eggsby

Article did not load on my tablet :sweat_smile:

I learned about this reading Ishtar Rising by Robert Anton Wilson

Seems to be enforcing ‘ubiquitous language’ at the machine level - not some kind of mathematical dual where one is invertible to the other - but enforcing soft skills as hard skills.

  ‘protobuf specs dont have enough information for us to codegen iceberg tables so we will write a new codegen spec language’

what makes a duck a duck? when we know which tables we can find it in

Sorry if my language was unclear, I meant to highlight this line from the wiki article:

"There is no proposed mechanism of action by which the positions and motions of stars and planets could affect people and events on Earth in the way astrologers say they do that does not contradict well-understood, basic aspects of biology and physics."

Relevant here because it essentially says "there is no empirical basis for spooky action at a distance" which has been grounds for dismissal of such action-at-a-distance claims like 'the relative positions of celestial bodies influence events on the earth'.

This kind of empiricism has been used as grounds to not critically evaluate these claims. Everyone is certainly free to have their own reasons around why they do not want to evaluate such claims. For example some people only want to consider things that are easily falsifiable and subject to particular scientific practices. The wiki article goes on to mention how Carl Sagan refused to disavow astrology on these grounds (i.e. gravity is weak so stellar influence writ large ought to weak) while still leaving room for a disavowal if it were on firmer grounds. I do think your point about simplicity is salient here.

I really doubt there was any human review of the targeting settings, likely just automated verification.

If I'm writing a tool that allows users to enter their own n-grams is there an expectation that I also need to write a tool that detects if they are inputting pernicious n-grams?

"You Cannot Have Exactly-Once Delivery" is not the same as "Exactly-Once Delivery is expensive".

Which is it?

I made oroboros for working with large collections of app and server configuration that need to be shared across many systems.

Previously I had been using an automation tool to template my configuration variables and found myself frustrated that I couldn't access their processed values outside of the tool.

Oroboros exposes all configuration as JSON over HTTP by recursively templating & assembling configuration files under some directory. This makes it simple to track & release changes through version control.

This project solves a particular need of mine, hopefully someone else find will find useful it as well.

Happy Hacking,

Thomas

Nice read, I had to think about a few of these.

In particular (a == 0) && (a == 1) && (a == 2).

Here's a fun value for a:

    var a = {
      called: -1,
      valueOf: function() {
        this.called += 1;
        return this.called;
      }
    };

I just started playing with docker and it is very cool, but I am not sold on complete isolation of every service in a stack.

I've taken to building my containers using packer & ansible, with docker acting as something between git and an lxc client.

If you run your tests against live instances do you deploy the exact same container? How do you control environmental or configuration drift?

I'd be interested in seeing a CI tool on top of docker :)