ajford

I always assumed it was the assumption that the user would be "remembered" by their auth token and that the user would be returning from an already auth'd device.

When your UI team is testing user flows, if they're not starting from a fresh/clean session, they can fall into that pattern and forget to walk the "existing user, no auth" path.

At a past employer we ran into that with regards to an Apple/Safari issue. It's been a while, but iirc it was something to do with scroll heights, modals, and a render engine update in Safari maybe 3-4 years ago. Nobody caught it during testing due to us using manual testing in a variety of browsers, and it was live for almost 3 weeks. It was annoying at normal use, but if you had fonts increased (i.e. eldery/vision impaired user) it meant the user would be unable to login.

After that incident we switched to a hosted browser testing solution (I think it was BrowserStack, but might have been something else) and we made sure to flush things and test on multiple browsers per platform. I don't remember the full list anymore but it was something like 20 different browser/platform/version tests each.

I sent the article to my spouse some time back and she's mad at me for ruining butterflies for her. She made the exact same Sixth Sense reference too!

We've now made it a challenge to spot "live" butterfly depictions everywhere.

I took it more as a comment on the sheer ineptitude and uselessness of TSA as anything beyond security theater.

I have no qualms about the change and think it's about damn time they made even this tiny modicum of effort to support non-binary people, but I'm almost certain that anyone with an X will be questioned about it regardless. Hell, even perfectly legit passports get questioned all the time.

Story time!

A buddy of mine flies with his US passport at all times, even for domestic flights, because he's of Arab decent and constantly gets hassled for it. One time we were flying together with a student group (we were in college at the time), he gets pulled aside by a couple of TSA agents as part of a "random inspection", out of a line of 20-30 people all mostly white or light complected.

He starts getting grilled about everything from destination, reason for the flight, bags, etc. Then they ask to see his ID and boarding pass. He hands over his passport and boarding pass and they start going off on him as to why he's got his passport for a domestic flight, how that looks suspicious and such.

He responds with "I get stopped for looking Arab every time I fly, so this just proves I'm a US citizen and is a federal ID." They stop and finally realize they've got everyone looking at them, hand back his passport and ticket and walk away.

I'd argue that you don't hear about failures as most people aren't experiencing them.

Out of at least a dozen people I know using them, every single person has multiple RPis that has been in pretty regular use. On top of that you have a plethora of documented uses and projects across the internet, ranging from home automation to arcade machines.

You seem to be painting an entire community with the same brush, but most of the people in this community would gladly help you solve the issues you seem to be facing. Perhaps the issue lies not with the RPis but with how you're using them....

I've had a lot of luck using the Samsung Evo Select micro sd cards, and use them almost exclusively with my RPis. I've recently been trying PNY Elite-X as well with good results, though I think I've only got one Pi running those (and two Wyze cams).

I've got multiple RasPis from 2nd, 3rd, and 4th gen that all have lasted years. I know of a Rpi 2b still running in a shed in the New Mexico desert that holds at about 90*F that's still going after 5 years.

I had a 3B running as a car infotainment system with hard shutdowns at least daily for three years. Sold that car and it's now in a drawer, but slated for use in an arcade perhaps.

I've got another 3b running as a Unifi controller that's been running for 2+ years.

I've got two 3b and one 4 running OctoPi (3d printer server) without issues for 3+ years.

Given the mass adoption and heavy usage of RPis, I would venture to day most people don't find them to be "horrible computers". I think in 5+ years the only time I've had an RPi fail at the hardware level and die was due to powering it via the 5V headers and accidentally reversing polarity. Might have also shorted one out when an enclosure was flooded due to bad weather seals.

All anecdotal, but I find most of the time when Pi's fail it's often either voltage sag, poor quality sd cards, or excessive writes to the SD. The SD cards are definitely a weak link in the RPi ecosystem.

Whenever possible, ensure your Pi isn't reporting undervoltage and look for ways to minimize writing to disk (or attach a USB drive for any heavy writing necessary).

You thought solving for X was hard... wait till you're solving for 你好!

I doubt there's much behind the joke beyond the stereotype of "Asians being good at math", tbh.

My Algebra teacher in school once solved a system of equations on the board using the Batman logo/symbol, a Peace sign, and a happy face. Just to illustrate the concept of variables since a few people were having issues grokking the concept.

Totally agreed. That's why I pointed out the difference. I've used many frameworks, and each brings it's own flavor and benefits to the table.

I wasn't disparaging Django, just commenting on my issue with it. I still used it for 2yrs, but the batteries included nature of it is something that most of my projects don't call for. When I was primarily working on web apps, I made heavy use of Flask for back-ends and APIs, with various flavors of JS for front-ends.

In one role however, we had an entire CMS/CRM/Subscription engine written in Django, and it was rather solid. Since everything was in Django, it all worked together well. When I started working with this app, it was in poor shape (and a couple versions out of date) but when we brought it up to date it was real nice. We could make good use of Django signals for all sorts of things with little extra effort, and the variety of plugins available was just fantastic.

I think the biggest skill a dev can develop is learning to recognize the pros and cons of a framework or language or tool without developing too much of a prejudice for any given one. Recognize that they are all just tools, and some are better suited for some tasks then others. Doesn't mean you can't press one into service if it's a poor fit, just means you'll need to do extra work.

Python might be a poorer fit for a highly parallel real-time task compared to something like Rust or Go. Doesn't mean you couldn't use it, but using a better option would make things easier. If you want something with an ORM, template engine, signaling system, and caching layer all built in with minimal fussing about with other libraries, Django is perfect. If you JUST want a rest api, perhaps FastAPI or Flask might be a better choice.

This seems like the kind of task that might be possible with a solar or renewable fuel.

Given the target speed of 1.5knots, I wonder if a large enough solar installation on a pair of modified tugs could achieve a decent runtime. Or perhaps focus on using the solar to generate hydrogen for a hydrogen burning engine given the availability of water.

That's my main issue with Django, and why I prefer a "bring your own modules" framework like Flask. Django is very "batteries included", but if you're not using the full Django paradigm, it gets more finicky.

That said, Django with DRF to float an API-only project/backend is almost standard at this point, and a great option if the devs are already familiar with Django. They can leverage the existing experience and get right to work without needing to get up to speed with a new framework.

In regards to the documentation, the module index can help a bit. If you know the module space you need to be in (i.e. you need to do something with the DB), just jumping into the module index and looking for entries in relation to django.db can get you in the ballpark real quick.

That's par for the course with JetBrains. I'm a big fan of PyCharm, don't get me wrong. It was the first IDE I've tried out of maybe a dozen that actually gave me a reason to convert from Vim with plugins and a number of self-written project mgmt/organization tools.

That said, JetBrains often rolls out new things while sitting on long-standing bugs. There's three or four issues in PyCharm and TeamCity that I've been tracking for a couple of years that are older than my kids. Hell, I think one of the TeamCity issues is almost old enough to get a learner's permit.

Are you talking about an API reference? Django has that: https://docs.djangoproject.com/en/3.2/ref/

They have an entire set of tutorial/getting started type docs that are definitely very wordy and based around getting you comfortable with Django's paradigm, but they do have a rather well done API/Module reference that does a good job getting to the meat of the topic imo.

Maybe, maybe not. The author has graduated from High School, meaning they're about to enter college or the workforce. I wouldn't be surprised to see this level of detail from someone at that level academically. Delighted, yes. Would I expect if from everyone? Hell no.

But surprised that a tech-enthusiast and eager learner might have put this much thought into this prank and it's potential consequences, not so much.

Teenagers/young adults tend to have different stressors and other things to occupy their time than the average adult in the workforce, meaning the author likely gave this prank a fair amount of their free time, and that dedication showed through in the amount of planning done.

Additionally it's likely, given they mentioned once or twice in the article they planned on posting a blog about the prank, that they might be hoping to use this on their resume or as a talking point in their career. If they're hoping to go into security or comp sci, this would be a decent feather in their cap and the amount of time spent is easily justified.

Glad to see a cooperative and supportive academic administration, and I'm sure the thoroughness and planning that the team demonstrated made it easier on the administration.

The sheer amount of testing and verifying no major impact to academic testing took place probably helped, and cleaning up after themselves and documenting their finding and reporting it to IT was a cherry on the top.

I like that the administration even requested that the team brief the district IT on the "attack".

I think the issue is not with supply of the chip, but Apple restricting API access to the chip on IOS devices. They're basically gatekeeping hw features, and preventing anyone else from providing the same level of features on IOS by software restriction.

Could be wrong though, as I haven't followed it closely, being an Android user.

Totally agree with you there. I really think they should start shipping with a second set of "inside" legs for those who want to use a tenting leg instead of using the thumb cluster.

I'll probably get to a smaller key count eventually. I'm almost finished adjusting to the Moonlander now, and it took me almost two months to get used to the loss of the separate F1-11 keys. They're now on the number row in a new layer, so I figure with enough time I'll get used to that and can toss more into other layers, but that's proving to be a slower process for me. I think I've had this board for almost 8 months now (had to look up my shipping info, as I thought it had been much less).

I really do love the fact that the mech keyboard community has spawned so many great keyboards, as it makes it so easy to find something that fits you best, and makes it so easy to customize.

Kinesis has also been teasing a new board that looks to be a split Advantage style board.

https://twitter.com/kinesisergo/status/1397335511694381058

With a programable mechanical kb you should be able to match the Mac layout. There are a few split versions that should have sufficient keys, especially if you're not using a ten-key frequently.

Try a short-travel or low profile mech switch. I believe both Cherry and Kailh have a Speed series with a short activation distance. The Kailh Choc line of low-profile switches has something like a 1-2mm travel which might work too.

One benefit with mechanical keys though is that there's a massive variety of types, so you can easily find something that works for you. You can get short-travel (i.e. Kailh speed series), soft bottom-outs via something like the Boba series or by using o-rings, high resistance or low resistance with varying spring weights, tactile vs clicky vs linear for feedback.

With a membrane keyboard you have zero choice beyond moving from one brand/model of keyboard to another, but with a hot-swappable keyboard you can try multiple switches until you find one that suits you and doesn't aggravate any existing RSI or trigger new forms of RSI.

Yeah, I can't imagine the Moonlander being a great fit for someone with normal/small hands. I got a Moonlander and happen to love it. Though, I really wish they hadn't decided to make the thumb cluster a tenting leg. I think that's my biggest complaint about it.

ZSA does sell a tenting leg kit that I believe should allow you to swap the thumb cluster screw for a leg, but I haven't tried it myself. It's only $24. Additionally, they have a 3d-printable file for a tenting leg that should work as well, if you happen to have a printer.

The Dactyl Manuform and Ergodox are also decent options for anyone looking for something like the ZSA Moonlander but don't want to deal with the Moonlander's issues. The Lily58 and Corne have far too few keys for my liking, but to each there own.

The pandemic was definitely extended in the US due to politics.

Every attempt at curtailing the spread was rolled back or defeated specifically due to politics at the highest levels of the US Gov't, from not implementing travel bans as appropriate to blocking masking mandates, to state level gov'ts re-opening economies and lifting lockdowns during surges. The current hot-spots in the US are almost perfectly aligned on US political boundaries.

Even now it continues as masking and vaccination continues to be politicized with one side spreading obviously unfounded FUD around the vaccine and inflammatory messaging that masking children is "child abuse".

I think New Zealand is rather famous for it's COVID response and recovery. Denmark and Finland also had great responses.

That said, I'm in the US and not well equipped to speak to their responses, having not experienced them first hand.

I think the black swan event is less of the pandemic itself and more of how it played out. If actions were taken and the pandemic was halted early, there would be less impact. At least speaking of the US.

The pandemic has gone on for far longer than it really needed to due to the politicization of safety measures. Several countries were able to nearly beat back the pandemic rather early, but the US is currently sitting in a third or fourth wave due to asinine behaviors of a few state governments.

Yes, a pandemic itself isn't black swan, but just how badly this was fumbled sure seems like it is (at least I damn well hope the next one isn't so badly managed).

No, but my tech-illiterate father is now setting up fire TV sticks with Kodi and plugins using Youtube guides for him and my step-siblings.

The whole "are the normies capable of this" argument has turned into an arms-race against Youtube/FB for easy setup guides.

That said, yes, the number of people they'd loose over shitty interfaces is probably kinda small, but I'd argue that as guides, youtube vids, and even all-in-one seedboxes become more available, that population is gonna grow. All it takes is some streamer to lay out the current process.

I mean, I have Paramount, Hulu, and HBO subscriptions, but I'm far more likely to fly the black flag and watch it from my Plex server instead of their apps due to how painful those apps are to use.

If I was a little less honest, it would be easy to skip paying the subscription entirely. I generally use their apps now to find and try new shows, then download them to Plex once I know I like a show (so I'm not burning bandwidth fetching shows I'm not gonna watch).

Buildings on land don't move. You don't have to plan for the scenario where your neighbor's house shifts it's location a few degrees and runs into your house, or where your neighbor's house will be in 2 years.

These constellations aren't stationary, they move based on their orbital parameters, and to properly predict their orbits, you'll have to account for their future movement for the next while (depending on expected lifetime and how close they come to your intended orbit).

Saying this is like buying a plot of land is missing the level of work and complexity that goes into orbital mechanics.

If you need to illuminate the same area of the earth, which would make sense if you're also looking to serve the same populations, then you'll need to either be at a different altitude or have neighboring or intersecting orbits, which will need to handle the orbits of all other objects. That's why the level of complexity increases a fair bit when you need to plan for two large-scale configurations.

Maybe for certain subsets of the user group. I don't like the dongle-hell that is the Apple laptop line. I'll take my Lenovo or Dell laptop that has RJ45, multiple usb ports, and more. I also prefer a physical laptop dock (not USB3), but I always had such a headache with them for linux support.

Apple laptops are always a fair bit pricier for a given set of hw specs than any other manufacturer would be.

Additionally, they lack any level of reparability and a trip to the genius bar will often be quite expensive. Hell, look at the debacle that was the butterfly keyboard they pushed for something like 5-6 years.

Apple's keyboards, even before/after the butterfly keyboard, are inferior (in my opinion) than the keyboard on the Thinkpad line, and feel quite mushy by comparison.

The aluminum construction is quite solid, but not without it's issues. They seem to get far hotter in my experience, but perhaps that's them acting as heat sink so I'll give that a pass. The hard edges on the case are a pain-point to me.

I know a lot of people seem to prefer the apple touchpads, but I've never liked them. I prefer physical mouse buttons, and the click never felt right to me on apple touchpads. Part of it might be that gesture inputs with the touchpads aggravated my RSI, so I was never able to make use of them.

All said, I don't agree with Apple making the best x64 laptops, but they fielded a competitor like many others. They have the advantage of being the only true Apple laptop though, so if you're all-in on the Apple ecosystem, they're definitely the best choice. Otherwise, there are many laptops that might surpass the macbooks, depending on your individual preferences.

But it does have material impact on others.

The Starlink constellations in this proposal are large, and any future development by others will have to account for the orbits of these potential constellations.

That's 30k new objects to plan for, and with two configurations at play, that means accounting for essentially 60k objects in motion, effectively doubling the work someone would have to do to plan their deployments.

I fully second the DNT header. Perhaps even some common set of headers, like DNT/FirstPartyOnly/Performance)

I like what GDPR did for privacy, but I absolutely hate it's impact on web browsing. Now _every_ damn site has a cookie banner covering the bottom third of the page. Plus many sites have a fixed nav or another damn banner at the top. Mobile browsing has gotten painful.