ossopite

I could relate to the claims in the article: for the last 6 months I've had soreness and pain in my left index fingertip that has confounded the doctors I've seen about it, and all that's helped is to avoid using it. Perhaps someone here has experienced something similar?

When typing I feel pain initially at the fingertip where nail meets skin, which worsens and radiates around to the middle finger side of the fingernail after more use. Even when typing without using the index finger, stretching the finger to keep it away from the keyboard induces some pain after a while. If I cut the nail very short, I think I notice some tenderness and loss of sensation in a spot near the middle of the skin just under the nail edge.

I think the pain developed over time while heavily using a split mechanical keyboard (kinesis freestyle edge) with poor typing technique and putting repeated pressure on the tip and side of the finger, but it has not gone away after switching to something more comfortable (kinesis 360). I don't remember any significant injury happening.

The only visual sign is that the skin seems strongly attached to the nail near its edge, there is minimal free edge compared to what my other fingernails (which are all short) have. Actually that is somewhat true of the other index finger, but to a much lesser extent. There is nothing apparently abnormal about the skin under the nail but perhaps any issue isn't visible.

Interestingly, the pain seems worse when my hands are warmer.

X-rays/MRI/ultrasound scans showed nothing abnormal apparently. All my internet searching for an explanation has yielded nothing, hence writing this comment to see if anyone can help.

for performance reasons, yes, I believe it's because the accumulation is over parallel computations so the ordering is at the mercy of the scheduler. but I'm not familiar with the precise details

edit: at 13:42 in https://www.youtube.com/watch?v=TB07_mUMt0U&t=13m42s there is an explanation of the phenomenon in the context of training but I suspect the same kind of operation is happening during inference

This reminds me rather of https://libraryofbabel.info/

The most important difference in my experience has been that pyright is a lot more rigorous and correct than mypy. I found that it was easy to confuse mypy, leaving large amounts of code unexpectedly not type checked. Put simply, I trust pyright and I do not trust mypy.

You also have a lot more control with pyright over where unknown types are permitted or what other checks should be performed. Also, it can scan library source code to infer types which I don't think mypy supports.

This site reminds me of http://rigaux.org/language-study/ which includes studies of where language concepts originated, lineage of languages, wide ranging syntax comparison (although missing some newer languages) and other interesting things

yes, postgres will add such a condition to the query and in simple cases like this is able to use a corresponding index

unfortunately this can break down in more complex cases. roughly postgres trusts a limited set of functions and operators not to leak information about rows (e.g. via error messages) that the RLS policy says a query should not be able to see. that set includes basic comparisons but not more esoteric operations like JSON lookups. at some point postgres will insist on checking the RLS policy result for a row before doing any further work, which can preclude the use of indexes

Yes, that's one of the most confusing. http://leewoodprojects.co.uk/wp-content/uploads/2010/04/2955... is a more useful diagram (unfortunately I can't find the original source)

A truly great comment. On the second observation, my thought on that debate is that we should consider the lifecycle of things made by artisans. Is it true that they are often made and never remade? Software at most companies is rather different. And how about the maintenance of software by people who did not write it? Its hidden complexity makes that a much harder proposition compared to the output of an artisan in a physical medium.

Maybe there is a place for programmers as skilled artisans, but maybe not in your typical software company today. If the artisan model has issues for such companies, maybe designing the system collaboratively in the abstract and considering code an "implementation detail" is indeed a better model. Or are there others?

This example gets trotted out a lot but I don't really understand it. Why do we assume cyclists have a conceptual mechanical understanding of a bicycle and can remember its exact appearance? If they build or repair bicycles, sure, but the majority of people don't do that. They just have learnt how to operate one by instinct.

> as long as there's an infinite number of possibilities in the distribution

how would one prove that claim? it's not obvious to me. it seems like the specific conditions under which the series of terms in the total expectation has a well defined sum is important, but I didn't fully grasp those conditions

that's an excellent treatment. I'm curious: this video picks a particular example of a distribution of the amounts and shows that the expected profit from switching is not defined. Can one somehow prove that for all possible distributions, either the expected profit from switching is 0 or it is undefined?

pyright is a command line program https://github.com/microsoft/pyright

in my experience it's a far better type checker than mypy, which tends to silently not check things without you ever realising

This sounds plausible but I'm not sure it's quite accurate, because in my experience, zooming in also affects the size of things that use px units in CSS.

yes, and in my experience UIs make this much easier than -p

but for me a bigger reason to prefer a git UI is that I can actually see the graph. I can do things like drag branches between commit nodes. I find this so much easier and safer than the cli

people talk a lot about MyPy but I must recommend Pyright as a better alternative. It's much more robust.

In my experience the main problem with MyPy is that when it's unsure about what's going on (especially due to imports it doesn't understand, sometimes because of misconfiguration) it just gives up on type checking with absolutely no warning. Pyright does not have this failure mode and handles Unknown types properly (with configurable strictness)

to my knowledge the easiest version of this manoeuvre, if you still have feature-1, is

    git checkout feature-2
    git rebase feature-1 --onto master

rather confusing. is there a better way?

edit: oh, you mentioned extra commits on master since the squash of feature-1. does that require two rebases?

    git rebase feature-1 --onto master-at-feature-1
    git rebase master

I agree with the sibling, I don't think it's _necessary_ for the contradictory aspects to have no casual link. There just has to be some kind of surprising contradiction, although maybe you tend to feel more entertained by the surprise when there is no obvious cause-and-effect.

To my mind a pardon that comes too late, or the lottery-bought racing car accident, are unfortunate but don't really hold a striking contradiction. That seems more significant than the causality.

Any business SaaS app where users like customer service representatives input data about their customers. Name, address, email, payment information and so on. Under no circumstances should these sort of input fields be autofilled.

Huh, I did wonder if there was an updated fork but I obviously didn't look hard enough. Thanks!

Clink [https://mridgers.github.io/clink/] adds this functionality and much more to good old fashioned command prompt (which I prefer to powershell because it's much faster and simpler)

I've played local multiplayer games over parsec (https://parsec.app/), it works great and is pretty much turnkey

That's absolutely right, at least in terms of the Chrome browser history. I always found this limitation astounding.

I use the History Trends Unlimited extension to work around this (https://chrome.google.com/webstore/detail/history-trends-unl...)

With 5 years of history the stats page is a bit slow to load but the search feature works great and it can export to CSV

in Verdana 4 space indentation is about 3 ordinary characters wide. I find that perfect. code indented with 2 spaces is more problematic though.

I'm confused. In every modern language I've used apart from go and terraform, only leading indentation matters and autoformatters do not apply mid-line alignment or align subsequent lines against characters in a previous line.

edit: and maybe Haskell? the layout rules are complicated

personally I don't think lining up e.g. variable or parameter names in c style declarations improves readability at all, and it pollutes git diffs.

but I do use a monospace font in the terminal because many command line applications align their output into columns using spaces.

I switched to coding in a proportional font, Verdana, a couple of years ago and never looked back.

That's the font you are reading this comment in, in fact.

Code is mostly words and words are much easier to read in a proportional font. Symbols are plenty clear enough in Verdana, all letters are distinguishable, and I don't use mid-line alignment so I don't see the point in a monospace font outside of the terminal.

This pretty much sums up my experience of poetry. Some great ideas, but an astoundingly buggy implementation that shows no sign of improving. We tried it at length in my day job and gave up in the end.

> You can do that, but you need to do it for every exception type ): I'd love it if Java had templated exceptions.

It does have them. You can write an interface with a type parameter `<E extends Throwable>` and declare functions inside it with `throws E`

Excel has no clamp function but a neat alternative is

    =MEDIAN(min, num, max)

I use hyperswitch (https://bahoom.com/hyperswitch) which emulates Windows alt-tab behaviour on MacOS. Not bulletproof but it mostly works.