HN user
I am a sysadmin and hobbyist game developer from Cincinnati.
Author writes an interesting, nuanced, wide-reaching essay about AI and society, with a main theme being about AI and its impact on our humanity.
All other top level arguments offer AI summaries that miss all of the interesting, nuanced, wide-reaching topics about AI and its impact on our humanity, and complain it was too long to read.
Truly a gem of irony.
SEEKING WORK, Senior Devops Engineer, Remote
Devops Engineer with a software quality focus. Well rounded in AWS. Jenkins and Github Actions CI/CD. Ansible, Terraform infrastructure as code. Large scale Apache Airflow installations. Experienced Python dev for scripting, web applications, and FaaS. Well versed in QA: test strategy, automated testing, automated infrastructure sandboxes.
SEEKING WORK, Senior Devops Engineer, Remote
Devops Engineer with a software quality focus. Well rounded in AWS. Jenkins and Github Actions CI/CD. Ansible, Terraform infrastructure as code. Large scale Apache Airflow installations. Experienced Python dev for scripting, web applications, and FaaS. Well versed in QA: test strategy, automated testing, automated infrastructure sandboxes.
It works for me, thanks!
I look at these while I'm daydreaming of building a cabin during work. https://www.ag.ndsu.edu/extension-aben/buildingplans/housing
Awesome, thanks for the links. I also tend to believe that it is a process over tooling question; you could even accomplish this with a text document and some checklists.
When I gave up social media sites, I created a little site that has some blogs and sites that are interesting to look at, and my new tab page is set to that.
Yahoo initially began as a links directory, and I thought that was a nice idea, so I made Yayhoo: https://bmsauer.github.io/yayhoo/
I like it a lot. I probably need to update it because I haven't done so in a while.
I'll preface this by saying this is all anecdotal evidence based on my personal observations, so it may not apply to everyone.
I often feel the same way, especially with tech. For those of us that remember before the internet was ubiquitous, the optimism and promise of the "information super highway" seems in stark contrast with what we see today. I try to keep in mind a few things:
First, we are living in a time that will be regarded as one of the most consequential in history. We're only ~20 years of nearly every person in the planet having access to all human information, instantly. Think of what people will say about this time period in 200 years. We are currently feeling the effects of growing pains.
Second, everything that embodies that early optimism is still there, its just harder to find. Which is related to my next point:
We are seeing diminishing returns in the benefits of constant consumption of media, energy, food, etc. There is so much choice out there, and the margins are so thin, that you need to consume more to be "satisfied". I often reflect on how many more full TV series we have all seen compared to a few generations ago. Or how much text we all read daily in the form of news, tweets, and forums, compared to the daily paper. Are we better for it? I think a lot of people don't feel better.
So that leads to the optimistic conclusions of this post. Generally speaking, we have more choice than ever before in history, across the board. But we have the burden of the responsibility of moderation and curation.
I find that when I feel this way, I try to shift my "consumption" mental state to "construction". We live in an amazing time to make things and distribute them. And because there's so much noise to compete with, you have to do it friend to friend, neighbor to neighbor. Its a glimpse of how the best parts of the "new world" can provide the best parts of the "old world".
How does Nix compare to Environment Modules: https://modules.sourceforge.net/ Am I correct in understanding they are similar, except Nix has the "pure" package building?
This reminds me of a technique I learned from this StackOverflow answer [1] that is about creating structs/records with TCL lists and dicts. It apparently comes from LISP philosophy:
proc mkFooBarRecord {foo bar} {
# Keep index #0 for a "type" for easier debugging
return [list "fooBarRecord" $foo $bar]
}
proc getFoo {fooBarRecord} {
if {[lindex $fooBarRecord 0] ne "fooBarRecord"} {error "not fooBarRecord"}
return [lindex $fooBarRecord 1]
}
I'm a new professor that was a devops engineer in the field until a year and a half ago. I understand that out in the field, we think professors are all just lazy out of touch academics, publishing papers to meet a quota; I used to think this myself. Some definitely are, but I think its important for the audience to understand that I am new to this, working very hard at being good at this, and primarily a teaching professor (I do not do research and teach a full load, no research or administrative duties). I teach computer science. These are my observations to the five points the author raises:
1. "First, attendance was sporadic with many of my students." I think this is true and always has been true for entry level college classes. You usually have a set of students who will never come, a set of students who will always come, and the middle is what you need to worry about. My solution to this is require attendance, which I first started this year. I've noticed remarkable differences in class; I'm more involved in learning, can fix problems earlier with less feedback time, and students are more engaged with the materials outside of class.
2. "Second, work outside of class was not done with consistency." This is a tough one. I was shocked to find how many students don't read anything. Not just the textbook; they don't read documentation or even the lab assignments. I'm sure this has been a thing since I was in school in the 2000's. I think this is a case of "students just don't know how to". I try to teach them how to read documentation, as it is a skill, not something natural to common life. I also think requiring short essays on reading material is superior over multiple choice because it makes "brute forcing" through the assignment less easy. You can just copy snippets from a web search, and that happens, but again, I'm just trying to move the "middle" here.
3. "Third, basic mathematics proficiency, whatever the claims of our admissions office, is greatly lacking." This is an interesting one too, especially in computer science. Most of my students in my entry level classes do not understand the concept of a filesystem. Using phones and chromebooks their entire life, they've never had to find files outside of a search field. Some know even less about how to use the computers we bring to class. I blame this on big tech making crappy products and driving people to appliance computers, but that's a rant for another day. This is surprising but not difficult to overcome; you just have to teach it in the first week. You lose valuable time you should be teaching the material to teaching "how to find your file to submit", but good pre-made materials can alleviate this somewhat.
4. "Fourth, students seemed reluctant to come to the office seeking help." This is definitely true, but also, they don't seek out reading the text/assignments for help or on campus tutoring. I also think this is something you have to teach; how to formulate a question based on something you don't know. I also want to mention here that cheating is rampant in computer science, so this is a more tempting approach for many students. Again, another issue for another discussion.
5. "Fifth, and finally, the introductory economics course, as stated in my 2014 Inside Higher Ed article, has remained somewhat constant." Other than the student learning outcomes (SLO's), courses in college are surprisingly flexible. I'm free to do pretty much what I want in the bounds of the SLO's. But: prepping a course is very work intensive, and it takes away from the student-centric work referenced in the other 4 points.
Anyway, in closing, just want to say that there are a lot of good hardworking, non-tenured, full time professors out there. Its the hardest I've ever worked in my entire career and the most challenging work I've ever done. A lot of the points raised in the article are true, and its not just "kids these days" but systemic problems ranging from early education, tech literacy, and motivations for even attending college.
My senior project in college was "Trombone Hero", which was a game based on guitar hero. I had an old trombone, and I took apart a cheap gamepad and put the sensors along a custom PVC pipe slide at positions 1 - 7. There was then a button attached to it that you would press for "blowing".
I then made a little tool that would play music and record inputs to generate the "levels". I used some Beatles songs I had on a CD, and then recorded each note individually on a real trombone to superimpose on the song.
I made the game engine with some Java graphics library that I don't remember. I was really proud of this when I made it, and is always a fun story to tell.
I found an old video from ~2012 when I demoed it to the class: https://www.youtube.com/watch?v=3rawTtdv50s
Frogatto and Friends is a good platformer not mentioned yet. Open source engine, not open source content. https://frogatto.com/
I'm a new computer science professor coming from "industry" (as the academics call it) and can confirm that cheating is rampant. I've also noticed that it is an arms race you'll never win; cheaters will always find a way to cheat, and you can try to subvert them as best you can but will eventually end up harming legitimate students, which is unacceptable.
What I wish existed was an independent board exam that students had to take for employment. Many professions have this, from medical professionals, accountants, engineers, and the trades. My thought is this would disincentivize cheating and encourage deep learning at the university level, while offloading the cost of cheat detection and enforcement to a board, whose job is to solely evaluate if someone is capable of entering the workforce. A separation of concerns.
Another reason for board exams: when I was working, I was asked to do a bunch of flimsy engineering to satisfy management. It would be nice if I could respond "If I do this, I could lose my license, so no." Just like if someone asked a carpenter to build a house not up to code, or a dentist told the hygienist to do shoddy work to improve throughput.
If anyone knows anything about this, I'd love to chat, because its something I've been thinking about for a long time and I hope there's a good reason why it doesn't exist.
No one has mentioned Linode yet, I had a good experience with them around 10 years ago, seems like they have more cloud product offerings now.
Makes sense. Maybe if you ask yourself why those with the existing infrastructure dont already have a version of your idea implemented, it can help you identify the important parts of your pitch.
The conventional wisdom here is that ideas are sellable because they are uniquely executed in a way that is both practical, cost effective, and solves a real problem.
Or, conversely, most great ideas seem great on the surface, but in reality, cannot be done in a way that makes business sense. Ive seen this a lot where great development practices to produce better code never seem to catch on due to business pressure, for example.
Taking at face value that you have a great idea, then your execution of the idea is the unique aspect that will prevent others from "stealing" it.
Or, from the other perspective, if the idea is so good and easy to execute that it can be "stolen", why doesnt it already exist?
What I always liked about SO jobs was it was tech centric, so you could filter on language or tool. It also seemed to have more "transparent" posts in terms of salary, benefits, and dev practices. What are good alternatives to SO jobs?
I had a Blackberry classic until a month ago when AT&T sent me a new phone in the mail with a letter saying they would deactivate my Blackberry automatically due to 5g.
I hate my new phone. There is absolutely no way I could effectively type without autocorrect. And because I won't sign in to Google, I can't install apps or use turn by turn navigation, so I'm essentially where I started with my blackberry. "Progress".
I learned about it reading the quake 2 sourcecode. In school or tutorial articles, they always teach the first technique, I find. "Delta timing" is what I knew it as.
Was the documentary "Tapped" by chance? I saw that one around that same timeframe you mentioned and it inspired me to get a reusable bottle I still use to this day.
Seems like it fails the Occam's Razor test, but I guess it isn't impossible.
I've read this before, and while I agree that there are tropes, like the techbabble mentioned in the article as well as creating suspense with arbitrary timelines ("we have a routine task BUT the core will explode in 30 minutes"), I think the fact that the specific tech is irrelevant is one of the greatest strengths of Star Trek.
The Star Trek universe is one where people live in a post-scarcity, post-war, and heavily automated earth and are free to pursue creative or explorative endeavors. The technology in the show is designed to be seamless; the computer responds to voice commands, doors open on their own, food is ordered from a machine. It is intentional that, in Star Trek, there aren't people huddled in front of a screen all day typing on a keyboard.
The reason for that is because a show that would be hyper focused on how tech works would be incredibly boring. And, at least in my opinion, it isn't important. The author says the characters "don't tell us anything interesting about the human condition under science fictional circumstances", but that isn't true. Many episodes revolve around what it means to be human and our role in the universe on a wider philosophical context. Whether the warp drive can realistically be snapped to warp 11 doesn't affect that at all.
To give a real world example, consider the Snowden revelations. I don't know the specifics on how the NSA algorithm worked, or what storage devices they used, or if it is possible to perform a certain operation on it. It isn't interesting nor does it speak to realities of a post 9/11 America. What is interesting are questions on the balance between privacy and security, the underwhelming public concern, and the lack of transparency in a government body. I feel like Star Trek would have covered those stories, techbabble or not.
Thanks for this tip, I need to do some updating.
I'm really glad I saw this. I have a ~2015 Blackberry Classic that I love, and when I got the texts, went to the store. I asked them that, since my phone supports 4GLTE (from the icon on the screen) would I be alright. The salesperson said it would work, so I forgot about it. But I didn't realize there was a different network for calls/data. And from how they are apparently forcing phones upon people, sounds like I need to be proactive in getting a phone I will like.
So yeah, thanks OP.
The one I always use is Kolourpaint. Installable via apt.
I currently do this, but I work all 5 weekdays. I just work shorter days Monday - Wednesday.
I love it, it is my favorite perk of the job I have. I'm able to use my afternoons to exercise, take care of chores around the house, and run errands if necessary. My night after dinner time is completely free.
Another perk that's similar but not exactly the same thing is flexible schedule, meaning if I have something to do during work, I can "make up the hours" later. This is especially powerful when you work 80% time normally.
I would say that the reason most companies don't want to do 4 day weeks is because of meeting schedules. This is why I'm here each work day; so I can be at my team standups every day.
I agree with your point that there is a need for advertisements to inform consumers about available options. I'd generally fall in the "dismissal of all things advertising" box, but I would add a nuance to it that it really depends if it was requested vs. forced upon you.
In both instances you mention as being useful advertising, shopping for furniture or how to sleep on an airplane, you are asking for advertisements. That makes sense. You are looking to solve a problem by purchasing a product.
From my perspective, there are two issues with the current climate of ads: First, that the overwhelming majority of ads are forced upon you. They track you, distract you, and have generally turned the internet into a wasteland. Second, that a search engine/social network/news site is the place to view ads. I would prefer a site dedicated to this use case, not have the use case tacked on to unrelated sites constantly in the way.
I feel the same way about physical ads, too. I don't want uninvited people knocking on my door to sell me their ISP. I don't want those terrible mailers with coupons in them. Billboards are ugly and distracting.
I also thought of this, but from the movie "WarGames".
The irony of this question is that the OP is concerned about the maximum optimization of "intellectual potential" wasted by people that are seeking maximum optimization of a game system.
There is more to life than a feverish march of scientific progress. There's value in wasted time, boredom, and frivolity.
Furthermore, I'm certain that current/historic systemic oppression of large swathes of humanity contributes more to "wasted intellectual potential of humanity". And in those cases, it is actually detrimental to the person, whereas a game is usually provides benefit to a person in happiness, relaxation, or challenge.