Dear Stack Overflow, I really want to talk but I don’t believe you’ll listen to me anymore.

I’m going to do my best to keep this as objective and as unbiased as I can, but…I’ve got something to get off of my chest.

Let’s clear the air first — the maintainer of curl, Daniel Stenberg, described the moderation action as thus:

This is a repost of the answer I posted on stackoverflow for this question. This answer immediately became my most ever upvoted answer on stackoverflow with 516 upvotes during the 48 hours it was up before a moderator deleted it for unspecified reasons. It had then already been marked “on hold” for being “primarily opinion- based” and then locked but kept: “exists because it has historical significance”. But apparently that wasn’t good enough.

To Daniel — know this; this was not directed at you. It isn’t the case that any of us are particularly upset at you. We’re primarily interested in preserving what quality we have on the site. I hope that you’ve now caught up on the Meta post to get a bit more context into why the post was originally deleted, but I also hope that you realize that an answer like this made for a better fit on your personal blog than Stack Overflow.

I was personally elated that you did reply — I happen to love and use curl every day. I can’t imagine a world without it. However, just because I happen to have a deep and abiding appreciation for your work, that doesn’t mean I don’t have obligations to Stack Overflow too.

I’m sorry you were inconvenienced, but I truly believe you were the victim of circumstance. The question — with all of its upvotes — sent you the wrong signal, and in an ideal world, you wouldn’t have answered that question.

But I suppose not everything is ideal. While I stand by my assertion that the post should’ve been removed, I hope then that you’re not so bitter towards the Meta-hounds who are trying to keep this site still relevant under the rush of poor questions.

And now, some background.

I consider myself to be a Meta-hound. I’m an active participant more on the Meta-side of the house than the main site these days. With that, I have an ear to the ground for a lot more of what happens on the site and what policies Stack Overflow has built up over a decade.

On Meta, one of my favorite pastimes is to help people their questions on the main site. It’s the reason why I’m one of the few non-moderators to have a gold badge in three of the four required Meta question badges. That’s not a feat many others have even come close to accomplishing unless they work here.

It goes without saying that, given my score on those posts, I know a thing or two about the site, and about content curation.

I understand the frustration of the new user who gets their question downvoted without an explanation (and no, I’m not suggesting we start doing that).

I get the annoyance behind someone just wanting their question answered.

I know what it’s like — firsthand, actually! — to feel like you’re not being listened to or you’re being ignored in a question.

Image for post
Image for post
A now-deleted response which indicates that the answerer just wasn’t interested in answering my question. Yeah, I know this pain.

Given this, I feel a lot more in-tune with the community and what is and is not acceptable.

However, I’ve noticed a recurring theme — I feel like my expertise is getting pushed aside, disregarded, or outright ignored. The feeling was subtle at first; it came in a burst of surprise and disappointment with what I believe is the quickest feature ever released — the “New Contributor badge”.

Image for post
Image for post

In this post history, we note that the post announcing this feature was followed up by its network-wide rollout a mere four days after it was announced. That’s hardly enough time to gather constructive criticism or feedback on the feature, which is what I thought the Meta post announcing it was about.

I only learned a few months later in an otherwise unreleated answer why the turnaround was so quick.

@Makoto: So, um, I hope you recognize that adding the new user indicator was trivially easy and fixing search is, well, not. It’s not as if that feature took away from anything else we were working on. Documentation is a much better example, I’d say.

(I mean…I had plenty to say about Documentation and what I believe was a reasonable suggestion about what could make it better. I’ll add that notch to the belt, too.)

I’m only trying to set the stage for what finally broke my back on this. The creator of curl recently responded to a question about why he released the code for free.

In all honesty, seeing maintainers, creators or architects respond to posts on Stack Overflow is rare, but awesome. It has a kind of buzz to it, and it can honestly have the chance to enhance the post.

Well, the chance.

This is the unredacted and full question posted on the site.

I recently started using libCurl for my VC++ project. I’ve been wondering: what is the incentive for the curl creators to release the entire library for free?

Is it purely to help their fellow developers? This is one of the best open source libraries I’ve used in recent times. (Other than Python and R programming languages).

Project cURL is entirely free and open. No person gets paid for developing curl full time. We do this voluntarily, mostly in our spare time. Occasionally companies pay individual developers to work on curl, but that’s up to each company and developer. This is not controlled by nor supervised in any way by the project.

My approach to questions like this is three-fold:

  • Take away context to remove any unintentional bias. I might gush a little bit if the question being asked is about something I am passionate about or I like, and I don’t want that influencing any moderation I do.
  • Distill the question to its core components. For this, the core components look like this: someone is using libcurl and wants to know the motivations behind why someone released it for free.
  • Evaluate that against the mass of known quality checks and constraints we have, evaluating for topicality and objectiveness. So um…this is where the flag went off; “Why did they do this”-type questions have prior art in several places; one of which is this particular Meta question which discusses language design decisions. I do not see it a quantum leap to consider a licensing decision question on Stack Overflow in the same light as a language design question; no random expert in the software could answer it satisfactorily.

By that, the question should have been closed.


Not only does it have a reasonable post score, it also got an answer from the principal developer of curl. It’s available on their blog, so you can peruse that at your leisure.

Like I said, this is actually cool. But it doesn’t mean Stack Overflow gets the permission to just “turn off” its quality control system, especially when we’re asking new users to go through a New Question Wizard.

But right now, we’re in a sticky situation here, folks; the automation which cleans up a lot of posts don’t function on a post which scores sufficiently high enough. This means that the community has to intervene, since the system is not capable of doing it on its own.

So…they did, in their specific way. A Meta post was created (as is the wont of someone asking about a post’s topicality), and the content was discussed.

(And yes, I am aware that the question linked to one of the “prior art” pieces I noticed. Take this as a formal position of me simply disagreeing with the assertion that this isn’t the same case.)

Community moderation is largely straightforward, save for two major cases:

  • A highly contentious and usually highly scored post which is closed and reopened repeatedly.
  • A post which has a historical lock attached to it.

This one has both.

This means that, in its undeleted form, no “mere mortal” may cast flags against it of any kind, including deletion. So, if a contentious situation were to arise, the only people who could actually fix this would be diamond-moderators.

I too lobbied for the deletion of the post, largely from an agnostic and neural approach based on my heuristic above. Know that my belief in the post’s topicality shouldn’t take away from the contributions made to open source, nor the opportunity to actually celebrate their accomplishments. Both things are awesome. Ideally, neither belongs on Stack Overflow.

And one did — one did delete the post, which I believe was the appropriate decision. Their justification can be found here.

Imagine my shock to wake up Tuesday morning to find out that it’s been undeleted!

This was overwhelmingly good content by our simple question grade until it was closed. As far as I can tell, the close voters acted because the question is not so much about programming as it is about software licensing. But the reason they chose is that it is “primarily opinion-based”, which does not seem to be true. Like questions about why language designers made certain choices, the question can be answered by the person who made the decision or by other evidence they left behind. In this case, the person who knew definitely actually answered the question. This is not a matter of opinion, but of historical fact.

We get an extraordinary volume of bad questions on this site. I don’t mean bad in the sense they should be closed, but bad in that they are incomprehensible and unanswerable. It’s frustrating for all involved. Closing this question helps that problem not at all. Instead it signals to a portion of the 34,348 people who have viewed the question so far that Stack Overflow does not value interesting content. This is fodder for yet more articles about how insular Stack Overflow is without discouraging questions asking for code and other such nuisances even a little.

So this is the part of my post where I drop the largely neutral tone and decide to just go right in.

First — and foremost — to everyone who believes Stack Overflow is insular, it has to be in order to survive. We’re having a really hard time keeping actual experts engaged — and not just the experts who know a language, I mean the ones who really dive deep — those contributions are starting to get noticeably rarer. The main thing that’s keeping experts engaged at all is how stringent we are on question quality. This means that we sacrifice a little bit of kindness for the benefit of utility; in this system, users who ask questions on the site which are suitable for the site can get the attention of experts who would be able to teach them a viable solution.

Second, if all we’re looking at is a “simple” question grade, my role is obsolete. I’ve spent years looking at questions on Stack Overflow, and I have spent lots and lots of time refining what kinds of questions would objectively be suitable and on-topic on the site. I feel like I do a pretty good job…and if we really wanted to use the simple question grade for Meta, it’d agree.

Third, and probably most critical, this answer assumes the response given by the maintainer was suitable for the question. It…really wasn’t, as I articulate.

Let’s take a look at the answer provided — specifically, does it actually answer the question?


Now, why do I and my fellow curl developers still continue to develop curl and give it away for free to the world?


2. I think it’s still the right thing to do. I’m proud of what we’ve accomplished and I truly want to make the world a better place and I think curl does its little part in this.

The rest of that answer is really just conversational. They’re admittedly gushing with pride over what they’ve done (and they have absolutely every right to be since I can’t imagine a world without curl), but the signal-to-noise ratio is pretty low here.

The response I get for this analysis is one I never expected in my life — one of me being “dismissive” towards the contributions of the question.

@Makoto: I re-read your answer. When I read it the first time, I found it extremely dismissal of the author of the answer. Re-reading your answer, I see you did address the content of the answer too. But the overwhelming sense I still get is dismissive. :-(

I don’t think there’s much I can do about this. This is totally a lose-lose scenario for me. A Community Manager — someone who has diamond moderator privileges — has overruled the community. On top of that, for my efforts in trying to help Stack Overflow, I’m apparently being dismissive of the author.

In this context, I have to take the L and move past this. Except this isn’t my loss. This is really Stack Overflow’s.

We’ve undone some established precedent on how we curate and moderate content based on a desire to cease the not-as-constructive criticism of Stack Overflow as a whole.

We’re confusing the heck out of the people who actually bother with content curation and give them an indicator that none of their signals are relied on; instead, it’s the metric of post score which is primarily looked at.

I personally feel snake bit and slighted given that my analysis was overlooked the first time, and then made to seem like I had something against the author. I’m not sure how it came across like I did, but I hope my prelude in this post makes it clear that I don’t.

But at the end of the day, perhaps I should just take some of my own advice.

If you do happen to feel demonized or vilified, then there’s no harm in you taking a break for a while.

Written by

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store