(semi unrelated) Does anyone else find them selves using SO less and less as you gain more experience? I'm over 10 years now and rarely find my self on SO.
I probably take more guidance from blog articles and GitHub issue comments than I do from StackOverflow. I think this started to happen for me around 2015/2016. I'm not sure though if that's a function of me requiring more niche/targeted guidance, or StackOverflow no longer capturing as much knowledge as it once did.
I was asking myself similar questions after finding that the survey results bore little resemblance to what I've encountered in the US software industry in the last few years (I've been at it for decades). My first hunch is that the results say more about SO survey takers than about the actual software industry - similar to how various political polls (e.g. FoxNews, Politico) seem skewed towards the bias of the polling organization. My second hunch is that it's a big world, and I and my colleagues may just inhabit a bubble far from the center of the bell curve.
These days, the only time I drop into SO is when a "how do you do that again?" search yields a relevant-looking question asked within the last year or so. As often as not, the search leads me to a blog post or primary source (e.g. mozilla javascript documentation). But when I do make it to SO, I find that it's still a bustling community full of legitimate expert guidance.
I believe that Stack Overflow belongs to a time when VCS was the exception rather than the rule, codebases were scattered across many different providers, discussions took place on dedicated channels or mailing lists, documentation was scarce and of poor quality, and tooling was limited to the most essential. All this added friction to any form of shared knowledge of being built by the general public.
I find this as well. Earlier in my career it was useful, but in the last bunch of years I rarely if ever am on SO, and if I am, I’m trying to contribute answers. Wouldn’t be surprised if the survey data is largely skewed to more junior people.
Anecdotally, I find that web search has been doing a much worse job of recognizing my query and presenting an SO result - which is why I personally use it less. Blogspam created from scraping SO has been completely pwning top results in Google and DDG(bing) and it’s really disheartening.
See for example “geeksforgeeks” and similar sites, which I find to be a poor and overly verbose resource, which now dominates the top position
> See for example “geeksforgeeks” and similar sites, which I find to be a poor and overly verbose resource, which now dominates the top position
Google Search's and Facebook Messenger's spam filters started letting tons of garbage through at roughly the same time. I wonder if the advances in generative deep learning helped with that.
This is so depressingly true. It's why most of my Google queries are affixed with `site:stackoverflow.com` or `site:reddit.com`. I lean on this so heavily that I've setup a couple Firefox keyword bookmarks so I can type `so <query>` or `r <query>` respectively into my address bar as a shortcut.
Doing this has vastly improved troubleshooting, finding recipes or product recommendations, etc. It slices through SEO blogspam. I get genuine results, fast.
1) My experience is greater. Many of the problems that I encountered before, I know how to solve.
2) I know what knowledge I'm looking for if I don't know the answer. Instead of searching for the problem I am having (and ending up on SO), I am searching for the specific part of the documentation that I need for solving the problem that I have.
3) I tend to bias to project specific knowledge now. The general pollution of search on Stack Overflow means that it is more productive to search Spring documentation or the project specific forums ( https://github.com/spring-projects/spring-kafka/discussions ) rather than Stack Overflow itself.
4) Stack Overflow content is having difficulty with the lack of curation of old answers and the deceased quality of people answering now. Finding an answer written in recent times with a few variations on "have you tried {x}?" without the material leading to how that suggestion was derived means that trying to apply the answer is shotgun "maybe this works" without an understanding for how it works or if that is the right solution.
4b) I try to avoid the "have you tried {x}" answers as that impacts the rate I grow my experience (and thus part 1) at. Likewise, in general, people who try the shotgun solutions (and I can see that in code reviews if they're not good about cleaning up (and they aren't) tend to continue to make the same type of errors again as they are following cookbook / paint by numbers approaches rather than understanding some basic food science or aesthetic theory.
I love that GitHub added discussions to repos directly now. You can ask a question without getting berated by some SO mod about how your question makes no sense because you didn’t use the technically correct term for something.
There always those really weird esoteric bugs that someone has already slogged through lots of trial and error to resolve. SO is great for those kinds of problems.
Yep - if you're getting a particular error code or message that makes no sense, googling as often as not finds it in a SO post, along with 20 ways to solve it, of which at least 1 usually gives you what you need to do so. It's hard to imagine how much time I'd have spent tracking it down otherwise.
StackOverflow has become nearly useless because they don't mandate version numbers on questions and answers. I never start searches on SO and walk away empty handed probably 95% of the time I find my way there.
I was actually part of a study on improving that. They selected for some criteria that I'm not privy to, but part of it was being active on the site. Iirc only a few hundred meeting invites went out in total. Some of the questions I was asked were also geared towards validating the problem so they could present it to management and get more manpower on solving it.
Since then I've seen sorting by 'recently most voted' being used by default sometimes, but I still have to read up on the actual results. They're somewhere on meta SO or meta SE.
Part of the problem is the attitude that questions are duplicates of questions posted 10 years ago, so they get closed without answers. I am not sure recently most voted is enough. I think the culture needs to change substantially, maybe figure out some way to discourage closing as duplicate (cost N*5 points where N is the age of the supposed duplicate in months or years). Make it really expensive.
Maybe my attitude is wrong, but I've always felt the onus is on the Question asker to dictate that they've found the old answers and provide explanation to why they're no longer relevant to the problem. "I've already tried solutions from <x>,<y>, and <z> but they use functions depreciated in 2.1 and I'm using 3.0."
I do recognize the problem you're describing though. I think I've developed this mindset because I stopped helping out in StackOverflow a lot due to the low quality of many questions. I did a bit of time moderating with the intent to teach new users how to improve their questions, but ultimately the amount of users who want quick answers outpaces my patience so I just moved on. I follow two niche categories now that I consider myself an expert in and that's about all I help out with anymore.
Yeah, questioners should be encouraged to include version numbers. Ideally there would be a specific mandatory field for a version number. I'm honestly shocked there isn't one, it seems so obvious there's no way I'm the first to think of it.
Absolutely. Two of the big reasons are improved understanding and problem solving skills on my part and improvements in the tech stack I use day to day making things more clear and covering more use cases cleanly.
The third is that I’ve come to recognize that the average quality of answers on SO is quite poor — while they technically serve as solutions to posted problems, they often come with big caveats… use of private/deprecated APIs, hackiness, and feature misuse abound.
As such when I use SO these days it’s usually not for wholesale solutions to problems but rather to add to a greater body of examples of APIs in action which I can then abstract and use as needed. It’s decent for this use case.
> The third is that I’ve come to recognize that the average quality of answers on SO is quite poor — while they technically serve as solutions to posted problems, they often come with big caveats… use of private/deprecated APIs, hackiness, and feature misuse abound.
The real clean answer is usually a mildly upvoted comment that scolds the answerer for violating a standard or using a bad practice. SO is still useful when that happens.
That would be great. I work in a generalist type of field where learning new tools is part of the job so I am still googling the same stupid questions almost every day
I google the same git commands three times a day and end up on stackoverflow. Everything else is answered by the github issue comment with 10 emojis under it.
I find that in recent years my best resources are official docs because that's where I find the "obvious" answers to the issue I'm having by not having previously read the docs. I also find lots of answers in gh issues (often unresolved) where I gain insight into what might be happening so I can come up with a solution. SO answers that I find useful are usually ones which link to related docs.
Why is that in recent years. The best resources have always been thofficial docs and that has not changed since before SO. The issue is that official docs are often not good and you need pointers too them.
In fact I would say that official docs are getting worse and you need more help from SO now.
Agreed that github issues are very useful for understanding. But even there for popular reps there are too many stupid questions to wade through.
It probably depends on the technology you are using. For example, the Android conceptual guides are very good, but the javadoc / references are abysmal. You simply can't get anything done without SO because its more authoritative than the Android javadoc.
I can confirm. Also the technology used really counts: with Javascript I got weird error and there are no standard in the ecosystem so I found my self looking on Google and SO all the time. With Ruby, on the other hand, there are better documentation, no weird errors and everything seems so smooth that I visit StackOverflow just once a month. I have been working with JS for 3 years and just 1 year with Rub
I find myself using it equally much, but maybe in a different way. There are always times I find my self wanting to do something weird in a framework a specific framework or a language so I go see if there are already solutions out there. There are also times where I find an existing question but I’m not happy with the answers so I write my own after some research—which I would have hoped to have been able to skip—then my answer serves as a reference for the next time I need to do something similar (which may be as long as a year or two, so I won’t remember).
I didn't start using SO till I had already been a professional programmer for 20+ years, so I have no idea how I might have used it had it existed when I was starting out.
After 35 years as a professional programmer, I find SO invaluable. I might not use it with any kind of regularity, but when I need the kind of thing SO delivers, SO delivers.
One thing I haven't seen mentioned so far is the community/moderation. Those who do "visit" frequently are quite the clique, much like Wikipedians, with many of the same kind of outcomes. I used to spend my time helping people on there several times a week but I've been put off.
That being said, unix.stackexchange.com, is great.
Every time I search on stackoverflow it thinks I am a robot and wants me to solve a capture. My workaround is to use "site:stackoverflow.com" on some search search engine. But this definitely made me use SO way less.
No, haha, I forgot the opposite of `include?` in Ruby yesterday and ended up on SO which promptly told me it was `exclude?`. Which I've known forever. Haha.
I stopped answering questions after their policy changed which states that you don't own your answers, they do.
I figured, what's the point in using SO if there's code beggars plaguing it and SO staff isn't taking proper steps to prevent it from happening so I stopped using it alltogether.
It's not the place it once was after it was ran over by horde of code plumbers who can't look for similar answers.
My SO usage decreased dramatically once I switched from a simple text editor (Sublime) to an IDE that can resolve symbols and drill down into their definitions.
A lot of my SO usage used to be just looking up API documentation/syntax (and SO would rank higher than the framework's own documentation) which I no longer need to do if I can just Cmd+Click into the symbol and see its internal implementation.
I answered "Very Favorable" to "How favorable are you about blockchain, crypto, and decentralization?"
I am indeed very favorable indeed to decentralization.
I also think blockchains are almost always a waste of electricity by computers pointlessly playing Numberwang all day. (I'm not quite sure if "almost" is needed there.)
I agree this is a very unfortunate grouping of topics. The irony is that there’s a lot of cryptocurrency stuff that isn’t even well decentralized, so it’s almost an objectively bad grouping.
> I also think blockchains are almost always a waste of electricity by computers pointlessly playing Numberwang all day. (I'm not quite sure if "almost" is needed there.)
In relation to what? Do you have an alternative way of maintaining blockchains security?
Rails is boring tech [0] now. Phoenix seems to top the "wanted" listed, but it only has about a ~1000 respondents. If I had to guess I'd say js/typescript will eventually eat rails, django etc. If I started coding today, or indeed within the last 10 years, it's not clear why I would choose anything other than Javascript.
If you look at most loved languages, most wanted languages and most popular languages among both students and professionals Python and Typescript are the best choices for new stuff. Of the newer languages it seems Rust has the most momentum behind it.
Been a Rails dev for my whole career and at this point my love for it is slowly sliding away.
It’s the most amazing framework ever for the first few years but after the app starts to grow older and larger it becomes almost unworkable. It’s virtually impossible to update even the smallest gem and know you haven’t caused issues. A rails upgrade is impossible without 100% test coverage of everything and even then you have to cross your fingers that nothing slipped through.
The current place I work at is investigating pulling things out of rails and in to a Rust macro service so we can have some confidence that changes actually work.
The investigation out-withstanding, do you actually know of any also amazing framework that has been used for more than a few years (which I guess is already pretty long on average for web tech stuff) that performs better in this department?
I wonder what we would even fairly compare Rails to. Django? Laravel is already quite a bit young but sure, why not. I guess finding someone who actually has deep, year long and equivalent experience on the candidates would be pretty hard.
I suspect junior developers care a lot more about Stack Overflow than more experienced developers. When I first started my career, I looked at SO at least daily. Now (10 years later), I look at it maybe monthly when I'm searching for a solution to a weird library/config issue.
If that hunch is correct, it would explain why the data are so skewed towards people with less than 5 years of professional experience.
In my case I am not sure if it's just more experience or search engine degradation of the past 15 years.
I used to search for solutions a lot more, nowadays, with the amount of content farming and spammy blog content, I feel a lot more averse to any search query at all.
The threshold is raised such that I really need to be stuck to go online and I will usually end up in vendor forums at some endpoint.
When you see (once more) that despite your 20 years of experience you're being paid less that what students are being paid in other countries :-) (yes people, I know it's my fault that I'm still staying at this job... no need for downvotes or negative comments).
Depending on your geography, it might be a fair price to pay for universal healthcare, free daycare and university for your kids, as well as having a notion of evening and weekend.
I'm still looking for a golden tool to convert total compensation plus geography into expected happiness.
The average American pays 10% of his income on healthcare expenses. [1] That would mean it would be around $5,000/year. The US charges maybe $30,000 more for college than some cheaper countries. With two kids, that's maybe $1,500/year amortized over a career.
Those aren't going to be worth it for the average person unless you live in Norway, Luxembourg, or the UAE. I doubt there's any country where those outweigh the average programmer salary difference, let alone after taxes. Even if you have a serious health condition and crappy health insurance, deductibles cap out at $7,000/year.
This feels a little reductive of the benefits that come from public education and healthcare.
Developers making 300,000$ a year don't make up the entirety of the US population. I think living somewhere where everyone gets an opportunity at life contributes greatly to overall happiness and helps reduce social tensions.
Good urban design, a nationalized electricity grid, or train system also contribute a lot to the disparity in taxes and salaries.
But there are of course other issues you don't see too much in other western countries - such as:
- rampant homelessness problem
- getting your car broken into, home burglarized, robbed, etc.
- police brutality
- gun crime
- higher rate of traffic accidents
- larger distances to travel / commute
- abortion rights
- political diversity
and so on.
I'm obviously not saying that those are at all unique to the US, but many of those things are more frequent in the US than comparable west-European countries.
Of course, if you live in a very nice area - those are things you might never experience in the US (at least as far as crime goes).
The 4.86 years of experience is the mean of the 677 respondents who indicated student.
Going back to school after a few years is one thing... (and I lack the median, but I'm going to guess that it isn't too far from the mean) but having most students go back to school after 4 years of professional experience seems a bit odd.
I really suspect that they're either counting their 2 months of summer internship as a year several times over, misreading the professional part of it and including hobby, or misreading the professional and considering that they're a senior with 3 years of academic experience which translates 1:1 with professional, so they'll put that down.
The issue is that type of "something is fishy here" without further drilling into it makes me more skeptical of other data outliers and suspect more reporting and analysis issues than a (self reported) poll can be trusted.
There are probably crypto-fintech-medtech jobs out there that pay 500k but we'd never get to the salary part of the conversation because the whole idea feels shady. Security might also pay well but there are too many personalities involved in the domain for me to find out.
Those are probably 16-month (effectively fulltime) coop students working at big tech/finance. By senior year some of my classmates were making 10-12k/month.
My guess is that there’s some sampling bias for the student population - if you’re plugged in enough to be doing this survey you’ve probably been hacking for a while.
Fwiw, I started getting paid for slinging code at 16, maintained the website for a university group in college, while still doing other work in the summers. I probably would have answered with “6 years of experience” if asked when I graduated.
Is that truly “professional experience”? In retrospect probably not, but 22yo me would have been too stubborn to answer otherwise.
then it'd be weird to say after year that you have 1 year of commercial experience, when your colleagues spent 40 hours / week and have "only" 1 year too
I believe that pay is not necessary, but working for somebody else or in team (e.g OSS) is necessary.
So yea, I wouldn't say that hacking something even cool, but alone is commercial xp.
Is it possible to find out the average salary of a specific country?
In the survey it's only possible to switch between US, India, Germany, UK and Canada.
2: SQL and Bash both went up a percentage point! Yay! The art of terse and efficient code is not yet dying.
3: Strange that "Web frameworks" mixes backend and frontend frameworks. Why is "node.js" listed here? Isn't node.js a runtime?
4: Docker made a big jump from 49% to 64%. I consider that a good thing. I don't believe in "containerizing" applications. But I prefer to have Docker stay around for the long term. Because containers on their own are so darn useful.
5: Git is at 94%! Really? 94% of developers use Git? Or do they use some tool that is using abstracting Git away, like GitHub? Anyhow. Thats great, because Git (like vim) is one of the best pieces of software out there and I love that it will be around for a very long time.
This fizzbuzz competition specifically requires the solution to work up to 2^63 which is not possible as the largest int in JS is 2^54-1. You would have to use BigInt which would greatly reduce your performance.
I'm not a programmer just a pedant ... can't you just concatenate two integers one as your bigend and one as your little end. Sure, probably not fast, but slow !== impossible.
This has piqued my interest, I'm slightly familiar with C and Python, do you (or does anyone) know of any sort of review of resource optimisation for a simple program (like fizzbuzz)?
To me, naively, it seems like the bigend would fit in a register and so memory use wouldn't increase noticeably if you used some sort of double integer type. So it's it processor bound?
That's not native 64-bit integers. That's BigInt, and it's already out everywhere, and unless the situation has changed, it's significantly slower than just 64-bit integers would be.
I'm not actually attacking JavaScript for this, note. I don't think it's really all that much of a problem given the purposes of JavaScript, and I think it could really be better fixed by doing something similar to what Lua did, and have all numbers transparently either a 64-bit floating point number or a 64-bit signed integer. If you are using JavaScript and something like specifically needing 64-bit integers is an issue for you, there's a good chance you should just be using WebAssembly.
I have no idea how one does I/O to stdout in Node, so in the following I'm just going to assume that we have a function printi() that takes an integer argument and prints it, without and padding and without a newline, we have a function prints() that takes a string and prints it, and that we have a function nl() that prints a newline.
Here's a program that would print the numbers from 1 to 9999999999999999999, which is larger than 2^63, without using BigInt.
for (let r = 1; r < 1000; ++r) {
printi(r); nl()
}
for (let l = 1; l < 10000000000000000; ++l) {
for (let r = 0; r < 10; ++r) {
printi(l); prints('00'); printi(r); nl()
}
for (let r = 10; r < 100; ++r) {
printi(l); prints('0'); printi(r); nl()
}
for (let r = 100; r < 1000; ++r) {
printi(l); printi(r); nl()
}
}
That's not yet FizzBuzz but it could be made so by wrapping the print lines with a conditional check to see if they should be replace with Fizz, Buzz, or FizzBuzz. Just keep a variable around the is the current line number mod 15, and use it for the FizzBuzz logic check.
But first let's take a closer look at the "count to 9999999999999999999" program and see if it can be sped up. The first thing to notice is that each time through the outer loop it calls printi(l) 1000 times. Printing integers is often slow so calling printi 1000 times on the same l is not good.
Instead, we should print l to a string at the top of the outer loop, and then prints that string in the inner loops. I'll assume there is an itos() function that takes an integer and return a string.
We could also precompute all the right side strings.
let right = []
for (let r = 0; r < 10; ++r) {
right.push('00' + itos(r) + '\n')
}
for (let r = 10; r < 100; ++r) {
right.push('0' + itos(r) + '\n')
}
for (let r = 100; r < 1000; ++r) {
right.push(itos(r) + '\n')
}
So now the counting parts would look something like this:
for (let r = 1; r < 1000; ++r) {
printi(r); nl()
}
for (let l = 1; l < 10000000000000000; ++l) {
left = itos(l)
for (r = 0; r < 1000; ++r) {
prints(left + right[r])
}
}
Now add in the FizzBuzz logic and put it all together:
let n = 1 // current count % 15
let right = []
for (let r = 0; r < 10; ++r) {
right.push('00' + itos(r) + '\n')
}
for (let r = 10; r < 100; ++r) {
right.push('0' + itos(r) + '\n')
}
for (let r = 100; r < 1000; ++r) {
right.push(itos(r) + '\n')
}
for (let r = 1; r < 1000; ++r) {
if (n == 0) {
prints('FizzBuzz\n')
} else if (n % 3 == 0) {
prints('Fizz\n')
} else if (n % 5 == 0) {
prints('Buzz\n')
} else {
printi(r); nl()
}
if (++n == 15) n = 0
}
for (let l = 1; l < 10000000000000000; ++l) {
left = itos(l)
for (r = 0; r < 1000; ++r) {
if (n == 0) {
prints('FizzBuzz\n')
} else if (n % 3 == 0) {
prints('Fizz\n')
} else if (n % 5 == 0) {
prints('Buzz\n')
} else {
prints(left + right[r])
}
if (++n == 15) n = 0
}
That's going to be slower than a similar simple FizzBuzz that just goes to 2^54-1, but I don't think it would be a lot slower.
Yes, in theory it could print all the numbers you said, but in practice it won't because of floats dropping precision at some point. For JavaScript that point is 2^53.
The easiest way to see this is by typing `9007199254740992 + 1` in your browser's dev console. It should spit out `9007199254740992`.
`(Math.pow(2, 53) + 1) == Math.pow(2, 53)` returns true.
That means your program would print invalid output (or rather start looping infinitely) starting at that number (which is much smaller than 2^63) thus disqualifying your solution.
When the person I was responding to said the largest JS int is 2^54-1 I assumed that meant I could use all integers in [0, 2^54-1], but you are right that I can only go up 2^53. That breaks the specific code I posted but not the underlying idea. Fixing it to deal with that lower upper bound is easy.
Just change the constant 10000000000000000 in the outer loop to 1000000000000000 (which is 2^49+437050046578688 and well below 2^53), change the filling of the right[] array to
for (let r = 0; r < 10; ++r) {
right.push('000' + itos(r) + '\n')
}
for (let r = 10; r < 100; ++r) {
right.push('00' + itos(r) + '\n')
}
for (let r = 100; r < 1000; ++r) {
right.push('0' + itos(r) + '\n')
}
for (let r = 1000; r < 10000; ++r) {
right.push(itos(r) + '\n')
}
and change the loop that starts "for (let r = 1; r < 1000; ++r) {" to go to 10000 instead of 1000.
> That's going to be slower than a similar simple FizzBuzz that just goes to 2^54-1, but I don't think it would be a lot slower.
OK, I've actually tried it, after fixing the issue Allypost identified with the original code (JavaScript integer arithmetic only works up to 2^53, not the 2^54-1 I had assumed).
Here's a simple FizzBuzz:
let n = 1
for (let r = 1; r < 9990000; ++r) {
if (n == 0) {
console.log('FizzBuzz')
} else if (n % 3 == 0) {
console.log('Fizz')
} else if (n % 5 == 0) {
console.log('Buzz')
} else {
console.log(r)
}
if (++n == 15) n = 0
}
I compared that to the the more complicated one, with the outer loop changed to just go to 999, so it would produce the same output as the simple FizzBuzz. (And with the more complicated one changed to use console.log for output instead of the dummy functions in the original, itos() replaces by letting JavaScript implicitly do the conversion, and getting rid of explicit newlines since console.log works in lines.
Running each a few times via node with stdout directed to /dev/null, the simpler might be slightly faster but there is enough variation run to run that there is overlap. E.g., I got 23.811 vs 23.979 when I ran the two once, and then 24.325 and 25.550 when I ran them again.
Similar relative results when directing output to a file. There I actually saw in most runs the complex one run slightly faster than the simple one, but there was enough variation that I can't really say that one was on average faster than the other. The runs saving to disk took about 2.3x as long as the ones that discarded output.
None of them were anywhere near the speed of the programs in other languages on the code golf page linked a fews comments upthread. We're looking at about 5.4 MB/second when not saving the output to disk. That's around 1/20th the speed of the slowest on on the code golf page.
I'm just running things with "node file.js". I don't know if there are options that could be set to make it faster.
Anyway, the assertion upthread that JavaScript is not fast for this seems plausible, though not for the reason given (that it would need BigInt).
I don't think it's a surprise honestly. It's the language used for creating programs that can be shared with anyone who has a browser. You don't share the code or the software and tell your friend to run it. You share a link to a webpage and it runs there. It's the language that underpins the most shareable form of software out there.
Re 2: Bash can be terse, yes. Efficient? Depends. It can be efficient in terms of developer time to get something running. Efficient in terms of CPU time? Maybe not.
> 1: Amazing that Javascript is the most popular language, despite nobody has yet managed to write a fast FizzBuzz in it :)
The reason for that is that standard output implementation in Node.js is very slow. When the challenge is to write as fast as possible to standard output that's not ideal.
Regarding Git being 94%, if you see the table just below about interacting with the version control system, you'd see 83.57% users using command-line. 28.44% are using version control hosting service web GUI. 83.57% is still a big number of people who are using git with fewer abstractions on top.
For me, important differences are how the cursor behaves when moving around, and the effect of backspace. I like consistent movements, i.e. always one character position per keypress, and therefore prefer spaces.
Udemy coming in at 66% is pretty surprising to me. I personally purchased some 50 courses on there, and am happy with the result (especially owning the course and not having to deal with a subscription), but it is surprising to me that it basically won that market. It isn't all that friendly to course maker iirc. And the course quality varies a lot. Its pricing model is also very shady, giving me the impression that it is a poorly run company.
Of course - youtube is not included; I imagine that'll really skew the graph
edit: shoot, I just checked and I actually purchased 98 courses.
Yeah, though I thought Udemy had the least favorable reputation just because of the weird pricing tactics and a lot of courses being glorified tutorials. There's some great stuff on there, but I didn't think it would rank that much higher than Coursera.
Interesting as always. It would be great to have a UX to easily compare the trends, such as 2022 vs 2021 vs 2020 though. Unless I missed it.
The "Web framework and technologies" section doesn't seem accurate to compare in a meaningful way.
For example, they have Node.js as a standalone option probably throwing off everything. Then these compare weird when put together as options: Next, Nuxt, React, Vue, etc.
I love Svelte and maintain some open source libraries for it, but in recent job search I was struck by a near total lack of jobs mentioning it being used.
Surprising for me: AWS's market share is still Azure + GCP combined amongst professional developers [1]. With various articles about Microsoft's success I was expecting lower. The same for loved / dreaded. AWS scores significantly better [2] in terms of loved over other cloud platforms. Being in the HN bubble I was expecting that to be lower also. The other items that scored high on loved / dreaded were mostly new technologies vs. dominant incumbents.
Regarding Azure, it is quite possible the primary developers are corporate/in-house types and not big on responding to this particular survey. About 70,000 developers (in total) responded which is a small fraction of the developer universe.
As an Android developer since 2009, I will not touch anything from Google ever again. Its a very asymmetrical relationship, and if something goes wrong, like your account gets suspended, its a Kafkaesque nightmare, with usually no recourse.
Sideloading is neither here nor there. Its about ruining a developer's livelihood, you know... like your account gets suspended (for right or wrong reasons), and ability to earn income stops. Not just because you can't distribute your app on Play Store, but because your ability to work for another organization is blunted because your addition to their account places them at risk too.
Can you explain further? It sounds like you're saying that if I work at company A, company A's Play store account gets banned, and then I go work at company B, that the Play store keeps track of this and then company B's account might get banned, just because I moved there.
There have been reports that if an individual's account has been banned, they can "poison" another company's account by being added to it. I have not heard of this happening across companies.
Happened to me, my account was banned because I was using it from a restricted country (I used a VPN), for some time it worked fine, then after that ban hammer, all other accounts that used my machine later on got banned. I recall that I read online that they track your MAC address that gets attached to the build artifacts when you upload them to the store and used to link the accounts together, it's even mentioned in their banning email "Your account is associated with another banned account...".
The web standard is powerful enough that native apps are unnecessary for most use cases. Hopefully the gap will be closed further in the next decade. API and documentation is better. Distribution is simpler and cheaper. Not having to interact with the Apple / Google monopolists is the cherry on top.
Common trope which if you'd actually read through any real-world data, is nowhere near to be true.
One basic example: https://jobs.braveclojure.com/, five pages of jobs that are all about Clojure (and some others languages + Clojure)
Not to mention when you're running a Clojure company and you go out to find candidates, most of them are really good and you won't have to look for very long, as long as you can compensate them well, as the competition between companies (for candidates) is fierce.
And few clojure shops will require previous experience with clojure. The ones I’ve worked on have all welcomed engineers new to the language, which on its own is a refreshing approach to hiring. I don’t think highly of companies that hire at the level of specificity of “X years in Y technology”.
Good engineering principles translate across stacks.
Many people using clojure were already senior level engineers. Few jump into clojure (or elixir or some of those other top languages) straight out of college. They make conscious choices to introduce these technologies to their companies or to seek out companies that utilize them, which implies they’re at a level where they’re able to exert more influence.
I've been a professional Clojure developer for some time now. My main observation when it comes to this is that Clojure is often picked to solve particular problems based on the capabilities it offers over other languages very deliberately. I've seen quite a few cases where an organization had attempted to solve a really high value problem with a more mainstream language, realized that it was next to impossible to succeed that way, and then tried Clojure with great results. In short, oftentimes Clojure makes it easy to solve really difficult problems and expert Clojure developers are in short supply.
Some examples of problems that I've seen Clojure excel in solving:
1. Complex rule modeling
2. Optimization systems
3. Big Data pipelines
4. Highly concurrent systems
5. Streaming systems
My hypothesis is that Clojure doesn't get used outside of the US, and so the median salary isn't pulled down by the comparatively lower salaries outside of the US
Tons of Clojure in Europe. Especially Northern Europe. Following the job boards and Clojurians Slack channel's #jobs channels I figure it's about 50/50 for US and EU.
I'm not knowledgeable of Norway, but I know of Sweden and Finland as having quite a few Clojure opportunities. Now keep in mind, "quite a few" is relative to the entire Clojure job market, which is very tiny.
I don't remember last years' but this was pretty mediocre indeed. E.g. after the first two graphs, I used autoscroll (middle click, move mouse) so that I 'saw' the whole page and all the graphs loaded and animations triggered. Then ctrl+up and read more comfortably.
How stackoverflow builds this in when flashing content is such a common annoyance, I don't know. Sure, you don't want to load a megabyte of data for when people open the page for just 2 seconds, but there are animations on top of the lazy loading, the actual data is not that large anyway, and the loading distance is not hard to increase either.
Breaking cmd+f functionality will immediately sour me on a site. It's surprising how many tech companies go out of their way to ruin basic behavior like that.
Unsurprising, really. The usefulness of it cannot be overstated.
Doesn't even matter if its used anywhere in the deployment chain; simply having the ability to pull up a replica of almost any *nix environment on my laptop in mere seconds, using it for tests, and then throwing it away resetting it again in mere seconds is beyond awesome, no matter if what I'm working on then goes into a huge complicated deployment chain, or is shoved onto some on-premises, zero-abstractions, baremetal server.
And how is it all configured? Plain text files. How is it controlled? Command Line. Meaning I can script it every way I want, using the tools I already have and use. Doesn't get in the way, doesn't demand that I work around it...it works with me and my tools in the same way they already work together.
Oh, and of course, good bye and good riddance to the days when I had to install and configure local RDMS for tests. Everything I use has an official image, so I just write some small setup script, a Dockerfile, knit everything together in a docker-compose.yml and presto, done: Application stack is up and running.
To me, Docker is as essential as my text editor these days.
And yet here I am having used Docker like 3 times. I've used Heroku, Nix, or even just versions through asdf if necessary and been fine—though Nix is the only truly reproducible option here. None of these options involved containers or overhead.
> ... unless you're running Windows containers (on Windows).
But aren't most Windows containers a bit on the heavier side? And don't you then need to also use the whole MS server setup for deploying your stuff to prod, which is a no-no in certain settings?
Edit: provided that you can even find an image for the software you need (from an official provider/latest versions/with proper instructions and source). Consider the following:
Then again, WSL2 is pretty okay for running *nix based OCI containers, apart from the file permissions (SSH keys and anything like that is a pain, especially with bind mounts).
Even a Hyper-V VM was a decent choice, though any sort of a performance overhead was also negligible - I've heard the story being worse on OS X in regards to disk performance, though not sure whether that's still relevant. The worst thing about Docker on Windows has generally been the weird bind mount syntax for the Windows file system paths (not too bad, to be honest) as well as the whole file permission thing, as well as the Hyper-V approach eating some of your RAM in the background.
Apart from that, it's mostly passable, though Docker/Podman on *nix is comparatively painless. Though I could say that about most development ecosystems, from PHP to Java. Windows is just generally better for certain classes of desktop software and gaming, *nix is generally better for most development related tasks and servers. /opinion
In all honesty, I've used Docker Desktop on WSL2 quite a bit too and the overhead is very minimal (additional memory consumption mostly).
Avoiding a transformational technology like containers (OCI is used by multiple implementations now, like containerd, podman etc) because of this seems a little silly to me.
> oh, and having to deal with docker desktop is not fun
To stay sane, just forget it exists until it crashes irretrievably, then reinstall it and repeat. Containers and images being disposable is the whole point anyway.
You don't even have to bother with Docker Desktop these days, you can start it in `/etc/wsl.conf` directly with `service start docker` and forget about it, it'll start with the VM. I think this is a rather new feature though.
Seriously. Look at how much blowback $10/month for Github Copilot got. $10! I don't know everyone's financial status, but from where I'm sitting, $10/month for that is cheap.
It's about the value, not the cost. You can get VSCode for free, or JetBrains Professional IDEs for about $10/month (after 2nd year). People just don't agree that Copilot has value comparable to those yet.
All the comments I've read complaining about the price were saying basically that. One even explicitly said that if it was cheaper he'd buy it for fun, but since it provided no value, he wouldn't... just to get a thread of people complaining that the price shouldn't change his opinion.
Back of the envelope this though. Lets say a programmer makes $60/hr, or $120k/yr. Lets say Copilot replaces 5 minutes of googling around to find the right stack overflow answer and copy and paste and adapt it to fit. Copilot just has to do that twice in a whole month to be worth $10/month.
I don't disagree, but as I said, it's about comparative value to other tools. Any first world developer can afford $10/month.
That being said, if you are a full time employee you won't magically earn an extra $10 a month for using Copilot, you will only be more productive. It will benefit your employer and Copilot doesn't have an enterprise subscription yet.
I think it’s largely because copilot is hardly worth it to use for free. I found it to provide negative value and waste my time with its correct looking but ultimately incorrect suggestions.
It is surprising how every other profession manages to pay for their work tools, only specific classes of software devs are so much against paying for anything, yet they will gladly have a means to also pay their own bills.
For me, it's due to the terrible process everywhere I've worked for getting approval to buy anything. In some cases, it's even against company policy to pay for a tool out of my own pocket. One product I needed was $10 but the person who was put in charge of all software purchasing decided to try to negotiate a site license with the vendor for a lower price and only for each seat used. She was basically trying to haggle over $10, on the off chance at some point in the future more people would want to use the tool and then the company could pay $8 or some other amount less than $10 per seat. It took months and the involvement of my manager, his manager, and a C level executive to get the $10 purchase approval to go through. I would have paid the $10 myself but that was a "zero tolerance" fireable offense.
I get why companies are concerned about having improperly licensed software on their machines, as the consequences can be great, but too many have gone too far in the opposite direction, making purchasing anything a long and convoluted process.
> Unsurprising, really. The usefulness of it cannot be overstated.
Furthermore, in addition to the points you mentioned, it also lets us:
- trivially remap ports to whatever we want (and expose whichever we'd like to the outside)
- easily set resource limits, so your instance of MariaDB doesn't bring the whole server down and make it unresponsive
- abstract away storage, in case you want a specific directory for your backups instead of following HFS (say, just have /app, treating the rest of the server as throwaway)
- perhaps most importantly, your host OS is now separate from the actual containers that you are running, updating/redeploying either becomes a breeze
Recently, I actually decided to build my own containers for all of my personal use cases and it's been an interesting experience: I base everything on a LTS version of Ubuntu and just use apt for getting all of the runtimes (Node, Java, .NET, PHP, Ruby, Python, ...) for my own software, and so far it's a nice experience.
All I need is the base Ubuntu image from Docker Hub and the rest is up to me and the regular mirrors/repositories for the software packages in question, most of which can be stored on my own Nexus instance as needed, as well as the whole build process is primarily driven through Gitea, Drone CI and a few "servers" (repurposed old computers with passive cooling) that I have on my desk.
Though for now I also use Bitnami images for databases and such, which are also decent and which I largely just cache on my end: https://bitnami.com/stacks/containers
So what I'm trying to say, is that there's a lot of flexibility that you can enjoy, both in making your own "templates" for web servers, programming languages, build toolchains etc., as well as you can grab pre-made stuff that other people have made, be it on Docker Hub, someone's Nexus/Artifactory/Harbor instance or another registry out there.
Docker and other OCI compatible tools have largely achieved the sort of adoption and widespread usage that projects like Nix and Guix could also benefit from.
I find for builds especially it's invaluable. I have a Makefile that builds Docker images and runs build processes for all 4 supported versions of Debian, 3 LTS versions of Ubuntu, RHEL7, and Fedora 35 and 36. It builds and packages the software on all of these, giving me debs and rpms for every relevant distro ("relevant" meaning distros used for servers or workstations for any of our employees and clients).
I then have another set of docker images for making package repositories and signing everything.
I do have a lot of mostly-redundant dockerfiles in some places, but in others I've managed to leverage m4 to reduce the redundancy (though I'm trying to keep it as slim as possible to avoid the pain of turning everything into a convoluted set of impossible-to-maintain macros).
Before this, I was using Vagrant and Ansible for builds, which was slow, memory hungry, and frustrating to debug. Not to mention that dependencies that needed to be built statically couldn't be easily cached, which comes out of the box with Docker.
I sorta learned a little about Docker+Kubernetes on a previous project, and it was a nightmare. I don't have a good mental delineation between the two products because the whole thing was a trainwreck, though I'm inclined to think that 95% of the horribleness was K8s.
That said, I also don't remember the Docker documentation to be very good. For someone that doesn't work with it professionally, what's a good starting point to learn Docker well?
Though k8s works with docker, you shouldnt need to use it off the bat, especially in the beginning without having lots of experience with docker itself. Most projects dont need k8s and i personally have never used it.
I agree the docker docs are lacking, i basically learned everything by googling stuff, stack overflow and reading some blog posts
Are you sure you need to learn Docker well? I ask this because I use docker on a daily basis, but I rarely actually “use” docker. What I mean by this is that we deploy everything using docker, but it’s handled by our DevOps pipeline and the “docker” part is really just a dockerfile that is typically given to us by our cloud vendor. I think the only thing I’ve changed in ours for nodejs, c#, Python and go images is the image version they get build with.
Lets say I want to write a typescript microservice and deploy it to azure as a serverless functions app. I’ll fork our bare metal nodejs project for azure functions, which is essentially the standard azure cli “create nodejs function -typescript -docker” (this isn’t correct syntax but you get it) with the linting and ts-compiling rules we use on all our projects (and an updated image version in the docker file as mentioned). While I build things and run them locally, I don’t use docker, it’s not until I actually want to deploy to azure and setup the release pipeline and trigger it that docker comes into play, but those DevOps steps (also streamlined) aren’t really docker heavy as they simply use the dockerfile that was mostly provided by Microsoft.
I can certainly build and run my micro service as a docker container locally, but I don’t need to. In fact the only times I did was when it failed to build during the azure pipeline but it turned out to be the azure container registry access controls every single time that happened, so these days I almost never “use” docker. In fact I use it so rarely I almost always have to Google command lines.
I know people use docker in many different ways, and that many use things like docker compose, but my point is that you can deploy everything you build with docker and never actually have to care much about docker itself. So maybe you should ask yourself if you really need to learn docker good before you spend too much time on it.
Use docker compose, it's simply composing docker run parameters as a configuration file. What I think game changing feature is docker exec -it, enable you to connect and execute commands inside container and debug from it.
One of my updated “Joel Test” job requirements is “do you give each developer access to a dev cloud account with fairly wide guardrails.”
I would much rather just spin up all my resources on a dev cloud account using CloudFormation/Terraform, and spend them down when I’m done.
Before I get (rightfully) called out. Yes I work at AWS now. But I also found that just as appealing three years ago when I worked at a 60 person company.
> simply having the ability to pull up a replica of almost any *nix environment on my laptop
Umm, I struggle with that. I have Ubuntu machines that run in production. They are VMs that run systemd for some stuff and Docker containers for others. The only way to (easily) replicate such machines on my laptop is via VMs (e.g., Vagrant + VMware), not via Docker only.
This is the opposite of my experience. Docker is a headache to work with for zero tangible benefit. It slows development down by adding an unnecessary layer to the stack.
You didn't write anything about your experience or your knowledge level, so it's impossible to conclude anything useful or tangible about your experience. Wouldn't you agree?
I’ve worked with full stack Rails for 10+ years. Built and sold two startups of my own, and wrote the code for two others that were acquired, among other things. I have a ton of practical experience on the topic.
Makes sense to me (but I'm of course biased) due to selection bias. My impression is that those for whom APL clicks, find the day-to-day APL experience very enjoyable, while those that just don't get it will tend to leave APL behind. Probably not so for more widely-used languages, where people will stay with them, even if it means daily frustration.
I was under the impression Solidity was a dumpster fire language. For example: https://news.ycombinator.com/item?id=14810008. It ranks surprisingly high. Is this riding off of Ethereum’s popularity? Perhaps in combination with the skewed sample (towards young/beginning devs).
> A majority of respondents (75%) have been working for 14 or fewer years as a professional developer, meaning they’ve never worked in a world without Stack Overflow.
Even worse -- the horrors of having to find a book in the library to answer your question.
When I started working all the senior devs had full bookshelves of O'Reilly books, and if you got stuck you went to them to either ask a question or use their books. Then as they left the company, all of us would fight over who got ownership of the O'Reilly book collection!
"Fully remote" is the most common work environment! Wasn't expecting that, even considering covid. I wonder if employers will pressure staff to return to the office (e.g. hybrid) and fully remote will lose that number 1 spot next year. Time will tell. https://survey.stackoverflow.co/2022/#section-employment-wor...
374 comments
[ 3.0 ms ] story [ 248 ms ] threadThese days, the only time I drop into SO is when a "how do you do that again?" search yields a relevant-looking question asked within the last year or so. As often as not, the search leads me to a blog post or primary source (e.g. mozilla javascript documentation). But when I do make it to SO, I find that it's still a bustling community full of legitimate expert guidance.
See for example “geeksforgeeks” and similar sites, which I find to be a poor and overly verbose resource, which now dominates the top position
Google Search's and Facebook Messenger's spam filters started letting tons of garbage through at roughly the same time. I wonder if the advances in generative deep learning helped with that.
Doing this has vastly improved troubleshooting, finding recipes or product recommendations, etc. It slices through SEO blogspam. I get genuine results, fast.
1) My experience is greater. Many of the problems that I encountered before, I know how to solve.
2) I know what knowledge I'm looking for if I don't know the answer. Instead of searching for the problem I am having (and ending up on SO), I am searching for the specific part of the documentation that I need for solving the problem that I have.
3) I tend to bias to project specific knowledge now. The general pollution of search on Stack Overflow means that it is more productive to search Spring documentation or the project specific forums ( https://github.com/spring-projects/spring-kafka/discussions ) rather than Stack Overflow itself.
4) Stack Overflow content is having difficulty with the lack of curation of old answers and the deceased quality of people answering now. Finding an answer written in recent times with a few variations on "have you tried {x}?" without the material leading to how that suggestion was derived means that trying to apply the answer is shotgun "maybe this works" without an understanding for how it works or if that is the right solution.
4b) I try to avoid the "have you tried {x}" answers as that impacts the rate I grow my experience (and thus part 1) at. Likewise, in general, people who try the shotgun solutions (and I can see that in code reviews if they're not good about cleaning up (and they aren't) tend to continue to make the same type of errors again as they are following cookbook / paint by numbers approaches rather than understanding some basic food science or aesthetic theory.
Since then I've seen sorting by 'recently most voted' being used by default sometimes, but I still have to read up on the actual results. They're somewhere on meta SO or meta SE.
I do recognize the problem you're describing though. I think I've developed this mindset because I stopped helping out in StackOverflow a lot due to the low quality of many questions. I did a bit of time moderating with the intent to teach new users how to improve their questions, but ultimately the amount of users who want quick answers outpaces my patience so I just moved on. I follow two niche categories now that I consider myself an expert in and that's about all I help out with anymore.
The third is that I’ve come to recognize that the average quality of answers on SO is quite poor — while they technically serve as solutions to posted problems, they often come with big caveats… use of private/deprecated APIs, hackiness, and feature misuse abound.
As such when I use SO these days it’s usually not for wholesale solutions to problems but rather to add to a greater body of examples of APIs in action which I can then abstract and use as needed. It’s decent for this use case.
The real clean answer is usually a mildly upvoted comment that scolds the answerer for violating a standard or using a bad practice. SO is still useful when that happens.
Agreed that github issues are very useful for understanding. But even there for popular reps there are too many stupid questions to wade through.
But now I can't remember the last time I visited SO and I stopped asking questions many years ago.
This is true even when learning new languages and other things.
- Documentation seems to have gotten way better overall.
- Github issues are often a good place to search for open/resolved problems.
- I've simply gotten better at reading and debugging code over time.
- There are often chat rooms for larger communities on Slack/Discord nowadays. You get a more fluent and direct form of communication there.
- I still sometimes reach for Stack Overflow. But it tends to be for things where I'm a total newbie.
After 35 years as a professional programmer, I find SO invaluable. I might not use it with any kind of regularity, but when I need the kind of thing SO delivers, SO delivers.
That being said, unix.stackexchange.com, is great.
12 years old account on SO.
Haven't used SO for anything for past 7 years.
I stopped answering questions after their policy changed which states that you don't own your answers, they do.
I figured, what's the point in using SO if there's code beggars plaguing it and SO staff isn't taking proper steps to prevent it from happening so I stopped using it alltogether.
It's not the place it once was after it was ran over by horde of code plumbers who can't look for similar answers.
A lot of my SO usage used to be just looking up API documentation/syntax (and SO would rank higher than the framework's own documentation) which I no longer need to do if I can just Cmd+Click into the symbol and see its internal implementation.
I am indeed very favorable indeed to decentralization.
I also think blockchains are almost always a waste of electricity by computers pointlessly playing Numberwang all day. (I'm not quite sure if "almost" is needed there.)
If my comment does get into a hall of fame, please edit out the repeated "indeed".
In relation to what? Do you have an alternative way of maintaining blockchains security?
(2) Those YoY salary increases are bonkers
(3) Poor Ruby on Rails. Seems to be slowly sliding into irrelevance in the face of stiff competition.
[0]: https://mcfunley.com/choose-boring-technology
It’s the most amazing framework ever for the first few years but after the app starts to grow older and larger it becomes almost unworkable. It’s virtually impossible to update even the smallest gem and know you haven’t caused issues. A rails upgrade is impossible without 100% test coverage of everything and even then you have to cross your fingers that nothing slipped through.
The current place I work at is investigating pulling things out of rails and in to a Rust macro service so we can have some confidence that changes actually work.
I wonder what we would even fairly compare Rails to. Django? Laravel is already quite a bit young but sure, why not. I guess finding someone who actually has deep, year long and equivalent experience on the candidates would be pretty hard.
- almost 33% of people have programmed professionally for less than 4 years
- just over 27% have programmed professionally between 5 and 9 years.
If that hunch is correct, it would explain why the data are so skewed towards people with less than 5 years of professional experience.
I used to search for solutions a lot more, nowadays, with the amount of content farming and spammy blog content, I feel a lot more averse to any search query at all.
The threshold is raised such that I really need to be stuck to go online and I will usually end up in vendor forums at some endpoint.
I'm still looking for a golden tool to convert total compensation plus geography into expected happiness.
Those aren't going to be worth it for the average person unless you live in Norway, Luxembourg, or the UAE. I doubt there's any country where those outweigh the average programmer salary difference, let alone after taxes. Even if you have a serious health condition and crappy health insurance, deductibles cap out at $7,000/year.
[1]: https://www.advisory.com/daily-briefing/2019/05/02/health-ca...
Developers making 300,000$ a year don't make up the entirety of the US population. I think living somewhere where everyone gets an opportunity at life contributes greatly to overall happiness and helps reduce social tensions.
Good urban design, a nationalized electricity grid, or train system also contribute a lot to the disparity in taxes and salaries.
- rampant homelessness problem
- getting your car broken into, home burglarized, robbed, etc.
- police brutality
- gun crime
- higher rate of traffic accidents
- larger distances to travel / commute
- abortion rights
- political diversity
and so on.
I'm obviously not saying that those are at all unique to the US, but many of those things are more frequent in the US than comparable west-European countries.
Of course, if you live in a very nice area - those are things you might never experience in the US (at least as far as crime goes).
The lowest years of experience of actual professionals belongs to... blockchain developers with 9.63 years (on average) of professional experience.
https://survey.stackoverflow.co/2022/#section-experience-yea...
BTW many people go (back) to school after obtaining some professional experience. Not sure about the context of this part of the poll though.
Going back to school after a few years is one thing... (and I lack the median, but I'm going to guess that it isn't too far from the mean) but having most students go back to school after 4 years of professional experience seems a bit odd.
I really suspect that they're either counting their 2 months of summer internship as a year several times over, misreading the professional part of it and including hobby, or misreading the professional and considering that they're a senior with 3 years of academic experience which translates 1:1 with professional, so they'll put that down.
The issue is that type of "something is fishy here" without further drilling into it makes me more skeptical of other data outliers and suspect more reporting and analysis issues than a (self reported) poll can be trusted.
You've heard of "5 years of experience" entry level jobs, now meet the "5 years of experience" students
https://survey.stackoverflow.co/2022/#salary-canada
It's their intern salary projected on a yearly basis so ~ $9737/month.
Fwiw, I started getting paid for slinging code at 16, maintained the website for a university group in college, while still doing other work in the summers. I probably would have answered with “6 years of experience” if asked when I graduated.
Is that truly “professional experience”? In retrospect probably not, but 22yo me would have been too stubborn to answer otherwise.
e.g if you've been working 4h a week,
then it'd be weird to say after year that you have 1 year of commercial experience, when your colleagues spent 40 hours / week and have "only" 1 year too
I believe that pay is not necessary, but working for somebody else or in team (e.g OSS) is necessary.
So yea, I wouldn't say that hacking something even cool, but alone is commercial xp.
That's because I've been working full time since 2nd semester
But I still do believe that it's weird.
Maybe people think about experience in general instead of commercial, full time?
1: Amazing that Javascript is the most popular language, despite nobody has yet managed to write a fast FizzBuzz in it :)
https://codegolf.stackexchange.com/questions/215216/high-thr...
2: SQL and Bash both went up a percentage point! Yay! The art of terse and efficient code is not yet dying.
3: Strange that "Web frameworks" mixes backend and frontend frameworks. Why is "node.js" listed here? Isn't node.js a runtime?
4: Docker made a big jump from 49% to 64%. I consider that a good thing. I don't believe in "containerizing" applications. But I prefer to have Docker stay around for the long term. Because containers on their own are so darn useful.
5: Git is at 94%! Really? 94% of developers use Git? Or do they use some tool that is using abstracting Git away, like GitHub? Anyhow. Thats great, because Git (like vim) is one of the best pieces of software out there and I love that it will be around for a very long time.
So not impossible but kinda pointless for JS
To me, naively, it seems like the bigend would fit in a register and so memory use wouldn't increase noticeably if you used some sort of double integer type. So it's it processor bound?
[0] https://chromestatus.com/feature/5371603852460032
I'm not actually attacking JavaScript for this, note. I don't think it's really all that much of a problem given the purposes of JavaScript, and I think it could really be better fixed by doing something similar to what Lua did, and have all numbers transparently either a 64-bit floating point number or a 64-bit signed integer. If you are using JavaScript and something like specifically needing 64-bit integers is an issue for you, there's a good chance you should just be using WebAssembly.
I have no idea how one does I/O to stdout in Node, so in the following I'm just going to assume that we have a function printi() that takes an integer argument and prints it, without and padding and without a newline, we have a function prints() that takes a string and prints it, and that we have a function nl() that prints a newline.
Here's a program that would print the numbers from 1 to 9999999999999999999, which is larger than 2^63, without using BigInt.
That's not yet FizzBuzz but it could be made so by wrapping the print lines with a conditional check to see if they should be replace with Fizz, Buzz, or FizzBuzz. Just keep a variable around the is the current line number mod 15, and use it for the FizzBuzz logic check.But first let's take a closer look at the "count to 9999999999999999999" program and see if it can be sped up. The first thing to notice is that each time through the outer loop it calls printi(l) 1000 times. Printing integers is often slow so calling printi 1000 times on the same l is not good.
Instead, we should print l to a string at the top of the outer loop, and then prints that string in the inner loops. I'll assume there is an itos() function that takes an integer and return a string.
We could also precompute all the right side strings.
So now the counting parts would look something like this: Now add in the FizzBuzz logic and put it all together: That's going to be slower than a similar simple FizzBuzz that just goes to 2^54-1, but I don't think it would be a lot slower.The easiest way to see this is by typing `9007199254740992 + 1` in your browser's dev console. It should spit out `9007199254740992`. `(Math.pow(2, 53) + 1) == Math.pow(2, 53)` returns true.
That means your program would print invalid output (or rather start looping infinitely) starting at that number (which is much smaller than 2^63) thus disqualifying your solution.
Just change the constant 10000000000000000 in the outer loop to 1000000000000000 (which is 2^49+437050046578688 and well below 2^53), change the filling of the right[] array to
and change the loop that starts "for (let r = 1; r < 1000; ++r) {" to go to 10000 instead of 1000.OK, I've actually tried it, after fixing the issue Allypost identified with the original code (JavaScript integer arithmetic only works up to 2^53, not the 2^54-1 I had assumed).
Here's a simple FizzBuzz:
I compared that to the the more complicated one, with the outer loop changed to just go to 999, so it would produce the same output as the simple FizzBuzz. (And with the more complicated one changed to use console.log for output instead of the dummy functions in the original, itos() replaces by letting JavaScript implicitly do the conversion, and getting rid of explicit newlines since console.log works in lines.Running each a few times via node with stdout directed to /dev/null, the simpler might be slightly faster but there is enough variation run to run that there is overlap. E.g., I got 23.811 vs 23.979 when I ran the two once, and then 24.325 and 25.550 when I ran them again.
Similar relative results when directing output to a file. There I actually saw in most runs the complex one run slightly faster than the simple one, but there was enough variation that I can't really say that one was on average faster than the other. The runs saving to disk took about 2.3x as long as the ones that discarded output.
None of them were anywhere near the speed of the programs in other languages on the code golf page linked a fews comments upthread. We're looking at about 5.4 MB/second when not saving the output to disk. That's around 1/20th the speed of the slowest on on the code golf page.
I'm just running things with "node file.js". I don't know if there are options that could be set to make it faster.
Anyway, the assertion upthread that JavaScript is not fast for this seems plausible, though not for the reason given (that it would need BigInt).
It's almost like "Best performance of all languages!" is not that important of an requirement for a language to become popular :)
In this context this almost always means node.js with express
I think this means that 94% of the people using Git have questions!!
IMHO, this means that a simple Distributed Version Control System is really required.
Right, it would be nice to have a simple one.
The reason for that is that standard output implementation in Node.js is very slow. When the challenge is to write as fast as possible to standard output that's not ideal.
Was it because of legacy editors or what?
Editor now helps formatting code, even when pressing enter button, it aligns entry on new line.
Of course - youtube is not included; I imagine that'll really skew the graph
edit: shoot, I just checked and I actually purchased 98 courses.
The "Web framework and technologies" section doesn't seem accurate to compare in a meaningful way.
For example, they have Node.js as a standalone option probably throwing off everything. Then these compare weird when put together as options: Next, Nuxt, React, Vue, etc.
Love it or hate it, jQuery was also #3.
First of all JS technologies. Second overall.
1 - https://survey.stackoverflow.co/2022/#most-popular-technolog...
2 - https://survey.stackoverflow.co/2022/#most-loved-dreaded-and...
The only advantage for Android is that you can theoretically also just distribute via sideloading
https://www.androidpolice.com/google-terminate-personal-acco...
Any ideas around why that could be?
One basic example: https://jobs.braveclojure.com/, five pages of jobs that are all about Clojure (and some others languages + Clojure)
Not to mention when you're running a Clojure company and you go out to find candidates, most of them are really good and you won't have to look for very long, as long as you can compensate them well, as the competition between companies (for candidates) is fierce.
Good engineering principles translate across stacks.
Some examples of problems that I've seen Clojure excel in solving: 1. Complex rule modeling 2. Optimization systems 3. Big Data pipelines 4. Highly concurrent systems 5. Streaming systems
For comparison:
Ruby - The meetup group of the tiny Ruby community (confirmed by experience) has 329 members: https://www.meetup.com/ruby-meetup-oslo/
Imba - The meetup group of the niche full-stack web programming language Imba, has 134 members. https://www.meetup.com/imba-oslo-meetup/
I tried looking for word with CMD+F and the section I was looking at completely disappeared!
How stackoverflow builds this in when flashing content is such a common annoyance, I don't know. Sure, you don't want to load a megabyte of data for when people open the page for just 2 seconds, but there are animations on top of the lazy loading, the actual data is not that large anyway, and the loading distance is not hard to increase either.
Unsurprising, really. The usefulness of it cannot be overstated.
Doesn't even matter if its used anywhere in the deployment chain; simply having the ability to pull up a replica of almost any *nix environment on my laptop in mere seconds, using it for tests, and then throwing it away resetting it again in mere seconds is beyond awesome, no matter if what I'm working on then goes into a huge complicated deployment chain, or is shoved onto some on-premises, zero-abstractions, baremetal server.
And how is it all configured? Plain text files. How is it controlled? Command Line. Meaning I can script it every way I want, using the tools I already have and use. Doesn't get in the way, doesn't demand that I work around it...it works with me and my tools in the same way they already work together.
Oh, and of course, good bye and good riddance to the days when I had to install and configure local RDMS for tests. Everything I use has an official image, so I just write some small setup script, a Dockerfile, knit everything together in a docker-compose.yml and presto, done: Application stack is up and running.
To me, Docker is as essential as my text editor these days.
But aren't most Windows containers a bit on the heavier side? And don't you then need to also use the whole MS server setup for deploying your stuff to prod, which is a no-no in certain settings?
Edit: provided that you can even find an image for the software you need (from an official provider/latest versions/with proper instructions and source). Consider the following:
Then again, WSL2 is pretty okay for running *nix based OCI containers, apart from the file permissions (SSH keys and anything like that is a pain, especially with bind mounts).Even a Hyper-V VM was a decent choice, though any sort of a performance overhead was also negligible - I've heard the story being worse on OS X in regards to disk performance, though not sure whether that's still relevant. The worst thing about Docker on Windows has generally been the weird bind mount syntax for the Windows file system paths (not too bad, to be honest) as well as the whole file permission thing, as well as the Hyper-V approach eating some of your RAM in the background.
Apart from that, it's mostly passable, though Docker/Podman on *nix is comparatively painless. Though I could say that about most development ecosystems, from PHP to Java. Windows is just generally better for certain classes of desktop software and gaming, *nix is generally better for most development related tasks and servers. /opinion
The caveat is that the images need to be in sync with the kernel version.
And they are still usefull, there is still plenty of Windows based servers that aren't going to be ported to UNIX environments anytime soon.
Well there's your problem..
In all honesty, I've used Docker Desktop on WSL2 quite a bit too and the overhead is very minimal (additional memory consumption mostly).
Avoiding a transformational technology like containers (OCI is used by multiple implementations now, like containerd, podman etc) because of this seems a little silly to me.
meh, I can count on my hand the times it was useful to me.
oh, and having to deal with docker desktop is not fun
To stay sane, just forget it exists until it crashes irretrievably, then reinstall it and repeat. Containers and images being disposable is the whole point anyway.
You can stop it there.
All the comments I've read complaining about the price were saying basically that. One even explicitly said that if it was cheaper he'd buy it for fun, but since it provided no value, he wouldn't... just to get a thread of people complaining that the price shouldn't change his opinion.
That being said, if you are a full time employee you won't magically earn an extra $10 a month for using Copilot, you will only be more productive. It will benefit your employer and Copilot doesn't have an enterprise subscription yet.
I get why companies are concerned about having improperly licensed software on their machines, as the consequences can be great, but too many have gone too far in the opposite direction, making purchasing anything a long and convoluted process.
Furthermore, in addition to the points you mentioned, it also lets us:
Recently, I actually decided to build my own containers for all of my personal use cases and it's been an interesting experience: I base everything on a LTS version of Ubuntu and just use apt for getting all of the runtimes (Node, Java, .NET, PHP, Ruby, Python, ...) for my own software, and so far it's a nice experience.All I need is the base Ubuntu image from Docker Hub and the rest is up to me and the regular mirrors/repositories for the software packages in question, most of which can be stored on my own Nexus instance as needed, as well as the whole build process is primarily driven through Gitea, Drone CI and a few "servers" (repurposed old computers with passive cooling) that I have on my desk.
Though for now I also use Bitnami images for databases and such, which are also decent and which I largely just cache on my end: https://bitnami.com/stacks/containers
So what I'm trying to say, is that there's a lot of flexibility that you can enjoy, both in making your own "templates" for web servers, programming languages, build toolchains etc., as well as you can grab pre-made stuff that other people have made, be it on Docker Hub, someone's Nexus/Artifactory/Harbor instance or another registry out there.
Docker and other OCI compatible tools have largely achieved the sort of adoption and widespread usage that projects like Nix and Guix could also benefit from.
I then have another set of docker images for making package repositories and signing everything.
I do have a lot of mostly-redundant dockerfiles in some places, but in others I've managed to leverage m4 to reduce the redundancy (though I'm trying to keep it as slim as possible to avoid the pain of turning everything into a convoluted set of impossible-to-maintain macros).
Before this, I was using Vagrant and Ansible for builds, which was slow, memory hungry, and frustrating to debug. Not to mention that dependencies that needed to be built statically couldn't be easily cached, which comes out of the box with Docker.
That said, I also don't remember the Docker documentation to be very good. For someone that doesn't work with it professionally, what's a good starting point to learn Docker well?
I agree the docker docs are lacking, i basically learned everything by googling stuff, stack overflow and reading some blog posts
Lets say I want to write a typescript microservice and deploy it to azure as a serverless functions app. I’ll fork our bare metal nodejs project for azure functions, which is essentially the standard azure cli “create nodejs function -typescript -docker” (this isn’t correct syntax but you get it) with the linting and ts-compiling rules we use on all our projects (and an updated image version in the docker file as mentioned). While I build things and run them locally, I don’t use docker, it’s not until I actually want to deploy to azure and setup the release pipeline and trigger it that docker comes into play, but those DevOps steps (also streamlined) aren’t really docker heavy as they simply use the dockerfile that was mostly provided by Microsoft.
I can certainly build and run my micro service as a docker container locally, but I don’t need to. In fact the only times I did was when it failed to build during the azure pipeline but it turned out to be the azure container registry access controls every single time that happened, so these days I almost never “use” docker. In fact I use it so rarely I almost always have to Google command lines.
I know people use docker in many different ways, and that many use things like docker compose, but my point is that you can deploy everything you build with docker and never actually have to care much about docker itself. So maybe you should ask yourself if you really need to learn docker good before you spend too much time on it.
I would much rather just spin up all my resources on a dev cloud account using CloudFormation/Terraform, and spend them down when I’m done.
Before I get (rightfully) called out. Yes I work at AWS now. But I also found that just as appealing three years ago when I worked at a 60 person company.
Umm, I struggle with that. I have Ubuntu machines that run in production. They are VMs that run systemd for some stuff and Docker containers for others. The only way to (easily) replicate such machines on my laptop is via VMs (e.g., Vagrant + VMware), not via Docker only.
I do use Docker a lot... Inside VMs.
Oh boy did this make me feel old.
When I started working all the senior devs had full bookshelves of O'Reilly books, and if you got stuck you went to them to either ask a question or use their books. Then as they left the company, all of us would fight over who got ownership of the O'Reilly book collection!