> This comes after an astonishing amount of 71,839 answers (and 0 questions!). He only joined in 2012, so that's an average of ~22.8 answers per day, every day, for the last 3144 days.To put perspective on the numbers, the second answerer on the site is Jon Skeet (our first millionaire) with 35K answers and then several others with 20k+
- person who is admired or idealized for courage, outstanding achievements, or noble qualities.
- the chief male character in a book, play, or movie, who is typically identified with good qualities, and with whom the reader is expected to sympathize.
- (in mythology and folklore) a person of superhuman qualities and often semidivine origin, in particular one whose exploits were the subject of ancient Greek myths.
Stack Exchange sites are not mere social media but more like an encyclopedia for specific topics. Answering as well as asking good questions requires extensive research and being mindful of future readers.
Sad to see SE contributors are compared with FB users.
Haha, yea, this isn't like having a snap streak of showing a funny face to your friend for 3,144 days. But alas, the internet at large probably thinks that is a harder feat than this.
[ example meant to highlight the difference between a low-impact, low-effort comment from a social media site like reddit versus sites like SO where some degree of technical expertise is a precondition ]
How is throwaway comments and opinions on generic social media sites in any way comparable to answering questions on a very technical subject in order to assist people working in said technical area?
I don’t disagree with your statement at all. However, people who contribute useful edits to Wikipedia or high quality answers to Stack Overflow are doing a bit more work than your average Facebook poster :) sorry you’re getting buried on this, I assume your point wasn’t necessarily that it wasn’t more useful than that.
The way I see it, there’s not as much of a gap as people think between the 99th percentile and 50th percentile users, because being active on one of these sites isn’t that difficult. However, I think this person does stick out. It seems there is a massive gap between them and even second place. I think that’s what really makes this astonishing.
The user has published numerous books on SQL and Data Mining. Makes me think he is an expert who spends time curating caveats in this ecosystem. This is a great way to build value for the community and self. https://www.amazon.com/s?i=stripbooks&rh=p_27%3AGordon+S.+Li...
Would be interesting to see the overlap in answers. It may be that instead of closing as duplicates, he’s just answering them and his SO cred keeps folks from chasing duplicates.
As an aside, he was also a big reason why I stopped participating in the network as we had a few run-ins over nuances where his support was “I wrote a book.”
If you're answering 23 questions on SO per day, you don't have time to maintain a website. Especially that initial time sink of converting from a Flash based site.
If it's occupying his life so much that he hasn't updated the homepage for his technology consulting firm in 5 to 10 years, then he seriously hope he's retired because the alternative is really not good.
Even at 10 minutes per answer that is roughly 3.7 hours a day or 26 hours a week. That’s more than half of a full 40 hour work week or like 23% of your waking hours each day!
I'm willing to bet this guy is so good at his craft that he bangs out quality answers in less than 10 min per question. That'd actually be a fun side project: for a given SO user, compute their average time between each of their answers and the corresponding question.
You are assuming that he answers questions every day. He could easily be spending weekends on SO and some evenings. It doesn't mean he's using work time to do this.
He may be a database engineer on call in case of problems, but not busy doing anything. Some people on call are doing crosswords, watching youtube, posting on facebook. He's posting on stackoverflow !
With SQL, there are many different ways to write a query and to get the correct result returned to you. Instead of "1+1=2" being the only answer, it'd be more like: "1+1=2", "2+0=2", and "0+2=2". All are correct.
In my experience on SO with both SQL and Python, there are usually a number of correct answers for every question. How correct it is depends on things that usually aren't in the question: size of the data set, dependencies or processes that may break as a result of answer, etc.
It’s easy to show that something is incorrect, because it doesn’t meet the base requirements of the question posed, or correct because it does.
It’s difficult to show a correct answer is good, or the best, for the question. Many answers you find on SO both provide a correct answer, as well as try to defend against and call out potential pitfalls and assumptions. Many answers also achieve at best “technically correct”, giving something that works, but slow, error-prone, inelegant, and really just a Bad Idea.
Technically correct is mostly a waste of time, unless nothing better exists. If this user is just technically correct (he’s probably not constantly wrong because he has more points than answers), then he’s ultimately not much better than your regular shitposter.
"1+1=3" is only incorrect in the standard accepted definition of arithmetic. That is inherently the subjective part. Nothing prevents people from using different math system. At the very least, I could say 1+1=10 in binary and you have to concede that you need to add more caveats to your statement. The other form of subjectivity is that style/performance/complexity is always considered for programming and the tradeoff is always subjective.
Math is objective, not subjective, which is why 1+1=3 is incorrect, not just a violation of my preferences.
Binary is not a different math system. It's the same math system with the same rules but a different radix. 1+1 does not equal ten in binary. It still equals 2. You're just using a different notation for the same concept of two. The substance of the thing is not identical to how you signify it, which is why you can write "two", "2", "II", "2r10", ...all of which signify the same abstraction.
You are kind of proving my point. You just defined 10 to be "two" by establishing which base you are using. Thus, notation is subjective based on the author's desires. But, my argument was not even that. My argument was that 2 comes before 3 is not a universal law of math. That was chosen (subjectively) by humans.
Solving problems for random people on StackOverflow helps them (the user, the corporation, and StackOverflow) make cash while not compensating you, and it also requires relevant knowledge and effort to communicate concisely and correctly.
On the other hand, people on Hacker News, Facebook, Tumblr, /. and so on don't actually have to know anything, and their contributions only make money for the host company, and barely so in some cases. Reddit, as an example, isn't really profitable because of its average user.
I contribute to similar things, myself, pretty frequently, but let's not pretend like social media and collaborative knowledge projects are completely equivalent and that effort spent is transitive.
You need to understand these are not only technical answers to technical questions, but also highly valuable ones because people must indicate that they "like" the answer.
Most even good answers get very little rep so the answers have to be pretty good to get that amount of rep on average.
I wonder how much rep, has to do with who decides to up vote? If you see Jon Skeet answer, people will up vote him more than the average user with 5k. This is a subjective opinion of course.
Gordon is amazing, and it's almost impossible to beat him to an answer for certain types of questions :) I had the chance to meet him a few years back to talk through some new SQL features coming to BigQuery (I'm at Snowflake now). He had a lot of great insights into what could make the product better, and what he thought was missing relative to other OLAP databases.
I like to answer sql questions too and it's like being in a quick draw match with Linoff around. he once blogged about how he's able to find all the new unanswered sql questions so quickly, iirc he made his own search
I remember John Skeet saying he would sometimes post a half finished answer just so he could get their first and deter other users then polish it later.
A Jon Skeet encounter made me quit Stack Overflow, never to look back. I asked a question, Jon Skeet came along with an answer that did not answer my question. But his non-answer got 200 upvotes.
I went to meta to talk about how the reputation system is broken if that could happen. Skeet came along and accused me of changing my question, and a bunch of SO fanboys mocked me and told me if I want more reputation I had to try harder.
That's when I knew SO had irreparably changed for the worse.
Oh I had some negative encounter with Jon Skeet as well. For a person that has so much notoriety seemed he was quite defensive about me pointing out his "non-answer" he gave to someone else.
My situation was that I noticed that person asking was on a very basic level and he gave answer that was not correct in context of experience or what was actually asked. Even though technically correct.
Every time I've asked a question on Stack Overflow, I include in the question what I've tried and hasn't worked, or that I'm specifically looking to do what I described and I'm not looking for a link to The XY Problem. It always gets edited out immediately so someone can post an unhelpful solution and get upvotes.
Yeah, see that's frustrating. The point isn't his answer. The point is that his non-answer got him upvotes from his fans, and that when I tried to address this broken reputation system in meta, people aggressively missed the point exactly like you did.
It's just weird how often a discussion on something on SO ends up with someone complaining about overzealous downvoting or closing of their perfectly reasonable question, but no one can ever actually post a link to it.
To me that says they're not telling the whole truth about their own behavior.
Given there could be any number of reasons that strangers on the internet do not supply to you that which you demand, especially given the down-votey tone you demand it; it is instructive that you narrow in on the least charitable.
I quit SO in 2014, and asked Stack Exchange to delete all data from my accounts across their properties. I doubt any link exists.
Heh, I had to go back through some of my old questions and sure enough Gordon has helped me before. I had to laugh because my comment below his answer was that he figured out my problem so fast that StackOverflow wouldn't even let me accept his answer as it was too soon after posting the question.
I've not seen Gordon Linoff, but usually when I have some sort of question about PostgreSQL minutia, I find that Erwin Brandstetter (https://meta.stackoverflow.com/users/939860/erwin-brandstett...) usually has posted very detailed and solid answers when it comes to Stack Overflow.
I have to say, I'm usually not searching Stack Overflow specifically when I'm looking for information, I'm usually in a general search engine that happens to show me StackOverflow answers. But the high frequency that I see and am in fact aided by Erwin Brandstetter's posting of good answers is certainly worthy of a call out here. My thanks to Erwin!
Couldn't agree more! Erwin's SO answers are incredible source of knowledge on Postgresql usage, especially things like plpgsql. I benefited from his answers a lot, thanks to Erwin.
This is insane. They must be super-efficient because I can't give an SO-quality answer to a question in under an hour.
I can see the appeal though. I used to spend my spare time in uni answering questions on SO. I actually learned a ton about how to conduct research, which is probably one the most useful skills I have in my toolbox.
I've noticed a trend, some people answer with what should be a comment and then progressively edit their answer while hogging the attention of the questioner and anyone coming to view the page. For some reason people are loath to downvote an early answer without competition, even if it's rubbish.
Yep. I Do this quite a lot in the Cross Validated site. It is the only way to get up votes or the correct answer check. If you take too much time to craft a well thought out answer, somebody will beat you to it and get all those.
So in order to get the best of both worlds, I anser fast and then, if needed, improve on it.
You can do it faster if you restrict yourself to questions that you already have researched. That is, to questions about bugs like ones you've fixed in your own code, or features you've found out how to use.
Voting by user is against the rules of Stack Overflow; if you intend to upvote this user's posts in order to reward them, the votes will be reverted and you might face disciplinary action.
As a network moderator, I actually know the answer to this, so I'm not going to openly speculate. Suffice to say, the rule is there for a reason (voting-by-user messes up the ordering of answers to each individual question), so please don't try a clever workaround just because you think you can.
At the level he is at, he is clearly doing it because he likes doing it. Maybe to build a little bit of recognition to help sell his books, but I doubt that it's a big factor.
I didn't read the article but this must be Gordon, the guy who answered every SQL question I asked, within very short period of time. This guy is a legend. Love him.
What kind of setup do you need to have in order to answer questions on StackOverflow?
From time to time I think it would be a useful exercise, but it's really hard to find questions that haven't already been answered. I'm thinking you probably need some form of alerting system to notify you as soon as a question comes in.
I have knowledge about an area of computing that very few people in the world know about, and very few people need to learn about it. Very rarely a questions gets asked, but when it does it's guaranteed the question may sit for months without an appropriate answer. When I joined SO in 2012 I went through all the unanswered questions on that topic and was able to give good answers to like 60% of them. Got a bunch of medals. But then never did it again.
I don't want to say it since this, along with the rest of my message history may be able to identify me. Maybe I need to switch to a fresh account again.
Anyway, it's a low-level library that is used by a small handful of libraries that a ton of people use. All hail the easy-to-use abstractions.
Most of the questions that get asked are homework questions that were poorly asked and lack much research. "What have you tried?" is a common first comment.
You can create saved searches, like sorting by new for a tag you're interested in. I don't have email notifications setup, but from time to time I visit my saved search for pytest questions -- which is a subject less inundated with homework questions and more with people trying to do their work, unsure of how to integrate their ideas with a testing framework.
Though, some of my most popular answers have come from digging a little deeper on an older question and adding a better write-up, or simply answering a question for a newer version of the technology at hand.
When it's a technology you use (or had used) everyday, it's easier to get a sense of the answers you would appreciate if you'd stumbled there from Google
Doesn't that depend on the mobile user and device?
Personally, a site like that doesn't impact me. I'm on a Note 8 and it's usable to a point where I will just do that with no worries.
Everything comes at a cost too. Mobile styling may avoid a zoom op or two, but also usually comes with more navigation elements, or leaner overall content presentation.
However, it also mean that during that time this individual might not be working or spending time with his/her family, etc. Extreme performance like what he is achieving always come to the expense of something else.
When I worked for a small television station I would pass the time answering questions about Linux, routers and other misc categories. Like building long haul 2.4ghz networks. Then I moved out of the control room and forgot I had an account.
It's funny I knew it was Gordon without even opening the article - quite the mad lad - there should be a special of honor for beating him to the draw on SO .. ! (I think I've got 3?)
He also has to be in the top 10 in terms of comments on questions and answers, he's everywhere :)
Impressive achievement. But for fun, as a skeptic, let's assume he outsourced some of the work, and then serves as an editor. How would we know the difference between this, and him answering the questions himself?
Hoping that Gordon sees this, he’s a wonderful person.
helpful to the nth degree as his dedication to answering questions as is so demonstrated here in earnest.
all your recognition is well-deserved. it’s been a pleasure to work with you (and nice job solving the interview qs i give to others in pure sql from the other side of my desk)
Not sure what his intention is/was, but I am assuming he also gets clients thru this. Many blog posts out there uses this technique to attract potential clients.
I love stackoverflow, use it all the time often from generic goggle searches, and I don't even have an account, its just that useful of a site. I suspect I would probably not answer very many questions due to others simply having better answers than me.
I don't get your point. Are you defending SO as some sort of gaming company? Even gaming companies have tournaments with enormous prize pools and usually award their patrons in some way. Tournaments, teams and sponsorships are the lifeblood of professional gaming.
If you go to his profile on SO, you will see that he advertises his company.
Also, I wasn't speaking about "professional gamers", I was speaking about every day people that forked over $100 for a game (or often $1,000 for many games) and then spend years playing it at home - unprofessionally, for no money.
It's just how some people chose to spend their time because they themselves find the experience gratifying. And in his case, there is at least a small pittance of financial gain through advertising and publicly verifiable reputation.
However unfathomable it might be, he may just genuinely like to spend his time here on earth knowing that he's helping people.
He clearly has expertise so I doubt he's poor or hurting for cash. Maybe an older fellow who has had enough of the grind and decides to spend his days sharing his knowledge with others.
Someone like this needs SO as much as they need him, it's mutually beneficial. He impacts millions, they make money. Both get what they came for.
All user content published is licensed under CC BY-SA 4.0 per [1] and backups/dumps of all this CC BY-SA 4.0 content are published every few months to the Internet Archive [2].
CC BY-SA means you could use that latest dump to create your own "SQL Q&A" site and charge users for the privilege of looking at Gordon's 76,000 answers to SQL related questions. You would need to attribute Gordon Linoff for the answers provided, amongst other obligations of CC BY-SA 4.0 including an inability to re-license the content.
Many would instinctively want to use a CC BY-NC license because of an unfounded fear of volunteering time to find out later that someone else is collecting a reward. There are many good reasons for avoiding CC NC (and even CC BY) licenses and choosing the least restrictive license possible. There is a good write up on Wikimedia's decision to prohibit CC NC licenses at [3] that explains some of the reasons.
The majority of sites relying on volunteer contributions including Hacker News, Experts Exchange, Quora, Reddit, social media sites, Flickr (mostly), etc do not enforce copyleft licensing and contributions made to these sites could easily be lost forever. Wikipedia and StackOverflow including their volunteers likely wouldn't have enjoyed the same level of success if they used or allowed CC NC (or more restrictive) licensing.
I kind of regret not joining stackoverflow I am sure alot of people get hired off of it instead of having to jump through hoops doing technical interviews. I am looking for work now in golang and i just hate having to do interviews over and over again with each new contract.
In my experience, it doesn't help in terms of interviews/hiring. It's often rewarding on its own right, but you're not missing out on jobs, anyway.
For context, I have a (now inactive) stackoverflow account with ~250,000 reputation. For a long time, I was the top answerer for many tags in the scientific python ecosystem (e.g. matplotlib, scipy, briefly top for numpy as well) and was in the top 10 for python as a whole at the time (i.e. showed up on the "top answers" page).
I was really proud of that once, and spent awhile going from "this is a fun distraction" to "oh, this could get me a job and help me switch careers, I should focus on it".
Not trying to brag there, just trying to put it in perspective. My SO track record was significant enough, particularly ~5 years ago when I was actively applying, that I really thought it would help me get jobs or at least give me an edge. It didn't.
I've never had my stackoverflow stuff asked about or mentioned in an interview before by the person interviewing. The couple of times I've incidentally brought it up during an interview (as an example of a specific problem they were asking about or as an example of teaching/training/communication), it went over as a lead balloon.
My OSS contributions (which are slim and were all a long time ago) have opened doors, but stackoverflow hasn't, at least that I know of.
I'm not saying it's not rewarding or that I regret it, just that, at least for me, it didn't seem to help in terms of job opportunities, and at the time, I was a bit disappointed in that.
As #1 (now #2) on ServerFault.com for many years, it opened a lot of doors for me as a consultant. I could reference my reputation as a "body of work" and it carried a lot of weight. Most employers, staff, recruiters are familiar with the sites, and it made things relatable.
I'd love to see actual data on this but I doubt that a lot of people get hired because of their online reputation. And by "hired" I mean it directly lead to a job offer instead of just getting your profile noticed and getting you started through the recruitment funnel. And of those that had an offer, how many actually took it?
Being top 1% contributor in Go would tell a recruiter you know Go but the job probably involves other technologies you need to prove yourself in. Even in the rare case that you are top contributor in every technology on their checklist, the job might still involve domain knowledge that isn't purely tech. Or they'd want to evaluate your personality.
Anecdote: not to brag given my skepticism above but so far the only online activity of mine that lead to a job offer was a Ruby Github repo on traditional AI. The catch: I am not a Rubyist and that repo was nothing more than a fork I'm studying. Long story short, I was looking for a job then, this consulting firm saw my Github profile so I got an interview. Onsite, the TL looked at my resume, visited my profile, saw the Ruby project, asked a few questions about it, and then recommended me for an offer with the hiring manager.
I did not take the job because (a) they looked desperate for anyone in the process and (b) I would've bailed anyway because I already found a company I'm set on joining. Nothing more came of it other than this HN comment.
I would hope that it also tells a recruiter that you are able to focus/persevere at something, that you have decent writing skills, and that you enjoy sharing your knowledge.
I am somewhat biased, I also have a high-rep account (mostly from C questions). I have not tried drumming up interest when interviewing, though.
to be fair to yourself, I am a member on Stack Overflow and in the past 13 years, 0 interviewers have asked about or referenced Stack Overflow (other than jokes about how we use it to cheat). I hear leetcode is all the rage these days, in prepping for technical interviews.
Are there any numbers available on that? Do people on SO really get job offers if they have a high enough reputation? If so, are the offers from high-quality companies doing interesting things?
All of my previous job changes have been due to "in real life" networking. In order words, a friend of a friend recommended me for the job. I'm not looking to change jobs anytime soon but I sometimes wonder if my lack of online presence would hurt my chances of finding another job if had to. I have a blog and and a github account but both are pretty bare. I don't have a LinkedIn account because they started spamming me 15 years ago and they have been blocked by my email server ever since.
Nah, hiring processes with technical steps will just happen regardless of one's standing on this or that site - after all, there is no guarantee you were the one posting... Unless you know someone inside who can vouch for you, you'll be tested.
It's only going to help with landing a first job, and even then, not much. Open source contributions are somewhat better, but you basically need to be heavy contributor or maintainer of a large project a company really cares about.
Unless you work for those companies and know what risks and attack vectors they are trying to mitigate, it’s really hard to criticize their placement of captchas.
> This comes after an astonishing amount of 71,839 answers (and 0 questions!). He only joined in 2012, so that's an average of ~22.8 answers per day, every day, for the last 3144 days.To put perspective on the numbers, the second answerer on the site is Jon Skeet (our first millionaire) with 35K answers and then several others with 20k+