Hacker News new | past | comments | ask | show | jobs | submit login
Congratulations for reaching a million, Gordon Linoff (stackoverflow.com)
172 points by brudgers 57 days ago | hide | past | favorite | 75 comments


Personally, I am a few thousand places behind, but still in the top 0.2%.

How? I asked and answered a few hundred questions early on, years ago, when SO was new and interesting...

Now those answers are old and, like most answers on SO, out of date. Usually when I google something technical and get seemingly the exact right question asked on SO, the answers are no longer correct. Software versions change.

And yet I still get a steady trickle of votes, forever increasing the gulf between outdated and no-longer-participating people like me and anyone starting on SO today.

I'm in the top 0.2% for the year, only started in earnest less than 2 years ago.

I got tired of always seeing outdated answers as the top, so I actively started searching them out (JS first, then Android) and answering with the newest variant.

I make sure to include browser/device support and update regularly.

How I wish mods could change the accepted answer (or let it be voted on like deletion) after original question asker has dropped off the face of the earth.

I wonder why they can't.

SO seems to be eternally undecided about whether it's trying to be an unconventionally structured forum, a massively multiplayer online game, or a jointly edited database of useful information.

I wish it would put a bit more emphasis on the last one of those.

SO needs to find a way to accommodate “Recommendations” (for tools, libs, docs, tuts) questions/answers. They rightfully banned them early-on because they get stale, fast - but as we’re seeing now, even “good” questions for a long-standing platform like Java, Android, .NET get old before too long - it’s especially annoying when someone asked a question back in 2011-2014 about something which you couldn’t do back then but that you can do now due to recent changes in the platform (e.g. using immutable types with object-initialiser syntax in C#) so right now the “No you can’t” answer from 6 years ago gets 100+ upvotes and a helpful beginner/new/low-ranking user posts a new reply but will get ignored by SO’s algorithm unless/until a higher-ranked user comes along and gives it a boost.

SO should add a “Vote to close because: Obsolete” option - which keeps the question visible but locks new answer replies and displays a very visible and prominent banner infotmjng users it’s for an old version (with the version range displayed and helpful information to find out what version you’re using).

It's simple, really - they're just trying to maximize the pageviews. If you have to click on 5 different Google search results to get the answer you needed, they've won. Getting the answer efficiently on the first click would be detrimental to their revenue model.

Wouldn't prioritizing short-term ad revenue be detrimental to the rest of their business model?

As someone currently utterly suffering trying to learn Android code via docs, google, and SO posts I want you to know that I am very thankful you are doing this. The state of online help for Android is on the edge of a comic tragedy.

I sincerely wanted to get into android development and this is why I didn’t. You’re totally right. I couldn’t tell if I was missing something. How can a major mobile platform (THE major platform?) be such a rats nest to begin developing for?

> How can a major mobile platform (THE major platform?) be such a rats nest to begin developing for?

The great mobile stagnation has begun. We're at least a few years into that now.

The mobile market is no longer growing (and or barely growing) based on usage and unit sales. The market has largely been saturated in terms of time and devices. Even India is heavily flooded with smartphones now, and it was one of the last great potential markets.

So what you have is a large group of existing developers with experience and a gradually reducing group of new inbound developers pursuing a shrinking number of free (widen open) opportunities (in jobs, in new apps that can be created, etc). It will end up mostly being a replacement system, where as developers slowly age out or leave, that's the max inbound new developers you can fit into the stagnate market (to have it be any other way, you need more growth than exists now, or is likely to exist going forward).

The existing developer base largely doesn't need the Stackoverflow information updated, and the demand from new developers isn't as high as it used to be. Most platforms start to stagnate in this manner after they pass their peak / prime days. Mobile is aging as a thing, in other words, and you're starting to see that show up in such ways as rotting Stackoverflow info.

What is the next big thing?

> How I wish mods could change the accepted answer


Doesn't have to be mods. Could be experts above a threshold via voting.

After a year, the owner should perhaps no longer own the question. They've already gotten the answer, and now it only matters to new viewers. If the answer changes, accepted answer needs to be updated.

This was addressed long ago https://meta.stackexchange.com/q/117379/155098 tl;dr: a) it's not impossible someone still uses the old tech b) it's easy to add a comment saying "this pertains to version X.Y, newer versions are different see [answer](link)" c) if the accepted answer is wrong then fix it, everyone can edit answers and people above 2000 can edit immediately, however note wrongness is very often subjective and expect pushback unless it is objectively, factually wrong.

For example, on my home turf, in PHP, the ancient create_function() has not been recommended to be used for many, many years now, it's deprecated now, and it'll be removed end of this year. But if you are diving into an old embedded system using PHP (owie, I know) then information about create_function() might be absolutely invaluable since noone has the slightest idea how to use it any more because it's already been out of use for a decade. A PHP 5.0 answer using create_function() doesn't become invalid just because PHP 5.3+ is using closures. For PHP 5.0 it is the correct answer.

This is the reason I'm not suggesting to edit or remove old answer, just down-rank them in results. If 99% of development uses a new standard, showing the deprecated one at the top by default on the post most developers will find via Google isn't good.

Even if you add a comment or an addendum with a link, it is only a band-aid that probably half of non-senior devs will still miss, trudging down the wrong path anyway

My friend, raise it on meta, here we are just yelling into the widn. That's where these discussions belong.

Perhaps these discussions I linked were too long ago. Perhaps after a set amount of time -- certainly measured in years -- people with a silver tag badge could vote to move the accepted checkmark and after say five such votes it would move over automatically and people with gold tag badges just could move them.

As someone who has written answers dating back more than a decade, if I see a comment along the lines you suggest I'll look into it and update the answer to try and keep it relevant.

There is an editor permission for this use case so that users besides the author can make updates to keep a popular answer up-to-date.

My experience is identical.

I answered a several dozen questions over a month, years ago. Now my answers have paid "dividends" for years, increasing my Stack Overflow Points. This is slightly odd, but I'll take it :)

But seriously, SO is a good tool for learning if you don't get excited by points, and just use it as a "this is what language/feature X is about" -- a method for zeroing in on your chosen topic.

Fellow 0.2%er here, I've had my account for over 11 years. I was very active early on, but can't remember the last time I answered a question (although I still get plenty of rep from those older answers). However, I look back fondly on that point in my career - answering questions made me a much better technical writer, and helped cement my own understanding of various topics. Not to mention the satisfaction that comes with crafting a clear, concise answer. If I had more spare time these days, maybe I'd get back into it.

So in the famous words of Drew Carey: "Everything is made up and the points don't matter"

Similar situation. I'm in the top 2% mainly because of a handful of question I asked 10+ years ago that keep bringing in points every single day.

One time I posted a fairly basic question and someone commented "why are you posting stupid questions if you have so many points?"

A coworker once saw the SO homepage on my machine with all my points and said "wow, what kind of answers are you posting?". I'm so embarrassed that I make a point of not being signed in at work.

If questions and answers carry the same weight in points assigned to profile (not sure they do?), that doesn't make much sense.

A vote on a question used to only get half as many points as a vote on an answer, 5 vs. 10. They changed that recently, don't remember why - now it's 10 for each.

For me, finding solutions for problems with older versions of software happens on Stack Overflow more often than anywhere else. Thanks for all your work!

Reminds me of investing... the people who invested earlier get more ROI over time.

More like investing every day, no?

Nope. Because of the rep cap, you can max out your reputation gain every day without ever answering another question. All you need is a significant enough collection of old useful answers that get upvotes every day. Jon Skeet will keep his rank at #1 even if he doesn't answer another question, because he gets 200 points every day from his old answers. The only way anyone can gain on him is by getting accepted answers or bounties, because those are not subject to the cap.

Similar experience here: I started early and played the game for a while, and years later with very little activity I'm still top 0.08%. My technique was to focus on speed. I wrote a tool to alert me to new questions and would often answer within one to two minutes. This wasn't as disruptive as it may sound to my real work, as I knew my domain well enough, so it was the mental equivalent of getting up for a quick walk.

The competitive bits are over the 200 points a day (you get passively from a certain point, I think it was about 150K for me).

If you answer a question the +15 points you get for a correct answer count towards the limit. Jon Skeet's and Gordon's game is very different from most users and it's on those margins.

> If you answer a question the +15 points you get for a correct answer count towards the limit.

I believe you meant it does NOT count.

And bounties do not count either.

https://i.imgur.com/fWpAT8B.png (I am slightly cheating with this screenshot because it's from Travel SE not SO.)

Long-long time ago I was posting about this on Meta: https://meta.stackexchange.com/questions/166952/old-timers-l...

I wonder how that could be prevented. Lock voting on old questions and answers automatically?

I'd probably just look at putting in a point value degradation over time mechanic, so a point x years ago is only worth a fraction of a point now

Sounds like inflation.

No, it's putting your point 'worth' on the same trajectory as the relevance of your code comment ie heading towards zero. There's more than a semantic difference between the two

You should edit the questions and answers that are outdated so other googlers can see it.

Yah I got a lucky answer on GIt before GIT was a popular thing. Been a sweet karma farm.

It is time to redistribute votes from old outdated answers to people creating new updated answers IMO.

To downvote on this is shortsighted. Some sort of karma rot (or whatever you want to call it) it will be necessary if SO wants to stay relevant. GP's comment implies that almost.

I understand that this might come across as envious, but it isn't on my part, and I don't see how else to deal with the problem.

What a radical left idea! Distribute the wealth!

People shouldn’t be farming karma forever just because they gave some answer years ago.

If someone finds an answer which helps them and so upvotes it, why should the reward depend on the age of the question?

I think you misunderstand the workings of true karma.

Likewise, my profile suggests I’m active and answering questions as my score steadily climbs, but I haven’t answered anything in a long time.

Actually, I’m a little ashamed to admit that my most recent points are for a typescript question I didn’t even answer totally correctly. It explains something useful about typescript, but I didn’t really address the question properly. People still upvote it, perhaps because it explains something people might need to know if they’re asking the original question anyway. I should edit the answer to address that now that I noticed.

One of my best answers stands the test of time. It just explains how to optimize images for the web, but explains it fairly agnostically by describing compression techniques and image content and how they combine well (or don’t). Not a lot has changed in that arena since 2010.

I would love to help people like I was helped early in my career, but I find SO fairly useless as a platform to do that these days. It doesn’t feel collaborative or constructive when I use it. It’s like a race to get the best answer to avoid throwing insight and knowledge into the void. Not to mention a plethora of other reasons it can be fairly miserable. I tried moderation for a while and I guess that was the final nail in the coffin.

Same, only I updated my answers when necessary.

> that's an average of ~22.8 answers per day, every day, for the last 3144 days

Gordon Linoff wrote a bit about his habits here[0] but I still can't fathom how he manages to maintain such a rate. Imagine just taking a day or two off – you would then have to write 45 (or 68 answers) the following day. So assuming that he does take a day off every once in a while, this means on non-vacation days his average is even higher.

[0] https://blog.data-miners.com/2014/08/an-achievement-on-stack... (I'm gettting an EOF / Connection Closed error in both FF and Chrome but the Internet Archive has it cached.)

The blog post describes him trying to catch up the deficit from vacation days away from the Internet.

StackOverflow has done a great job with the points incentives. As noted in this blog, the daily cap on points is important - it's valuable to put the brakes on people who are too into it and in danger of burning out. Candy Crush does the same thing.

A few years ago, I asked Joel Spolsky what he might do differently: he mentioned that getting to a million points is a magical number for many people but it's too tough on StackOverflow. If the point system were set up to be more generous, more people could achieve this milestone and more valuable contributors would have this somewhat arbitrary but enjoyable recognition

Oh, it was "HTTPS Everywhere" which kept on redirecting me to the https version without me noticing! Thanks a lot!

Its like sports. You ask someone at the top of their game why they keep doing it and all of them wont have any great answer.

Its like once your brain gets used to doing anything well it just keeps doing it.

That's certainly a good point and might explain how Linoff has managed to keep up his motivation over the years.

Nevertheless, I wasn't asking about the why, I was asking how. :)

Both him and the other "millionaire" are book authors. It makes sense: on one hand you increase awareness and recognition of yourself and on the other you get to know what are common questions and doubts of your potential readers and how to explain things better.

>He only joined in 2012, so that's an average of ~22.8 answers per day, every day, for the last 3144 days.

Hope that's not his real name because if it is his boss might think he needs more work to do.

I think he has more than enough skills to be his own boss and charge by the hour.

It's suprising somebody gives 70k answers, but asks no questions. I wonder why that is. Does he know everything about his area and not interested in other areas?

What I've found over the years is as I get more experienced, I can cobble together whatever I want from existing SO answers plus the rest of the internet. My volume of asks has declined enormously over the years. He might simply already have been at that end of the experience curve when he joined.

Same for me. I only ask questions now if there is no documentation for something or it is a bleeding edge technology.

I haven't asked a python or C++ question in probably 5 years because the existing answers are all excellent and I can glean what I need from one or more of them.

SO really shines once you've cleared the minimum bar in terms of computing knowledge. From what I hear it's not super great for people that haven't cleared the minimum bar yet (students, absolute beginners, etc.), but it's truly a gem of the internet (I didn't use it as a student because I didn't learn about its existence until I already graduated college - plus, it launched when I was a sophomore in college so it really hadn't built up a good answer base until later).

SO can also really shine given how it illuminates how you can rarely go deep enough. Even after 35 years as a highly technical programmer, I still regularly read stuff on SO and have an "oh really?" reaction over and over again.

I've found that SO is great for questions that aren't too basic, but aren't too hard either. Overly basic questions get downvoted. Really hard questions don't get answered (not in the Spark niche at least). Questions that people with 2-4 years of experience ask seem to be the bread & butter of SO.

One of the most valuable lessons you learn with experience is how to be resourceful. If you look at what's being asked on SO, it's mostly beginners who didn't know how to divine a solution nor read source code.

I have left stackoverflow around 2011, still top 4% with - ~300 answers with a lots of community wiki posts and even more comments, some giveaway through bounty on random questions. I have not been compelled to ask anything. Stackoverflow format is not suited for hard questions, that would require personal research. It's not suitable format for a topic discussed in a mail list either. It's a nice tool for quick answers on an easy ones that would score well via quick searches.

In other words. To post a question: I'd have to do my own research, followed by prototyping, testing, debugging. Be unable to reach anywhere, unable to get help by talking to peers directly. The topic should not be domain specific. Gather my thoughts, experience, etc. write them down and post a question. There has been a single case where a colleague of mine asked a question and it never got any attention whatsoever. We found the issue/solution few days later.

For areas that aren't ultra new the space of good questions is kinda saturated.

This is less true for answers - even with old questions with accepted answers you can gain points by providing a better answer. Not necessarily better in terms of pure information but maybe better edited or explained more clearly.

And obviously you can still rack points by quickly answering the bad questions (before they get deduped or closed).

Disclaimer: this is only about my experience, not implying anything about Gordon Linoff - I have zero knowledge about his record on SO.

> And obviously you can still rack points by quickly answering the bad questions

and your old very simple answers to common questions continue to rack up the most points

you get points on stackoverflow by answering common problems. not much from contributing more nuanced indepth things

I've pretty much found github issues to have the newer up to date discussions

I've never been compelled to ask questions on SO. Often, I find a question similar enough to mine, but it wasn't answered, or the answer is clearly wrong, or sadly outdated. If I would ask, I would expect the same result, plus I'd have to wait around for it.

Usually, I can just dig in and find the answer myself, but I was trying to get a solution quicker with search.

Of course, I've only answered one question too, so maybe it's just not my kind of site.

People who ask questions are usually different from people who answer questions. I have significant presence on Quora where I wrote 8300+ answers and asked 98 questions in 10 years with 10057 followers. My monthly views are about 150000.

Quora has different incentives for asking questions and answering them. I am also connected to a community of people who mostly ask questions and don't write answers.

It is hard to ask a good question to which I can't find an answer. I ask questions for which there are no good existing answers and usually outside my realm of expertise. This is true for most other top contributors on Quora. Most of us ask very few questions.

Good question.

Mere speculation: Maybe he enjoys the asymmetry, and has some other accounts on which he asks questions.

What surprises me is that SO claims I'm not allowed to answer questions until i get points and suggest I ask questions... so how can he have not asked any and get started?

That has always been a real put off at SO. They actively don't want me to answer.

I thought there wasn't a reputation requirement for writing an answer?

A lot of people seem to get their start by suggesting edits that later get accepted.

Is it plausible that someone would use a different account to ask questions?

Too busy answering to ask questions.

Genuine question for Stack overflow users with high points: Aside from the improvement in your own skills, do you find it opens any other doors for you?

I think it got me an interview once, but nothing ever came of it. In retrospect one of the questions I was grilled on in the interview was based on a SO answer I had given; it was so long ago that I didn't remember the specific point the interviewer was looking for. I didn't discover the connection until something made me revisit the answer years later.

I got an email from an Amazon recruiter through it.

Why is this an achievement? In the past the creators of Stackoverflow have come out and they completely devalued the points as a meaningful indicator of comparison. (There was the global recalc and that was one of the things mentioned in the announcement)

In any sort of gamification there is a reward for reaching an arbitrary threshold. It's all part of the game.

I agree with you.. however in the global recalc they intentionally ignored the people who were taking the points seriously (for rewards and the like) and said "well they don't matter anyways". We can't reward them for taking it seriously, and excuse them when they claim the opposite.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact