A ‘frighteningly ambitious’ way to improve email

I've read about a very similar phenomena over at http://en.wikipedia.org/wiki/Attention_economy.

"One way this might be implemented is by charging senders a small fee per e-mail sent, often referred to as a "Sender Bond." It might be close to free for an advertiser to send a single e-mail message to a single recipient, but sending that same e-mail to 1000 recipients would cost him 1000 times as much. A 2002 experiment with this kind of usage-based e-mail pricing found that it caused senders to spend more effort targeting their messages to recipients who would find them relevant, thus shifting the cost of deciding whether a given e-mail message is relevant from the recipient to the sender (Kraut 2002)."

--Please feel free to ignore the following email rant.--

Email is a pretty impressive digital translation of the written letter. And for cases where a letter written by hand on paper would work well, email works even better. As long as by better you mean faster and without paper or stamps or penmanship. The SMTP protocol is over 30 years old and still going strong; that's pretty amazing. But email kind of sucks, too.

Frankly, most people aren't good at writing letters. Simply creating a meaningful subject line is difficult. And keeping a conversation germane to the stated subject is also difficult; folks often talk about multiple topics in a typical conversation. To review long conversations, one must read awkwardly from bottom to top, opposite normal English reading direction. Adding someone to a conversation in progress is also awkward. Sending another message (a correction or addition to your last message) before the person (or people) you're talking with replies creates confusion as a reply can only be made to one or the other message, dropping parts of the conversation. Attachments can be problematic. Clearly, spam the likes of which we see today was never anticipated by SMTP's authors, and neither has anyone come up with an especially successful solution to it. We've all just come to accept spam as part of the background noise, filtering the polluted stream, rather than removing the decomposing carcasses from the upstream source.

Creating an alternative that addresses its shortcomings would be ambitious. And I think many people were frightened by Wave, which addressed many of email's shortcomings but was disruptive in the way so many people like to claim they admire but will almost always shun in practice (kind of like how folks always claim to love to root for the underdog, unless it's Haiti). I'm not saying Wave was perfect, but it had great potential. And as far as genuinely ambitious attempts at improving email (or the kind of conversing we typically do with email), I can't think of another example.

I wonder if this could be useful here.

http://www.telescopictext.com/ (Context: https://twitter.com/rgherrmann/status/300311324905324545)

It would definitely have to be automatically generated to be viable, as in no more effort than to write a normal long email. But doing that automatically is a hard problem.

This is pretty much what Inbox Pro already does, no? I tested it by sending an email, and got an automated response:

> Hi [my name],

> Your email is #85 in Boris Veldhuijzen van Zanten's inbox. You can __optimize your message__ so it will be easier to answer.

Clicking the link leads to a webpage showing average response times for Yes/No questions (5 days), Short questions (I think 14 hours?), Long questions (3 days), and FYIs (3 days). I converted my message into an FYI and submitted it.

I think the only difference in what you proposed is your version limits it to "the first time a sender emails the recipient", while this system would presumably give the same treatment to future emails from me.