Dan Luu
patreon
An unintended consequence of company propoganda
🕑 Added 2017-10-29 16:05:22 +0000 UTC I once met an employee from Blackboard. Without thinking, I stupidly blurted out “What’s it like working on software that no one likes?” Surprisingly, rather than being offended, the person was just confused. They thought that the software was widely loved and couldn’t believe that anyone liked the software. This struck me as pretty strange since, in my social circles, Blackboard is the most disliked software in existence. There’s more disliked software, such as Epic medical records software or SharePoint, but while those are hated, most people don’t have exposure to them. Blackboard has achieved such dominance in the educational market that most people I know have used the software; no one I know likes it.Blackboard (as well as Epic and SharePoint) are classic examples of software where the user and the purchaser are different and the purchaser basically doesn’t care about the user, so you don’t expect the software to be good and it would be a bit of a surprised if the software didn’t deliver a bad user experience. But it was even more surprising to me that someone working on some of the most reviled software out there could think that the software was widely loved. It hadn’t occurred to me that the software was also bad because the people writing the software thought the software was great and therefore didn’t need improvement, but this turned out to be the case (this episode was before I worked at Microsoft and met people who worked on widely hated Microsoft products).
I was recently reminded of this Blackboard employee by a discussion on lobsters moderation (lobsters is an HN or reddit like forum). The moderator deleted a short, content-free, negative comment, which caused an uproar in the community -- many people are against any sort moderation except spam removal -- which led to a discussion on what sort of moderation should be used. Part of the moderator’s response was that unmoderated communities, like Usenet, 4chan, and YouTube, have terrible discussions.
A tumblr employee responded with:
At Tumblr I think we solved it (mostly) mechanically. I left in 2015 so some of this may have changed, but we felt very strongly that “sunlight is the best disinfectant” and we did not delete content unless it was child porn, self-harm, or gore. You can still go see terrorist cat blogs posting under #catsofjihad. The community was great about shutting down those who were abusive, and the mechanics of response were important for that. Once you created a post, anyone could reference your original, even if you deleted it. Tumblr threads are not trees, but the particular chain of responses that are favored by whoever decided to respond, giving responders full control over the context, making responses far more clear. You lose the full-picture, but gain clarity of context. This was backed up by a support team that was usually great about taking action when abuse was happening. The clarity of context is what I believe gave the community the tools it needed to effectively self-police. This is something I believe is sorely missed in other platforms
I find this response to be truly extraordinary, up there with the response from the Blackboard employee I met. In the communities I know of, tumblr is considered to be roughly as bad as twitter (there’s often good discussion, but it’s in spite of, not because of tumblr’s mechanism design, and terrible discussions are everywhere and often bleed into good discussions). “Shitposting” is much more common than on twitter and quoting people out of context to create a giant pile-on is roughly as common as on twitter. Tumblr’s longer length limit allows a sort of nuanced discussion that’s impossible on twitter, but it’s not clear which is better (or worse) overall.
The situation is so bad that Scott Alexander’s canonical examples of bad mechanism design leading to outrage-porn instead of real discussion are twitter and tumblr. People have some objections to Scott’s post, but you basically never see “tumblr’s mechanism design leads to good discussions” among the objections.
Given that tumblr’s mechanism design is often considered to be so bad that it’s a canonical example of bad design, how can it be that tumblr employees think that they’ve “solved it (mostly) mechanically”?
I think part of it is that people tend to think that what they’re working on is pretty good, even when it’s not particularly good. Companies then encourage this tendency as much as possible. Every company I’ve worked at has spent a fair amount of effort convincing employees that the product is amazing, that the product is about to blow up and become a huge success, and that users love the product. This seems to pretty well. I’d say that, independent of the company, roughly half of employees buy into this and believe that their product is widely loved and about to take off, even when it’s clear that the product is widely reviled and going nowhere.
This kind of internal propaganda probably has benefits for the company -- it seems plausible that people are more likely to stay on products if they’re true believers, leading to lower attrition and it’s also plausible that companies can pay people less if their employees really believe in the mission and the product.
One thing I find interesting about this is that the internal propaganda is pretty clearly aimed at line engineers. As you go up the management chain, fewer and fewer people buy into the propaganda, and if you have a candid conversation with someone at or near the top, it’s clear that they don’t buy the propaganda and are well aware of the problems in the product and its place in the market.
Maybe this actually works ok because the people making the decisions aren’t blinded by propaganda, and the company ends up being more profitable because employing true believers is cheaper than employing mercenaries. However, in most places I’ve worked, line engineers have a decent amount of freedom and can significantly affect the product. If you convince them that the product has no major problems, the problems are much less likely to get fixed. Maybe this is the right decision from a business standpoint -- after all, all of the products mentioned in this post seem to be doing pretty well. But I can’t say that I like it.
