I'm going to take the opposite of this argument, and for bonafides, I run the product team for a 1000+ person company.
I'm going to assume that this thing was tested out the wazoo, both qualitatively and behind enough feature-flags and buckets to keep Optimizely in business for a year. So I have no doubt that this wasn't just a hunch on their part, but something that they meticulously tracked. If I'm wrong, definitely let me know.
Anyway, the main point, what percentage of people do we think knows markdown to such an extent where it's easier for them to use than a WYSIWYG editor would be? I'd venture on HN alone that number is <10%, and considering Slack is used by a far wider and less-savvy audience than on here, I don't have much of a problem with the reasoning behind the change.
I think the lack of backwards-compatibility is a legit criticism, and sure, an opt-out would be nice, but I don't think for a second that this is as egregious as people are making it out to be.
As someone who runs the product team for a 1000+ person company, the concept of intransigent minorities should be of interest to you. [1]
The gist of it is that an intransigent minority can often dictate the choices of a flexible majority. It is unwise to make decisions based solely on majority rule, as stubborn minorities can affect complex systems in ways that you do not expect. The phenomenon can be verified in a variety of situations across history.
To put this into context for Slack, I believe technical users have been particularly important for the app. Tech users were early adopters for Slack - I'm sure a lot of corporate accounts started out with engineering teams. Tech users wrote the chat bots and integrations that Slack is known for. They got people to use Slack, and they can get people to stop using it.
Let's say you're changing the way a certain feature in your application works. The majority of users are happy or indifferent about the change (but weren't particularly bothered by the old behaviour). A minority of users not only reject the change, but are vocal in how much they hate it. They will not tolerate the new behaviour, and will try to get other users to abandon your product unless you resolve the situation. What do you do? You can ignore them for now, but can you predict the long term effects of turning your back to the intransigent minority?
And I'd add that inclusion is especially important for a tool that's supposed to connect everybody in the company. Thinking that one can ignore a chunk of the audience because they have no choice but to keep using your product is at best confused, and at worse displaying enormous arrogance. It's the sort of enterprise-grade nonsense that Slack was initially seen as an antidote to.
I get that things that want to be popular quite often shiv the early adopters as they become mainstream, as a) that particular orange has been well juiced, and b) satisfying early-adopter needs really can get in the way of that holy grail, maximizing revenue.
But here I don't think there's a big conflict. There are a bunch of ways this could have been done to make everybody happy. It could have been opt-in for a while and then opt out. Or as others point out, there are apparently good WYSIWYG experiences already out there. With a market cap of $12 billion, Slack could afford to buy one, or at least the team behind it.
The only explanation I see is that it was rushed out the door to meet some executive's goals. Which suggests that Slack is becoming the sort of clueless, sales-driven, user-contemptuous company the initially aimed to overthrow.
Everyone that was using slack was clearly aware of backticks, underscores, and stars, which is really the most you need, and probably 99% of everything used.
Surrounding some text with these marks to achieve some formatting is hardly rocket science. If someone doesn’t do it it is really because they don’t want to understand.
If you however make an editor that fucks up using any of these marks? Well, that’s a big time failure.
Exactly, all the cool kids were using backticks, underscores and stars. Naturally the lesser users would figure it out just to fit in. Everyone feels mildly powerful being able to wield a tool like this as trivial as it is. Now we're all just back to being bottom feeding piker fish toiling away on word docs.
I work with a bunch of non-tech people. I get what they're doing. The techy people are the only ones who bother using ` for anything. I have no idea if non-techies will start now that there's a button for it, but typing it is now really annoying for everyone who has been using it.
I kinda reminds me when Confluence added their WYSIWYG editor--or using Frontpage. I basically /had/ to go to code view to get the page to look consistent. So far Discourse has been my favorite implementation, with preview being off to the side. I'm sure that would eat up too much real estate for Slack.
>Anyway, the main point, what percentage of people do we think knows markdown to such an extent where it's easier for them to use than a WYSIWYG editor would be?
Well, Slack supported so little of Markdown that this is a moot point.
If they had all those metrics and that many test users, how could such obvious backtick issues exist?
Even simple dog-fooding by their own internal coding team should have revealed that issue.
If they had a test group, it was definitely a group that didn't use those features to begin with and basically didn't use them in the new UI either. Meticulously tracking the wrong sample group simply leads to reams of bad data.
It's logical from a short-sighted perspective as I'm sure it looks good to investors to have a longer checklist of features. But shipping a broken feature is eroding customer trust.
Anyway, the main point, what percentage of people do we think knows markdown to such an extent where it's easier for them to use than a WYSIWYG editor would be? I'd venture on HN alone that number is <10%, and considering Slack is used by a far wider and less-savvy audience than on here, I don't have much of a problem with the reasoning behind the change.
I think the lack of backwards-compatibility is a legit criticism, and sure, an opt-out would be nice, but I don't think for a second that this is as egregious as people are making it out to be.