I don't think the premise is right. From the beginning to now, Gmail evolved quite a bit. For one, they used their analytical knowledge to prioritize emails, which I found to be very helpful. Two, they started filtering newsletters and advertisements automatically. Third, the UI got revamped quite a bit. Add to that all the basics Gmail does right (spamfiltering, tag+-syntax, …), and you see why it is hard to compete.

That said, I don't use gmail anymore because I don't want my emails to be directly in the hands of the NSA.

But let's focus on their list for a bit. Because innovating a sector is not a goal. Having a goal and therefore innovating, that is a goal. So, does the list they formulated hold up? I don't think so:

better interfaces: Better than gmail? Good luck. Besides, even old-style programs like claws-mail, sylpheed and thunderbird have really good interfaces, for their purpose.

better email management: Gmail does that. And one has to be really careful with that: It is not something every user need, sure not something every user wants, especially if it even once sorts something the wrong way.

better workflow integration: Emails are not tasks. They don't need task management in general. Emails are a communication medium, and sometimes, they contain or become tasks. but that is not a general requirement for users.

better attachment handling: What does that mean besides searching for attachments, maybe including their title, and browsing through the photos? Like attachments.me tried to do. But all of that is not innovation…

better social integration: The action one might want to do, like tweetbacks, should be linked in the email. Isn't that already the case? Blogs do that (ok, serendipity does, don't know about the others).

better prioritisation and analytic: To general. Besides, Gmail tries that already.

So, hmpf. Though I want to add that their general product idea (shared inboxes for taskmanagement) doesn't look bad, for a very specific usecase.

I agree with you almost entirely. Although I do have one specific minor off-topic feature I'd love to have in email that I haven't seen anywhere.

I want the ability to "like" an email, and have the sender know that I "liked" it. I don't want to write "yes, I agree", or even worse, a terse response of "like" as I feel that's a waste of my time, the sender's time, an unnecessary "+1" in their inbox.

I just want a way to say "I read your email and it sounds good to me" without having to actually write anything. Just click the "heart" or whatever symbol, and the sender is notified next time they open their email client with the heart showing next to the thread. Would work out nicely for lists and group emails as well.

I suppose there are clunky workarounds that would allow such a thing dependent upon client implementation, but it's just not that significant of a thing.

As for shared inboxen for task management, I'd recommend Asana (with whom I've no affiliation, besides being a happy customer).

What about an UI which mimic a social network in the Facebook style but which uses email as backend? It would be decentralized for free, and entirely interoperable with people using plain email (those who uses this UI would see that you liked there message, others would see a reply that says "I liked your email", for instance). Comments would be responses to email (using the In-reply-to header). The interface could easily creates groups of contact or post to all of them (equivalent to posting on your "wall"), etc. The action of "friending" someone on this UI could optionally be mutual GPG key signing behind the scene which would enable encryption of messages between friends.

I think there is a lot to do with this idea.

I have wondered if you could build a federated social network (or something that is more like an inter-social-network) with email as the underlying transport. You'd need to layer a lot of top of this, of course, but it would be kind of fun.

I like the idea. There is many ways to extend e-mail functionality by building additional layers on top of it. But imagine having a group e-mail conversation for a large number of people where just a fraction of them is using the featured e-mail client. How would you communicate this additional information to the other? More e-mails? Or just completely skip it?

At http://www.mailcloudapp.com We're currently working on new ways to use e-mail. Always hungry for feedback and new ideas! ;)

My first attempt at a solution, were I to actually try something like this - say, with my hypothetical chrome/firefox extension called LikeMail - would be to send out a canned email with custom headers.:

"enobrev has read and 'liked' your message with the LikeMail chrome extension! Download it _here_ so you can do the same with your emails."

Then on the recipient's client, if they have LikeMail, consume that message and hide it, while showing the appropriate signaling in the inbox, etc.

Definitely a clunky workaround, and can easily be considered spammy, I suppose, although if that automated message were customizable, maybe less so. I think the reason for the email (as opposed to skipping it) is to convey the info the action of "liking" is meant to convey. Essentially if I hit "like" button in LikeMail I'm trying to tell the sender "I read and liked your email". Sending nothing would go against that intention.

This in general is one of those things that I thing's been growing in the background of the internet: the utility of fake Internet points as social feedback. This includes karma, likes, +1's, retweets and more.

It's the ability to give pure positive reinforcement without actually using words, and it's one of the most powerful ideas to come out of the Web.

To a lot of corporations email is more like version control. The attachments are the source code, and the text of the message is like comments in code. The stream of information is a fundamental output of the business, and is definitely not just a communication medium. Managing that stream is not something that can be fully automated; it just needs better tools (like Github achieves for Version control).

True, in businesses humans agent are the programs (doing the work) and discussions are meta-level decisions aka source code.

Emails are not tasks? I disagree. You're right that they are a communication medium but it not incompatible. From my point of view most of the time they are both.

Sometimes it is just a question of definitions. If I get an email from my mum with an ebook attached, I can regard it as a task (to add that ebook to my reader, to write back later). I'd argue that this is not the definition of a task, the writing back part alone especially not. That is just communication.

But sure, if you use emails primarily to collaborate with others, for you the emails are task thinking might very well be true.

> I don't want my emails to be directly in the hands of the NSA

Then I hope you don't send any messages in plaintext, and good luck with the headers.

I sure try to encrypt as much as possible.

The #1 difficulty with email startups is that they have to invent a market that doesn't exist and doesn't want to exist.

Few people even know that they can access Gmail anywhere other than at mail.google.com or via the Gmail app. Ditto for hotmail/outlook.com and pretty much every other freemail service out there. And of course ad-supported freemail providers have no incentive to let their users access email with a third-party app.

The distinction between an email service and an email app is so murky in the minds of most people, that Mozilla added a dialog box to Thunderbird where they offer to create a new email account for the user. Seriously, a lot of people were wondering why Thunderbird didn't come with its own email address! Instead of trying to dispel this myth, Mozilla just let people keep believing it.

You're ignoring enterprise, there is a huge industry around business email hosting which more often than not uses an email client (and more often than not that client is outlook or the iphone mail app).

Google Apps was certainly taking a huge chunk out of enterprise, but I don't know if that's still happening now that you have to pay for it from employee #1.

Outlook usage is so embedded and big enough that Google built a sync app, but moderately funnily, the 'how to integrate google apps with Outlook' video on this page is missing, not sure if that's a reflection of how few people use it?

https://support.google.com/a/answer/33322?hl=en

Here is an idea for something that has been in my head for some time.

All that would be required to jumpstart the innovation process is a developer-friendly API and standardised events for extending email:

  onReceive(headers, body);
  onOpen(...);
  onTagged(...);
  ...

That would allow developers to do creative things with email. It may even give way to niceties such as:

  npm install email-itinerary-parser

These events could be standardised across the board. A plugin ecosystem would allow emails to operate much like blogging platforms do today. You could have a self-hosted email platform with a plugin dashboard where you would pick and choose your plugins. The UI would be the hard part but is a tractable problem given the state of client-side MVC libraries that didn't exist until a few years ago.

A service like Akismet would provide SPAM filtering and other services could be built around it. The trust factor would be an issue but it might be even possible to send these services only extrapolated data instead of the entire contents of the email.

It would even be possible to build atypical interfaces around your email. You could for example build a blogs UI that shows all emails sent to [email protected]. The blogs Controller could automatically reply to that email with the published status.

In general, things could move in a progressive direction much faster in the hands of the developer community.

Email didn't evolve with Gmail. Only the UI evolved.

It could be suggested that Wave was a revolution rather than an evolution, and that was its greatest weakness. I'm sad that it's gone.

Not completely gone, it still exists as Apache incubator project : http://incubator.apache.org/wave/

True. But for it to succeed and be widely adopted, it really needs an organization that is able to host it for the masses.

Yet another case of a blog where I click the logo at the top and I don't get taken to the product, I get taken to the blog index. Infuriating!

It's fixed, sorry for the infuriation :)

Interactive email Web UI is around since 1997 when Microsoft introduced Outlook Web Access. GMail is not really that special.

I think largest email innovation is right now happening at KDE with integration of KMail into rest of the desktop.

I thought the first webmail was HoTMaiL (notice the original capitalization) in 1996 and that is was only later bought by Microsoft.

I love it when the Microsoft apologists come out :)

Outlook Web Access... yeah that changed the world, gmail was but a shadow in its glory.

I use Linux since 2002 and actually hate M$.

But yes, it introduced Ajax and changed the world. Gmail was revolutionary because it offered 1GB when 10MB was the norm. Not because of UI.

Microsoft introduced a proprietary technology only IE5 could use and changed the world?

Google implemented in both a maps app (completely new UI, and WAY more novel use of Ajax than outlook) and a mail app a standard that worked across all browsers (also something missing in outlook).

Google showed us that you can use a new piece of tech across all browsers. Microsoft showed us you could do something in IE, and only IE.

Your statement is equivalent to saying "ActiveX changed the world". You maybe right, but no one cared about the change, and the change was probably negative.

> completely new UI, and WAY more novel use of Ajax than outlook

As to novelty: XMLHttpRequest was invented by Microsoft for Outlook Web Access. "Ajax" was invented by Jesse James Garret as a new name for XMLHttpRequest.

I remember the first time I saw someone use OWA on IE, thinking: HTF does that work? Then I remembered he was using IE and chalked it up to proprietary browser extensions, which was correct. Proprietary but useful enough that the other browsers eventually had to duplicate the feature, with different syntax.

That duplication is what made Google Maps (and later, Gmail) possible.

So, yes. Microsoft gets credit for this one. Not for the generosity of their hearts, but for doing it and making it work and dragging the rest of the world along behind them. That had been the tradition with browser tech since the beginning, with positive and negative effects. This was a positive one.

Definitely not a Microsoft apologist here. And XHR could have been better. But no one else cooked it up, so, they get some credit.

This is like saying we should thank Sun for Java Applets, because they gave us web sockets, 3d graphics, etc in the browser. They dragged the whole industry into the future!

The could do what ajax does before microsoft "invented" XMLHttpRequest...

Or we could admit, that coming up with proprietary crap on the web doesn't mean much, the hard part is making it work for everyone.

This thread is one discussion where I hardly see any coherence or modality in opinions (not that it's bad). To me, this indicates that email innovation has a lot of latent potential. For example, email could have evolved to encompass messaging as a first class citizen rather than an add-on widget.

I see this as similar to Blueray, a solution searching for a problem. The vast majority of people don't have a significant problem with the current implementation of email, and those that do develop workflows. It's like a toaster, it works fine as it.

The next step in the evolution should be end to end encryption that is (yet) impossible to crack

Email in its existence is a medium of communication and there can be innovation in the the modes of communication but the service in its own concept is difficult to be changed. Its same as saying to innovate on letter writing.

The innovation was Google Wave. And it was pretty much ignored.

Because it's based on extremely complex ideas. Multi-dimensional emails are great but they are really hard.

Since Gmail? What new did Gmail bring to email users exactly?

Gmail introduced conversation threading, gigabyte storage, speed, powerful search, and lots more no?

And decent spam blocking. I switched to Gmail after 10 years with Yahoo because I was fed up with 15 spams a day in the inbox.

> conversation threading

I believe this existed before gmail

> gigabyte storage

was not a tech problem before

> speed

in what way? the ui was and still is as fast as most lightweight desktop clients

> powerful search

maybe, although some clients supported powerful search with regex before

I think gmail was not that innovative, but it did bring everything together very well, in a very user-friendly way, along with some new-ish stuff like tags and the no need to delete anything.

Actually - when was conversation threading introduced in GMail? I know the Apple's Mail.app had it in 2011.

It was part of the initial launch in 2004.

for me: what is that "a lot more"? I just want to learn something new (btw, labels/tags are not really new, just redesigned imap folders)

Tags have an important difference from folders in that messages can have multiple tags and you don't think of messages as being in a folder.

Personally I find this quite useful and it definitely improved the way I use email.

Opera Mail had tags quite early on, not sure if it was before Gmail though.

Non trivial statistical analysis of email contents and your interactions with your email to effect decent prioritisation?

That would be my top feature in gmail.

s/gmail/mutt/

FTFY

Analyzing the copyrighted contents of third party non-Gmail users - who did not consent to such an analysis - for profit.

Free email hosting for multi-GB mailboxes with a threaded web client.

I had 2GB mailbox in 2006 and an unlimited a year later. And I was never fan of threaded views, so I guess I didn't notice, heh.

I've been testing Front and it's very promising.

Why do we always need to innovate?

" Why is it so hard to innovate? " Hmmm ... because mail just works since 20+ years And most of us just need grep command you IOS fanboy

I was going to say much the same that it works fine already. Although thinking about it there have been real improvements like being able to send 2GB files with gmail which was darn hard 20 years ago. And presently if I compose a reply with gmail it does it in a stupid little pop up window while if I fire up Outlook it takes like 5 minutes to come to life. So maybe it's not all perfect yet.

Actually what still bugs me is that I use email as my default record of all conversations I've had since 94 or so but I have not found a way for it to include SMS messages. Facebook you can forward to email but not SMS as far as I know. I want to be able to say to the program show me all the conversations with Bob last year and see everything, SMS, WhatsApp etc included. So come on startup dudes, build away! I daresay much of the problem is the likes of Apple being precious about letting you download the SMSs, but anyway.