This thread is about Slack, but it’s also a glimpse into how and why email slowly destroyed itself. 1
There was quite a bit of innovation in the email world 20 years ago, but Microsoft—as they do—stomped the shit out of the competition. 2
Then there was one: Outlook. Email was Outlook and Outlook was email. (Yes, I know there were others (don’t DM me), but, hell, there’s still Bing) 3
The innovators at Microsoft were willing to change a lot of things about Outlook in a quest to improve email, but the one thing they never changed was the Inbox. 4
The inbox is a perfect illustration of the concept of implementation model, where the interaction model presented to the user is a literal rendering of the underlying technical structure. 5
The obsolete “Save” and "Save As…” menu items are examples of implementation model. They are a literal rendering of the shape and function of executable main memory and secondary disk storage. 6
The inbox is pure implementation model: A literal representation of the big FIFO stack of ASCII texts that was how email was actually constructed under the hood. 7
The core problem is that, while email technology is about FIFO stacks, that isn’t how people think. That is, email clients were ALWAYS based on a non-existent and problematic mental model. 8
Virtually every innovation in email software was a band-aid over that mortal wound. 9
Every other subsequent text-based communications system was a response to that fatal design trope: SMS, text, Messenger, you-name-it. 10
Threading was the big leap forward. It recognized that there was something…the industry never quite seemed to grasp what it was…beyond just a sequential list of stuff. 11
There were many problems with that: threading isn’t really a commonly held mental model. Threading was a layer of shit added on top of the inbox. conversations are more complex than simple threading models. 12
And two other huge problems with the inbox model dragged it down. 1) there was no way to deal with last month’s emails, and 2) spam. 13
Recall that Microsoft never effectively addressed either of these problems. Nor has Google’s Gmail. They remain significant issues. 14
Gmail actually implemented a solution to the “last month’s email” problem: they just keep adding more servers and keep everything. This is of course problematic, and I’ll leave the understanding of that as a problem for the reader to solve (Yay Knuth!). 15
Despite all of the effort to defeat spam, not much has changed and everyone’s inbox is full of shit all of the time, and that’s likely the main cause for why everyone hates email and refuses to use it. 16
By the way, Slack is an inbox system, as is Messenger and SMS. It’s just that they are closed systems, so spam is a controllable problem. We have SO thrown the baby out with the bathwater on these shitty substitutes for well-designed email. 17
In 2007 I founded a company (called Citizen Software) to address the problem. As usual, I was very lucky and failed to attract venture capital (long story). Our slogan was THINK OUTSIDE THE INBOX. 18
Our product was an email client that solved the big problems of email by getting rid of the inbox. We instead used a visual conversation space, where you could stroll from one thread to another spatially, rather than temporally. 19
The “last months email” problem was solved by the simple expedient of automatically deleting ALL emails after, say, 6 months, unless you mark them for permanence. 20
The spam problem was solved by the equally simple expedient of only showing conversations with people you knew (like a closed system) and spam fell into the bilge for you to peruse in your spare time. 21
None of these solutions were fully worked out of course, because we never shipped. However, they were representations of email based on human mental models rather than implementation models. 22
The @jensenharris thread makes clear that Microsoft’s research was based not on what people wanted, but on what people wanted from existing interaction paradigms: "pilers and filers.” 23
And isn’t it funny (but not ha-ha funny) that the challenges of Slack are the same ones that faced email designers 20 years ago: disorganization drowns us, and organization is not a useful tool for dealing with email/tweets/slacks/texts. 24
Search is a good solution, but ONLY IF you are a computer software geek. Normal humans still seek some human-scale order in their digital communications, and fail to find it. 25
So, there will emerge a replacement for Slack, just as Slack replaced Outlook. For a brief shining moment, the new platforms are empty and appear to be a heavenly salvation to the problems of the previous platform. 26
But the problem is identical to the one faced by traffic planners: adding new roads only briefly eases traffic, as traffic density is based on the human tolerance for frustration and delay, not an absolute quantity of vehicles (texts). 27
So, like travel, the solution is multi-modes of communication/transportation, each one based on human mental models rather than on implementation models. Good luck! 28/
