Tuesday 12 April 2016

To: the World
From: Deevybee
Re: We have a problem

Everyone I know is exhausted by email. You can spend a day battling back the incoming tide of messages, but next morning when you wake up, there it is again. Much of it is spam and can be deleted without reading, but it still absorbs attention and energy. But there's plenty of other stuff that sits there in your inbox eyeing you balefully until you respond. People I know divide into two classes: those who have given in and just live with an oppressive burden of 500 unread messages, and those who destroy themselves trying to keep on top of it. The best advice I've read on how to manage the situation is this by Tim Harford, but even obeying his rules only reduces the pain, but does not eliminate it.

Is there a solution? I've thought of one. It probably is impossible but I'm going to put it out there and see what you all say.

The idea is that there should be a cost to sending email. This seems weird for two reasons:  when email first came out we all loved it precisely because it was free: adding a cost seems perverse. Second, how on earth could it be made to work?

Well, suppose big organisations, such as Universities, could set up their systems so as to filter incoming emails and check whether the sender was registered. I assume that since filtering already occurs at some level for spam, this should not be impossible. If the sender is not registered, they get a bounce inviting them to open an account. Once they have an account, then a v small charge is made for each email that is delivered. The charge should be adequate to cover the cost of running the filtering and billing operation, but no more.

This could be arranged so that communications within a domain would be free, so it would not save us from mass communications from admin – but given that in my department most of these have recently been about the serious matter of closure of toilets, and even gender reassignment of toilets, perhaps that is as well. I can also envisage institutions have reciprocal arrangements so that all university domains, for instance, would agree not to charge each other. We would also need to be able to set up 'whitelists' of addresses that would be exempt from a charge: in my line of work we use email to communicate with volunteers and organisations who help our research, so we'd have to find a way to indicate that if we initiate the email exchange, the recipient is not charged for replying.

I've been trying to decide whether such a set-up would be good on balance, or whether it would create more problems than it would solve. There is no doubt that it would lead to an initial period of havoc, but it would wipe out spam at a stroke. The fear is that it could also prevent genuine and important messages getting through. Would I miss the opportunity of a lifetime to collaborate with a colleague, to go to a marvellous conference, or to take on an outstanding overseas student? The answer is probably yes, though, of course, if people were really unmotivated or unable to register, there is always snail mail.

Well, this is just an early morning thought, prompted by the daily routine of deleting the mass invitations to meet a sexy lady, deliver keynote at a conference on sludge disposal in China or be the recipient of a huge donation from a distressed oligarch. There must be a better way, but what is it?


  1. It's a nice idea, though I suspect it's unworkable in practice.

    For me, I think the problem would be much reduced if everyone were familiar with the Email Charter:


    Quite how you persuade everyone to follow it I'm not sure, but if everyone did I think we'd all be a lot better off.

    FWIW, my method for dealing with emails works as follows:

    1. Don't keep emails open all the time.

    I close my email program most of the time, and only open it up to check my emails 3-4 times per day. This means not only that I am not constantly interrupted, but if I receive an email that requires some action on my part followed by another email 10 minutes later saying that my action is not needed after all, then I've save myself a task (this happens quite often).

    2. Process emails quickly

    This approach is largely taken from the book "Getting Things Done". When I look at my emails, I don't spend too long on each one. Is it relevant to me at all? If not, delete it. Do I need to be aware of it but not act on it? Read it (maybe only skim read it if it's long and I probably don't need to know about it in detail right now), then file it. If it needs action, then move to step 3.

    3. Choose when to act on emails

    For emails that need some action on my part, I first figure out if I can act on it quickly (how quickly depends on how busy I am, but typically in 2-5 minutes). If I can, I act on it right away. If not, I move it into my "@Action" email folder (the @ symbol makes sure it's at the top of my list of email folders) and add it to my to-do list. If it's urgent and I have to make time to act on it today then it will go on today's list, otherwise by default it goes on tomorrow's list. If it's not so urgent or I know I'm going to be busy tomorrow I'll schedule it for another day.

  2. I try to use email like I use Twitter, so that my inbox works as a feed.
    How to do it:

    1. Make sure the subject line (title/header) of each email contains the message eg 'New paper on xxx you might want to read' or 're YYY: can you comment by Fri 15 April?'

    2. Make emails short - 140 words should be enough

    3. Put further details in an attachment or accessible link eg to a Google Doc. Much easier for the recipient to save, share, and use.

    Trish Groves

  3. Unfortunately, this would not "wipe out spam at a stroke", as email headers can easily be forged, senders spoofed, and academics move around too much to be able to pin them to an institutional IP address that might serve as an added check.

    I initially thought your post was going in another direction: a social or personal cost attached to sending an email (perhaps relative to the number of recipients), which would make people think twice before sending. A bit like the bitcoin mining system, which takes a non-trivial amount of computing power and thereby creates value.

    That said, unread items are an inevitable side effect of asynchronous communication, which I think everybody agrees is pretty useful overall. I think we'll have to live with it.

  4. It's not just a matter of how to minimise time spent on e-mail, of course. Most answers to e-mails will involve you taking on yet another thing, expanding your workload exponentially. So why do we do it? An interesting aspect of e-mail is the moral or character judgements people make about those who do or do not respond promptly to e-mail. "S/he is SO efficient. They reply to all e-mails within minutes of my sending". "S/he doesn't reply to e-mails!! What on earth does s/he Do all day?"

    Does this matter? Probably not in practical terms, but it is a conversation opener to more interesting discussions on the influence of e-mail on our social interactions, concepts of self and others etc.

    So, be brave. For those nagging e-mails asking you to take on something you could really do without, don't answer. They'll ring you eventually if they really can't find someone else to do it, and it is much more difficult for them to persuade you by flattery over the phone than by e-mail.

  5. This is not directly relevant to your suggestion, Dorothy, but sometimes when I am looking in despair at my inbox, I try and remind myself of the annoying and time-consuming things that used to suck up my work time in the pre-email days - but that now don't.

    Exhibit A: the photocopier. Honestly, I feel like I spent most of the first year of my lectureship in the photocopy room. We did have email then, but not much in the way of group mailing lists etc, and most communication was still paper based. So sending a simple message to my department colleagues involved writing the message using a "memo" template in word, printing it out, making 40 photocopies and then putting one in everyone's pigeon hole. Definitely a 20 minute job compared to the 2 minute task that would now be. Likewise, circulating a journal club article - at least 20 minutes at the photocopier printing out a copy for everyone (and swearing at the paper jams/collation failures etc), whereas that is now done in no time. And submitting a paper! At least three hard copies - that had to be successfully printed, photocopied, collated, stapled (endless stapler failures) and then packaged up to send by international mail. Admittedly electronic paper submission can be frustrating too, but I think certainly less time-consuming. I feel better about email when I think about the photocopy room.

    Exhibit B: the phone. In those same days, when I walked into my office in the morning my phone would inevitably be flashing with 6, 7 or 8 messages. And my pigeon hole would be filled with little cards with phone messages for me to respond to that had come through the office. I can't exactly remember, but I'd say I spent at least the first 1/2 - 1 hour most mornings following up on those phone messages. Now, I rarely see my phone flashing more than once a fortnight.

    So I find that reminding myself of these time-savings - and having a good spam filter - is quite therapeutic.

  6. I don't find "genuine" spam to be much of a problem. First, there is less of it than a few years ago, and second, it only takes a second or two to identify and delete it.

    Much more time-consuming are mails that don't *really* require a reply, but you feel ought to get one for the sake of basic politeness (and/or not thinking that people out there are saying, "Well, who does that stuck-up Nick Brown guy think he is, I sent him an e-mail and he didn't even bother to reply"). These often involve telling someone that you don't really have time to look at their issue (or they've just asked the wrong person), which in turn means that you have to be extra-careful about the wording. Six of these at five minutes each and that's half an hour you're never going to get back.

  7. I'd second Nick Brown's sentiment that spam isn't really much of an issue these days. I hardly see any of it, spam filters have gotten so good.
    I've also set up email filters that put direct emails to only me in a special place and all other emails in various folders that I can choose to look at or not. That narrows down the 'must read & reply' emails per day to a manageable size.

    Obviously, the more prominent people are, the more direct emails they get :-)