Blurts

Several Senate Democrats have proposed a Data Care Act. It seems obvious that there is pressure for politicians to impose expectations on online companies to address concerns for the use and protection of user information. I have concerns that Republicans whether they object to this proposal or note will refuse to consider legislation offered by Democrats. However, the public should be made aware of the specifics of this proposal.

I see the proposal as consisting of two components – a) what are the responsibilities of providers to protect data and to inform users when protective measures fail and b) what are the responsibilities for how companies use the data collected.

I have minor concerns about both provisions. First, I think it impossible to absoluately protect data. Hence, the proposed penalties when some data are compromised (e.g., medical records) are possibly unfair. The way I read experts on hacking it is impossible to protect data that are accessible on the Internet and what responsible efforts at protection should be specified.

An important section of the proposed legislation related to the use of data follows.

shall take reasonable steps to ensure
6 that the practices of any person to whom the
7 online service provider discloses or sells, or with
8 whom the online service provider shares, indi-
9 vidual identifying data fulfill the duties of care,
10 loyalty, and confidentiality assumed by the per-
11 son under the contract described in subpara-
12 graph (B), including by auditing, on a regular
13 basis, the data security and data information
14 practices of any such person.

I guess a strict interpretation of this section would satisfy my basic concern, but I am against the sharing of personal data from one company to another. My logic goes something like this. I assume I enter into an implied contract when I use a service that collects my data. I assume that I am “paying” for the service by accepting the visibility of ads and the collection of my data to personalize these ads. I am not agreeing to having my data shared with unknown others even if this process is intended to improve the personalization of the ads I receive. It is my lack of awareness of just who these other companies are that is at issue. I can deal with ads I know are ads (I guess this is also my assumption) and with my interpretation of the motives of the company accepting my payment for the free accesss to a service provided.

This perspective concerns my right to informed choice. I can make a decision not to spend money in Hobby Lobby or Chick-fil-A as a function of my opinion regarding how a part of my money might be spent. I assume the same right when it comes to Google or Facebook. I cannot exercise this right if Google or Facebook has shared my data (my payment) with organizations not made known to me.

Dean Shareski posed a question on Twitter about the lack of critical reviews of education books. The tweet generated an interesting discussion which you can probably still read if you search Twitter for Shareski soon after this post appears.

Just to offer a criticism so at least one will exist I will provide the following. Last evening my wife was reading “The hungry caterpillar” to our two youngest grandkids and happened to notice what seemed an error in the book. The author, Eric Carle, ends the book by describing the emergence of a beautiful butterfly from a cocoon. We have reared and released monarchs for several years with these same preschool kids and even they know that butterflies emerge from a chrysalis and not a cocoon. Whatever happened to promoting  STEM across the curriculum, etc.?

Just because I wanted to know I did a search on “The hungry caterpillar” and cocoon and found that others had spotted the same issue. Carle even offers a defense. He claims that there are some special butterflies that emerge from a cocoon. I doubt he took this position until after the statement in his popular book was challenged. He also claims the book claims this is a special caterpillar and if you want to be concerned about anything you should be bothered by the claim the caterpillar eats lollipops and ice cream. OK – in fiction, you are allowed to make whatever claim you want. Not sure I would try to explain this distinction to my grandkids. 

Back to the question of commenting on education books within Twitter. As I thought about the question, I decided Shareski probably was correct. While one can find all kinds of critical evaluations elsewhere, you don’t see this on Twitter. No one suggests that they bought xxxx because educators kept recommending it, but you thought it was mostly fluff. 

I have certain standards I adhere to when I suggest a book on educational practice for educators. I look for concrete suggestions for practice, but I also want to be informed of the science the author argues to justify a specific practices. It bothers me when authors are willing to make specific suggestions, but I feel that there is no credible basis for such classroom activities. Even if I end up disagreeing with the justification offered, I want to know that the author has struggled with the issue of evidence and has something specific to propose. Else and especially if I think the evidence says otherwise, I am inclined to regard the book as potentially dangerous fluff.

Why don’t I just say this? Perhaps, something like “interesting and creative idea, but based on flawed logic and a lack of sound evidence”. I think if you say things like this you end up being labeled an elitist who lacks real classroom experience. You might note that this claim seems now to be quite commonly used by politicians who want the arguments against their positions to be dismissed. This position seems to appeal to true believers and seems to work better than actual evidence. It seems to play well with groups who feel underappreciated and fits with the self-perception that they understand things others who are not in their position do not. Blog posts or other “long form” refutations offer the opportunity to challenge such positions with more detail that cannot be provided in a tweet. Of course, there is always the problem of others being willing to read something that is more than 280 characters in order to consider the evidence provided. 

Can I offer examples of education books and articles that offer specific suggestions backed by quality data? Sure. For example, search for the books and articles of Deanna Kuhn on the topics of problem based learning and on argumentation. In both cases, the suggestions and examples for the classroom are specific and replicated. The evidence is carefully collected and demonstrates significant advantages over both traditional strategies and shoddy applications of similar sounding approaches. I think I have written about these examples here and in another blog. Just search for Kuhn.

Maybe, it is just easier to write for those with similar values and similar preferences for how positions taken are justified. This would seem a version of the filter bubble, but there are reasons filter bubbles exist.

I came across an article written from a brain-based model of learning that makes the point I have also made many times. A model of how learning happens is not intended as a recipe for instruction. To me, a description of how learning happens covers the ways in which we learn – period. We learn from external circumstances that fit many different approaches based on our internal behavior as learners. The learning theory is about the internal behaviors or structures enabling learning and the method of instruction is about the external circumstances. Educators can manipulate the external circumstances. Learners can manipulate both the external circumstances and sometimes the internal processes.

My training was as a cognitive psychologist and while I assume that the models cognitive psychologists proposed for how learning happens must somehow be based in biology I don’t see inconsistencies in these approaches. I find the cognitive model more complete and useful and hence a better way to propose specific interpretations of learning struggles and a better guide for designing external activities more likely to encourage effective internal processes. Constructs such as the limitations of working memory, metacognitive recognition of cognitive failures that can be used to adjust processing activities, the use of existing knowledge to offer meaning (whether accurate or note) to new experience, etc. seem a useful framework. Learners don’t have or recognize existing knowledge (information or structures) – provide experiences or help learners recognize. Learners can’t meet the processing demands because the tasks that are expected overload working memory – find ways to off load tasks, slow the input of new demands, Learners are unable to identify their own failures as they occur in real time – find ways to provide external feedback. These fixes are difficult to make work. Often, one fix creates another problem (e.g., an external task designed to fix one problem increases processing demands, developing background knowledge and structures that allow understanding take time that may seem impractical within a group-based approach), but this is the reality of accomplishing difficult things.

Perhaps cognitive models of learning might be considered a short hand or higher level language (to use a technologuy analogy) for biological explanations. This is where I think we may end up when our understanding of brain biology becomes more sophisticated.

The issue of whether brain-based or cognitive models of mental activity are most useful is not really the issue here. The issue raised in the article I cite and in my personal experience is the misrepresentation of what models of learning are. As the author of the article notes – constructivism as a model of learning does not equate to the promotion of hands-on learning activities. Constructivism, cognitive models, or brain based models have to be able explain how learning happens via a variety of life experiences. Educators and those who prepare and support educators need to understand this difference.

I would make one further suggestion – research on the effectiveness of specific learning experiences does not really require an underlying learning theory. The learning experiences work or not or more accurately are more or less effective than a contrasting learning experience. The models of learning are valuable when it comes to proposing learning experiences to be tested and sometimes to debug less than ideal outcomes resulting from a specific experience – this has to happen and this seems a way to encourage this to happen or this did not seem to work and this may be why.

Google claims that it is motivated to speed up the internet. Among the projects is has initiated to this end is Squoosh. This online service (you can also install the code yourself) drastically compresses images with little noticeable decline in image quality.

The tool is intuitive to use and perfect for chromebooks. You will note that my sample image has been compressed from 16MB to 904 kB. The left and right hand portions of the image show the contrasting appearances of the before and after