rablackburn

You mean if there is a positive causation ;)

But yeah, I think the difficulty of sociology makes it tough to reason about these issues in a comprehensive, logical way. And it's so hard to experimentally prove _anything_ when proving the counterfactual is "well, we'd need to split reality into two and wait 50 years".

> The results are not achieved by different standards for graduation, but by different standards for admission.

This is the key callout I think. It allows you to address representation problems while sidestepping opposing arguments about graduate quality.

Do it in rounds, unfilled spots open up to anyone.

If your child really wants to be a doctor but can’t get a spot, well, I’m sure that increased demand for medical spots means you’re willing to pay more, and the University can use that to expand their medical program. Or you can send them to another university.

This seems like a reasonable approach to improving systemic inequalities without causing systemic harm.

Yeah, the random straw-man divergence in the middle of the article was a bit off putting.

I like framing climate change in a further reaching time context, but if anything I think the author has reached too far. The point I took away was something like “it’s all going to happen anyway, and we’ll all be dead, so why bother”.

I think the sweet spot of forward-thinking is somewhere between next financial year and the next geological epoch; I propose around 5 generations (150 years?).

I think if we spent more time imagining and talking about the world our grandchildren’s grandchildren will inherit, it’s close enough to feel a personal connection, and far enough away we can meaningfully direct our destiny.

Not the OP, but I'd just wrap it up like this:

  <div aria-role="img" aria-label="{your alt-text here}">
    <pre>...</pre>
  </div>

A simple approach is best. I don't have a screenreader on my current machine (as far as I know) so I can't verify actual behaviour right now, but according to MDN:

> ...most screenreaders will consider the element with role="img" set on it to be to be like a black box, and not access the individual elements inside it.

(https://developer.mozilla.org/en-US/docs/Web/Accessibility/A...)

That sounds great, I don't suppose you have anything of that project that's ready to share (even just to look at, not use)?

It’s like conflating palm reading with astrology.

Irrespective of whether you believe the methods have any value, we should at least make sure we’re talking about the same thing.

> The true cost of mining with all the regulation and corruption is making imports cheaper for many plants.

Where has so much less regulation around coal mining that local production is being undercut?

It sure is useful when hacking together “always-square”, responsive elements.

But otherwise you’re right, it doesn’t really come up.

Slightly relevant story time:

On one of the better teams I’ve worked with the water-cooler activity was trying to come up with an algorithm to describe the behaviour of the Coca Cola machine in the kitchen.

It was one of those clear plexiglass front machines where after punching in the coordinates of the item you wanted, a mechanical arm would move to the coordinates, take a drink, and place it in the delivery bay at the bottom.

While it would always get the drink you wanted, it would rarely go to the exact coordinates you specified. Ie, if there were three columns filled with coke, and you punched in the coordinates for the one on the far right, the arm may take a can from the center, or left column.

We would often wager a can of coke (“if I can’t find anything wrong in your PR, I’ll buy you a coke”), so we were perhaps drinking more soft drink than was medically advisable, but in our defence:

a) the machine was really cheap (AUD$1 or $1.5)

b) it was an excellent 10min break game

Eventually we thought they had it figured out we would gather and make our predictions, but occasionally there would be an upset that would throw a wrench into the model.

We got to a point where we just couldn’t reconcile the machine behaviour with any kind of coherent set of rules, then one morning we saw the delivery guy stocking it.

After chatting with him we learned our algorithm was more or less correct, but the internal state of the machine was prone to getting out of sync with the actual stock levels, so it would make the “wrong” choice near the end of the refill cycle. Then he gave the engineer talking to him a free energy drink (source of stock problems right there haha)

While we were no Knuths, I love that these kinds of games are so universal among engineers/devs. In fact, if I can get someone to tell me a similar story of theirs in an interview (for a technical role) I’m much more likely to consider them for the position. Curiosity is a powerful trait for a developer.

…and the (simple) algorithm for the machine is: Take from the column with the most stock, if there are multiple columns with the same stock level, take from the column furthest to the left.

> RSS is increasing popularly again.

Is this true?

Every time RSS comes up we have this same discussion

A: “RSS is dead”

B: “well, I’m still using it and it’s great. Clearly it’s not dead!”

But A and B are having two different conversations.

For broad public consumption the only success RSS had was podcasts, and Spotify is actively trying to capture that open ecosystem and make it their cash cow.

Kids aren’t sitting around sharing the RSS feeds they like, it’s YouTube, TikTok, Discord; all proprietary walled gardens of content.

So yes, the technology isn’t dead, and the world is big enough there’s even a self-sustaining (probably) community of RSS users, but it still hasn’t quite lived up to it’s promise.

Streambus is exciting because the cost of hosting content keeps dropping, maybe we can reproduce the podcast model with video now. But as always the network effect and discoverability are big hurdles to overcome.

Podcasts succeeded not just because of RSS, they succeeded because Apple supported them in the iTunes Store (which is _still_ the most important place to get reviewed as a podcast) solving the discoverability problem.

My understanding of this is that it’s done deliberately to prevent listeners from recording the full track. Made more sense when the radio was _the_ way to listen to new music. Less so now.

There are places in the world where the DJ will literally come on in the middle of the song and play their station ident for the same reason.

I don’t know if I agree with this, but I think it depends what you mean when you say “programming”.

I use programming as one of the examples of intrinsically human work.

You have a (general computing) machine that can do anything(), it only has meaning and utility from humans thinking about problems they have and how to solve them.

The hard part of programming is analysing what problems you have, and what you want to do about them.

This is why one of the distinguishing features between junior and senior programmers is that seniors tend to think about the (human) problem they’re solving a lot more.

Tools like copilot help with the physical interfacing between programmer and machine, but they don’t eliminate programming.

But a recurring worry people are expressing in this thread is: some people think they _do_, and how insidious that idea is.

The tool is great, the managers deciding that time spent thinking about its output, or developing their staff’s skills, is a waste because “just use what the AI says”, are the danger.

As long as people and computers exist, there will be programming.

not actually “anything”

I wish I worked at a place that thought that much about their efficiency, at least it kind of makes sense.

Instead it seems like no one thinks about the business, but about the social politics of the organisation.

I think that’s probably more aptly described as serfdom than as a welfare state

Huh, why does the device “have to be left switched on” for then? If it’s non-volatile it doesn’t matter right?

Or is the point that the counter only counts time when the device is on? Kinda of like anacron vs cron

> …they can work or go to Polytech

Sincere question: can they work? Does anywhere hire high-school aged kids for work that isn’t non-trivial/grunt work?

If the goal is to not waste their time I can’t imagine working a casual job is more stimulating than being in a classroom. Maybe a year working casually would inspire them back to school the next year?

Interesting, I was wondering why I was never really seeing green PCBs anymore except for “boring” pci cards like USB hubs, or wireless cards. Fashion makes sense.

Kind of makes me nostalgic and want to build a high-end PC with only “retro” green PCB components

I rarely have to think about stuff at this level, but am I interpreting how this works right?

When a remote login fails on these devices, it kills its ssh server process, waits a certain amount of (time * how many failed attempts)ms, then starts it back up?

But if you truly hammer at one of these devices you could force it to reboot, which makes it forget the counter, so an attackers gets an unlimited number of the “cheapest” guesses (though it’d still slow it down a bit).

The solution is the machine physically writes the memory from RAM into onboard FLASH mem, and is reading/writing the count variable to a file instead of just using a variable.

FLASH memory is the type that keeps state as long as it has a trickle of power, so as long as the device is plugged in, this is essentially “long term” storage (but not truly like a hdd would be).

A reboot will wipe the systems RAM (because you don’t want any bad state from behaviour lurking around), but not the FLASH. And because it doesn’t ever physically cut the power the FLASH persists and the intended increasingly harsh punishment for failed logins works.

Lower level stuff is awesome.

Not the poster you were replying to, but I’ve come to the same conclusion and here’s my not-particularly-rigorous reasoning:

It seems to be equal parts users stuck in a local maxima of computing skill and how that enables lax software engineering standards.

When’s the last time you’ve sat down with a user who isn’t remotely interested in tech and watch them work/use a computer? Most of the population’s mental model of a computer is starkly different to the average hacker news reader. You can still hear the same complaints about how computers “don’t do what i want it to do” that i remember my parents generation saying, and they were experiencing the first waves of computerisation in their offices.

The story became that the older generation just couldn’t understand the new generation, but kids are amazing with computers because they’re growing up with them. Well, some of those kids are just as hopeless. It’s partly an education problem (hard to learn computing from a teacher who doesn’t understand it themselves), and partly because UI design trended to simplifying everything as much as possible so that users who don’t understand computing can still enjoy and use their devices. Now there’s not a great incentive to learn more than you need to just use the UI you’re given, and computing skill tends to get stuck in this local maxima.

I won’t go on about my other point in detail as it’s a perennial favourite for hacker news discussion. But hardware gets faster so quickly, but our software is so hastily thrown together that it eats up all the gains. Users don’t notice that software they’re using is crap because their mental model of computing isn’t developed enough to know what’s happening. Instead we get this casting of devices as somewhat malevolent entities (“ugh, my stupid computer keeps losing my stuff. I need to buy a new one that isn’t so dumb”)

We use to think this would be resolved with time and generational change, but it seems like there’s just a more-or-less static percentage of the population that just doesn’t get computers. (Which is completely understandable, people have different interests, it’s hard to inculcate an appreciation of something in your entire population, look at peoples relationships with mathematics)

Part of the reason UiPath is doing so well is because it raised the bar when compared to its competitors from 'the worst' to 'usable'. Blue Prism is alright, but they're so closed off and proprietary it gets in the way of actually solving the problem. AA is just, as you say, 'the worst'.

I'm not the biggest fan of RPA but it definitely has a use case. Though my hunch is that UiPath is going to be something like the Slack of RPA soon if there isn't any progress in the anti-trust front.

Microsoft is starting a big push with Flow/PowerAutomate, and Microsoft _owns_ the legacy corporate environment where RPA finds all its customers. RPA makes money charging expensive licenses per 'bot', and MS just gave 1000 free licenses of power automate to MS 365 customers (at least, the small NPO I just migrated got them). Good luck to UiPaths sales team going forward.

I’ve noticed this too. I’ve found myself iterating on a nice extended form control library where I try to extend the native inputs and forms in the most minimal, yet complete, way I can.

There are actually a lot of nice improvements being discussed in the CSS level 4 specs, but some of the issues are last tagged as updated in 2012...

But basically we just need a consistent value interface (two-way data binding), and a more finely-grained validation/state lifecycle where we can plug in our own custom validation functions.

Colleagues and I always joke that every web developer gets the idea to write a form library at some point, and suffering through trying to implement it is the only way to dissuade them of the idea. Now the joke is on me as I’ve come full circle back to writing one.

I’ve spent so much time on it now I’ve been considering reaching out to help contribute to the spec. But I have no idea how receptive the working group is. This may just become my eternal project, there’s always more edge cases to find.

None of OP's criteria for unfollowing/muting people was them saying things they disagree with.

It's the difference between not watching NBA games because you're not interested in basketball, and refusing to watch the local news because "they have a (left|right) bias".

There is more content out there competing for your attention than you have time in your life to consume it. Rather than letting twitter's default algorithm feed you the most "engaging", OP is saying a more mindful approach works for them.

odd, I'm in Aus too and I watched the entire thing live with no issues (assuming you're in QLD on EST)

Are SSR/SPA solutions really that much more expensive than local apps to develop?

The start of my ladder looks much the same as yours except I jump from local files straight to single database SSR/SPA. Maybe slightly more complicated than a local app because you need auth, but when prototyping something I just use basic auth to start.

Chuck it up on a $5 DO box and you instantly get:

- Multidevice access, desktop and mobile (write mobile-first css and the tiny screens aren’t a problem)

- a UI that’s trivial to hack on (HTML and css are easy)

- likewise, a technical base that’s easy to extend. Drop in react, build out the api side and use it as a source for other projects, etc

All this assuming you have reasonably reliable network access, and even then you can build it as a clever PWA falling back to local storage if you need to (I think, prototyping that is still on my todo list)

Heck, save yourself $5/month and stick it on a raspberry pi at home.

It’s truly a golden sweet-spot for personal software tools. The only external dependency that irks me is needing a domain name.

This was my biggest take-away from the article too. It never occurred to me to even try interacting with a CLI application with my mouse.

I'd appreciate it if more developers considered 'discoverability' as a core part of their design philosophy. I think the author is close, but when you need to ask someone/search 'why is this green?' that's enough of a barrier that most people won't bother asking (or it'll be a small enough issue that they don't consciously notice it).

There's even a generally accepted solution for explaining semantic colours, a legend. Might be nice to have one down with the F-keys.

It's been awhile since I've tried it, but the first thing that comes to mind is not being able to put a file on the desktop.

I completely understand and even vaguely agree with their reasoning, I stopped putting icons on my desktops maybe 10 years ago?

But sometimes I just want to put a temporary file on my desktop as a visual reminder to finish something up later.

In this case it's not the same class of 'bug' as the missing thumbnails in gnome's file picker, it was a deliberate design decision, and I respect that the project thought about the experience they want to create and executed on that vision. It's just not the experience for me.

This is fantastic. You know you’ve come up with a good visualisation when it feels so instantly familiar.

I’m trying to think of how you can convert some ancillary information with color (with accessible fallbacks of course). Maybe color coded users? So you could see the same colors gravitating around threads.

I’m not sure how many unique users comment on an average thread though and how useful that would be. Maybe only showing unique colors for users with more than 2 posts in a thread?

I love it though. Great job

I don't disagree with anything you've just said. I think I'm just focusing on the present and future, while you are talking about how we got here.

I find it much easier these days to get people to consider trying a non-Windows environment but it never goes well because of these small issues.

Of course people expect things to behave in certain ways because Microsoft had an anti-competitive monopoly and most people learned about computing using their platform.

It's an unfair, shitty situation.

But the point remains that desktop GNU/Linux has a lot of these little UX bugs, and we're not even paying attention to them. The experience isn't as smooth as Windows because we don't have dedicated UI/UX designers thinking over these issues and trying to come up with cohesive solutions.

When's the last time a new consumer-facing UX paradigm that 'just worked' came out of the open-source world? Apple churns them out, oftentimes created from scratch, and it extends to their physical hardware design (ie the ipod scroll wheel was amazing for its time, airpods are a joy to use).

The open source world makes amazing things but the UX quality _is_ a major issue in the uptake of new users today.

While I appreciate you taking the time to write this note, I don't think I will edit the word.

The fact that it sometimes is being used by racists gives me a little pause (that's basically how semantic shift happens, and there comes a point where the new meaning eclipses the old), but I have never come across it in that context, whereas I see it used in it's original meaning commonly (and in professional contexts).

If someone tries to cast a comment of mine in a technical thread as racist I don't think changing my language is going to help.

Similarly I'm not so invested in my imaginary internet points that being flagged by such a poor future algorithm concerns me. If HN got to that point, I would probably have lost interest in it long before.

I'm here for interesting, technical discussion. Not for the zeitgeist's culture wars that are engulfing all the other platforms. With the exception of recent threads (which considering the tumultuous events that are happening, I understand), I find HN to be a lovely corner of the old web where users try to read each other's content in good faith.

I don't think the author attributes universal consistency to Windows? They just say that this particular issue has been 'solved' by Windows since 1998. It's been raised as an issue in Gnome since 2004 and still hasn't been fixed.

I agree with most of what they say, and I think the point is that we're uniquely bad at addressing these issues in the open-source world.

We accept a degraded UX experience (which I think this unarguably is from a normal user's point of view) because at least it's not proprietary lock-in. But it's a big reason why desktop GNU/Linux doesn't crack 2% usage. Inversely, I think this is _the_ core value proposition of Apple's ecosystem. They pay particular attention to consistent, unsurprising integration, and the Apple experience offsets their iron-grip over your device for most users. It's easier to convince someone to change how they do a task if you consistently deliver pleasant UX.

I recently had the opportunity to help a _very_ 'normal' user with their computing, and it was remarkably eye-opening watching them interact with their machine (Windows at the time, but they also use a Mac). Indeed, they hated Windows because it wasn't as consistent as a Mac, but Windows was dealing with their ambiguous interaction remarkably well I thought. Much better than a GNU/Linux DE would have.

If we want a world where normal people can use free software for their everyday computing needs this is something we need to get much better at. Forget 'normal' users even, these projects (and threads on HN) are full of _developers_ arguing back and forth that they find it infuriating/unusable. And unfortunately not many of us have the time and money to contribute code or fork projects to increase competition and spur improvements.

(also, hi again Jedd :) )