[godelski]

While I generally agree with the conclusion of that, I think it might be a bit too naive.

The quantity group has a trivial way to "hack" the metric. I can just sit there snapping photos of everything. I could just set up a camera to automatically snap photos all day and night. To be honest, if I'm not doing this at a stationary wall there's probably a good chance I get a good photo since even a tiny probability can be an expected result given enough samples.

But I think the real magic ingredient comes from the explanation

  > The group never worried about the quality of their work, so they spent time experimenting with lighting, composition, and such.

The way I read this is "The quantity group felt assured in their grade, so used the time to be creative and without pressure." But I think if you modified the experiment so that you'd grade students on a curve and in proportion to the number of photos they took then the results might differ. Their grade might not feel secure as it could take just one person to communicate that they're just snapping photos all day as fast as they can. In this setting I think you'd have even less ability to explore and experiment than the quality group.

I do think the message is right though and I think this is the right strategy in any creative or primarily mental endeavor (including coding). The more the process depends on creativity the more time needs to be allocated to this type of exploration and freedom. I think in jobs like research that this should be the basis for how they are structured and effectively you should mostly remove evaluation metrics and embrace the fundamentally ad hoc nature. In things like coding I think you need more of a mix and the right mix depends highly on the actual objectives. But I wanted to make the above distinction because I think it is important if we're trying to figure out what those objectives are.

[d-us-vb]

This is an interesting take. It maps decently to the "first to market" mentality of a lot of programming these days. For sure, we'd get better programmers if they were judged by the quantity of code they produced in their personal projects, but if "quantity learners" are simply forced to churn out bad code without any reflection or the time to experiment, then I would agree that it seems pretty naive to think they'll ever improve... except at a handful of patterns that make bad code sort of work.

[godelski]

Also think about software recently. Is it actually better? IME I face more bugs than ever. Trivial ones to that are clearly being deprioritized but relatively easily solvable.[0]

Do an experiment for me. Write down every bug you face today. Even small. I think you'll be surprised at how many there are and even more at how many are likely simple to solve.

I know it's not just me as so many around me are getting increasingly frustrated with their devices. It's not the big things, it's a thousand paper cuts. But if you just you look at a paper cut in isolation, it isn't meaningful. That's the problem and why they get ignored. But they get created because we care more about speed than direction. I'd argue that's a good way to end up over a cliff

[0] https://news.ycombinator.com/item?id=45363210

[pixl97]

>I face more bugs than ever

I mean, lets leave phones out of this for a moment and look at PCs...

When was the last time your PC operating system crashed?

When was the last time your applications you use on your PC crashed?

When was the last time you could not find an application for your PC that did what you needed to accomplish?

The early days of software absolutly sucked. Crashes, data loss, and limitations were the name of the game. The big things like data corruption were constantly problems. You didn't notice the small problems because you were fighting the big ones. The problem spaces software was solving for people was also relatively small. Now it's easy to find small bugs everywhere because software ate the world. It's harder to name something software has not expanded into than what it has, and yet we are still not near the boundaries of exploration of what software can do. When a system has not discovered its boundaries speed will almost always win over direction.

[rkomorn]

Yeah. The BSoD wasn't dreaded because it was an infrequent surprise visitor.

[godelski]

  > When was the last time your PC operating system crashed?

I daily drive both Linux and Mac. Last year I had a Windows laptop from work.

For Linux? I'm not sure, but definitely longer than 6 mo. Last crash I remember was my fault and quite awhile ago.

For OSX? Last week. The most common one of when I close my laptop, go "ops, need to do X real quick", and when I open it up the screen doesn't come up. Not sure if a backlight issue but I don't think it is. It doesn't seem to be logging in and I can't get keys like volume to respond where I'd have feedback. Flashlight trick doesn't work. If I wait the laptop reboots itself and half the time I do get a crash report (I suspect this happens more than I think too since occasionally I'll come to my laptop and it was rebooted. Like back from lunch or a quick break, not overnight. Less frequently I find crash reports) Happened since day 1 and even on the last MacBook I had. I get this error at least twice a month. Also can happen when disconnecting my monitor.

Windows? Jesus Christ, how do people live like that? Arch was more stable a decade ago.

  > When was the last time your applications you use on your PC crashed?

On Linux, two weeks ago I had a crash while playing cyberpunk (the most optimized game there ever was... except maybe StarField). Last week Silksong had a soft error where the joystick stopped responding when my wireless controller issued the low battery warning. Outside that, I can't think of anything other than when I accidentally run a sim calling too many resources, but that's not common and I'm not sure it counts.

On Mac, at least every two weeks. I feel like 6mo ago it was more like once a month though. I've been interviewing lately and Teams is definitely a bigger problem than Zoom. I think Firefox has crashed twice in the last 6 months? I'm also a tab hoarder. But mail crashed way more early on so I switched back to Thunderbird and while it doesn't crash I'm sure there is a small memory leak. I'll restart like once every 2 weeks because it'll start pushing a gig of ram. I know, I'm picky, but my email client shouldn't ever use a gig of ram. And I was writing my PhD thesis a few months back and preview would occasionally crash. Zathura had no issues. Slack definitely more frequently than FF.

Windows? Just about every day. I was able to reduce errors once the IT guy informed me that Windows Hello being used for the login was why Outlook was constantly crashing. Switching to a typed password did a lot. But after that it still reminded me of the days where I was learning Linux and distro hopping.

  > When was the last time you could not find an application for your PC that did what you needed to accomplish?

Daily? Okay, but probably weekly?

Less of a problem on Linux. Often it's small things so I can write a quick shell script. I can find most things I want on the AUR but often I'll build from source.

OSX? That depends. Do I have to pay? If so, IDK. I'm not willing to pay a subscription fee for what's a glorified shell script.

Windows? Work laptop so wasn't really looking.

I generally agree, things are getting better but the problem wasn't that big 10-20 years ago ime. I can only think of 2 instances where I had data losses and both were on Linux laptops while distro hopping over 15 years ago. In both cases I was able to recover too. Yeah maybe the issue was bigger in the pre Windows 95 days but that's way back and hardware has also made significant strides. Don't give software the credit for better hardware.

But I think you're missing an important question:

  How often do programs have unintended or unexpected behavior?

That is not a crash but is a bug. The calendar issue I mentioned above is not unique to my phone. Later on I also mention how I had essentially no means to merge contacts despite identical names, nicknames, phone numbers, and birthdays (differing only on email and notes). *WHO THE FUCK THINKS "FIND CONTRACTS" IS A BETTER SOLUTION THAN CLICKING TWO ENTIRES AND THEN CLICK "MERGE"? And guess what, it couldn't find the duplicates. This is a trivial database problem! I was able to eventually find a way to merge after some googling. But I discovered the duplication because my gf had 3 entries on my calendar for her birthday. When I merged, she ended up with 4! Again, that is a trivial database problem.

This category of problem is greatly increasing in frequency. I know you might think it's a comparison bias, and I think that's a reasonable guess, but it isn't. I've definitely "infected" my friends and family so they're more "sensitive" to bugs but years after that they agree that their devices are becoming harder to use for the same tasks they did before. Either through little bugs like this piling up or having some new features being pushed on them that they don't want and disrupts their workflow. That last one is really common.

So I don't want to dismiss you, because I don't think you're entirely wrong. But also I think you're being too quick to dismiss me. Your justification isn't a complete explanation of my experience. Nor does it account for how programs and getting slower and heavier. I mean God damn, how many seconds does it take for Microsoft Word to open (cached? Not cached?) and how many gigabytes of ram does it use? Those shouldn't even be the units of measurement we should be discussing! Both are at least a magnitude too large. I'm absolutely certain programs are bloating. And I think it should be unsurprising that with bloated comes more opportunities for errors.

This is also the type of problem I expect average people to not notice as it happens more slowly and if you don't understand computers it's easy to believe it's just the way it is. But we're on HN, we do. We know better. We also know how the sausage is made and we've experienced the increasing pressure for speed and seen the change in how programs and programmers are evaluated. I do not think you can discount these effects.

It shouldn't be a surprise that moving faster correlates strongly with making more mistakes. You don't deny the optimization for speed, but do you really think you can accelerate for free? There's a balance and I'm confident we've sped through that equilibrium.