[AnimalMuppet]

Paul Graham said that email was horrible, and it shouldn't be hard to come up with something better. But he didn't actually put forward a concrete proposal.

So: What would be better? Given that the world currently runs on email, what would be enough better that it would be worth it to switch? I don't know. I don't think Paul Graham knew then, or knows now. And I think that the reason people haven't switched is that, so far, there isn't an alternative that's enough better. And the reason there isn't one is because, contrary to Paul Graham's expectation, creating something better is actually pretty hard.

[clusmore]

>So: What would be better? Given that the world currently runs on email, what would be enough better that it would be worth it to switch? I don't know.

This is something I've thought about quite a bit recently. For one, I absolutely hate the fact that when I give out my email address, the receiver can give my address to a third party. The third party has full capability to contact me, I can never know who gave up my email address, and I can never revoke the capability. At a minimum, an email replacement would need to address this.

You're right, replacing email would be really hard. There are a few things I'd like to see that could at least improve the current situation.

1. Push companies/organisations that request your email address to provide you in advance with the sender address they will use in the future to contact you, so that if we wanted to we could maintain a whitelist of accepted addresses to prevent all unsolicited mail.

2. Come up with a service that makes it easy to manage throw-away addresses that redirect to the same inbox so that I can essentially revoke the address that I hand out to somebody.

[stephenr]

> when I give out my email address, the receiver can give my address to a third party

Use sub-addressing [1]

So, if you signup on example.com, and your email is "[e-mail]" you give them "[e-mail]" or whatever format works for your mailbox provider/software.

If you wanted to go one step further, a filter could be configured to look at the sub-address component, and compare it with the sender's domain and mark it as spam if there isn't a match.

1. https://en.wikipedia.org/wiki/Email_address#Sub-addressing

[clusmore]

This can work for online services during signup. I haven't yet seen this, but I wouldn't be surprised if some services were smart enough to drop the sub-address before on-selling your email.

In face-to-face exchanges (for me recently, this came up a lot while giving out my address to real estate agents), anything involving '+' causes a huge amount of confusion and they start asking questions. I'd much rather a "normal" looking address, where I'm perfectly happy for normal to be say base64 with '.', '_' and '-' and as the three special characters (at least people will think its valid).

[stephenr]

+ doesn't have to be the separator. Gmail supports a '.' for example, and if you run the mail server yourself, its just limited by what characters the software you run supports.

If you have your own domain, you can of course also do [e-mail] and have it go to a catch-all style mailbox.

[LorenzoLlamas]

That's true of your name, your address, your DOB, SS, and your phone numbers. Nothing stopped those things from being given out and collated by third parties. Nothing will stop email being given the same treatment.

All the tech tricks to try and figure out who gave who my email address are pointless exercises in time-wasting. Adding various words to your email (like with Gmail aliases) does no good because anyone can figure out your real original email. Only machines are (somewhat) fooled by this.

Of course, if you owned your own domain, or paid for quality email, you could have many multiple email addresses to partially address this, but since everyone switched over to so-called free email with social graph companies, what's the point.

Email will never die. Why should it? It's fast (instant, nearly) and you can send nearly anything to anyone - limited only by size (and with a few tricks, not even limited there). It's worldwide (not limited by who's on what 'network' or what is popular in whatever country).

If anything, I think we are seeing (or will see) a resurgence of email as the 'de facto' way of contacting and communicating with people. It even holds up pretty well in court. Email newsletters have made a come-back (a little bit) and with the stagnation of FB and G+ (did it ever grow?) and Twitter on the verge of Ch 11, how will you communicate with others? How will companies reach you?

Or did you want a fax machine?

[clusmore]

>That's true of your name, your address, your DOB, SS, and your phone numbers. Nothing stopped those things from being given out and collated by third parties. Nothing will stop email being given the same treatment.

That's why I'm suggesting that an email replacement not suffer from this same flaw. Just because it affects other things important to me, doesn't make it not a flaw.

>Email will never die. Why should it? It's fast (instant, nearly) and you can send nearly anything to anyone

This is why I'm suggesting that the best replacement to email might turn out to just be a wrapper around email so that you get full compatibility. As others have suggested in their replies, you can work around some of the flaws in email simply by using it differently.

If I ran my own server, I could create a new address per contact and then I could "revoke" the address by instructing my server to trash anything sent to that address. I'd do this if I ever noticed unsolicited email come in and I would know immediately who gave it up based on the receiving address. The contact would never need to know that I was doing any of this, they'd just be emailing me as normal. The only problem with this at the moment is that a lot of email servers will outright block all incoming mail from unknown servers.

[toepitt]

There are workarounds to both 1 and 2. A company or person with a legit message can vet the message without the need to whitelist the sender address.

If anyone powerful is serious about trying a new twist email me.

[dwe3000]

I think part of the problem with getting rid of email is that it has proven itself flexible enough to handle a variety of communication needs, so to replace it, several tasks/apps need to work in combination. For example, an archived text messaging/IM handles part of it, possibly including identity, but something would also need to stand in for the long-format or large file transfer (like Dropbox, Google Docs, etc.).

[tgragnato]

IM apps are not a feasible plan until they will provide a "no lock-in" interoperability (adding to archivability).

> You would think it was crazy if your cell phone could call only people with phones on the same network. But we put up with that absurd situation when it comes to instant messaging -- and have for years. Worse, there's little sign of change anytime soon.

The premises to accomplish this are: protocol standardisation, dissemination among users, collaboration between app makers.

> https://en.wikipedia.org/wiki/Comparison_of_instant_messagin...

The decentralised protocols listed are: Bitmessage, Ricochet, Ring and TOX. None of them is widely deployed.

Email is decentralized, that means I can quit Gmail any time I want and use another service. Or even start my own.

[dwe3000]

Agreed. I would want "No lock-in" for IM to be a requirement. I wonder whether or not pure volume - e.g., Facebook Messenger - would be enough, however, to drive people in that direction even without the interoperability with other systems.

(I am not a Facebook user; just recognizing that they do have market share on their side.)