Mail Pilot: Email Reimagined | Hacker News Reader

Mail Pilot: Email Reimagined

kickstarter.com

113 pointsalexobenauer69 comments

"Your email is your to-do list".

This is exactly the OPPOSITE of what I want. I consider myself a reasonably high-volume email recipient (>400 actionable emails a day on average).

The problem with "email as todo", which is a habit I had to work very hard to train myself out of, is that it turns your inbox into a to-do list that others can put things on.

This becomes the WORST to-do list ever.

With credit to Amy Hoy (Freckle, 30x500, etc): I believe that the problem is that every email that comes into your inbox is painful because it is associated with an unknown amount of work, and you don't know the amount of work until you open and read it.

Amy's newest product, Charm, is a help desk tool built to solve THAT problem - fundamentally decoupling the processing of email from the work that those emails contain.

My workflow right now includes:

1) first line of defense assistant does triage, cleaning out things that I don't ever need to see and either need deleting, or don't need my input. This includes anything from FAQ to scheduling meetings, etc. 2) second line of defense is answering emails that only need an answer - usually nothing more than a sentence or two. 3) everything else that's left in my inbox requires work. turning emails that do need my attention into todo's means that I can work from a to-do list, or ideally the current day's to-do list with items prioritized onto it with the help of an assistant.

I'm pretty sure that @sivers has a similar workflow that I've read about here on HN, he may be able to comment more.

This makes email 10000000x less painful, and makes my to-do list less like playing whack-a-mole. End of story: it's easier to get more work done. Email is working for me instead of against me. That is email nirvana, or as close as I'm willing to accept today :)

Of course - the problem with this workflow is that it requires another human. I don't know if that's avoidable. Truthfully, the people who I've hired who work with my inbox I trust deeply. My inbox is a strange dichotomy of a place that is private, but that anybody can send something into. He needs a lot of context for making smart calls, and for me, I need to trust him to make those calls.

Email clients need work - a lot of work - but this is more than renaming the statuses of an email and showing me who I've emailed most recently with.

Take the work out of email, and put it where it belongs.

a to-do list that others can put things on. This becomes the WORST to-do list ever.

If you still do the work that emails contain, you still have a todo list other people can put things on, don't you?

So what's the real problem with inbox-as-todo-list? That new work goes at the top? That work and non-work are mixed? That one TODO item can be in several revisions over many visually separated emails? That you can't track partially done things?

> If you still do the work that emails contain, you still have a todo list other people can put things on, don't you?

Sometimes, but not always. As your inbox volume increases, the % of things that come in that you don't want to become work for you also increases.

Note: I don't consider myself "famous". That said, I've gotten some good mainstream press related to my projects and I've also gone out of my way to make myself available to people by sharing my time, experiences, and lessons learned. This has paid off MASSIVELY and I wouldn't change that. What it's lead to, though, is the impact of a lot of people who don't think about the work they're creating for someone before they hit send. They're not forming complete thoughts, really asking for what they want/need, and in some cases, even asking anything at all.

I wouldn't give up the 20% of high quality email I get for the 80% of difficult to process email I get, but the 80% is still there.

That means that a pretty significant % of the non-spam emails I get are emails that need more work done to them before they're complete enough for me to do anything with them.

If I think of my inbox and assistant as a worker queue, it's their job is to keep incoming messages from being lost and disorganized, and in many cases, preparing the information/requests in the email for being worked on by me.

Maybe the issue is that the person adding things to my to-do list knows my workload, my priorities. The person emailing me doesn't have that context, and rarely has the consideration of it in the first place.

There is not a one-to-one correspondence between items in my inbox and items on my todo list, and items in inbox don't necessarily say what I need to do.

e.g. an e-mail in my inbox asking me to clean the frobulator might translate to one of the following todo lists:

a) 1. clean the frobulator

b) 1. buy frobulator cleaning fluid 2. empty the frobulator 3. clean the frobulator 4. refill the frobulator 5. reply to sender informing them that the frobulator is now clean

c) 1. reply to sender telling them why I can't/won't clean the frobulator

d) (no action necessary)

But my inbox won't help me remember which of these is the case. And that's ok, because that's not what it's for. That's what my todo list is for.

That's part of the issue.

But this application proposal - and many peoples' workflows for everything from day-to-day email to customer support treat their inboxes as to-do lists. It's a bad habit, one I had myself and had to work HARD to break. An email interface that trains people out of that habit, or keeps people from forming it, would improve productivity for a lot of people without them even realizing it.

> My workflow right now includes:

What you described looks exactly like the "zero inbox" approach, which incidentally builds off the "your email is your to-do list" idea.

What am I missing? Why do you say yours is an exact opposite?

This is a good point, anything that they create should make life easier for people who have a wide range of email volumes. That is easy to forget when designing an application because the fastest user persona to consult for design decisions is yourself.

This is well-articulated, but as you say yourself ("high-volume recipient"), you're probably an outlier. The problems and solutions that the kickstarter project talks about really resonated with me (I probably only receive about 50 emails per day).

I could use a little clarification on your exact point though -- sounds like you're saying that your email application isn't your todo list, but your emails become your todo list (just in some other silo than your inbox)? Seems like the same thing to me -- what if the email application was more like that other silo, then you don't need 2 silos.

Finally, in the interest of full disclosure, it's probably fair to mention that you have been Amy Hoy's business partner in the past (not that this implies any dishonesty or discreditation of your point... just sayin').

I might be an outlier - but I know that these bad habits formed BEFORE I was an outlier.

As I mention in another comment, the important part is the transition from inbox to to-do list having a level of intelligence to it that considers context, the relationship with the person, current workload and priorities, etc. Right now a human does that. I'd pay for a product that did it, if it didn't feel to the sender like a robot was working for me.

I've collaborated with Amy, yes, but we're not nor have we ever been business partners. Good friends and co-conspirators for sure, like many of you on here. Her husband Thomas is her only business partner.

So 3 years ago I had exactly the same idea. I executed on it, built an email client and it was basically a commercial failure (although we ended up licensing some of our stuff).

So here is my advice: don't even bother unless you want to build a small lifestyle business (and don't even bother dreaming of a subscription model).

Email has been around for 20 something years and the usage patterns are so engraved in users mindsets there is no way you can get them out of that experience. The whole idea around productivity will only appeal to a small set of power users.

If you really are set on building this thing then here is a howto: 1. go here: https://github.com/waseems/inbox2_desktop 2. fork the code 3. add a todo list 4. compile and profit (well probably not)

Sorry for being so cynical but I have been there :-)

Does anyone feel like there's something really odd about using Kickstarter to fund proprietary software? Like some of the other Kickstarter projects that fund patented inventions. Like, if you want to launch a proprietary product, _do that_. The way of making money doing that would be to _do that_, not to raise it in some odd fashion, from people who won't actually benefit from it, or own any stake in the idea or profits, beyond 'above a level', a 'copy' of the application or an account?

Kickstarter is presale marketing venue to raise startup funds "kickstarter". It doesn't matter to Kickstarter what the seller offers in exchange for those funds.

This is just a more honest approach than vendors who lie about already creating a product before they sell it. (Hello, 1977 Bill Gates and most startups ever.)

I don't think it's odd, but I'm much more likely to contribute to something open source than proprietary on Kickstarter. With open source, if the project fails, the parts of the project still be useful to the world. Not so much with proprietary stuff.

I don't think it's any more odd than the cool gadget-themed kickstarter projects, or music, video, etc. Basically you're paying in advance for a product/service, is how they're usually framed. What's wrong with that? It seems like a great way to start a company.

I'd say the last thing you mention is how most projects on Kickstarter work, at least those I've seen. A common project type is for someone to produce an album or a movie, usually with no open license or special licensing privileges for you, which seems along the same lines as proprietary software.

It's a pretty neat version of lean startup, in my opinion. Rather than just gathering email addresses from people that might vaguely be interested in your product, you get cold hard cash from people that are definitely interested enough in your product to buy it. You know for sure you have a market, because they already bought your thing.

The concept looks interesting. I was sold on it and then, "backers will get a steeply marked down subscription." Wait, a what? Subscription? Sorry, you lost me. No way will I pay a subscription fee to use an email program. Apart from that, as a standalone app, it sounds compelling.

Same here! Very interesting until I realized it's subscription based. Apps should have a one time cost and maybe costs for major updates. I can't even see the reason to have a recurring fee, why?

I assume their justification are the push notification servers, though that's a bit thin for my taste.

Because it's not an app in the regular sense, it's an email filtering and tagging web service. You'd be paying for their servers crunching your emails over the time you were subscribed to them.

Edit: From their FAQ:

  Mail Pilot is more than an app; it's a comprehensive online software
  service. It runs on servers, and delivers and syncs content between
  every Mail Pilot app that you might use.

  You can access Mail Pilot through a number of desktop, native
  mobile, and web applications. To make this possible, Mail Pilot adds
  services to your existing mail accounts that it syncs between all of
  your devices. Our servers store metadata about your messages (not
  the messages themselves, however!) to bring to you the reimagination
  of email that so many have come to love.

  Mail clients can get away with charging one time. Our apps are all
  free, but the service is paid for, much like other online
  productivity and email enhancement services.

I already pay a subscription fee for my email, in the form of a $50 per year GAFYD account (of which I use only the email). Paying a subscription for an email program isn't my favorite idea in the world, but I'm willing to try it out in the name of Great Email Justice.

How many computers do you have? I'm starting my career as a small business and already have 3 + Smartphone. I would be happy to pay for a subscription rather than 3 licenses + Phone App. It helps to have my account available on the net, so I can check my email from any computer with an Internet connection.

It's time we move forward, I think. I used not to like the idea of subscription, but I'm finding it cheaper now.

Or you could just choose software that doesn't require a license per machine. Of course open source software doesn't have this issue but I also run proprietary software that says I can install it on as many machines as want as long as it's only used by 1 person at a time.

Sending/receiving data (in this case email) uses energy. Since energy is not currently free, someone has to pay for it.

A service that involves the repeated sending/receiving of data, (in this case email) incurs a cost over time that someone has to pay for.

There are three solutions to this issue:

A. Run a free service until you run out of money. (not sustainable)

B. Run a free service subsidized by ads. (user-hostile and possibly unsustainable)

C. Charge the user a subscription fee. (fair and sustainable)

Unfortunately, the perception that email and other services should be free has permeated society. The internet made the transfer of information cheaper by many orders of magnitude, but it is still not free.

You're ignoring the fact that the $35,000 they're trying to raise is listed as being for exactly that:

> Where the money will go... Hosting. We need some serious hosting space to deploy our storage-intensive services.

I'm not ignoring that. If they only raise $35,000 then they'd fall into category A. The cost of hosting is ongoing as long as a service is being used, therefore they will need more than $35,000 if they plan to be around for any length of time, hence subscriptions.

Yeah, its actually pretty dumb IMO that they say the money will go to hosting. With AWS etc you only use what you pay for and the costs should be pretty minimal for beta testing and then covered by subscription fees once they launch.

Money going towards mobile dev makes more sense although is much more risky to back given the short timeframe and limited budget even if they meet their goal.

But you have to provide your own e-mail service and e-mail hosting. They aren't hosting your e-mail[0]. How do they justify a recurring annual subscription?

0: "Your email is never stored on our servers, and is left on your current email servers."

They are still providing a service. Creating and maintaining an application takes time and time is money.

I think my point still stands as a general principle though, its applicability (or the lack thereof) to this particular service notwithstanding.

Axiom: If you don't pay them to keep the lights on, they will go out of business. Corollary: If it isn't worth it to you to pay for their added value, they don't have a real business.

That is a very good idea. I do something very similar with Gmail Priority Inbox + Google Calendar already but it's very slightly more work. I have for categories in my Inbox: Starred, Calendar, Todo, Everything else.

* Starred are things I'm following until they are done - similar to the complete/incomplete feature they mention. It includes emails like Amazon shipments, work-in-process items, long email threads etc. The label doesn't matter.

* Calendar section is automated deadline-driven todos that must be done - paying bills, saying Happy B'days etc. It's similar to their timelines feature but I don't have an easy way to say 'Remind me about this email in June'. Instead, I use the label 'Someday' and once a week or so, go through my Someday emails and star them if necessary. If Priority Inbox allowed me to add 5 sections instead of 4, I would just make Someday one of the section too and be done with it.

* Todo section is very similar to Mark for Review and I just add the 'Todo' label to emails when I can't deal with them immediately but plan on doing so in the near future. 'Someday' is for when I want to push it off for a few weeks or months.

* Everything else is all other incoming emails. If I archive them, they're gone. If I star them, it means I want to track their progress. If I add 'Todo' / 'Someday' label, it means I intend to work on them soon or someday.

Gmail offers almost all the other benefits they mention, including multiple-mailbox, autopilot (filters/rules), and tons of intelligence. Personally, I don't mind sticking to Gmail for now but this type of approach to email validates my current practices and hopefully improves how all of us manage with email.

Oh. This is about an email _program_. I thought they would try to attack smtp+pop+imap and try to solve some of the problems with how email works. That would be a really hard and interesting (albeit probably fruitless) challenge!

Edit: Anyway, this project looks pretty cool - even with the subscription. I've backed it.

Delivery date of June, 2012?

I know you probably think 4 months is a long time, but it is a probably not near long enough to develop something this complex.

Be prepared for that schedule to slip.

mixed reactions to this.

admirable to want to tackle a big problem, and 'make your own dream job'.

'if email actually worked'.

"worked" for whom? The fundamental problem as I see it with approaches like this is that email is a multiparty system, and it won't "work" for your life because other people don't share your views of how email should "work".

"For example, send all messages you've been CC'd in to be reviewed when you have more time"

You're assuming that someone who cc'ed me on something doesn't want a response right now.

"automatically have responses to a job opening put into a specific review list"

Assuming people email in with appropriate subject headings or specific text.

There's as many different ways people use email as there are people just about, and email is almost always a two-or-more-way street. Modifying how I consume and use email on my end can be helpful, but only to the extent that it doesn't violate the process on the other end.

I didn't see anything about an API in there. Having an API to program against would be useful, and an API that would be able to work against the multiple aggregated providers would be really useful. Scriptable incoming email processing (not necessarily realtime, but close) that could move your mail around and integrate with other services would be really useful too.

"Email requires action"

If that's your fundamental premise, I can see how that would shape your thinking and design. It might not actually be wrong, but I'm not sure it's necessarily right, or at least right for a lot of people. Unless you count "reading" and "delete" as actions too.

Subscription software - not necessarily against that, but I'm presuming this will be 'on the web' hosted? Not local to my desktop? Hrm...

Something I'd like to see is good search and ability to search through related messages while I'm reading a specific message already. This is broken in every mail client I've tried. I can read, or I can search, but not both (without multiple tabs in web-based mail clients). Reading a message and not being able to get more context and info from previous conversations is frustrating.

Good luck!

> You're assuming that someone who cc'ed me on something doesn't want a response right now.

I'm just focusing on this, not to argue with you, but to point out something that irks me.

If you want a reply from me, send an email TO me. If you simply want me to be aware of something at some point, CC is fine. But if you want a reply, and you CC me, you've given me no indication without me having to read the email.

I despise the practice of using CC as if the TO field can only contain one email. I make it a point to send an email TO people I want a response from, and CC those who don't need to respond.

I know many people operate this way, but it's not standard enough. I get people who CC me on stuff because they want the primary person to feel like they're the real decision maker, and the CCed people are supposed to chime in if we want to support or vote down the email's content. Who is CCed and who is TOed can be something of a political game. Not saying it should, but it is.

I understand. =)

I've fought back against this very thing (and was fairly successful with the results). It required a lot of force and careful wording.

"No, you didn't send anything _to_ me. You did copy me a message you sent to someone else, however."

I went out of my way to filter all emails CC'ed to me to another folder called review, and I would review it at the end of the day, or maybe the next day. It took a lot of standing ground and being blunt with people, but eventually it started to work. Before I left, there was discussion on improving email communication, and making it clear to everyone in the company the difference between CC and TO, and how to use them, when to use them, and what to expect.

I'd hate to think what I'd say if I was in the middle of your situation. I don't play the political game well. =)

Re: the API - have you seen http://context.io/? It addresses both of your issues:

- Ability to work against the multiple aggregated providers (indexes any IMAP-enabled mailbox, regardless of provider)

- Scriptable incoming email processing (not necessarily realtime, but close) that could move your mail around and integrate with other services

Always cool to see more Hokies doing cool things :-). Turns out there are a lot of resources in town for this sort of thing:

Dr. Perez in Computer Science, he actually has done a ton of research in how individuals use information (PIM) including email.

It might be useful to do some competitive analysis as well: http://www.boomeranggmail.com/ (attempts solves the 'do this later problem')

As your probably aware MailTrust created by Hokie Pat Mathews and bought by Rackspace still has a pretty big office. I'd track him down and get some feedback from him.

My two cents, feel free to reach out ben at olark.

Hey Hokie! Thanks for the post.

Pat Matthews actually emailed me this afternoon, which I was honored to be on the receiving end of.

I know of Dr. Perez, and definitely need to reach out.

Boomerang is interesting, and validated our need to go to subscription billing. Thanks for sending along that link.

"If email actually worked" then there would be some way to know whether email was sent and delivered--something like software transactional memory for email with logs that could be exchanged and mutually verified for consistency among emailing parties. (I'm supposed to be in a research position, but I have other academics nagging me about email delivery. There are constant problems with dropped email, mysterious rules that "helpful" email clients silently add that trash incoming messages, and no accountability from our local IT dept.)

I've run into this problem when setting up email servers before, between DKIM, SPF, Various blacklists and Spam filters on the servers and clients that may catch important (non-spam) email it's hard to keep up with even when you just want a small personal server.

Email is just fine for me, I don't understand why people need to pretend they reinvent it every other week (social email, todo list email...).

The only thing I'm missing is on the technical side, a transaction system like you describe.

How great would it be if all the servers could reply to my mail client : I have received your email it isn't considered spam and the end user got it.

You get upvoted for agreeing with me that some sort of end-to-end verification service protocol is missing. Now there's a schlep if ever there was.

Email works fine, the problem is that lots of software implements it incorrectly.

Take spam filters for example: The correct way for this (and what I also do on my own private mailserver) is classifying mail during the SMTP session. Thus, if my mailserver considers a particular message as spam the sender receives an error message stating that his message could not be delivered.

And now take Gmail: Happily accepts most messages, but labels some of them as Spam. Of course, no-one ever checks the spam folder. A few misclassified mails and people will start complaining that email is broken.

Email clients have to do this because they use Bayesian filtering to get rid of Spam.

It turns out, if the sender of spam messages can find out which messages got through and which didn't, then they could build up the same bayesian filter on their computer, and figure out exactly what email they can send to not get marked as spam. Bayesian filtering only works if the spammer doesn't know which messages get through and which don't. I believe this is also one of the reasons images are disabled by default in gmail.

I'm not sure how this would work if a spammer targeted their own gmail inbox to discover which messages succeeded, but maybe spam filtering in gmail is only partially collective, so that individual inboxes build up slightly different bayes nets.

Source: http://www.codinghorror.com/blog/2004/09/popfile-vs-popfile....

My issue with email is that it is hard to quickly extract the information that I need from it. This issue does not usually stem from what the sender has written, but how it has been sent, and how various email clients interpret the message.

Threaded email conversations were a step in the right direction, but I've yet to come across an implementation that just works. I wish email clients were designed to look more like instant messaging applications by abstracting away everything in the body apart from the actual content.

Most clients handle replies acceptably most of the time, but I also want to see signatures gone (perhaps recognising phone numbers and adding them to my address book) and automatic re-formatting forwarded emails so that it looks like it was sent directly to me.

Of course sitting back and complaining is lazy and unproductive, perhaps I should have a go at building an email client that abstracts away the frustrating parts of email.

Edit: I'm envious of a lot of collaborative communication apps that exist but unfortunately everyone has to be using it which is a barrier to entry that it some circumstances, is just too high. I envision a similar tool that is built on top of email as a standalone application (web and/or desktop). This would afford the benefits of making communication a much quicker and easier to manage process, while not requiring anyone you need to communicate with to change their process.

Subscription-based consumer desktop program. Wow. It'd be very interesting to see how this unfolds.

(edit) Just read the comments on Kickstarter, and it is a service with a desktop interface, meaning proxying all emails through a two-person company. Privacy, security and reliability are just some issues that they have to address.

> At it's core email is a list of messages that require further actions.

That's not always the case, in fact I read a lot more than I write emails.

I feel they took one use case of email and tried to solve it. Email is complicated because it has plenty of use cases and everyone uses it differently.

Email basically is other peoples TODO list for YOU.

Product look kinds of cool but the statement that every email needs to be actioned turns me wayyyy off.

Email is pretty much the worst productivity tool ever. Mainly because the second you freaking open it you hit the Inbox, which can contain stuff from relatives, spam or assorted marketing messages. Lots of stuff that should never have made it there in the first place. Even if something useful arrives it's so unlikely that it is the number 1 thing you should be working on.

But yeah will be interesting to see how they look at it.

One of the things that looks interesting to me is the aggregation function. I currently have multiple email accounts and I have to have 3 (or more) browser tabs open to deal with them all. It would be very handy to be able to go to one place and get all of my mail.

One downside I can see with this is making sure that I'm responding to the correct message from the correct email address. Mixing up work and home accounts would not be good.

Sparrow (http://sparrowmailapp.com/) does this really well.

I needed this a few years ago badly. I ended up forwarding all email to my gmail. Then... And this is considered excessive by some... I forwarded all those to a MobileMe account and check it all in Apple Mail. It automatically detects what email address my email came to and when I reply that's the email it will show from. Works great and haven't changed my setup since.

From there you can create rules and whatever else you need to.

Having worked at Microsoft, I saw several PMs attempt to reimagine email over 4 years there--eventually all the radical changes get cut time and time again. MS employees might as well just submit kickstarter projects-- get outside funding and maybe even reintegrate into MS products :)

I'm sure some people will find this useful, but it seems to be the way I already use email. I don't have to deal with a lot of email so maybe this has more use for those who do.

If it's in my inbox it needs my attention. If not it's archived or deleted. To enter review mode I go to my inbox.

This sounds like a lot of clicking. I don't want to manage my email even more than I already do.

Maybe have email be read by default and have the ability to mark it as a task... or make a gmail addon that lets you import an email into your GTD app.

Wanting to get email to "work" is a noble effort, but that means different things to different people.

A more intuitive user interface will sadly not magically transform email into something usable.

Receiving fewer emails, might.

I might pay a one-time amount for this kind of application, but a subscription service to manage email ? Not going to happen.

The idea of email not going away after I've read it and being a to-do list sounds slightly terrifying.

I'll stick to my Gmail for now.

I'm sick and tired of to-do list apps...

Weird my email seems to be working fine :), checking it out though!

Interesting to see a startup funded by Kickstart rather than VC.

So basically GMail with elaborate filters and better search?

Feels like an app, not a monthly subscription.

email sucks. collaborative tools are what you should be backing VCs!

Is it cross-platform?

They say so.

By the time we're out of beta, we will be launching on Mac / Windows / Linux, 32 & 64 bit. Mobile will initially include iOS, Android, and some Blackberry models. A subsequent release will hopefully also include Windows Phone. Finally, we will also launch with a web and a mobile web version.

Thanks for taking the time to check it out!

Other than introducing a new interface I don't understand what new concept is being introduced here. It doesn't sound like they're putting hooks from the email client into other programs, which is really what needs to happen. Email itself could generally go away if instead of asking questions people could send little "applets" that would open inline with the message, and can directly accomplish the task being asked of you.

So instead of sending an email to manually write contact info, someone could send me a "request: contact info" message. Then when I receive that message, I just click one button called "Activate" which would automatically fill out my contact info and fire off an auto-generated response.

The number of times in a year that I have write my name, email, address, phone etc is mindnumbingly stupid. I should be able to have an email program that can auto-fetch all this type of information and more.

Reply on news.ycombinator.com