Hacker News Reader

Comment by mrb | Hacker News Reader

The fact "email will last forever" is precisely why I believe it is absolutely the best protocol to build automatic services to access and share personal information. I have spoken about it here: http://blog.zorinaq.com/?e=76

Edit: Someone1234: you don't need to wait for companies to support my proposal. It already works today with Gmail: try my live demo on pm33at@gmail.com - it works by leveraging standard auto-responder features. As to the comparison with XKCD 927, it is invalid: my proposal is radically different from anything that currently exists, because it is decentralized while all current personal information sharing platforms are centralized.

Edit: rakoo: yes, XMPP would work too, and would probably be a bit better than SMTP because it is real time. However in practice building something over SMTP has more chances of seeing success (because not everybody uses XMPP, but everybody has email).

View on HN

That sounds like a really cool idea. Have you submitted it here? Got me thinking of pluses and minuses, and ways to bootstrap. Probably somebody would have to put up a free email service that implemented these hidden auto-replies, and maybe submit some pull requests to open-source projects to use it. Maybe Android and iPhone apps for the service that handled the phone info side. Sounds a bit big for an individual side project to me, but maybe it's a startup. It would probably help to get input from a bunch of people/projects on what they'd want for specific use cases.

Seconded. This is a great idea, please submit it here for discussion.

Alright, I submitted it again:

https://news.ycombinator.com/item?id=8263848

I had submitted it twice before, but it got few upvotes: https://news.ycombinator.com/item?id=8157645 and https://news.ycombinator.com/item?id=8030337 It is hard to find a short title that succinctly describes the concept.

ufmace: I think writing a browser extension that configures filters + auto-responders on some popular webmails (Gmail, Hotmail, etc) would work best to entice users to adopt the technology. As I explained, many webmails already have all the necessary features to implement the concept (see FAQ 1 at http://blog.zorinaq.com/?e=76). Also, writing a smartphone app to do the same for mobile users would help. There is really no need to launch and run a full-blown webmail service.

I upvoted it, though I feel the pain on getting upvotes and reads on here. I think the last article I wrote got more upvotes on Lobsters than here. You gotta get just the right title at the right time, or it slides off the front page before anybody sees it.

I'm not sure what the best way to get going is. A webmail service dedicated to it would get the best UI, but there's the overhead of getting a webmail service with competitive features to Gmail, Yahoo, etc going and keeping it running. Or you could do a plugin or something, where it might be a challenge to put together a good UI and keep it in sync with Gmail's updates.

Upvoted and commented. Thanks for thinking good ideas.

I like the proposal, but rather than using special (unfriendly) subjects, I'd use custom email headers. eg: X-Autorespond: name, age, location

Why? So that clients that do not support this protocol will display a normal email to the end-user, instead of something not-so-friendly.

This also means that the body can be free-formed text explaining to the user what this message represents, and how to support this protocol.

It strikes me that this is all available with XMPP, which also has user@domain address; If we build the correct tools, XMPP can very well supplant SMTP.

http://sat.goffi.org/features.html

Which makes it even sadder that XMPP is being dropped by so many. XMPP is awesome.

Indeed, if XMPP gained enough popularity, it would solve this problem, since it already implements several of these things (personal information, location, etc), AND is extensible.

Does this have the classic XKCD 927 problem (https://xkcd.com/927/)? Also wasn't this what XML was meant to solve (a single unified standard for interop)?

It could "work" but unless you got at least Microsoft (Exchange server, Hotmail) and Google (Gmail) to sign up then it won't go anywhere (which is largely the issue with all email momentum, Microsoft and Google have no interest so it cannot move forward).

XML is a markup to establish protocol, not a standard by itself (other than the basic syntactical standard). A standard would be something like MathML.

Google did try to evolve email with Wave... it didn't take, for the reasons described in the article (among others, perhaps).

I tried emailing that address and didn't get a reply.

But I think that could be a rather cool idea - something like REST over SMTP? Know a single email address, send it a blank email get back and kick off the whole HATEOS process - which actually might more sense for something like this rather than over HTTP.

Put "autoresponder:online_identity" in the subject, you will get a reply.

I do like the blank email + HATEOAS idea.

Ah yes - I was being a bit thick and I hadn't read your blog post at that point.

NB I left a comment there but it didn't format correctly. What I suggested was having the subject line be similar to the first line of an HTTP request i.e. <verb> <path>

Just read your writeup and tested it. Holy hell, this is an awesome idea! I would like seeing it adopted just like social sign-ins.

This is a very simple, yet elegant idea. I'm beating myself with a stick not to have thought of it earlier!

Reply on news.ycombinator.com