time management and more about understanding the problem spaces you are exploring and the "niches" that you are creating.

    10. If I see one or two low-progress days in a row, I realise it’s time to make a conscious effort to improve my productivity.

      For me I think this could be a sign that I was getting sick and needed to rest!

    11. Right after I complete a task worth writing down, I spend approximately five seconds adding it to my log and then move on.

      The benefits of this way of working could be partly due to positive psychology: you acknowledge things you have been successful at, and this gives you a boost.

    12. I am a big advocate of keeping a log of what I have accomplished throughout the day.

      Given how much data we produce, there already is a log. E.g., consider git commits as a source of data as well as text messages or emails. Pulling these things into a unified tracker would be a nice thing to automate.

    13. Over time, my estimates became closer and closer to how much time I really needed–minus the occasional outlier, obviously.

      I ran into huge conflicts with my boss when, in my second postdoc, I went over my initial planned timeline, and didn't deliver quite what she had in mind. The problem there wasn't necessarily my time management (although she claimed it was), but also the degree to which the project was well specified (something that she had some responsibility for). I imagine part of making good effort estimates is a good understanding of the "domain" you are working in and a reasonable way to break tasks down into subtasks. If there are multiple stakeholders (as in most workplace situations) it seems important for all of these stakeholders to have at least some grasp of the common "language" describing the domain. Without these kinds of resources, I think there's no such thing as good time management (unless time is spent producing them!).

    14. I actually am productive when I finish a handful of well-defined tasks throughout the day.

      This sounds like the epitome of self-management thinking.

      I'm also interested in productivity, but I have some doubts about how that relates to self-management. This is complex and I don't know if there are any "right answers" but, coming back to Banathy for example, it quite depends on what you're aiming to produce, doesn't it?

      At the top of this page of notes, I included a scan of a "spiral" process from Banathy, where he talks about all the different levels of a system that need to be produced to bring about major change. Self-management might be the answer at some but not all of these stages.

    15. Tracking my time helps me stay on track.

      Do you think that this is relevant at a certain level of granularity? For example, if you put "scratch nose" or "scratch head" every time you scratched an itch, that would be time consuming and probably a non-useful overhead because the information you logged would not be actionable.

    16. I open the Time Tracker tool and jot down what I’m going to work on next.

      At one point I tried to set up a stock tool that would automatically record how I was spending time, but I didn't use it or its reporting functions very much. It would be interesting to see how the output of Time Tracker corresponds to your plan for the day.

    17. watching obscure YouTube video

      There's a book called Wasting Time on the Internet that might be relevant here.

    18. Having a daily plan in front of my nose serves as a reminder that the perfect is the enemy of the good

      The "daily plan" might also include (or imply) success criteria, so, giving some pointers about when something is good enough to ship.

      Yes, I suppose I do wonder how much the plan would impede creativity, even while it frees cognitive load. I saw an interview with a Shaolin guy where he said that freedom only comes from discipline. But what about serendipitous unplanned encounters? That's another category?

    19. By spending a few minutes planning your day, you reduce your cognitive load throughout the day.

      This seems like good advice and could potentially be set in bold face.

    20. you notice that the function that contains your bug has poor documentation. So you spend a moment updating the documentation.

      Funny! Yes this sort of thing has happened to me in the past, especially when I was focusing on self-directed projects. I think this is one reason why collaborative projects can sometimse go better, and why in collaborations we have things like standups. It's also true that teams can waste time in much the same way, but perhaps it's slightly less likely?

      At a higher level of abstraction, I think it is a shame that PhD researchers (your target audience) are so often stuck doing "individual" research, when "collaborative" research can be much more powerful and effective. This doesn't invalidate the need/use for individually tailored methods of organisation, but sometimes the main limiting factor is not having people to talk to. Vide, you asked for comments on Twitter presumably because other people would have ideas on this material that you didn't have.

    21. When I finish a task, I move it from my todo list to my work log, which is in the same file, so it’s a matter of literal seconds

      This seems to replicate Org Mode's mode change and archiving actions.

    22. freeing you of the cognitive load of having to remember the task.

      For me I have made many attempts to use Org Mode as intended (as a task tracker) and it never quite clicks. I'm not sure where it breaks down. The system has a bunch of complex options that need to be configured. At present the only task tracking tool that kind of works for me is Google Calendar, since it is linked to events that will happen whether or not I show up (and it is much better if I do remember to look at it and show up). Org Mode's agenda could be used similarly but since my colleagues use Google Calendar, that's preferable. I don't think Org Mode's google calendar integration is that robust at the moment, but I could be wrong.

    23. For example, configure a keyboard shortcut that opens your todo file.

      Here a sidebar or endnote that shows what this might mean in different organisational systems could be helpful. For example, in C-c a a opens the agenda in Org Mode, C-c n f gives a list of files to open in Org Roam... I'm sure there are ways to do it within a windowing system too (e.g., a key command that will run inside Ubuntu, hit emacsclient and open a new file in Org Roam sounds nice, maybe a good exercise for me for later today!)

    24. tools

      "Task tracking tools" ?

    25. Once I transitioned from my postdoc into the “real world”, I quickly realised that this had to change.

      The book has begun kind of in media res so by this point the reader does not know who you are or why they should take your word for it that the ideas you are developing here are bona fide best practices. It might be nice to have a paragraph earlier on that says who you are and why your views should be given consideration!

    26. Atomic Habits [1], James Clear

      Might be nice to make full use of the Tufte style here and put the ref in the margin!

    27. After all, research is a marathon and not a sprint.

      So far it is implicit, but many people don't go on to academic careers in research. Nevertheless, things like "working in a startup" are also described as marathons! You could potentially widen out your "net" a little bit and point out that research skills are relevant even for people who don't have an academic career. Hopefully that's obvious but it may not be!

    28. I spell out aspects of research that don’t see much elaboration elsewhere

      It might be good to have a set of pointers to books on other aspects of the "new graduate student" journey, of which there are many.

    29. the reputation of conferences

      And even the relative importance of journal papers! During my first postdoc I just did what my line manager asked us to do, and wrote a bunch of conference papers (which was his preference). I didn't realise that these papers had effectively no relevance for the REF.

    30. reading hundreds of papers

      Earlier this week, someone pointed me to org-roam, which is a clone of Roam Research, which itself looks like a clone of Notion. I started using org-roam and it seems pretty natural. There is an org-roam-bibtex package as well, but I haven't got that set up yet.

      Of course, Org mode comes with the learning curve for Emacs too, so it's not clear this solves the problem you were talking about in this paragraph.

    31. new graduate students in computer systems.

      Oh, when I clicked through from Twitter I thought you were talking about "systems" studies in general, not just with computers (e.g., in the school of Banathy). Well, computer systems are good too! But of course they bring a whole set of further problems -- writing and maintaining code. Golly!

    32. research

      reach?

    33. The resulting research paper is originally called paper.doc, then paper-new.doc, paper-new-2.doc, paper-final.doc, and eventually paper-final-FINAL.doc

      Well, I tend to write academic papers in LaTeX, and often do the collaborative ones on Overleaf, so versioning is taken care of by Git.

      But not all of my collaborators are academics, and we've used Google Docs, Github Pages, hypothes.is, Etherpad, HackMD, Floobits, wikis both classic and modern... there is a real chance of exhaustion due to switching between all of these different formats, but none of them is 100% satisfactory.

    1. The real test of meaningfulness will come in the use of the extracted information, via evaluation of agent performance on both synthetic tasks and human-in-the-loop applications.

      This sentence kicks off a paragraph with a grab-bag of different methods: all interesting, but from different domains. As such they seem likely to confuse the reader. Don't get me wrong, TPGs, PP/AI, and IAD are all really cool, but their application here remains speculative! The actual problem being addressed exists at a higher level of abstraction.

    2. “complex feed-backs present between individuals and their environments”

      Clearly this wasn't a watertight case for getting the job. Nevertheless I think there are interesting connections between this way of thinking and the Andy Clark way of thinking about niche construction.

    3. Helena Miton’s work on “the role of institutions in generating and transmitting technical knowledge and practices” would find concrete analogues within the work I have proposed.

      I wonder if we could set up an interview to talk about this!

    4. koans

      As people learn programming, what else do they need besides koans?

    5. In order for agents to explore this domain, they will need stepping stones, starting from the simplest possible tasks and growing in complexity.

      This is a worthwhile principle to keep in mind.

    1. We're beginning by marking up these old proposals with new commentary.

      Annotations are via the Hypothes.is sidebar. Anyone is welcome to add them!

    1. rationale

      Instead of a Problem, Solution, Rationale triple, the Y Combinator folks talk about a Problem, Solution, Insight triple. The insight concerns unfair advantages related to growth. It could come from across several "suits": Founders, Market, Product, Acquisition, and Monopoly.

      Whereas the rationale concerns repeatable patterns, startup ideas concern more time-, location- and context-specific opportunities. So, OK, there are some differences: but it seems to me that the common features between these two design languages is worth keeping in mind!

      The breakdown across different "suits" reminds me also of the multiple capitals theory used by XinX.

    2. We can build computational models of social processes and research heuristics using a formal variant of the design pattern methodology

      What I think I have only just realised today is that the process based material could be combined with Monocl and properly formalised inside of category theory.

    3. The latter will draw on direct observation, interviews, “instrumentation” of the social media accounts of researchers who agree to participate in the study, and software integration work as relevant.

      This epistemic aspect is preserved in many of the proposals and seems to be a key aspect of the offering.

    4. enrich the inquiry with both technical and “common sense” features

      I think the "impact" again falls flat in this paragraph.

      The last bit here harkens back to a disagreement that I had with Bob Boyer at UT Austin circa 2003. I felt it would be worthwhile to model mathematics in more common sense terms, but he did not see the value of that pursuit and wanted me to work on a project more oriented towards theorem proving. But all of the interest (and difficulty) of that proposal is implicit here.

      Naturally people will be interested in the adventure but once people begin to grasp the technical ideas they ask "Is it possible? What's really required? What are the stepping stones or proof points?"

    5. a content-oriented model of technical domains, which will be directly useful for education and research

      It's as if the big ideas here are buried (perhaps because I didn't want to be too bold, but in that case it's strange to write a paragraph about 'importance')! To be clear, if it actually worked this idea would transform the way we do knowledge work. That's a bit more bold! The 2019 whitepaper "Imagination Unleashed: Democratising the Knowledge Economy" might say some of this better than I can, even now! (I have only had a look so far.)

    6. The next step in understanding the relationship between content and process could be made using simulation studies

      This should probably have been pulled to the top of the document, since it is an overall thesis for the proposal.

    7. One example application is a recommender system that shows questions and answers from Stack Exchange to a programmer at work.

      I think this is referring "Seahawk: Stack Overflow in the IDE" by Luca Ponzanelli et al. (https://lucaponzanelli.gitlab.io/).

    8. The people who visit the library can teach you where the data that surrounds you comes from.

      This is basically the thesis statement of this paragraph, embedded here in the middle.

    9. ContentMine

      http://contentmine.org/ "We offer a broad range of text mining services for small, medium size and large projects. Our mission is to give researchers an easy-to-use open source text mining code allowing them to find, download, analyse and extract knowledge from academic papers."

    1. Admas Kanyagia

      It would be great to talk with her about the project of building a 'talent incubator' for engineers!

  15. Aug 2018
    1. # pdfnup-2x4 slides.pdf '1-8' | only pages 1 to 8

      This is a way to add comments on a line.

    2. This is a way to add comments on a file

  16. Mar 2017
    1. What the statement should say is that if HHH is a proper finite index subset of GGG then G≠⋃g∈GgHg−1G≠⋃g∈GgHg−1G\ne\bigcup_{g\in G}gHg^{-1}

      reform

    2. Does this remain true for the infinite case also?

      The main Question.

  17. Jan 2016
  18. Dec 2015
    1.                                                                 

      This underlining should be longer!

    1. Simulation comes into its own when the phenomena to be studied is either not directly accessible or difficult to observe directly.

      But it also seems to be a useful method when the phenomena is observable, but its workings are not understood. The point is that simulation gives hypotheses about what's going on inside a black box, I suppose.

  19. Nov 2015
    1. Their trust in writing, produced by external characters which are no part of themselves, will discourage the use of their own memory within them.

      How should we respond to this criticism, considering the ubiquity of writing these days?

    1. Motivation

      Maybe this needs a short section intro, since, actually, the topic under discussion here is Chapter Summaries, not Motivation.

    1. Technology Features/Functions

      This could be a good place to add links to interesting websites @daytripper

  20. Aug 2015
    1. We want to kick things off with a candid confession: we’re not going to pretend that this book is perfect.

      I'm interested in understanding how hypothes.is handles changing content. I made a comment on an earlier version of this page, and the text I highlighted no longer exists. But what happens if the text only moves around? ... OK, here's the technical description: https://hypothes.is/blog/fuzzy-anchoring/

      Interesting. I might like to be able to see a pool of "orphaned" annotations, like the one I made before, which don't reattach anywhere. But current behavior is certainly OK for now!

  21. Jul 2015
    1. we think will be helpful as you get started.

      It's worth pointing out the ways in which the handbook is not perfect. I've added some discussion about that in https://github.com/Peeragogy/Peeragogy.github.io/issues/1