benvolio

The bill must of course be read in its wider context. The Supreme Court of Canada cases of R v Bykovets [1] and R v Spencer [2] are essential context for interpreting these amendments.

From R v Bykovets, paras 6-8:

[6] I respectfully disagree. This analysis runs counter to this Court’s jurisprudence under s. 8 of the Charter. We have never approached privacy piecemeal, based on police’s stated intention to use the information they gather in only one way. The right against unreasonable search and seizure, like all Charter rights, must receive a broad and purposive interpretation, reflective of its constitutional source. Since Hunter v. Southam Inc., [1984] 2 S.C.R. 145, we have held that s. 8 seeks to prevent breaches of privacy, rather than to condemn or condone breaches based on the state’s ultimate use of that information. Privacy, once breached, cannot be restored.

[7] To that end, our Court has applied a normative standard to reasonable expectations of privacy. We have defined s. 8 in terms of what privacy should be — in a free, democratic, and open society — balancing the individual’s right to be left alone against the community’s insistence on protection. This normative standard demands we take a broad, functional approach to the subject matter of the search and that we focus on its potential to reveal personal or biographical core information (R. v. Marakah, 2017 SCC 59, [2017] 2 S.C.R. 608, at para. 32).

[8] Informational privacy is particularly critical — and particularly challenging. Our jurisprudence recognizes that computers are unique and present privacy risks that differ from s. 8’s traditional objects. Thus, this Court has determined that s. 8 generally prevents police from seizing a computer without a warrant — even though the device itself provides no information without judicial permission to search its contents — because seizing the computer gives the state the means through which to access its content (R. v. Reeves, 2018 SCC 56, [2018] 3 S.C.R. 531, at para. 34). (emphasis mine)

[1] https://decisions.scc-csc.ca/scc-csc/scc-csc/en/item/20302/i... [2] https://decisions.scc-csc.ca/scc-csc/scc-csc/en/item/14233/i...

Seems like this is referring to Clayton Christensen’s Where Does Growth Come From? talk:

https://m.youtube.com/watch?v=rHdS_4GsKmg

I would guess that it would be unlikely for any ICC case to have significant standing in any American courthouse. The US is no longer a signatory to the Rome Statute which was the treaty which established the ICC and I don’t believe there is any interest in being bound to their rulings.

>The Code Llama models provide stable generations with up to 100,000 tokens of context.

Not a bad context window, but makes me wonder how embedded code models would pick that context when dealing with a codebase larger than 100K tokens.

And this makes me further wonder if, when coding with such a tool (or at least a knowledge that they’re becoming more widely used and leaned on), are there some new considerations that we should be applying (or at least starting to think about) when programming? Perhaps having more or fewer comments, perhaps more terse and less readable code that would consume fewer tokens, perhaps different file structures, or even more deliberate naming conventions (like Hungarian notation but for code models) to facilitate searching or token pattern matching of some kind. Ultimately, in what ways could (or should) we adapt to make the most of these tools?

A relatively comprehensive look into PFAS can be gained by reading the Draft state of per- and polyfluoroalkyl substances (PFAS) report published by Health Canada. [1] The Existing Substances Risk Assessment Bureau at Health Canada (which assesses risks of chemicals that are in commerce in Canada), is increasingly moving towards new analysis techniques broadly characterized as New Approach Methodologies (NAMs) (see section 6.2.5 and 7.6 in the report). So, to my understanding, part of the apparent gap in research is being plugged by the application of models which infer chemicals properties, interactions, and effects without needing to study each of them individually to gain at least an initial idea of how they might harm organisms and environments.

See:

> The evolving landscape of chemical production has rendered toxicological testing using traditional models (i.e., live animals) impractical, and advances in science coupled with ethical concerns have resulted in government agencies, including the United States (US EPA 2021b), European Union, and Canada (Bhuller et al. 2021), committing to reduce, refine, and potentially eliminate the use of mammalian models from certain regulatory testing requirements, where scientifically justified. New approach methodologies (NAMs) are broadly described by the international risk assessment community as any technology, method, approach, or a combination of these that can be used to reduce, refine, or replace animal testing and allow more rapid and effective screening of chemicals. These methods may include the use of computer models or assays with biological molecules, cells, tissues, or organs as well as exposure measurement approaches. [1]

[1] https://www.canada.ca/en/environment-climate-change/services...

P.S. CTRL+F "Health effects" in that report to get a cursory idea of the risks.

Source: worked as a data analyst at the Existing Substances Risk Assessment Bureau

  Location: Canada
  Remote: Yes or hybrid or onsite
  Willing to relocate: Yes
  Technologies: Python, C++, MATLAB, Java, PostgreSQL, HTML, CSS, AWS, Django.
  Résumé/CV: martyrudolf.com/resume/
  Email: martywrudolf+hn at gmail dot com

Newly graduated Computer Engineer with specialization in software. 2 years of professional programming experience including positions in government, industry, and independent consulting work.

It's been my view that everything that one chooses to write is generally chosen from the space of all other possible things that could be said. Particularly, I think making the first exposure to your article a clickbait headline says some important things about the contents of the article and the approach of the author. Those are important signals that help those with a keen eye to weed out low signal-to-noise ratio articles.

On the flip side, there seems to be a lot of noise lately and it would be nice to have a break from it, so I'd almost want to be spared from being served these articles entirely. But then that's taking editorial powers to another level.

  Location: Edmonton, Alberta, Canada
  Remote: Yes or hybrid or onsite
  Willing to relocate: Yes
  Technologies: Python, C++, MATLAB, Java, PostgreSQL, HTML, CSS, AWS, Django.
  Résumé/CV: martyrudolf.com/resume/
  Email: martywrudolf+hn at gmail dot com

Newly graduated Computer Engineer with specialization in software. 2 years of professional programming experience including positions in government, industry, and independent consulting work.

David Autor of MIT is not a bad place to start.