The History of Email | Hacker News Reader

The History of Email

eager.io

111 pointszackbloom21 comments

If you're enjoying this series[1] as much as I am, let me highly recommend the book "Where Wizards Stay Up Late - The Origins of The Internet"[2].

1. https://news.ycombinator.com/submitted?id=zackbloom

2. http://katiehafner.com/books/where-wizards-stay-up-late/

I enjoyed that book. I've been surprised it doesn't get mentioned more often.

Obligatory link to "the case of the 500-mile email":

http://www.ibiblio.org/harris/500milemail.html

This is all wrong! If you search "Inventor of email" on Google you are shown the one-true email god is VA Shiva Ayyadurai. Stop denying him the title of inventor of email with all this propaganda.

Seriously though... I thought Google was going for accuracy in their search results. It's amazing that this idiots website is listed at the top when searching for this information.

I'd argue that Google achieved maximum accuracy with that result. Most people searching for the "Inventor of email" seem to be looking for that person, because the story of his controversy seems to be way more popular than the actual history of email.

That guy is good for a laugh if you know a network engineer who for some reason has never heard of him... Send a URL and in about 60 seconds get a reply of "holy shit this guy is delusional".

Alleged "History of email" and no mention of bang-addresses? For years we were all routing our emails via uunet with bang addresses before @ became a thing.

I actually left them out only because I had talked about them in the context of URLs: https://eager.io/blog/the-history-of-the-url-domain-and-prot...

Some more info, posted only a year ago, saving you all that sweet, precious searching time:

http://www.somebits.com/weblog/tech/bang-path-addressing-uuc...

We created a visualization called, A Brief history Of Email Apps : http://email-apps-timeline.missiveapp.com

We highlight:

- All email client ever built on major platforms on a timeline (GH issues to add more https://github.com/missive/email-apps-timeline/issues)

- Acquisition + discontinuation

As you can see last few years have been interesting!

We are still using email to this day as a relied-upon means of communication -- easy to spoof, difficult to secure, and commonly exploited for spam.

I don't see this as a case of "if it ain't broke, don't fix it".

>I don't see this as a case of "if it ain't broke, don't fix it"

It's just that every time someone attempts to "reinvent email" or replace email, they end up creating something worse or at least something less flexible.

Attempts to deal with the existing state of affairs, such as Lavabit, Protonmail who offer security, cannot offer enough features due to limitations in how email works.

What big proposals have been made to revamp email entirely?

This makes me a little speechless. If a system has these attributes that was carefully crafted, complex, proprietary, then I would agree it's a bad thing. But email is basically putting text in a certain format and transporting it from A to B. It's simple and open, of course people use it to do bad things as well.

Side note: Do you really still have problems with easy to spoof messages? Or with spam? Especially the last one is a non-issue nowadays, I think. In the past we had like hundreds of spam mails unfiltered in our inboxes. Now I have a really good spam filter and checking my spam box from time to time I don't even see that many messages popping up.

Haha, ever tried to set up your own mail server?

Nowadays you are de facto bound to mega providers to get reasonable mail service. It's just not an open platform anymore if you periodically can't send mail to 50% user base silos (yahoo, gmail, live) because of unpredictable spam filtering and no support from them to fix it.

And you need like 1GiB of RAM just to run the spam filtering monster software patchwork of perl spaghetti. Not even mentioning the maintanance work of updating this stuff.

Mail is forever broken and converges to a marked with high entry barriers in which only google and microsoft play.

I've got one Barracuda left in production. It has 512 MB of RAM but handles tens of thousands of messages a day.

Hosts running FreeBSD with 1 GB of RAM handle a few times that with ease.

I think my point was not just from the perspective of a user of email. There are entire research thrusts just in identifying spam, nowadays using machine learning, even. Much bandwidth is wasted, and if you set up your own boxes, you'll have to deal with it yourself, minimally with something like SpamAssasin.

I'd suggest that a majority of users do not value those characteristics - email is "good enough" for its intended purpose.

When you think about it, real letters and the postal system have the exact same issues.

As others have said, people keep trying to 'fix' email and failing. The problems of email are fundamentally very hard to solve without centralisation:

- in order to define "spoofing" you need an identity agency (or signing)

- in order to secure email (and signing) you need a key distribution system, preferably with an identity system so you can get the right keys for someone you're trying to communicate with. You also need to solve "local" key distribution among the user's computers, phones, fridges etc.

- in order to deal with spam you need someone to identify it and filter it for you, which inherently gives them access privileges to your mail.

The feeling when you overwrite the spool instead of appending to it.

Reply on news.ycombinator.com