leobakerhytch

Regrettably, we humans just don’t have a good sense of scale.

See Slaughterbots [0] for a dramatization of this scenario.

[0] https://youtu.be/O-2tpwW0kmU

And a fourth time on a per-article basis if you, as an individual, non-academic taxpayer, would like you read the very research your taxes have funded.

When you say historic buildings, I suppose (hope) you don’t mean listed ones, since they’d be breaking the law? Regardless, it’s a crying shame what people have ripped out of even very ordinary Victorian and Edwardian houses. Decorative mouldings, cast iron fireplaces, original doors, geometric tiling.

You might like to try Write yourself a Git [0] (discussed here previously [1]). YMMV, but I find the best way to learn something deeply is to get hands on. Less of a chance of convincing yourself you understand something that you really don’t.

For less of a time commitment, Git from the inside out [2] is a really nice explanation of the internals, from initializing a repo and the files that creates in the .git directory, all the way to pulling from and pushing to remotes.

[0] https://wyag.thb.lt/

[1] https://news.ycombinator.com/item?id=19386141

[2] https://maryrosecook.com/blog/post/git-from-the-inside-out

Even though the full-size filter would surely work with a phone camera, if you could scale it down to a form factor that clips onto the phone, I can see this selling like hot cakes. Might be one of the rare cases where influencer marketing is really effective too. Is it feasible to grind the glass at a physical scale to make that work?

All in all, very cool product, and congrats on making a living from it!

Clickbait title, but I wholeheartedly agree with the article’s main assertion, that startups need product engineers who think about more than just code. Engineers who can make decisions for the benefit of the customer, and ultimately the business.

This is all the more reason to delegate responsibilities, rather than tasks (as argued in Little Tasks, Little Trust [0]), so programmers actually get the necessary experience with UX, design, interacting with customers, etc., and exposure to the consequences of what they build, in order to grow into full-fledged product engineers.

I fear though that such roles, and consequently such engineers, are few and far between.

[0] https://news.ycombinator.com/item?id=25346460

From the Twitter thread linked by yoquan:

Explosives were used to deploy the parachute, so care had to be taken that there were no late detonations. [0]

Precisely what you guessed it to guessed it to be.

[0] https://mobile.twitter.com/girlandkat/status/133551478006118...

Absolutely. I think that’s why it’s so vital to combine both qualitative and quantitative data.

I doubt there are too many people in tech making this mistake, but without numbers there’s a good chance you fall prey to people’s inaccurate perceptions or explanations for their own behavior.

On the other hand, without those first-hand accounts, it’s all too easy to tell a mistaken story about your numbers. In particular, your users’ sentiment towards their time spent in your app is a guess, unless they tell you.

I think timescale is another crucial dimension to evaluating time-in-app, and whether it’s a positive indicator for your business.

For instance, driving up engagement has doubtless been good for Facebook’s financials in the short- to medium-term. Arguably though, that relentless focus has led to the present political climate where they’re fighting off regulation. Whether that’s an existential threat is yet to be seen (one can only hope), but it certainly casts the metric in a different light.

As you say, it comes down to the fact that there is no silver bullet metric; there’s no substitute for thinking about and dissecting the data you collect.

That’s what qualitative data is for. At the very least, talking to customers and reading their feedback. Preferably followed by organizing that data somehow, and using it to better understand the quantitative data you have.

It’s fair comment, though, that undifferentiated ‘engagement’ is rarely a good metric.

Rather than building dedicated integrations (which can always come later), a relatively quick win that’s tool-agnostic would be to implement webhooks.

I’ve used Typeform’s webhook feature a number of times to good effect, whereby you receive one POST request to whatever endpoint you specify upon each form submission. This may have changed, but when I last used it that was the only way of getting your data out of their system, which was admittedly frustrating. A lot better than nothing though.

Used it for all kinds of things: quizzes hooked up to a loyalty program, to award the user points for all correct answers; satisfaction surveys after ordering a product; writing and photography competitions (again hooked up to the loyalty program); an in-depth annual user survey; etc.

Except I think there’s convincing argument to make that engagement will go down over time, if the algorithm makes no attempt to prioritize or suggest novel content.

The rare occasions I discover a new channel, it’s almost always from some source other than the algorithm: a referral from a friend, this site, another YouTuber, etc. My viewership of the same repetitive roster of videos absolutely tails off until I find something new from elsewhere.

For example, in months of being subscribed to my mechanics [0] (who does incredibly engrossing and relaxing restorations of mechanical stuff), not once was I suggested a video from Baumgartner Restoration [1], an art conservator who produces videos with a similar attention to detail and high production value.

Thematically this should be an easy recommendation for YouTube to make, but evidently the content is just different enough that it scores as a false-negative. After finding the latter channel independently, my viewing time absolutely rose for a while.

In theory, YouTube ought to be able to detect and learn from this signal of non-algorithmic discovery of new content. Yet, here we are.

[0] https://www.youtube.com/c/mymechanics

[1] https://www.youtube.com/c/BaumgartnerRestoration

To add to that list: abstract syntax trees.

It would take a pretty monumental effort to build, but I think a huge step forward could be made with a VCS that has knowledge of code structure, and the relationships between source files.

Large scale refactoring is such a fragile affair with git. Conceptually though, it should be possible to cleanly and concisely represent such operations as renaming functions, reordering arguments, widening type parameters, etc.

Stability of the programming language in question is of course highly desirable, if you were to build such a system. The pace of change of Go would be much more conducive to success than that of JavaScript (or TypeScript).

I realized in the last year or two that I was never really bad with names, I just didn’t give myself the chance to be good with them. For someone I’ve just met, it takes a little effort to use their name at least once or twice soon after I learn it, and it’s much more likely to stick.

I’m also much stricter with going through the slight awkwardness if I didn’t catch the name first time (even asking to spell it out to be 100% sure). Then afterwards, I’ll write down the name in Notes with something memorable about the person, and where I met them. 99% of the time, I never look at this again. Just like that, name remembered.

Then, I’ll manage not to use a person’s name next time I see them, even when I know it. Oops.

I fully agree that the default porcelain has poor UX for managing the staging area, but doesn’t a porcelain without it encourage overly large commits?

All sorts of workflows become substantially more difficult (if not impossible) with kitchen sink commits. Undoing a single-line change, for instance. I dislike large pull requests, let alone commits that introduce a half dozen different changes.

Sure, if you’re really disciplined you can produce small commits without staging, by committing as you go along, but you’re breaking flow every time you commit in this style.

I say all this because I would love a CLI that has half-way sane handling of (e.g.) restoring staged, deleted files; doesn’t conflate restoring files with switching branches; has consistency of flags between commands (such as commit vs stash message); etc., etc. I should try gitless ;)

Just to bring to levity to this discussion, your quote very much brings to mind the hilarious and ever-relevant Four Yorkshiremen[0] sketch from Monty Python.

Right. I had to get up in the morning at ten o'clock at night, half an hour before I went to bed, drink a cup of sulphuric acid, work twenty-nine hours a day down mill, and pay mill owner for permission to come to work, and when we got home, our Dad and our mother would kill us, and dance about on our graves singing 'Hallelujah.'

[0] https://youtu.be/ue7wM0QC5LE

I can assure you 130km/h is treated as merely advisory. That is, you are advised to go at least 130.

I highly recommend the browser extension Session Buddy [0], if you, like me, regularly find you’ve opened tens to hundreds of tabs.

It’s quite cathartic to save out my tabs into topic-based sessions, close everything, then reopen just one session on whatever I want to focus on after an exploratory phase.

[0] https://sessionbuddy.com/

You say this was your first time outside India. May I ask, what was your path to setting up a company in the US and moving there? Was this via an L1 intra-company transfer, an E2 investor visa, or something else?

I couldn’t get this to work, but setting back my clock to October 20th did the trick (along with deleting the modal’s element via dev tools as jonahx mentions).

GMBN has a couple of great videos on buying a mountain bike [0][1], but I can really recommend a direct to consumer brand such as Canyon or YT — especially if you’re not sure whether you’ll commit to the sport. You get a lot more bike for your money compared to the store brands, and you’ll have an easier time getting into the sport the better your first bike is.

The major choice to make, irrespective of brand, is what type of bike — which is dictated by the terrain you have available and would like to ride. There are subcategories, but a useful rough split would be: a cross-country (XC) bike for good pedaling performance both uphill and on the flat, but not so capable going downhill and in the rough stuff; an enduro if you want to ride rougher downhill terrain (roots, rock gardens, etc.), but still have a bike that pedals well enough up technical climbs and fire roads; and a downhill bike if you want to ride the roughest, most technical terrain, with the steepest descents, and you have access to lift-served mountains (not recommended to start with ;).

FWIW, I’ve had a such a great time riding the Canyon Torque this summer. It’s a long-travel enduro, really capable on the descents, and perfectly fine to get back to the top.

As for motivation, I would echo the sibling comment. Time in nature, good exercise, a ton of techniques to learn and master, a really sociable scene, such a rush riding tough and steep stuff. You won’t regret it.

[0] https://youtu.be/EX1Kb6TEyn8 [1]: https://youtu.be/Y_xiq2mczX4A

A lesson from history shows this not to be the case: hand-wrought nails were such a precious and militarily significant commodity in Roman England, that when the legion left, they painstakingly concealed their hoard to prevent the locals gaining access to them.

Fast forward to the 1860s, and nail production is automated to the extent that wire nails can be produced by the tens of millions with almost no human intervention.

The worker who forges nails by hand and the inventor who automates their production are undoubtedly orders of magnitude apart in productivity.

Turning off viewing history and search history puts the recommendations much more under your own control — they’re then (AFAICT) only based on likes and subscriptions, and you still have the option to reject recommendations, as sbierwagen says.

I’m not sure I would call it quitting tech, but after studying computer science at university and dabbling in some start-ups that went nowhere, I was a property developer for a while.

Learnt an incredible amount from it, not least how punishing physical labour can be (demoliton, being the first and probably least-skilled job on a project, made up most of the hands-on work I did — the rest was left to professionals).

I’m not a founder (yet), but now working at an early stage startup, I’m finding so much that carries over to the business of tech. Assessing the market potential of properties available for auction, the poker game of the auction itself, making profit projections on the back of extremely shaky estimates of the time and resources needed to complete a project, the difficulty of scheduling myriad interdependent tasks, to be carried out by a small army of subcontractors.

On that last point, an especially memorable lesson was learned after uncovering hidden structural instabilities that needed immediate repair (huge cracks in the back wall of a 100 year old row house, caused by an undersized window lintel, covered by a suspiciously low false ceiling — I’m sure there’s a lesson in there about hiding mistakes). Dealing with the cascade of jobs that had to be rebooked was extremely instructive in the need for slack in a schedule.

Clear specification is another thing I now value highly. I taught myself to use architectural CAD software in frustration at a number costly mistakes, which traced back to misinterpretation of imprecise verbal instructions or freehand sketches. Especially when people are proud of their abilities, they’re inclined to read between the lines of the directions they’re given, which won’t necessarily line up with your vision — especially if you haven’t actually communicated that vision.

I could go on, but in short: absolutely no regrets. And something I’ll almost certainly do again, when I hopefully have more capital to work with, and can hire a project manager to execute my vision!

Nice video on the design and manufacturing of aluminum drinks cans here: https://youtu.be/hUhisi2FBuw

The whole thing is pretty ingenious, but the way the internal pressure helps vent the can while the tab is acting as a second class lever, then as soon as it’s vented becomes a first class lever to push the scored section of the top into the can — that’s great engineering.

I was hoping someone would mention this. Have you seen the product in person? I’d be very curious whether it lives up to how good it looks on video.

For those unfamiliar, the panel simulates Rayleigh scattering to give the appearance of a blue sky, with a simulated sun and close to incandescent color rendering. Too bad the price tag still seems to be in the tens of thousands though.

Thanks for this suggestion — will try this as a way to escape the ‘filter bubble’. To clarify: do you reset completely, retaining absolutely no likes, subscriptions, etc., or do you find it sufficient to clear your watched videos?

Link to https://data.gov.uk on that page in fact points to ’file:///C:/Users/mcallister_mj2/AppData/Local/Microsoft/Windows/Temporary%20Internet%20Files/Content.Outlook/5P3C0P9L/data.gov.uk’

Oh dear.

If you look at the corners of the plot, you can see the short sides of the octagon that the completed building will eventually become.

Overlaid on the financial district of New York, for comparison: http://mapfrappe.com/?show=39101

Is that… Jesus, on a pogo stick?

http://i.imgur.com/f8EMwTP.png