Eternal September was never a problem for SO, it was an asset. Duplicate questions was never a problem for those asking or answering questions (I did both), only for a relatively small group of loudmouthed moderators. Now they hardly have any dupes to worry about but they also have no content to moderate!
The reason that duplicates were treated as dangerous was that SO viewed their most important user not as anyone you have mentioned but instead they prized the lurker most- the person who typed their problem into Google and got brought to SO, and never asked or answered a question because they got what they wanted from that one page load. The entire structure of SO was built around this user.
So why does that mean that duplicates are dangerous? Because of updates. When someone answered a question about how to do something in Python (but it was 2008 so it was written in Python2) SO had ways to get a more correct, up-to-date answer to that question written in 2015 (and then again in 2019) and get that upvoted, and moderators could reward that new answer by editing the original etc.
That is why duplicates were a major threat: if the same question is asked and answered thousands of times, no one is going to go do the work to update all of those answers all across the site. Those lurkers are now dependent on the whims of Google as to which of the many answers you get taken to, and whether it has the latest answer or some answer that stopped working years ago.
And that is why they were so hostile to duplicate questions.
To extend this, the eventual problem is that you eventually lose all of your new users, which means you lose the extremely valuable intermediate users (ones that know enough to ask complex questions, but not so much that they can figure out the answer).
But in trying to solve that problem, they threw the baby out with the bathwater!
They were so fixated on solving that that they failed to realize: training all their power users to grief anyone who didn't behave like a power user off the site was detrimental (to having a site with non-power users). Everyone came for a question and answer site, but when it transformed it into a "question and get downvoted and modded into oblivion" site, everyone left.
AI put the final nail in the coffin, but SO was dead before AI arrived ... from this self-inflicted wound.
It doesn't matter because the duplicate is linked to the original, so most visitors coming from Google would still view the actual answer. It works like a soft HTTP 302 response.
Ime, math.SE had a much friendlier vibe than most other SE sites. Primarily because you could ask about a problem you were struggling with and get help. No moderator would instantly show up and close the question as a dupe of a ten-year-old question about double integration techniques or some such.
People asking questions mostly wanted help, but most moderators thought they were curating some kind of question-answer form encyclopaedia. Very different perspectives.
And those that believe unions benefit workers, particularly in the tech space, are the same people that think life would be better under a collectivist form of government and that the minimum wage needs to be raised continuously. Unions had their place in early 20th century. Those conditions are in the past. If your "conditions" in a tech job needs improvement, get a new job. Anyone that went into tech expecting it to be like a 7-3 manufacturing job, went into the wrong business. I spent years working 60+ hours a week and loved it. If you don't, change careers.
Working 60+ hours a week for years is nothing to brag about. It just means your projects were mismanaged, done poorly and you were absolutely taken advantage of. You also mentioned in another comment that you are disposable if you are "slinging code" 40+ hours a week.
No, we weren’t being taken advantage of. We were working a project that changed an industry dramatically. The projects weren’t mismanaged. They were very aggressive to be sure, but that’s sometimes what it takes to be first and change things. We did it voluntarily and were well compensated for the times. As for disposable code slingers? That’s been true for the last 30+ years.
So which industry was changed dramatically by your team working 60+ hours a week for years? If you we’re compensated for overtime you can thank unions for that, if you worked voluntarily overtime you were by definition taken advantage of.
> As for disposable code slingers? That’s been true for the last 30+ years.
The last 30 years saw massive growth in SWE salaries. That doesn’t happen if SWEs are disposable.
We aren't talking true software engineers. We're talking about people that think their job is to write code and nothing else. Yes, SWE salaries have gone up a good deal in recent years. However, we heard the same outcries about unions in the late 80s-90s when outsourcing and offshoring began to flourish. As for compensation for voluntary overtime? It had nothing to do with unions and everything to do with visionary management and aggressive, committed team members. The average person in the industry today doesn't have the same drive. I saw that repeatedly the last few years I managed. Their expectations were based on what they had read about FAANG jobs, which didn't reflect the typical commercial situation. So while salaries did rise, they weren't competitive with the Big Tech firms. That's just reality. But those applying to the jobs thought differently. It was common to get college graduates from middle and lower tier schools or junior developers with 1-2 years experience expecting to make $150K+ and get 4 weeks of vacation. That's just not realistic at most companies. And if you were able to reach a mutual agreement, they seldom stayed more than a year or two before trying to leverage up another notch.
Brother, you voluntarily worked overtime without compensation because you "loved it" and thought you were already highly compensated. All this means is that your company took advantage of free labor and you missed out on being paid for your work. Whatever your compensation levels are, giving free labor to your employer is a foolish game. 60+ hours a week for years, come on...those projects are absolutely mismanaged if the work can’t be done without that much overtime.
Python, strip ANSI escape codes regexp. Copy-paste regexp. Bam, done. I'm comfortable with delegating such menial, low-value tasks. I can focus on the important tasks. I'm a much stronger coder than the bot but also lazy.
The average bystander might want to write high-performance code for their risc-v cpu. Then they must know precisely which instructions are available and what the performance implications of using them are. E.g., the difference between a shared and non-shared fp register file is huge.
For the "average bystander" they're going to buy an OS and compatible hardware, or if they're the average programmer they're going to use a compiler and libraries that solve the problem already for them. Very very few people need to worry about the details.
If you want to get the absolute most out of a specific CPU that is in your hands then you of course have to refer to the documentation for that specific CPU.
That process doesn't depend on whether it's an x86 or an Arm or a RISC-V.
That's why x86 people refer to the HUGE document maintained by Agner Fog.
If you want your code to run well on all standards-compliant implementations then you write according to the ISA documentation, in this case RVA23. Or ARMv9-A. Or x86_64 v3.
Nope. I want to get the most out of all cpus that will run my code. This is a combinatorial problem that grows exponentially by the number of relevant extensions. So, yes, you need to know the hardware, but accounting for combinations of 5 features is way easier than accounting for combinations of 10 features.
Riscv is basically repeating the same mistake X11 did. A minimal base that could be varied endlessly by combining extensions. I didn't work for X11. Some extensions became de facto mandatory (shm) while others fell by the wayside. But you could never rely on the availability of a given extension because someone somewhere might not have had it or disabled it. Then Wayland came along and cleaned up the gazillion extensions mis-design because it was a huge PITA. Riscv will get there too, sooner or later.
You think the average person writes performance optmized code?
If you are on that level then you know pretty well what you are targeting. And even then in 99% of cases you just look at the top level profile.
If you do performance analysis for some specific embeded project that is not using a standard profile, then its a bit more work, but hardly impossible.
Bruh, the "average person" won't buy a riscv-based computer in decades. The average bystander to the riscv project indeed writes high-performance code for their, so far, mostly non-existent or emulated riscv processors.
Your seriously arguing the the avg person write performance code so critical that minor difference in hardware implementation are relevant? Most people write code that isn't that performance critical, fireware or they are porting existing software over. A extreme minority of people that interact with an ISA is hand optimizing code.
Lol... the RISC-V ecosystem has loooong passed that stage. RISC-V is eating into markets from deeply embedded to automotive, high-end server cpu's to specialized accelerators. That's mass-produced hard silicon.
It's here to stay, coming to a device near you Real Soon Now (tm).
Do high-performance RISC-V CPUs (that you can actually buy) still exist? SiFive Unleashed was great but IIRC it was a single batch that never returned.
I have in my hands one of the new SpacemiT K3 machines. It arrived today. I'm comparing it to several other things, and finding that it is pretty comparable to a "late 2012" Mac Mini with a i7-3720QM with base 2.6 GHz turbo 3.6 GHz running Ubuntu 24.04. They are quite close in feel for general use, web browsing, code editing, watching YouTube etc. The Mac is a little faster on many things, a LOT slower on others (anything that can use 8 cores, obviously).
You might say that's not "high performance" but we thought it was pretty good a dozen years ago.
The previous SpacemiT K1 chip two years ago was more like one of the last Pentium IIIs or PowerpC G4s, except with a lot more cores.
SpacemiT have a next generation K5 coming out, they say, at the end of the year. Tenstorrent have their new Ascalon-X core comparable to Apple's late 2020 M1 — and designed by the same guy who designed the M1. They've taped out a chip using that and say they'll be selling a dev board in Q2 or Q3. For now the first version is using an old chip process and it will be running at half the clock speed of the M1, but that's still going to be a very decent machine.
The HiFive Unleashed was of course 8 years ago. Since then there have been the HiFive Unmatched (roughly like Cortex A55) and the HiFive Premier P550 (a bit better than Cortex A72, other than no SIMD).
> You might say that's not "high performance" but we thought it was pretty good a dozen years ago.
Definitely sounds pretty high-performance compared to basically every RISC I've seen (and including nearly every cell phone I've ever owned with the exception of the Apple ones).
Tenstorrent is awesome, can't wait to see if I can afford any of their hardware in 5ish years. I miss when you could buy TPUs as a consumer (Coral etc.)
I have not! Don’t get me wrong, I have great respect for people who have spent extreme time and energy into researching anything, including works of fiction. I’m sure they are not dumb and are immensely more well read and focused on their area than I am!
The joke was meant to poke a little fun at superfans and not to belittle.
reply