Skip to main content

Maps are not the Territory

A map functions as a reduced, simplified, streamlined representation of the real world. We all function in maps, or a simplified heuristic models of different parts of our reality. Heuristics, the simplification of the world representation is an example of a ‘map’. Anything really, which simplifies the world to a more digestible, parseable system is a map.

With this context, we can see how there are maps all over our world, everyday. GPS directions, configuration files, electrical diagrams, software architecture, so on and so on.

The problem with any maps, or simplification systems is that it hides away a lot of the complexities, gives false confidence to the reader, and makes things much simpler sounding that it really is. Take for example, an electrician’s diagram of your house. They have cable lines, junction boxes, and switches and outlets. It hides away the amperage requirements, the wire thickness and resistance of those wires which are needed to be taken into account when building this out. Also, how these lines are NOT really straight, and they have to interact with the wood and foundations as well as the pipes for HVAC and plumbing; those are on different maps, and different guides for wire usage.

So in summary:
  • Maps are representation or a simulation of the real-world in a simplified form for a specific purpose and audience.

  • Those maps are NOT a 1:1 representation of reality.

  • Experience and knowledge is required to fully use maps; and true experience and knowledge comes knowing what those maps do NOT represent.

Mental Models

I have always had a vague recollection of mental models, perhaps on thinking about HOW to think.

Recently, I was pointed to a book, called The Great Mental Models by Shane Perris , and it put into words the vague recollection and grasp I had on mental models, which I relied on reflexivly and intuitively.

I find it, often, its hard to think! You are lost in the details, and consequences of your decision and the follow up actions. I have to get away from these compunding factors to think clearly about the problem or task at hand.

I realize, I often fall back on first principal thinking, but in reality that's only one way of many, in how to approach a problem, and other models may work better to certain situations.

The models:
  • Inversion Principal

  • Always invert

  • First Principals

  • What is the thing you can't reduce?

  • 2nd (and perhaps, third) order affects

  • What are the affects of ths your action, and what are the affects of those affects?

  • Circle of Confidence

  • Probablistic Thinking

  • Occam's razor

  • Simplest systems, with the least moving parts, are the most likely to be true

  • Hanlon's razor

  • Never attribute to malice that which is adequately explained by stupidity

  • Thought Experiments

  • The Map is not the Territory

  • Our representations (maps, models, words, beliefs) of reality are not reality itself (the territory); they are simplifications that can be useful but are incomplete and can lead to errors if mistaken for the complex, actual world

I am going to be writing more in depth about this.

Systems Design Interviews

This is a little reflection on problems with system design interviews, the ends and means of executing the proper answering framework.

You may notice that this is targeted to a generalized audience, however it is the of the topic which is important to me.

The Mistakes

Initially, upon starting the interview

  1. Not asking enough questions

  2. Not reading the questions carefully

  3. Not quantifying the problem, best guess (with some assumptions) is OK

  4. Not writing down the Functional and Non-Functional Requirements

During the Actual Design

  1. Not talking about tradeoffs - i.e. Why am I choosing this? What are the alternatives?

  2. Starting with too much complexity, and/or pre-mature optimization

  3. If you are thinking about problems, and applying solutions, make sure you justify/talk out loud about them.

The Solution

The Framework

  1. Talk with the interviewer, have an authentic conversation during this process

  2. Read the damned question

  3. Ask clarifying questions

  4. Write down the Functional and Non-Functional Requirements

  5. Design small and simple

  6. Design for scale if necessary, TALK about tradeoffs

Helpful Notes

  1. Think in Frameworks and First Principal (First principal is, you are trying to communicate that you know how to design this)

  2. Communicate, and over communicate

  3. Practice, practice, practice

  4. Mock interviews with people you trust or professionals HelloInterview is a service I used

  5. Notes, or systems to help you remember the framework. Sticky notes around your screen might help.

On Being a Warrior

To be a warrior, means to do what is necessary and good, in your judgement, despite fear, doubt, difficulties and uncertainty.

Being a warrior does not have much to do with overcoming your external enemies. It has a lot to overcoming your own self.

It might mean following your own path, it might mean joining a group of like minded people.

It might mean that you take a huge risk and put your life and other things on the line, or it might mean taking the safer path and taking the 'cowards' way out. (Only you whether or not it was a judgement based on fear)

It also means accepting your own failures, and learning from it; not dweling in your past mistakes as if it defines you.

It might mean seeking help when needed, it might mean giving help when un-wanted.

It might mean you pass harsh judgement or be harshly judgement, or just the opposite, be praised by others. That is not the end goal.

It'll most likely require making difficult decisions and commiting; trusting your own judgement.

My symptoms and experience with the Covid-19.

I just wanted to log my experience with my exposure to the COVID-19. As far as I know, I am in the 'recovered' group, and able to move about. However, there's no way to test this since as of write now, there's no antibody or immunoglobin testing is Los Angeles.

Timeline

  • 3-10: Initial Exposure (through work) in Boston (Start of 2 week)
  • 3-11: Friend(G) staying in hotel with me for a visit.
  • 3-15: First alert that I was exposed via email, Landed in LA. Started very light coughing. 4 days in of isolation.
  • 3-16: Staying at G's apartment to avoid infecting people at home. G has already been exposed so he's in his own isolation timeline.
  • 3-17: Got my campervan initial load of gear. Started feeling flash of fever. But it's so short that I can't tell if I actually had fever.
  • 3-18: Left G's apartment. More flashes of fever, but still indiscernible from allergies.
  • 3-19: Remote Ventura. Remote campground living; isolated. Still just light coughs and flash of fever.
  • 3-19: Remote Ventura. Remote campground living; isolated. Still just light coughs and flash of fever.
  • 3-20: Remote Bishop area. By hot springs. There are others camping out around here, but avoiding them. Light coughs, but worst fever to date. In the evening leading to shivering. Couldn't tell if it was from cold weather.
  • 3-21: Remote Bishop area. By hot springs. There are others camping out around here, but avoiding them. Light coughs, flashes of fever.
  • 3-22: Remote Mammoth Lakes area, another remote location; Too many people around so left for even more remote location after ski tour (telebowl). Ended up really cold; but the symptoms lessened.
  • 3-23: Remote Neveda, middle of nowhere, not sure where. Passed through Benton crossing, awesome climbing area, worth a revisit. Saw White Mountain, beatiful. Worth re-visiting. No changes in symptoms.
  • 3-24: Delta, Utah. Rented a motel to keep out of the wet and cold. (End of 14 days Quarantine)
  • 3-25: Delta, Utah. Rented a motel to keep out of the wet and cold. Met up with friend (A) for hiking. No symptoms.
  • 3-26: Wedge Overlook, Utah. More hiking in Utah with friend (A). No symptoms.
  • 3-27: Started snowing, Stayed at A's house. Hint of fever, couldn't really tell. Cough worsened, thinking it's cold-dry-high-altitude effect. BUT lost sense of smell. Went to get tested at U of U.
  • 3-28: Started snowing. No fever. Same coughs. Did Circle Peak ski tour and left for LA. Slept in middle of nowhere.
  • 3-29: Driving back to LA, got a call saying I was positive for the Corona. Told GF, and she left before I entered the house.
  • 3-30: At home in Los Angeles. No symptoms.
  • 4-3: Got a letter from LA Public Health about being quarantined at home for 3 days post symptoms. This was already 6 days post symptoms so a little late.
  • 4-5: Friend A went to get tested, was negative for Corona.
  • 4-7: (Day of writing) At Home in Los Angeles. No Symptoms.

Splitboarding Gear

Generic Quick List

  • Splitboard
  • Splitboard specific binding
  • Skins that fit your splitboard
  • Poles with powder baskets
  • Backpack with avalanche gear support
  • Ski Crampons that fit your bindings
  • Helmet
  • Sunglasses AND Goggles
  • Advanced: Air Bag
  • Advanced: Ice axe
  • Advanced: Crampons for your boot
  • Advanced: Crevice Rescue gear

My Current Setup

  • Smith Wildcats as my uphill glasses
  • M3 Goggles
  • Smith Helmet (straps neatly wth poacher backpack)
  • Karakoram Prime
  • Karakoram flex-strap
  • Karakoram crampons
  • Black Diamond Carbon whippets. Poles are reliable, the axe is only needed in certain circumstance.
  • DAKINE Poacher 32L
  • MAMMUT R.A.S. airbag, which fits inside the Poacher.
  • Jones Solution (166, its too big. Go small as possible)
  • Jones Skins, size D
  • Petzl Ice axe
  • Petzel Crampons

Things I learned

  • Getting skins wet will lead some big foobar. Don't get skins wet, but if you do, keep it inside your jacket to melt and dry
  • Don't get snow on the glue side of skin!
  • Really, don't!
  • I am not ready for Crevice leading
  • Avy Courses are important, take at least avy 1, but if you continue into bigger missions, take more advanced classes
  • Try it once with a guide see if its your taste, its hard.
  • You don't really need ice axe and crampons all the time; for missions where you need those, lots of planning is required.
  • Ice crampons for your ski are small, easy to carry, but might save your butt once in s awhile
  • Changing your binding height during approach is pretty key to managing fatigue
  • That said, your technique will count more for efficiency and fatigue
  • Heat management, with layers, for me is pretty important. Its the difference between enjoying my outing vs slogging through suffering. I get into this a little more below.
  • Avy gear (probes and beacon) is pretty key, practice and check on them before the season. And again during the season.
  • Avalanche air bags are a good extra insurance, but they are only useful for subcases of all avalanches, and you might still bne injuered and stuck. So don't rely on it.
  • That said, once you buy one, always go out with one. Not only on days you think avy will happen. Its insurance, and safety gear. Its like not wearing a helmet, cause you think you won't get injured on this day.

Heat management

  • In most conditions, this is what I carry. A thin base layer, a Patagonia R1, a light insulation layer (usually packable down), and a soft shell. I wear normal shell pants, with light insulation, somtimes bring a insulation layer I can put on just in case.
  • When you combine a light insulation layer with soft shell, it gets SUPER warm.
  • When you wear a R1 with a soft shell, again, it gets pretty warm.
  • I usually only wear the R1 + base layer when I start out the approach. Its hard work and most people will sweat balls. I do too. I need that to evaporate as fast as I can.
  • As you go up, you change the layers depending on conditions, if I stop to break, all the layers go on, when I start riding down, its usualy just a soft shell over the R1, since riding is also hard work, but a lot of wind will get me cold.
  • A face mask of sorts are always useful, Sunglasses for going and goggles for coming down are also useful, and different types of gloves for going up and waterproof warm for coming down is quite sueful as well.

## Skinning Its a movement pattern unique to you and something you will have to learn to figure out. Its is very unintuitive. For example, standing on a slope, you lean DOWNHILL to maximize skin contact, not lean uphill to use an edge. I will write in the future about skinning techniques. But, to say, you are gonna have to practice it and find your own efficiency.

Pelican Autodeploy to S3 from Github

Requirements

  • Use pelican-quickstart to setup your pelican environment. This will ask you and setup most of the basic variables you need for your automation.
  • Also setup a Github repo for the blog you are going to be hosting.
  • You should be able to test your blog using make devserver.
  • Get your secret and access key for that IAM account; save this you are gonna need to add it as env variables in your CircleCI.
  • Before you can do this, you should be able to already do make s3_upload on your local environment.
  • If you don't know how to point your DNS to a s3 bucket, that's in another post.  

Step-by-step

  1. Log into you github account.
  2. Goto the marketplace to install CirclCI
  3. Add your project to CircleCI Projects
  4. Go to your "Setting->projects->(your repo)" and then find the "Environment Variables" tab, and add 'AWS_ACCESS_KEY_ID' and 'AWS_SECRET_ACCESS_KEY'.
  5. Make a ".circleci" folder, and inside that add a "config.yml" file.  

Setting up your config.yml for CircleCI

version: 2
jobs:
  build:
    docker:
      - image: circleci/python:3.6.1

    working_directory: ~/repo

    steps:
      - checkout

      - run:
          name: install dependencies
          command: |
            sudo pip install pelican markdown awscli
            git clone https://github.com/Pelican-Elegant/elegant.git && pelican-themes --install elegant

      - deploy:
          name: deploying to s3
          command: |
            make html
            make s3_upload
  • I have added the git clone line for installing themes, but you are free to use your own, or if you are having weird issues, try starting without a theme.