daralthus

Just to elaborate:

State is a snapshot of what is deployed at a given time, it is used at the next run to compare if there were any drifts since, eg.: "Did anyone delete a server manually?" etc.

By default it is stored with Pulumi, but if you are on a budget you can just use S3 with a few tradeoffs (concurrent deployments would need a lock that you implement yourself):

https://www.pulumi.com/docs/intro/concepts/state/

You can get rid of toil, automate the whole world and the day's needs, but it will be just used as a leverage for power. People don't not only have material needs.

Is this accreditation accepted to apply to Georgia Tech's OMS CS?

Agreed, it's a shame they couldn't settle with a business model. On the positive: the source is available and their descriptions of the distortions are still the the best reference.

Let us know if you make something ;)

Here's a paper on depth: http://groups.csail.mit.edu/graphics/CodedAperture/

Stanford has a class on computational imaging: http://stanford.edu/class/ee367/

I think these slides should give you a pointer to enough papers to fill a lifetime: https://web.stanford.edu/class/ee367/slides/lecture9.pdf

Georgia Tech also has a class, that's on youtube and Udacity: https://www.udacity.com/course/computational-photography--ud...

Likely need to tackle from multiple sides:

1. Physical: Sleep more. No screen days. Diet changes. Sport.

2. Emotional: [1] CBT. Get rid of perfectionism and huge romantic /heroic ideas that make you suffer. "burnout is heartbreak". Choose way way smaller projects: You commit, finish and get a good feeling from feedback within a day.

[1] https://www.quirk.fyi/distortions

The only thing better than a pinhole camera is a pinhole camera with multiple holes.

Let me explain:

When you have two holes (apertures), their images would likely overlap (depending on the arrangement).

Imagine you took two photos from a few millimetres apart to the left and to the right, then you superimposed them. Some areas would look ok (as both views saw the same distant object) and some would be less ok (of objects that are closer).

If you have a DSLR take the lens off and put a sheet with holes in front. Just try it.

Now as you add more holes you get more and more overlapping images on the same area of the sensor and eventually end up with a blur.

This blur is the same blur you see with a single large hole. It's just that with a large hole you had infinitely many overlaps!

Even cooler, if you happen to arrange the holes in a specific pattern you could capture images with different combinations of perspectives from different holes and you may even undo the overlaps. This is called coded aperture imaging:

https://www.paulcarlisle.net/codedaperture/

This doesn't just solve the biggest problem (limited light) of a single hole, but also captures depth information and you can use it for 3d reconstruction, refocusing etc.

One final bit, with a warning of a deep rabbit hole:

That "infinitely many overlaps" I was talking about happens with lenses too and is essentially a convolution where you convolve the image with itself (actually many different perspectives of itself if I am correct). Which is just the Fourier transform.

Not sure what to make of this. Billionaire companies != seed funded startups. The exploitation really kicks into gear after you have 5000 employees and there are many inflection points of power differentials until then. Each not as a big deal, but slowly eroding some rights the farther and lower you go.

I wonder what will happen to TiltBrush after this?

Thank you for the paper.

Of course proofs can be still correct, it's just that we cannot cope with infinites very well without talking about computational limits.

My argument was hyperbolic, I regret, but it was less about what's incorrect and more about what's incomplete. The aspirations of early mathematicians was to discover a platonic ideal. Instead of that we only got useful tools that are always* up for re-interpretation depending on context.

An example of this is that Euclid's fifth postulate, which be consistent with many other interpretations of geometry too, not just the single one originally intended. It turned into a tool of formal scaffolding instead of an omnipotent "truth". Going from absolute to relative in power.

*In cases where they involve infinites. Including simple expressions like "take N to be an integer"

> They were probably amazing (and much simpler) tools back in the day, but the world has moved and we're somewhat constrained by what's familiar.

Exactly my feelings about Phabricator. GitLab simply runs circles around it if you consider the issue tracking, CI etc. as part of the problem.

Because of the halting issue you cannot summarise reality into a set of axioms and rules. Which is the original purpose of mathematics. Therefore you cannot say mathematics is right or true without adding at the end: "this actually could all be wrong".

Can someone please give us a better understanding in what's meant by this?

> Duality in mathematics has been a profound tool for theoretical understanding. Can it be extended to develop principled computational techniques where duality and geometry are the basis for novel algorithms?

Async Google Docs > async Slack > meetings

You can setup a template with all they main points and checklists already in place, then get people to fill it out and collaborate on it.

Here's an example with the headlines, each would become it's own section:

    - Problem summary
    - Impact
    - Measure of success
    - “Why does this happen?”
    - Current solutions
    - Ideal scenario
    - Constraints and needs
    - Unknowns
    - Proposed change
    - Alternatives

And checklists for the writers:

    - On what level, and how detailed I want to describe things?
    - Did I write details into unrelated sections?
    - Who is this document for? Do they really know all that stuff I assume?
    - What impact would this change have on the business?
    - Can we quantify it?
    - How do we measure that the change solved the problem? How does winning look like? (ROI, proof of impact)
    - What is the root cause? (“Why? Why? Why? Why? Why?”) Are we solving the right problem?
    - Can we break up the solution into smaller chunks to reach the goal faster?
    - Are we trying to solve a general problem before something specific?
    - How do people solve this already?
    - Did we talk to them yet?
    etc.

Then you offloaded all the questions you would have asked anyway to a document.

SEEKING WORK | London | REMOTE

https://stackoverflow.com/story/peterkadlot

Full Stack Developer with 8+ years experience at startups and agencies.

Regularly work with:

- Languages: Javascript, Typescript, Python

- Backend: GraphQL, PostgreSQL, MongoDB

- Frontend: React

- Mobile: React Native

- Infra: AWS, GCP, Kubernetes, Serverless, Terraform, Pulumi

peter (at) makethink.com

> Use subtyping in Julia and enums in Rust. One caveat though: Enums can not be extended from the “outside”. If you plan to let users of your library add new type variants, you will have to use some kind of trait approach.

This extensibility from the outside is the most important idea behind Julia!

For example this is why a generic deep learning library doesn't need any extra implementation details to be able to run on GPU as well. https://fluxml.ai/Flux.jl/stable/gpu/

They are not competitors. Antitrust law might not apply.

free != quality

ads == worse quality

L.S. and S.V. Penrose did some experiments around this: https://www.youtube.com/watch?v=2_9ohFWR0Vs

Use Pulumi or Terraform.

There's a slight misunderstanding here: the main purpose of tax is not revenue for government spending, but to impose a policy on the actors in the economy. It's a way to take out the excess money from circulation that central and retail banks printed, effectively controlling inflation.

Conversely, I seem to have the opposite problem and not being able to work on anything in the concrete. Always just expanding onto different perspectives and revelations about the systems I am supposed to work with.

Any advice for that?

Are you kidding, these bikes are strictly superior to Lime's in every perceivable way like robustness, speed, brakes, ergonomy, maintenance etc.

I also found this helps.

Basically hold your breath till you feel it becoming uncomfortable. Hold for just a few more seconds (but no need to pass out). A reflex will kick in and open your airways.

https://endlesss.fm/ gives you a shared loop pedal. It's not real time, but just as social and the looping means you don't have to worry about lag.

Is go really high level? Imho the type system is so botched you cannot really say that.

One cannot even write a generic `reduce` function...

This is the same Lidl who spent €500m on a failed SAP rollout, right?

In moo.do this is the default.

You may find yourself wondering: what would happen if the gears are not circular?

The answer is then they can implement a wide range of mathematical functions like reciprocal, tangent, square etc. and you will get a kick out watching an hour long video about a mechanical "computer" with cams building up differential calculations: https://www.youtube.com/watch?v=s1i-dnAH9Y4

Enjoy!