HNHacker News
TopNewBestAskShowJobs

austinl

2,694 karma · joined December 7, 2012

submissionscomments
austinl··on Thanks HN: You helped save a company that now helps thousands make a living
I agree in the sense that the HN community cannot save a bad product. And sometimes it wouldn't even save a good product.

But I think the thanks here is for the community that was willing to give a new product a try, provide feedback, and help the founders build momentum towards what it ultimately became — especially when it was clear there was no traction on other sites. I think that's something to be thankful for, and shows that HN can be different from other communities in meaningful ways.

austinl··on 'Megascale' structures that humans could one day build
Wikipedia has some great content on megastructures. It's a good place to find some sci-fi book recommendations as well. I've ended up there more than once on late nights.

Here are a few good starting points:

https://en.wikipedia.org/wiki/Megastructure

https://en.wikipedia.org/wiki/Astronomical_engineering

https://en.wikipedia.org/wiki/Kardashev_scale

austinl··on Quitting a New Job
I can appreciate this — I stayed with my previous company for 5 years, which seems like an eternity to most folks I talk to, but I worked on 3 very different teams during that time period. It felt like joining a brand new company each time with new challenges and things to learn. I'd recommend exploring that route before leaving if it's open to you.
austinl··on Ask HN: Is HN slow today?
I use Hacker News as a quick benchmark for "how fast is the Internet here?" because the speed is so consistently good. Typically slow Hacker News load = slow Internet.
austinl··on Cyberpunk 2077: How 2020’s biggest video game launch turned into a shambles
Strongly agree with this. I’ve been playing on Nvidia GeForce Now (so essentially a high-end PC), and haven’t encountered any major issues. I’d say the game sets a new standard in terms of the level of detail in the environment. The first time you leave your apartment in Night City is simply stunning, probably the only comparison would be the Uncharted series (which was much more linear).

Cyberpunk was perhaps the most hyped game of the last year, so no matter what CD released, I expected a strong and polarizing reaction. I agree with their decision to offer refunds, but I think they don’t need to be as apologetic about what was released. The same issues came up years ago with publishers who targeted the PS4/Xbox One — Cyberpunk just drew a lot more attention this time around.

austinl··on No Cookie for You
Can anyone familiar with the topic explain what distinguishes essential from non-essential cookies?

GitHub gives the example of "those used by third-party analytics, tracking, and advertising services", but curious if the law defines some sort of bright line here.

austinl··on Type in the exact number of machines to proceed
I like this format in general, since it communicates the command is severe/irreversible. Heroku implements a similar confirmation when performing destructive actions. Commands require your to pass a `--confirm ${APP NAME}` flag, so the original command itself does nothing. Of course, this doesn't prevent you including those flags in makefiles, etc. I once dropped a table in a side project by accident because I took the wrong tab autocomplete suggestion in a makefile.
austinl··on On Our Abusive Relationship with Mozilla’s Firefox
I came to the comments specifically curious about this. I'm currently running Firefox with multi-container, and a couple of the about:config tweaks from https://www.privacytools.io/browsers/#about_config

Obviously it's all relative (e.g. compared to Tor), but is the general consensus that this is good enough?

austinl··on Swift Algorithms
As an iOS engineer, I'm a bit biased, but I'd still say it's currently not practical to learn it if you're not doing Apple development.

That said, I love the language, and it's a big part of what motivates me to continue being an iOS engineer (I'm not sure I'd still be at if we had kept on with Objective-C).

I would love to see server-side Swift take off, but the ecosystem is still fairly new, so any large project would require you to roll your own solutions more often that you would need to in other languages. I was also excited to see https://www.tensorflow.org/swift — Swift being used for ML. Swift is working on support for differential programming, which is a cool development in the space. More info here: https://github.com/apple/swift/blob/main/docs/Differentiable...

Perhaps eventually there will be enough applications outside of the Apple ecosystem to reach a critical mass.

austinl··on The case for comments in code
We have a lint rule that will fail if your TODO comment doesn't have a link to a ticket for that task in it. It's not perfect, but I think this has reduced TODOs in our codebase to tasks that will be more or less done in the immediate future, e.g. follow-up PRs.

Because of this, I think these TODOs are pretty helpful in code review. Reviewers often leave comments like, "while you're at it, why don't you do this too?". But with a TODO and a ticket, you can indicate that you're about to do that thing, but are breaking things up into smaller PRs to make things easier to review. Since there's a ticket associated with the TODO, it's a bit more likely to actually get done.

austinl··on Why books don't work (2019)
People also expect too much after reading a book just once. Reading a book once isn't enough to make a lasting impression, if you're able to remember anything at all — especially for a complex book like Thinking, Fast and Slow.

The books I truly love, I've read at least three times, with pages full of highlights and notes. Re-reading a great book is worth more than reading the latest trends. Seneca wrote a bit about this nearly 2000 years ago:

Be careful, however, that there is no element of discursiveness and desultoriness about this reading you refer to, this reading of many different authors and books of every description. You should be extending your stay among writers whose genius is unquestionable, deriving constant nourishment from them if you wish to gain anything from your reading that will find a lasting place in your mind.

austinl··on What I Learned Going Through YC as a 17 Year Old
This reminds of startups founded by folks that were former executives at BigCos. Of course, it varies from person to person, but many of them reek of big business formalism that's kind of off-putting. Certainly there are pros and cons that come with that experience.
austinl··on Deep learning job postings have collapsed in the past six months
I've heard this happen in a lot of places — companies want to be "data-driven", but then leadership simply ignores the data. I think being data-driven is something that is built into company culture, or otherwise it's too easy to just ignore the results and ship.

The place I currently work is data-driven (perhaps to a fault). Every change is wrapped behind an experiment and analyzed. Engineers play a major role in this process (responsible for analysis of simple experiments), whereas the data org owns more thorough, long-term analysis. This means there are a significant number of people invested in making numbers go up. It also means we're very good at finding local maxima, but struggle greatly shipping larger changes that land somewhere else on the graph.

Some of the best advice I've heard related to this is for leadership to be honest about the "why". Sometimes we just want to ship a redesign to eventually find a new maximum, even through we know it will hurt metrics for a while.

austinl··on Wind and solar are cheaper than thought, admits UK government
The book The Grid by Gretchen Bakke is a great primer on why this is such an important problem. Energy is strange, in the sense that everything produced needs to be consumed more or less immediately — it's not like other resources that can be easily stockpiled (even with batteries). Our infrastructure, at least the antiquated American power grid infrastructure, is built with the assumption that production and consumption is balanced.

Maintaining the energy grid is a careful balancing act between preventing surges and blackouts. There are a lot of grid-related engineering problems that come from increasing decentralization (e.g. wind, solar, hydro) because, as you mentioned, it increases unpredictability in supply.

austinl··on Efficiency is dangerous and slowing down makes life better
as by a man of leisure is my favorite phrase in the passage. I think it's the key to the whole thing. It's implies what kind of mindset you should try to adopt: relax, enjoy what you're doing and appreciate the moment (if you're able to).

Aurelius' adopted father was a consul three times, which is certainly not a stress-free job! But he apparently was able to keep cool by the way he approached his work.

https://en.wikipedia.org/wiki/Roman_consul

austinl··on Efficiency is dangerous and slowing down makes life better
One of my favorite quotes on this topic comes from Aurelius' description of his adopted father in Meditations.

[Y]ou would never say of him that he "broke out a sweat": but everything was allotted its own time and thought, as by a man of leisure - his way was unhurried, organized, vigorous, consistent in all.

I feel like I spend a lot of time rushing from one thing to the next, constantly questioning whether I'm spending time wisely. And then I end up accomplishing less because I lack focus in one area. I've instead been trying to relax, slow down, and take tasks one at a time until completion. I'd also recommend Cal Newport's book, Deep Work, on this.

austinl··on New academic journal only publishes 'unsurprising' research rejected by others
I'd also love to see a journal that agreed to publish work before any results are known. Researchers would submit hypotheses and methodology for review, and the journal would publish the results after the experiments were conducted, regardless of their outcome.

It would still incentivize interesting hypotheses, but wouldn't lead to results-biased publications.

austinl··on The dual PhD problem of today’s startups
Another reason we see duplicate startups within a few domains is that there are many VCs, and each wants to have a horse in the race.

Perhaps they missed on one fintech company, and now it's grown enough to make the market. So they lead an investment in a competitor, or a company in a similar space. This significantly de-risks the investment vs. allocating money towards something entirely new.

austinl··on That coworker who never stops refactoring
This is me as well. It really hit home when I was talking with a friend about not joining a company that's just trying to get stuff out the door quickly. She said "you're an artisan, you just want to practice your art! And that's hard to do at some places."

I do really like that analogy, of artisanship or craftsmanship. I understand the business need to move quickly, but over time I've found myself more intrinsically motivated by building what I think of as well-designed systems. That, to me, is art. That's what I get excited about. I'd also like to think that some of the extra time I spend upfront is saved down the road when things don't break. I think great engineering cultures respect that balance.

austinl··on The Clean Network – United States Department of State
This section on Wikipedia covers the history of the concerns. I wouldn't say there's a definitive smoking gun that's public knowledge. Instead, there are several instances of vulnerabilities discovered in Huawei systems that may have been taken advantage of (by the PRC or other parties). My guess is that there's more to a few of these stories that only the intelligence community knows about.

One other point of concern from Western governments is that Chinese law compels Huawei to cooperate with Chinese intelligence services, which presents a vulnerability whether that law has been taken advantage of or not. This is much different than say, Apple, which has refused (or it has been simply impossible due to E2E) for them to cooperate.

https://en.wikipedia.org/wiki/Criticism_of_Huawei#Espionage_...

austinl··on Debt collector profits soar with people flush and easy to find
Great advice. Honestly, things like this are half the reason I don't pick up the phone any more if I don't recognize the number. I used to enable the feature that sent all unrecognized numbers to voicemail, but found this was too aggressive for occasional calls (delivery drivers, etc). Probably 4 out of 5 of the calls I used to get were either scams or robocall telemarketing.
austinl··on Why Should You Learn Vim in 2020
People who are extremely comfortable with vim + tmux look like they're doing straight up wizardry (probably the closest to what people imagine programmers do in movies). I remember seeing one of these folks during my first engineering internship and I was sold. Now I can't imagine doing most things without vim + tmux + the_silver_searcher + fzf (vim plugins).

The biggest downside for me is I don't think autocomplete is on par with other IDEs (even with YouCompleteMe, which is great). To be fair, this would be a massive challenge, considering how generalized vim is vs. an IDE like Xcode — which is purpose-built for writing Swift/ObjC. I feel like half of the time in Xcode I'm just hitting tab, and the code writes itself. I want that same experience in vim.

austinl··on A Close Call: How a Near Failure Propelled Me to Succeed [pdf]
I'm sure it depends on the challenge. Most folks that I'd consider geniuses just relentlessly pursue the things they're interested in. Depending on the topic, this gives them enough ability to get by the basic concepts tested in school. And then later in life, a good career if they're lucky (definitely lucked out being interested in computers).

I think these people are still working hard in a sense, they just have a low tolerance for work that doesn't interest them.

austinl··on A Close Call: How a Near Failure Propelled Me to Succeed [pdf]
This is something that afflicts a lot of younger, talented students. I was able to breeze through K-8 without studying, then attended a challenging high school where I went from a straight A student to a B/C student. It took me at least two years to realize that I could no longer succeed by simply showing up. Fortunately, this was great preparation for college. I saw many straight-A high school devastated after their first round of college engineering exams.

I think this is an unfortunate consequence of our education system at-scale. It's difficult for advanced students to be properly challenged until they reach a certain reckoning point, which for Terence was impressively during his grad school exams.

austinl··on The Case Against OOP Is Wildly Overstated
I like the theory behind app architecture, but I haven't seen a complex design pattern work well in practice (unless the team is small, and everyone has a solid understanding of how it works).

I worked on a mobile app once that really tried to force one of the fancier design patterns — it required something like six files to handle one screen (because, of course, it's best practice to separate your view, view model, controller, etc.) However, a good part of the org didn't understand the nuances of what code should go where, and found it absurdly tedious to manage all of these files to build one minor screen. Velocity plummeted.

I think a good design pattern allows you to scale as complexity increases. Let folks start simple, then provide a clear path to a consistent design pattern that handles more complexity as it's needed.

austinl··on Australia is doubling fees for arts degrees
> "Rather than saying 'the government provides subsidies / loans' -- it's preferable to think of this as the population does this, with the government merely administrating."

I think this is true of all public spending in some sense. Governments don't have money. Rather, the population does, and via taxation, the government is able to administer it how it chooses, for better or worse. Hopefully for the better, of course.

austinl··on To Get More Replies, Say Less (2017)
I should add that the key to good writing (in most domains) is to be concise. Editing is primarily about cutting what's unnecessary, but it's often treated like the part where some words are swapped out with fancier ones (e.g. "use" to "utilize").

I'm a big fan of PG's essay, Write Like You Talk, on this topic.

http://www.paulgraham.com/talk.html

austinl··on To Get More Replies, Say Less (2017)
The most effective recruiting emails I receive (in terms of how likely I am to reply) are:

* Plain text

* Directly addressed to me from name@company.com

* 3-4 sentences max, slightly personalized

* Optionally have a link to the company career page/job listing (better if the raw link is pasted instead of embedded in text, to save me the time hover-checking it).

I'm more likely to reply because it feels like there's a real person on the other end that "hand-wrote" this email. Not some amorphous system that crawled my LinkedIn and sent me an automated message. Kudos to recruiters that are doing this — doing the thing that doesn't scale to get an edge.

austinl··on The Navy Says It's Not Recruiting on Twitch. This Handbook Shows It Is
Yeah, this 100% semantics, with a misleading headline.

A better story would have read "How the U.S. Navy leverages Twitch and esports as a marketing channel". Instead, it takes a quote from one of the players, Crosswhite, then labels Crosswhite as representing the whole Navy with the headline: "The Navy Says It's Not Recruiting on Twitch".

What's worse is it tries to sell this controversy, as if the Navy is denying its involvement and Vice unearthed some secret documents, which is just not true.

Here's the quote from the article:

Machinist's mate First Class Andrew “Saltysn1pe” Crosswhite answered a question from the audience about why the Navy was streaming video games. “We’re here to show that we play video games, literally,” Crosswhite said. “We’re not here to recruit. That is not the point of this.” According to Navy documents viewed by Motherboard, that’s not true.

Crosswhite obviously meant that they are not actually recruiting on the channel, but the purpose is undoubtedly marketing for the Navy. I don't think any of the players on the team are that naive — they're supported by the Navy because of the marketing opportunity.

austinl··on The Guns of Bitcoin (2017)
You mentioned expensive art, which is a pretty fascinating field because the value of individual pieces can be fairly subjective and difficult to appraise.

This has massive tax implications in the United States. Imagine donating a piece of art, which you claim is worth $1 million. But the question of who is allowed to make that claim is quite complex, and often goes unnoticed unless the IRS chooses to investigate.

A couple of articles on this:

Inflated art appraisals cost U.S. government untold millions https://www.latimes.com/local/la-me-irs2mar02-story.html

The secretive panel of art experts that tells the IRS how much art is worth https://www.washingtonpost.com/lifestyle/magazine/the-secret...

← PreviousPage 4 of 9Next →