Lxr

I imagine it’s extremely difficult to come up with a metric that captures everything about the case. BMI, age, tumor type, tumor shape, history of kidney disease, diabetes, where do you stop?

It’s also a trap if you forget you’re logged in and the user immediately gets notified you viewed their profile.

Neliti| Full-stack developer | Sydney, Jakarta | REMOTE

Neliti is a funded startup based in Sydney & Jakarta building software to change the way academic research is published and accessed. Our goal is to remove all paywalls from scientific work. We launched last year and have grown to over 2 million users per month and we're now hiring our first employee!

We're looking for smart people who love writing code and share our passion for open knowledge. Some cool things we're doing: designing an ultra-user-friendly publishing web app, mining data from academic papers with ML, implementing new ranking algorithms for authors and journals. Our main product runs on Django but we're pretty language/framework agnostic. We are currently focused on the Indonesian market but have plans to expand.

Compensation would be a combination of salary and equity, with a significant equity component. You'll be joining us at ground zero with plenty of freedom and opportunity for personal growth.

Contact email in my profile. More info: https://www.neliti.com/careers

The link in your profile is broken, FYI.

This looks awesome, can’t wait to try it. Last time I used pyquery though it was considerably more limited than CSS and jQuery syntax, and I reverted to XPath. Has it improved recently?

To further elaborate on 2, it solves the problem of maintaining loose version ranges in your requirements.txt file, but keeping the versions pinned when you deploy. For example if you put `foo>=2` in your requirements.txt, this is dangerous without some way of pinning e.g. `foo==2.18.2` and running your tests against that before you deploy. But you obviously don't want to manually edit requirements.txt with minor version numbers every time you update. In the past I've maintained a separate file with loose versions and then updated packages with

  pip install -r requirements-to-freeze.txt --upgrade && pip freeze -l -r requirements-to-freeze.txt > requirements.txt

Pipenv makes this much nicer.

I agree you need to catch it early. I often wonder what the most time-efficient form of mitigation is.

Forgive my ignorance but why is submitting a hash a problem? Because Troy knows which passwords have been checked? Why should I care about that? I get that it’s like submitting your password in the clear if it’s in the DB, but in that case surely you have bigger problems.

Can you clarify what problem this solves?

That's a poor argument in my opinion. Should my washing machine have no serviceable parts because I might electrocute myself?

I believe this is why companies often go downhill after the original founders leave and “professional management” takes over - founders generally take a less short-sighted view.

Randomised intervention - i.e. randomly selecting a group of people to change their behaviour and comparing to the group that doesn’t change, who ideally receives a placebo. The random part is important, any other type of selection (e.g. observing people who choose to change their own behaviour) doesn’t work, in theory at least.

I’ve also wondered, because it’s really helpful. I always prefer to read the original study and author’s own analysis, caveats etc.

Highly, highly recommend iTerm2, it’s a great piece of software and there’s really no reason to use the default terminal over it.

> I fear that the most probable hypothesis is that nobody has been sufficiently motivated to check these things out carefully as yet.

For some motivation, you might still get a check from Knuth if you find an error. https://en.wikipedia.org/wiki/Knuth_reward_check

Honestly, I believe the exact opposite is true. Why should you make your users and their batteries work harder to get what they want? Somebody comes to our website with a task in their head, and fundamentally our goal is to help them complete that task as quickly as possible. "Offload as much work to the client as possible" does not help our users complete their tasks faster. It makes life easier for dev teams by having servers that focus on APIs, reusing js code to create native apps, etc. But in the vast majority of cases it does not help the user, even if done perfectly. No, it's not faster because "no whole page reload" and yes, it does break expected browser behaviour unless you are extremely, unreasonably careful. The only exception is if you're building Google Maps or something extremely "appy" and need to have an alternative to the app store, for some reason.

My vote for law 11 is “don’t use JavaScript unless it’s absolutely necessary”. That way the browser actually behaves the way you expect.

Why doesn’t this happen with linux, in your opinion? Is it the license?

It taught me so many things as well. One that comes to mind is after they introduced the Grand Exchange, “merch clans” became popular running pump and dump schemes on random items. Unsurprisingly the only people to actually get rich were the people who bought before the next item was announced, i.e. the insiders who actually decided the items. This has many parallels in the real world!

Was it Lego Mindstorms? Man I had so much fun with that as a kid. I would totally recommend it for learning to code too.

Is there a reason the proof of work can’t do something more useful, like protein folding?

> Things that are actually health promoting, like having a full-time parent to care for the kids and primary breadwinner, eating right, exercise etc are boring and don't make VCs rich.

Fair enough, but there are also plenty of conditions for which these things don’t work adequately, and we absolutely do need to encourage research. I think this is a great move from YC.

Can you point to some specific articles by Paul Graham to explain what you mean?

Just to let you know, the fixed width formatting is very hard to read on mobile.

On average, people actually do have fewer friends than their friends have (also works for sexual partners).

https://en.m.wikipedia.org/wiki/Friendship_paradox

Is their goal to annotate everything or just the non-obvious things? Also how would a tool like this handle cases where the “correct” type is a generic base class but at runtime it only sees a certain subclass? To be pythonic, a function that accepts a tuple should usually also accept a list for example, but at runtime that may never happen.

Instagram would disagree with you. Seriously, Python is an absolute pleasure for managing large web projects and the lack of static typing has never been an issue for my company (spend the time writing tests instead!)

Related: http://www.paulgraham.com/say.html

This is exactly the problem with advertising as a business model. You necessarily have to make your product worse to make more money, and it becomes a balancing act. Subscription SaaS services on the other hand are free to make the product as clean and easy to use as possible.