Create a gist now

Instantly share code, notes, and snippets.

What would you like to do?
[RC Diary] SICP 1.2 and a bit of rage (-56)

[RC Diary] Raging over SICP 1.2 and winston-cloudwatch (-56)

SICP

I've struggled a lot with SICP today. I don't see the point of teaching something using difficult concepts as examples.

Why would you, for example, give an exercise like Exercise 1.13 in this version of SICP, just search for "Exercise 1.13" in the page to find it.

I am trying to learn how to program, so my focus is on that. I don't even have a valid point here maybe, because you might argue that by giving the reader problems (whatever nature they are) to solve, then they are going to get better.

I will stop ranting, it's too late for this. I might just have a look at the next chapters and see what I can take from those, probably just reading through without doing the exercises. But that's just too bad for me, I learn by doing, I am going through The Little Schemer by doing every single exercise in Clojure.

I think the same approach can't be followed with SICP, at least not by me.

winston-cloudwatch problem

A user reported a bug, AWS docs on the matter are lacking useful information:

DataAlreadyAcceptedException The event was already logged.

from what I've understood they have 4 servers starting together, each of them writing to the same group and stream. So concurrency is definitely a big hint here, I think I will proceed like this to identify the problem:

  • create a shell script that runs 4 processes
  • all processes should use the same group and stream
  • start the shell script using the debugger flag on
  • watch for that error

If the error does not happen then I have to think about something else. But probably the error won't come up right away, I will probably need to run it few times to get it to show its ugly face.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment