I spoke before about why email sucks, and concluded the problem is that current email clients revolve around the technical mechanism of email, not how we use it. They focus on messages, and only indirectly let us deal with the things we really care about: who we’re talking to, what we’re talking about, and when. I intend to discuss these essential dimensions of email from the point of view of human beings, look at how current software addresses them, and suggest improvements. But before that, I would like to go back and explain a couple of important aspects of the current situation – and why they’re a problem – in more detail.
The Inbox Problem
My complaints in are chiefly about the inbox. Simply put, it has too much in it and it requires too much effort to manage. New mail messages arrive faster than I deal with the old ones. Soon enough messages start scrolling out of view, after which I tend to forget about them. Ideally, I’d like to have high-priority items at the top of the list, but at the same time I don’t want to miss new messages.
Flags can help by keeping important messages near the top of the list. Personally, I haven’t the patience to flag and unflag every email I receive. If I did I would run the risk of pushing important new messages out of view. But even when flags work, the inbox continues to grow. The only way to deal with this is to move messages elsewhere – usually to folders. But this requires too much fiddly mousing. Clicking on a message, then dragging it and depositing it in the right folder require a fair amount of precision. Done repeatedly with many messages, this is just too much effort.
Filter rules are supposed to solve this problem. According to these rules, email is deposited in a folder according to the sender, recipient, subject text, etc. But now, instead of looking in one place for current email, the user must now look in multiple places. This is fine for high-volume mailing lists, but for most email it only makes the problem worse. But that’s not the only problem with folders.
Folders Don’t Scale
The real problem with folders is that they don’t scale. Some people delete messages once they’re done with them, but for those of us who do not, our mail archives are always growing. This leaves us with two options. The first is to add folders – for example, to cover new topics or contacts. But then instead of having too many messages, we end up with too many folders! The other solution is to do nothing – keep a small set of folders and don’t add new ones. In this case, those folders themselves become overloaded with messages to the point where they gain us little.
Getting mail into the right folder is always a problem, as I describe above. Filter rules have two further defects: first, they are a pain to set up, and second, if the placement of messages in a folder can be so simply described in a rule then it’s not really adding any information or structure. We might as well just use a search instead. At this point, we’re back to the whole raft of fancy techniques and controls for sorting and grouping messages based on message details like the sender, recipient, time, and subject keywords. In the end, folders haven’t gained much.
Now the funny thing about folders is that they’re actually proxies for people, topics, or a period of time (I actually have folders for all of these). But these are precisely the three dimensions I mentioned earlier. Why not redesign the client to actually focus on these things directly?