Hacker News new | past | comments | ask | show | jobs | submit login
Dropbox closing Carousel and Mailbox (dropbox.com)
624 points by _freu on Dec 7, 2015 | hide | past | favorite | 421 comments

Ugh. Yet another intriguing email startup being acquihired and killed off by a more-established tech company (see also: Sparrow). It's 2015 and I still bounce around email clients every couple months because all of the major options have substantial flaws.

I co-founded ITA Software and have been using some of the proceeds from the sale to Google to fund an email startup. (http://inky.com) The challenges outlined in this thread are all accurate. If you're thinking about putting effort (or money) behind an email startup, think again. I simply don't believe it's doable for a normal startup (i.e., one without a crazy rich guy plowing money into it until it's good enough). The VC-backed "winners" will all get aqui-hired, and then the acquirers will be unhappy because what they bought wasn't actually good enough for prime time.

Here are just a handful of the reasons:

- It is deceptively hard. Like airfare pricing, most people think it's a couple weekends of work, when in fact it's at least 5 years and millions of lines of code to get to a real MVP that someone could use instead of Outlook, Thunderbird or Apple Mail, etc... The set of abstractions required to do it right are unobvious and nontrivial, and you end up with unfixable problems if you don't get them all right. We've rewritten Inky essentially from scratch twice, and only in the current version 3 (circa October 2015) do we finally have the abstractions correct. Look at the Thunderbird bugzilla logs and you will find thousands of bug entries, many of which go on for pages and pages. ITA was similar, by the way: 1996-2001 was basically iteration to get to an MVP.

And because it's not just hard, but deceptively hard, founders and their teams systematically underestimate the effort required, which leads to all kinds of failure modes. Even in the success stores like Mailbox, the teams basically punt on a number of hard problems that really are required for an MVP (like properly sanitizing HTML, supporting Exchange and IMAP "for real", proper support for all the weirdly encoded mail you see in the wild, encrypted mail via S/MIME, etc.)

- The other entities you have to work with are a complete nightmare, and as the client app everything appears to be your fault. Google, Apple, Yahoo, etc. just do whatever the hell they want, randomly, have broken servers they never fix, and place arbitrary problematic restrictions on clients: all limit clients to a small number of connections, and Apple won't let an email app run in the background, so you can't support "real" push notifications without keeping the user's credentials unencrypted in the cloud (which I refuse to do). Google, out of the blue, started banning clients that didn't use oAuth, sending users scary messages about unsafe software. oAuth is great, but just breaking everyone who doesn't support it yet is really, really nasty.

- VCs absolutely hate the space. Email startups have been dogs forever, and every name-brand VC has been burned on a crappy email investment.

- Consumers simply don't care about usability, privacy, or anything else enough to not use what comes with their phones.

- Security-conscious people who might pay for something that's genuinely secure never believe anything is genuinely secure in practice unless it's open source... in which case there is no business model.

- Existing mainstream clients are ironically sticky, because once users develop coping strategies to deal with the quirks of the client they're forced to use, they never want to invest the effort into learning the quirks of a new client.

So why do I work on it? Because in my opinion it's a really important problem that affects billions of people -- and one that needs to be solved properly by folks who are doing it not just as a data harvesting operation, loss-leader to sell devices, or as a way to build a moat around a core offering (search).

> It is deceptively hard [...] The set of abstractions required to do it right are unobvious and nontrivial ...

I would be very interested to hear more about this aspect, if you would be willing to share. I'm always curious about things that sound like they should be simple but end up being surprisingly complicated if you want do do a good job :)

Fascinating! I'd love to read more about the challenges involved in both airfare pricing, and email! Can you link to any posts that go into greater technical detail?

Also, regarding Inky-- I just spent a few minutes browsing your site, and it didn't answer my first question: why should I switch to your product from gmail?

For the airfare-pricing bit, see "Computational Complexity Of Air Travel Planning" [2003] by Carl de Marcken of ITA Software:


I tried Inky when it was a "show HN" thread two(?) years ago. It was solid then, and I'm sure it's only gotten better now.

Thanks. It is vastly better, in pretty much every way. We also support Android and Exchange now, which we didn't back then.

Do you plan to add PGP support?

Maybe, but not any time soon. S/MIME solves the same problem and is what enterprises and governments want. It's incredibly complicated and difficult to implement properly, but it does include solutions to some security-related issues that PGP doesn't (e.g., security labeling).

In 1996 there was a huge need for airfare search. Look at the success you've had.

On the contrary email in 2015 isn't a huge need or a pain point for most people.

Since you're entering a crowded market your product has to be 10X better. Currently it's probably not which is why most people perceive Inky as a marginal improvement.

Still I'd love to see other projects you're working on Dave. Clearly you're talented and with the right idea I could see you getting back to where ITA & Crash was.

Are you a venture capitalist? Because if not, you should be!

Not yet but I appreciate the compliment.

Probably because there simply isn't money to be made in being an e-mail startup.

Or, more correctly, there isn't enough money to be made to satisfy VC investors. Mailbox and the like could have lived quite comfortably as paid-for apps with comfortable but small levels of profit, but that isn't what Silicon Valley rewards.

EDIT: also, IIRC Apple's restrictions on background processing meant that they needed to have every user's account set up on their servers to process push notifications. Did that ever change? If not that's a non trivial expense/complication to scaling, especially when you're up against an OS-integrated app that doesn't have those restrictions.

That's because making a big pile of money the Silicon Valley way requires two things: a product, and a moat. It's not hard to create an email product, but email moats are hard because it's an anarchist technology by design.

Would you mind explaining what does "a moat" mean in this context?

Warren Buffet talks about "moats" all the time. Basically a business with a moat is just a business with a defensible advantage.

If you think of your business as a castle, a moat is something that stops others from taking your castle from you. This could be anything from a network effect (Facebook, LinkedIn), economies of scale (Amazon, Walmart), to patents (most biotech companies) to brand (Coke).

The way he put it was something like, "If you gave me 100 billion dollars to take Coke's market, I can't. They've been investing in infrastructure and brand all over the world for decades. Even if I could make a better tasting beverage, people would still buy Coke. They have a moat".


Platform lock-in.

It's a ditch dug around a castle to stop people getting in. i.e. the 'wall' in the walled garden.

So, "wall" != "moat" Some castles had a wall and a moat. Anywho. I think there is a differences in kind here that is material to the discussion.

A moat is a competitive advantage that's hard for competitors reproduce or otherwise overcome. Coke's moat is its near universal brand awareness and affinity. Comcast's moat is its infrastructure and political clout. Crossing a moat in any conventional way is very resource intensive and often leaves the challenger vulnerable.

You can make an alright pile of money without the moat. Perhaps the problem is the kool-aid level in the companies making this stuff...

Though they are making good things, so it's hard to be too critical.

I thought the business proposition of Mailbox and Carousel was providing more utility/centralization around Dropbox services/storage (via file attachments and photo storage). Which never was a great idea, but still.

Yet that same Silicon Valley rewarded Mailbox with $100MM in an acquisition...

That's exactly my point. Silicon Valley rewards "go big or go bust", it doesn't reward "small, sustainable business".

I don't for a second blame the founders of Mailbox for doing what they did. When the market creates a strong set of incentives like this you can't really expect anyone to do anything other than take the $100m. But this is the side-effect.

VC is generally designed for big bets and big payoffs (invest in 10 risky bets in the hope that 1 pays off to cover your losses on the other 9). Less risky ventures with smaller payoffs take longer to grow, and don't satisfy the payoff requirements of the VC model.

That doesn't mean they're bad companies, but there's definitely a funding gap for the type of company that will require $3M in investment up front and be worth $10M in 5 years. That's still a great return, but right now it's the domain of private equity.

VCs themselves aren't the problem. Big bets are fine. The problem is acquihiring. It's:

- good for founders - they get a big fat paycheck and a chance for a job in a big company

- good for VCs - a payoff, even if small, is better than no payoff at all, so they want their 9 not-so-good startups to be bought

- good for the buying company - they can literally just kill the competition by throwing money at them, + they get some proven and experienced employees, which may be worth it given the unreliability of the standard hiring process

Care to guess who this deal is not good for?

Startups are (and are being heavily marketed as) a get-rich-quick scheme. Most of the people involved are self-selected to follow this process. And all this talk about providing value to the users? That's the part of the scheme where you get used.

This may be a contrarian view, but: I'm not convinced it's good for the buying company. Good engineers are hard to find, but this is an awfully expensive way to hire; you may not keep the good people for very long; and if you're the kind of company that has a really hard time hiring good people, you may not also be very good at getting engagement and productivity out of the people you acquihired.

My view on acquihires is that to some extent it's a trick the acquiring company plays on itself. Some manager is getting to hire a few good people using money (the acquisitions budget) they wouldn't otherwise have access to.

Acquihires are more about hiring functioning teams composed of engineers and executives who are used to working with each other. That's a lot more expensive to build than just the salaries; and you can generally take that team and throw them at a problem with a reasonable expectation they'll accomplish something.

Also, there's generally some amount of IP and customer data involved as well.

"you can generally take that team and throw them at a problem with a reasonable expectation they'll accomplish something"

For about 6 months to a year, after which time most of them have left, or have gone into "waiting for the golden handcuffs to come off" zombie-worker mode on a project they have no inherent interest in.

Totally anecdotal but the couple of acquihires I've seen from the inside have gone like this and it seems to be the natural state of them based on what I've heard second-hand from others.

I agree with this based on the 20 or so people I've known to go through this process to join a large company (amazon, google, etc). I've heard first-hand accounts of people even re-negotiating deals just to get out of the golden handcuffs they thought they could tolerate but couldn't. I understand this sounds a bit hoaky and I would love to give names and companies here, but obviously can't.

Agreed; they tend not to be great deals for the acquiring companies long-term. But short term you can rally that team to build something else, and your likelihood of finding the mythical 10x developer (or manager) is statistically higher in a startup that successfully launched a product in a compressed time frame (even if the market didn't work out for them).

Any acquisition involves a lot of turnover. Guaranteed that this turnover is built into the models they use when evaluating acquihires. Consider how expensive most company's talent acquisition costs are though (upwards of $100k+ per candidate hired for top talent) and it starts to make sense.

VCs are totally the problem here. It's not that VCs are bad people who want to screw over users, but their business model requires them to get payback on their bets, even if it's small. It can be hard to know up front if an idea is viable as a hockey-stick software company or if customer acquisition is much, much harder - often because you don't know who your target customer will be specifically until you've iterated a few times.

And many times, there's no other exit option for the VCs. Does it suck for users? Absolutely - which is why companies often have lots of apprehension about being early adopters of a startup's technology. It's why B2B startups are really hard, and why they tend not to hockey-stick because the customer acquisition costs are high and don't shrink with scale.

It's up to the founders to convince their VCs to pass on the acquihire because they think the company still has hockey-stick potential. Zuckerberg did this a number of times - it was still really arrogant, but it paid off for Facebook and shows why founders need a bit of arrogance to be successful. It's a lot harder to do this if you're running a startup that doesn't have a lot of market traction.

I think most VCs would rather see the company stick it out and go for the hockey-stick & IPO. An acquihire is a blip on their returns; it's basically worthless. They're usually arranged as a courtesy to founders, so that if you work your butt off for 3 years, do everything right, and still don't find success, you don't leave with hard feelings toward the other people you worked with.

You get acquihires when the assumptions behind the company's business model have been proven false and there's basically no way to solve it other than rebooting the company. Arguably this is an argument to not take VC until you've got the hockey-stick trajectory, so you can pivot & reboot at will without dissolving an existing organization. But even this sucks for customers, if you have any.

>VCs are totally the problem here. It's not that VCs are bad people who want to screw over users, but their business model requires them to get payback on their bets, even if it's small.

Doesn't that make them bad though? I mean it's like saying "It's not that the mafia are bad people who want to screw over people, but their business model requires them to blackmail and occasionally kill" (not comparing levels of "badness" of course, merely showing the problem in said reasoning).

It's not like following a particular "business model" is not a decision people get to make. And if people take the VC model that favours acquihires, then they very much have decided to do stuff that screws with users.

Another thing that should be mentioned is that most VC's have a close relationship with bigger tech companies. Its a way to move money out of the public markets and back into private. There are multiple implications with this type of behavior.

But this totally negates your point about "there isn't enough money in an email app". Clearly Dropbox thought there was at least a $100MM opportunity...either talent to produce $100MM+ in value, customers to convert into $100MM+ value, or staving off competition that would result in $100MM in potential loss.

> "Clearly Dropbox thought there was at least a $100MM opportunity"

The opportunity was the team, not the product they were producing.

Say you had a room full of math prodigies, but all they were doing was watching TV all day. You hire these math prodigies for $X and put them to work solving math problems - this doesn't mean that watching TV is worth $X, in fact it likely means the opposite, that their value is underutilized by whatever they're doing right now.

So there isn't enough money in making email apps, but a team capable of producing a great email app can be put to use doing things that do make a lot of money - but it sure isn't making email apps.

And that's unfortunate, because as users we're missing out on a lot of great products. The dark side of the immense growth in software (and immense growth in software salaries) is that entire products and services are no longer economical, even though we'd all love to see them. The VC system that drives these early acquisitions and acquihires is just one more factor among many that limit the field of what can be economically produced.

Yeah but that's ridiculous, they had only 13 employees that's 7.6MM per employee over the span of 2.5 years, some of them I'm sure they left already... You tell me when this starts making sense because it just doesn't right now.

Too much money! I don't care how big of a genius you are you're not worth that as a developer of an email client, and honestly to write an Email client is not genius level shit. The UI was dope, and their invite process was funny, that's all.

Looks like Dropbox fucked up, that's all, bad business decision, trying to copy Google and do better than them in the end they did the same.

The deal was stock and cash, so it is pretty much impossible to judge how good a deal this was for Dropbox[1]. Stock is a very cheap way for a company to acquire another.

[1] http://techcrunch.com/2013/03/15/mailbox-cost-dropbox-around...

"Stock is a very cheap way for a company to acquire another."

Not exactly. Could be the other way round too. It depends on whether the stock of the company is worth more or less than cash in the long run.


Except since it is the company issuing stock in itself you can't really think of it as fungible. If it ends up being worth more than the strike price then literally one of the events that led to that price was the purchase stock being issued. Of course that doesn't mean the is a causal link, but in a very real sense the stock was free for the company itself.

The only cost to the company is opportunity cost.

"in a very real sense the stock was free for the company itself"

Stock always has value, especially to management.

Unless the company has poor self-esteem. :)

The reason so many technology companies today make acquisitions using stock is because they (correctly) understand their stock to be inflated, relative to intrinsic value. If that were not the case, if they understood their stock to be undervalued, then they would definitely not be giving it away in transactions. They would be making all-cash offers instead.

> " I don't care how big of a genius you are you're not worth that as a developer of an email client, and honestly to write an Email client is not genius level shit. The UI was dope"

But that's the key isn't it? I think you're looking at this from too narrow a point of view.

Yeah, email clients aren't worth much money, so its development isn't worth much either (when's the last time anyone you know paid for an email client?).

The point isn't to acquire a "developer of email clients" it's to acquire a "developer of top-of-class mobile apps", which is considerably valuable. Think of the average quality of a mobile app you use on your phone, the bar is pretty low. Producing truly amazing mobile apps (like you said, the UI was dope) is hard, and requires the flawless, cohesive operation of a lot of facets from engineering to UX design.

Truly excellent mobile teams are very rare, and worth a lot of money. To the tune of $7.6MM over 2 years? Beats me, but it doesn't seem completely out of the question.

Talented developers are worth a lot of money - teams of talented developers, designers, and product people who have a track record of producing stellar work together are worth a multiple of their individual values.

> " and honestly to write an Email client is not genius level shit."

It's not genius level shit, but it's a surprisingly rare skill, and by that I mean writing very, very good frontends. Most people here can write a functional email client if they had to, but how many of them can write a really fucking good one, so good and so easy to use that people - and not just the technorati - sing its praises from the rooftops and get their friends and family to download it?

The Mailbox team was operating at a level well above nearly all mobile app teams, and the quality of their work was easily an order of magnitude better than the industry norm of its time. This may not justify their acquisition price, but I think you're selling them short a bit. They're a team of people known for exceptional quality work, and command a high price for it.

I'm sure some of them are still there, however. They shutdown the product and put the team members to work elsewhere. So it could be prorated over a longer period depending on the acquisition deal.

This seems like a too-generous take.

Dropbox could have thought that and just been wrong. Even in the normal economy, per The Economist about half of all acquisitions destroy value. In our little hothouse, I'd expect a much larger percentage of value-destroying takeovers.

But there are plenty of other reasons deals get done than a company-wide coherent calculation of economic value. E.g., I've heard of acquisitions happening because a CEO needs to demonstrate to the board that they're doing something. Or incentive structures set up to reward an M&A team based on how many deals they do. Or executives wanting to bring in their old buddies from another company. Or, "gosh, money is so cheap right now we'd better spend it because our competitors are spending."

It's pretty clear that Dropbox valued Mailbox for its potential to be something other than the standalone email app it was. And you might be right - there might well be enough talent on the Mailbox team to produce $100M+ in value - but not making a standalone email app.

If we're being ultimately generous with Dropbox then what they are doing is exactly correct - taking a team that can create a $100M+ product and reassigning them somewhere that they can fulfill that potential. But that's of scant consolation to all the people that want to continue using the email app.

By the same logic not shutting it down would result in greater than a $100M loss.

"That's exactly my point. Silicon Valley rewards "go big or go bust", it doesn't reward "small, sustainable business"."

Because if that's your goal, you clearly aren't passionate about what you're doing. You might as well go work for a 9-5 company doing CRUD apps. /s

That went to founders and investors. Mailbox has clearly received a bullet.

> there isn't enough money to be made to satisfy VC investors. Mailbox and the like could have lived quite comfortably as paid-for apps with comfortable

I'm not sure that's true (see: Sparrow). I just don't think there is enough demand for people to want to pay for an email client. I imagine for a huge majority of people, default mail client's work fine enough for personal use.

Not to mention, if you're old like me, you remember paying (I think?) $100 and then eventually $49 for eudora / upgrades for eudora. In approx 1990 dollars. We now live in a world where people pitch tantrums about paying $10 for an app, and expect perpetual upgrades for that price! So simultaneously expectations have skyrocketed and prices have plummeted.

Part of the problem is, unlike say Pinboard, an email client doesn't feel like 1-3 devs worth of work. The protocols are horrendous, you have to understand the quirks of lots of servers, etc. Plus all the UI and backend and search work. I'm not sure it's approachable as a small indie company.

There's also, of course, bad behavior by large companies such as mozilla: Thunderbird made it very difficult to build an indie email client because you have to compete on merits and against free, but they eventually got bored and just quit making it. Not to mention competing with free/ad-supported. And the semi-annual YC/vc supported email client company.

Not to mention what is coming close to active sabotage of productivity apps by the ios and mac app stores (lack of trials, lack of upgrades, etc).

I remember paying for The Bat! and loving it.

Pegasus mail was pretty good, even if a bit enterprisy. I just searched and wow... it's still being developed! The website even still lists the msdos version. http://www.pmail.com/

I'm sorry, I apologise in advance, but this irks me.

"We now live in a world where people pitch tantrums about paying $10 for an app"

Where do users/customers get these expectations? Call it market forces, race to the bottom, whatever... do a little root cause and not blame joe public please.

That's a long running debate and for our purposes unimportant.

It's still a prominent, maybe even the most prominent, item on my personal list of "reasons I shouldn't build an email app."

point taken.

Yup. I still don't see why everyone complains about current mail clients. Mail.app on OS X / iOS, Outlook on Windows, works for me.

To make me switch, you would have to show me why the new thing was so much better. Like how Google search results were always better than Yahoo! (IMHO, no objective measurements used) so I switched. You would also have to explain it in a sentence I could immediately understand, whereas all the recent clients I've ready about either (a) claim to make things better but don't tell/show me how before I get bored of the page, or (b) use manager speak that I can't understand or is vague and abstract promises to increase collaboration or something. Therefore I've given almost no attention to any recent attempt to change email.

There's a big distance between "works for me" and "delights me."

I use Thunderbird on Linux and Windows. It "works for me," in the sense that it's stable and predictable and doesn't lose my data. But it also has lots of niggly little problems that annoy me, like UI decisions that made more sense ten years ago than they do today and limited tools to support people like me who have a ton of mail.

So while I use and recommend Thunderbird, I also would love to see something come along that is so much better that it makes me fall in love with it, instead of just shrugging my shoulders and resigning myself to this being as good as things get.

How much are you looking to pay for this? How many people are looking for delight in their email client? Do these numbers multiplied add up to a few tens of millions?

"I'm not sure that's true (see: Sparrow). I just don't think there is enough demand for people to want to pay for an email client."

Yeah, this is definitely what I saw with Sparrow. For it to become popular it has to be adopted by a substantial number of people. However, I find that only really tech savvy people cared about Sparrow and its streamlined interface. When I showed it to other people, they were reluctant to use anything else because they were afraid to move from their Apple Mail. I'm guessing if Sparrow had existed for windows it would've been the same thing with people using Outlook.

And then there's people who don't even use e-mail, who think it's outdated and prefer PM messaging as a form of communication. It's easy to see why a lot of these promising mail client apps don't really take off the way other apps and services do.

Nevertheless, Sparrow will always have a home in my Mac, as long as it keeps working on OS X.

Same. Sparrow is still my favorite email app on the Mac. Sadly had to give up on it with iOS as it just got super wonky. Will use it as long as it works on OSX.

Still using Sparrow, but wondering if it works on El Capitan? Does anybody use it on El Capitan?

I use it with El Capitan. Still works fine.

Better than fine. Works great. Still the best email client, even after not being updated for years. That's saying something.

And you are limited by a 30 year old standard (IMAP) with an insane message format. Building a truly new email experience would require a new protocol as well.

There's JMAP protocol (from fastmail.com): http://jmap.io/

I really hope this protocol takes off and gains widespread adoption.

Me too :) Hoping to announce a bunch more support in the Cyrus IMAP server very soon!

It could be worse. I still have relatives using POP--across multiple devices!

Do you think Nylas is headed towards this eventual doom too?

I guess Mailpile isn't trying to be commercial, so perhaps it will keep soldiering on the power of donations for as long as that keeps up.

Hey there-- I work at Nylas.

We've structured our business to... actually have a business. Our team spent the first year selling APIs to developers [1] which already power lots of paid apps. The N1 mail app we released two months ago[2] follows this and will have paid features next year.

Plus both are open source free software, and quite popular on GitHub. [3]

[1] https://nylas.com/features/

[2] https://nylas.com/n1

[3] http://github.com/nylas/

Can you compare yourself to Mailpile? I want an email client (and maybe server) that is privacy-conscious and easy to configure. Are you two completely different projects?

We're two totally different projects. We focus on a lot of stuff in addition to privacy (like great design).

I understand what Mailpile is. It's a self-hosted email web client.

What is Nylas? The website doesn't seem to be written for me, but for PHBs. Apparently you have great business, design, and APIs, but what is it that those things do? Are you an email client? A server? A webmail service?

What makes you different from Gmail or from Mailpile?

Nylas is two things:

1. APIs for developers that replace IMAP/SMTP/Exchange. Use this to build any app that integrates with a user's mailbox, calendar, or address book.

2. A new extensible email app, built on web technologies (Chromium+NodeJS). This is called N1 and was released about 2 months ago.

We sell the APIs as an infrastructure business (like AWS).

This is the first I've heard of N1. It looks appealing, but after following the links and looking at it on GitHub, it seems like installing the sync server for self hosting is still non-trivial.

I went to the "releases" tab on GitHub and it looks like the last one was in 2014. I see things in the GitHub repo that suggest there are Dockerfiles and Debian packages, but I don't see an actual Debian package repository or an entry on the Docker Hub. Is the recommended way to use it today just cloning the git repo? Or is there a stable release of the sync server that's tested/known to work with the binaries of the client?

Yep, just clone the git repo and set up on Debian (like Ubuntu 12.04) and run setup.sh. Most developers run it in a VM.

We don't have "releases" because we are constantly releasing to production. Check out the `production` branch on GitHub for what we're running right now. Usually it's just a handful of hours behind master (at most).

I’d pay to have delete actually delete email completely (not just from my inbox, but from all as well).

@softawre: He's asking for it to delete on the receivers side too.

I think certain enterprise-y email solutions have this, but there's no way to have that happen in the distributed email ecosystem unless all your messages were just links to a website that could delete the content centrally, or similarly, jpg's of rendered text that were somehow forced to be remotely fetched each time.

And, indeed, that sort of functionality runs counter to the idea that once something has been sent to me, it's "mine" to keep/delete/whatever without someone else having the ability to change or delete it without my permission.

You have no idea how much some managers love this idea.

I know I had a former boss that loved to try using the outlook "unsend the stupid thing i sent" feature.

Little did he know the cients could turn that bit of functionality off. He hated when we replied to the emails he had "purged".

in order to do that, you can send an unique link as content so you can have total control over what you sent. doubt that many would click but i couldn't think of any other way.

other than that, i don't think receivers like the idea that others can control what's in their inbox.


I think he means he'd like to be able to irretrievably destroy all records of an email.

Yeah right now it labels the email with “trash” in the all folder instead of actually deleting it.

You wouldn't expect Slack to be able to make so much off of messaging - there's got to be some way to be better enough to make money.

Slack is a corporate product that sells subscriptions to their native-only service. An e-mail client that only let you see your last 30 days of e-mail without paying for it likely wouldn't catch on because people have different expectations for how e-mail should work/openness than they do with a corporate communications platform.

Side effect of this is that Mailbox was the first to support out of the box notifications on my Apple Watch + gmail accounts. Apple mail doesn't monitor/push emails from Gmail so you don't get notified on the watch.

Also, it's not 'disruptive' which isn't very sexy. At the same time, no one's really disrupted email (probably since gmail) and even sms for the most part, as a form of communication because they are the only cross platform with a very low barrier to entry. Note that all the sms replacement apps still use sms as a fall back since it's a shit show of walled garden apps.

I'm starting to believe that any idea of "disrupting" communications via a private company should be culled in its infancy. Market can't do infrastructure. Most companies will do their utmost to lock you in, and then abandon the product as soon as it's even marginally more profitable to do something else. Doubly so with Internet startups, for which the barrier to exit is incredibly low.

Yeah, it's frustrating. I was using Sparrow until it got acquired and then shut down. Then switched to Mailbox... which then got acquired... Whoops.

I ended up so frustrated I created my own email client in my spare time and half accidentally got into YC S14 haha. You can check it out here: http://www.slidemailapp.com

I'll admit, as a previous poster said, it's not without it's flaws but we try hard and we really do spend all our waking hours improving it since it's our baby and we love it to death. We've been beta testing the private current release with a lot of Hacker News users and I think we're fast approaching something really awesome!

Feel free to email me at vu0tran@gmail.com for a beta invite and would LOVE to hear your suggestions of what you'd like to see in an email client!!


What's your end goal with this? Surely it must be getting acquired by one of the big email providers.

Yup. Acquired for $101M then subsequently shut down in 1-2 years (just kidding of course).

I don't know if it's in my best interest to answer this publicly, but screw it. The honest answer is I'm not entirely too sure, but an acquisition is always on the table for any startup IMO. I think whenever you ask a founder, "If someone offered you $100M, would you take it?", it's a pretty loaded question.

There's really three outcomes to a startup: They die, they get acquired, they IPO. The thing is, if you think your company is valued at $1M and someone offers you $100M, of course you're most likely going to take it. It's 100x what you were expecting and it would be unwise not to take it. However, if you deep down believe that your company is worth more or can be worth more than the offering price and you have the proof to back it up, then it'd be unwise to take the offer.

For me, I strongly believe that email has A LOT to go. I mean, it's nearly 2016 and I still can't connect my Dropbox to my Gmail app client or send stuff into Evernote / Trello? Given all the time we spend on email, shouldn't we at least have some better automation / AI integration by now? Something smarter and more contextually aware that can make my life a bit easier?

So many people use email in the world. Literally everyone that has a Facebook account has to have an email account. I think the reason why we haven't seen a billion dollar plus company yet is that it's very ubiquitous and it's hard to build solutions for the masses. So companies give up, sell out, close down, and then we have to start again at ground zero.

So to your question, what's the end goal? Well, if we're right about our beliefs, hopefully we can become the next billion dollar company. If not, then it's the other two options: get acquired or die. Which one will it be? No idea, but we're just working our butts off right now.

> There's really three outcomes to a startup: They die, they get acquired, they IPO.

There's a fourth: they serve the needs of their customers over the course of years and decades. which, as someone in the market for an email client I can fall in love with and use for the rest of my life; is exactly what I want.

It's interesting that this option wasn't even considered.

Why does every startup need to aim to take over the world overnight? Is it not feasible to first take external funding to grow quickly in early stages, and then transition to a more organic growth model once you reach a point of profitability and sustainability (and maybe also slowly buy back shares in the company to eventually become fully autonomous)?

If that was an option, he wouldn't be in YC, half-accidentally or otherwise. Nothing against YC here, but long-term steady-state isn't what they're looking for. Remember that next time you start using a VC-funded tool.

Again, how was that not an option? Why did he have to get into YC? What's wrong with telling them that the type of unsustainable growth isn't right for their business now that they've had a chance to look at things?

> Why does every startup need to aim to take over the world overnight?

Its preferable to work on a project for a few years instead of half/full decades.

It's perfectly feasible to build a smaller, sustainable business then sell to someone who does want to have that lifestyle.

Of course its feasible! But is it preferable? No. I want to work on on something for 2-3 years, tops, and then cash out and or move on. I don't want to grind a project for the next 10 years. Life is too short for that.

thats certainly your choice to make, but my point was simply that there is an alternative way to do things. For me, if I could work on my best idea for the rest of my life, and it could go and serve the needs of millions of people (or even just a few thousand) across the globe, that would be pretty cool.

I don't disagree with you at all, but the idea I want to work on isn't going to make me wealthy. I just need enough money in the bank so that basic needs are met, rent is taken care of, and I can go work on fixing homelessness, dysfunctional healthcare systems, etc.

Those are really the only three outcomes for a venture-backed startup

Once you raise venture capital, you have to go big or go home, in the span of a decade. You don't have the "grow slow and stay private" option after raising a Series A.

That being said, this is just the current, common state in VC-backed companies. With the weak IPO market and hostile public environment that forces short-term quarterly thinking, we may see alternatives like formalized secondary markets. Perhaps companies like Uber will be able to avoid going public and still provide liquidity options for their early investors & employees.

> Yup. Acquired for $101M then subsequently shut down in 1-2 years (just kidding of course).

> hopefully we can become the next billion dollar company. If not, then it's the other two options: get acquired or die.

No offense, but it doesn't really sound like you were kidding. I have no skin in this game because I never used Mailbox and am content to continue using GMail, but if I was upset about this shutdown and in the market for a new email client that would be around long-term, your answer wouldn't exactly inspire confidence. You've basically said you're going to shut down (or get acquired and then, inevitably shut down) unless you hit the lottery ticket of a billion dollar company.

Speaking more generally, I think this sort of attitude is going to catch up with VC-backed startups sooner or later. Why would I make my workflow depend on a startup that is, in all likelihood, going to either shut down or get acquired and shut down? For anything with a high switching cost, using a product from a VC-backed startup that has this outlook is just going to cause me problems later if I get dependent on it.

I think a lot of companies would do better by their customers (and themselves) if they instead looked for a more sustainable business model. But there isn't as much of a lottery ticket in those businesses, so I can see why some founders avoid it.

You can sign up for Facebook with just a mobile number through the app.

"So to your question, what's the end goal? Well, if we're right about our beliefs, hopefully we can become the next billion dollar company. If not, then it's the other two options: get acquired or die."

What's wrong with just having a nice business that just makes some money?

Without joining the rush to judgement, allow me to say that I find this to be a legitimate inquiry, especially given the nature of your appeal in this thread. Care to respond?

Exactly. By joining YC, the goal is automatically getting acquired. So the parent is solving the problem temporarily, until they become the problem.

...and then get shutdown?

Nitpick: "If you delete the application from your phone, your data is gone for good" is the worst possible way to phrase what is otherwise an excellent feature :-)

Haha, thanks for the feedback. Copywriting was never my strong suit ;) Any tips on how I can improve it?

"We don't own your data - Your emails never leave your phone."

Devil(Luddite)'s advocate: "But how do I send emails to my friend's phone then?!"

"Your inbox never leaves your phone", copy isn't my strong suit either.

Thanks for the tip!

vu0tran, I'm almost sold. It looks great, and I love the private part (which was the one thing I could barely live with in Mailbox).

If you do snooze[1], I'll throw all my money at you :)

[1]: which of course either requires server side (e.g. Mailbox) or (preferably) client side logic. I'd be fine with the latter, but I agree it's a non-trivial problem.

Webmail is way too ubiquitous for desktop email software to be worth the million dollars a year it costs to hire five bay-area software developers to build and maintain it. I imagine Dropbox would have kept it around if it saw more user traction, but it didn't.

Inbox by GMail has some of the ideas from the Sparrow team incorporated in it, but it's only really usable on mobile. (The desktop Inbox web app is awfully slow compared to, say, GMail or Outlook.com.)

When Hotmail came out, who still paid for Eudora licenses?

That said, if a decent email client is that important to you, don't underestimate the impact you could have building it yourself as open source on evenings and weekends. And if you don't think it's worth your time to do that much, why would you expect anyone else to?

HotMaiL didn't make a dent in Eudora market share. What killed Eudora was Outlook Express, bundled with the operating system for free. Incidentally, the same strategy was used against Netscape (who then proceeded to quickly kill itself when faced with that prospect).

> And if you don't think it's worth your time to do that much, why would you expect anyone else to?

I don't expect anyone else to build me a FOSS email client. I do expect that by 2015 somebody would have figured out a way for me to give them money for a good email experience. :)

I did expect that by 2015 somebody would have figured out a way to make a decent kettle that lasts a lifetime, or a similarly decent fridge, or washing machine, or even a watch. Sadly, I was disappointed.

If you own something, you're not buying it - so the seller doesn't get the money. Software figured this out quickly - that's why everything that has not moved to cloud yet is being sold in form of licenses - a temporary grant to use some software. Rest of the world did it with planned obsolescence, and now the trend is to actually make you license hardware instead of buying it. It's sick. But that's where the money is, that's what the Market[0] says to do, therefore it's happening.

[0] - or Moloch.

There's a difference - software also has unplanned obsolescence. For instance, I've been working on a non-profit website that has a fair amount of QuickTime video from way back when that was a good choice. If you don't do something about it, it won't play anymore on modern browsers. It's hard to align the interests of software developers with the interests of customers using a one-time purchase, but with work that continues indefinitely. Appliances are different since I don't need their features to change long after the sale. In this case, if they can charge me more at the time of the sale for an appliance that lasts longer, we can align interests.

I think the guy behind MailMate¹ has done just that. It’s definitely the most powerful mail client on OS X. I ended up buying it, at least!


¹ — http://freron.com/

I still love Apple Mail... and it's native. In fact I think it's the best email platform out there currently, although I haven't tried N1 yet.

I'd like Inbox-like features theoretically, but the fundamental UX/UI is done better than anyone else IMHO.

If you mean OSX do this:

- Open Mail, roughly time how long it takes from left pane click to message fully rendered on the right (~2-3 seconds)

- Open Dropbox (fast before it goes away!) and do the same (~250ms)

I get roughly 100-150 emails overnight that I go through in the morning. At 2-3 seconds PER THREAD render this is insanely unproductive use of my time.

Every time I open OSX Mail I have this exact process in my head and I promptly close it.

Not sure what's happening with your setup. For me it's nearly instant. Some long long threads or heavy HTML can be up to about 800ms.

I'm also talking about my iOS experience. There are some perf issues that would be nice, but for me that's mostly in search and initial sync than stuff that's already downloaded.

The inbox-like features ARE the fundamenatal UX/UI. I don't want a mail client, I want a gmail client, and IMAP/POP3 are terrible replacements.

FFS, just hand the product over to the community. Or have someone manage the project (security/features/quality) and accept patches from the community. Or use ANY of the revenue to hire someone outside the aqui-hire part-time to work on it.

It's such a shitty, disrespectful, dangerous pattern to buy and shutdown products. The community knows it and bemoans it every time. Blogs have been set up just to highlight how shitty it is. Countless words written about how shitty it is. This isn't like giving away free cookies and then stopping, it's like standing someone up on a date.

"—and we’re sorry. It’s not easy to say goodbye to products we all love..." but it's just business move, and business only leverages emotions and feelings when it's profitable for us.

I'm genuinely curious as to what it is that Gmail (for example) doesn't provide yet that is so essential to dealing with emails. I gave a try to Mailbox but didn't see what the added value was, other than a couple UX gimmicks (some of which are now part of many email clients).

Was there a killer feature that I missed?

Off the top of my head - offline support, PGP support, multi-account support, decent mail-list/threading support, separation of client from service provider..

I'm actually quite a fan of the gmail interface; it is one of the most usable webmail experiences in my opinion (I use google apps both at work and for personal use). For day-long usage, I'd rather stick with mutt/vim.

Folders. Didn't Gmail kill folders?

It has labels, of which you can apply one or more of to a message. When represented in IMAP, it appears as folders (though you get duplicate mail objects for every additional folder that a mail is tagged with).

If you only apply only one label to a mail at a given time, it works as a folder would do just fine.

When Mailbox first came out, there was no Google Inbox, no snoozing, and the ios e-mail experience was lacking and sluggish in general. Mailbox fixed all that, and it was the first to properly do so that we trusted. We trusted it as it was by the well established Dropbox.

But since Inbox came out, I have seen no reason to open Mailbox, and I would say that Mailbox actually inspired Inbox.

Multiple accounts. I have multiple work accounts (several companies) and a personal one. Having each open in tabs/window would be too much. I use Apple's mail apps on Mac and iOS, with a single unified inbox.

Yeah, after Sparrow was killed off I switched to Airmail for OS X (currently using Airmail 2). It's not without flaws and I'm always keeping an eye on the options.

On iOS I'm pretty happy with Microsoft Outlook (can't believe I just said that), I haven't found anything better yet.

I do use multiple accounts extensively so that limits my choices to just those clients that support multiple accounts well (unified inbox, etc).

If you're a desktop email aficionado, I'd love to hear your feedback on N1. It's kind of similar to Mailbox desktop, but works cross-platform and is open source. https://nylas.com/n1

(Also feel free to write me directly. mg@nylas.com)

I tried it with a single account, the UI is nice but I kind of prefer a more dense UI for email. I'm probably just really used to Airmail's UI, it's similar to what Sparrow used to be and I felt that was approaching perfection at the time.

Also the memory footprint was a little large but that's understandable since it's built on Chrome/Node. Web browsers and development related processes (mostly Ruby) are my top memory consumers right now, I'd rather not have my email client join their ranks.

Does it have a "snooze email" feature, though? The page suggests it doesn't, and that's (IMO) the important part of Mailbox, not just the general interface of it.

I asked this a while back when it came out, sounds like it doesn't.

[1]: https://news.ycombinator.com/item?id=10334154

Update: It looks like they have a public Trello board for voting on features. Go vote for snoozed emails!

[2]: https://trello.com/b/hxsqB6vx/n1-open-source-roadmap

Is the email stored on your servers ? Do we get contacts and calendars inside the app as well ? If yes, on which email providers/account types ? Why isn't this info on the website ? Thanks.

If you look on this page it talks about the mail being hosted on their servers: https://nylas.com/N1/faq

Everyone want to "fix" email, and replace it with some proprietary bullshit. Email isn't broken.

Like many things, clients could do with some usability enhancements. E.g. encrypted mail. Without which, I find it off-putting.

Encryption is really an identity problem.

For intra-corporate mail, email works just great, because the identity problem is solved. Trusted communications between 3rd parties just isn't a natural mail use case.

> For intra-corporate mail, email works just great

Not really. Hence the rise of intra-corporate chat software.

People instead prefer to join focused, well defined groups to send and receive intra-corporate communication.

Not bee CC'ed spammed all to h*ll about every little office detail that has zero impact on you and your job.

Email, unfortunately, is very broken.

Spam, in the I'm getting random Viagra emails sense, is being handle very well lately.

That said... We have a very large industry that easily helps you get your "ham" (the mail in-between Spam and mail you really want to receive) inboxed.

This is email from legit companies. Maybe you signed up to a newsletter, maybe you didn't. Maybe you're just getting a cold email. But those emails are inboxed.

People are still complaining about being overwhelmed by their email, even though our Spam filters have improved greatly.

Email is plenty broken. Spam, for example.

Spam is pretty much addressed these days.

You need to consider what is more valuable, the ability to universally send arbitrary content to arbitrary end points without intermediaries .... Or no spam, and perfect security.

As I said, it's all about identity.

Proprietary tools just aren't worth the turnover for me.

It's untenable to depend on developer tools that a company owns. I love free software, but in all honesty the reason I incorporate it is more selfish. My workstation is exclusively free software so I don't have to be abreast of business news to get my job done.

I've basically settled on my emacs + notmuch + offlineimap solution not because I think it's better than all email clients -- but it's the one that if all parties involved quit, I could probably use it for another 10-ish years.

For example - offlineimap right now still has this old bug where sometime it uses all my cpu (and a bunch of memory) in some sync case. Ok, I don't have time to debug this. I set limits and have it be killed when that state occurs. Annoying? Sure. Have to configure whole new mail client? Nope!

I moved from offlineimap to mbsync (part of the iSync project) a few years ago and it has fixed most of the issues I had with offlineimap (mainly high resource usage).

Several people have suggested similar. I should do it, but the nice part of OSS: I can do it whenever it bothers me enough :)

Do you have a config posted anywhere? I'd love to see exactly how hard it would be to switch.

It's pretty straightforward to configure - Arch have a decent guide on their wiki: https://wiki.archlinux.org/index.php/Isync

Edit: Arch seems to suggest running mbsync under a global systemd unit - user units are more appropriate for this task - my units can be found here: https://github.com/rdark/systemd-user-units

How do you send email with your pipeline? msmtp?

emacs has built in smtp support, so I just use that.

Notably, I use the smtpmail-multi package along with smtpmail in emacs to support sending from whatever address I put in the To: line[0]

[0]: https://github.com/codemac/config/blob/master/emacs.d/boot.o...

That is, a $100m acqui-"hire". And the problem is, probably even Dropbox thinks that way. Am I the only one who thinks this is just a bit bonkers?

> I still bounce around email clients every couple months

Between two-factor authentication, application specific passwords and the notification center, configuring a new mail client can be a lot of work to have to redo every couple months. I did this for years before finally settling on the default clients for both iOS and OS X, and I've noticed a lot of my coworkers have done the same.

What do you not like about Outlook or Gmail (I think the market leaders)?

They don't work with other people's IMAP/POP3 servers?

(Well, the Gmail iOS app certainly doesn't; I haven't tried the Outlook client but the description in the app store suggests a very limited set of providers.)

The GMail Android app does work with other people's IMAP and POP3 servers, for what it's worth. But then you have to use Android.


Gmail for Android supports third party pop\imap servers quite well.

Ah, good to know, thanks.

Outlook (I assume you're referring to the mobile app, not the desktop client) puts all of your emails on MS's servers, so that's a no go for lots of people.

Last I checked, the gmail app only worked with gmail accounts.

The Gmail app actually allows 3rd-party IMAP connections.

Do you mean gmail's web interface? That works by downloading your emails from your 3rd-party IMAP connections into Gmail's server.

No, he means the 'Gmail' app (on Android phones at least, not sure about iPhones) allows third part servers. They rolled the Mail client into it.

The "delay for later" feature is tremendously useful compared to traditional inbox handling.

Of course, to trust that feature, it also has to be tremendously reliable, and that's the biggest thing that Mailbox got right in the first place. That's also why it very handily beat out Mail Pilot, which had the same idea and got started earlier but to this day is still filled with terrible bugs.

I'm not the parent but Outlook's IMAP support is quirky at best and I very strongly prefer native applications to web applications. I also host my own email which marks gmail right out.

Email should be simple but there has been essentially no lasting innovation in email clients in a very long time unless you really want Google to host it for you

Gmail on mobile still doesn't have a unified inbox, and it also doesn't have features I've come to depend on such as snoozing messages. I usually use Inbox by Gmail on mobile, which also lacks a unified inbox and (bizarrely enough) support for HTML signatures but does have pretty good message snoozing, and I like how it handles bundling.

Outlook (incidentally, another example of my original complaint, as it's just former email startup Acompli with a fresh skin although obviously it's probably not getting shut down anytime too soon) doesn't have very good message searching for Gmail, which is a common complaint I have for non-Google mail apps. Obviously, I get that Google has a huge built-in advantage here; but if your email app is going to support the "Mailbox paradigm" of abandoning folders/traditional labels in favor of just marking things as done, it also needs to have search functionality that can back it up. Outlook was also really buggy for me when I last tried it, although that was a little while ago so it may have improved.

The new Outlook app is actually surprisingly good. I think the Sunrise guys (sunrise.am) are running that team and it shows. I laughed the first 2 times friends recommended me the app, but I tried it and now also highly recommend it.

I used Acompli and now I use Outlook.

My favorite tiny detail is that I can tune the notifications to push only unread counts to the icon badge and only for the "focused" inbox. That means I can avoid constant push notices for email yet still be able to glance down when I want to and see if I have any unread emails that are actual human communications.

For certain. I just gave it a try a couple of weeks ago and it's now my default client as a gmail user.

Gmail on mobile does have a unified inbox:


Gmail for Android does, I suspect the OP was talking about iOS.

It only came out 6 months ago or so, I honestly thought I was doing him a favour letting him know about the new functionality.

Instead I get downvoted. Gee, thanks. Since when has downvote come to mean disagree on HN?

Well, I didn't down vote you so I can't say for sure. But "Gmail on mobile does have a unified inbox" isn't a true statement if you are an iOS user.

It wasn't said with ill intent. A reply explaining would have been much better than a downvote.

Message searching for Gmail in Outlook iOS/Android is piped directly from Gmail itself. Your search is forwarded on to them and results are displayed back in the app. Immediate search results are from on-device and then filled with the Gmail results after they are received. So if the results are bad, then maybe Gmail has a secret API they use to make their app better.

Seriously how can you even like Outlook? Yes people need to use it, but want? Does anybody really know other mail programs and wants to use Outlook?

It's such a pain to process hundreds of emails in Outlook, and there's no snoozing like there is with Mailbox or Inbox.

Outlook does have snooze/scheduling.

Just curious, how is processing hundreds of emails in Mailbox or Inbox better than Outlook? Seems like trying to do that on a mobile device would be hard no matter what app.

Am I the only one still using Sparrow to this day?

Sure the build is 2+ years old, but the usability is just incredible. I haven't found anything better on desktop since.

Airmail is quite good, and AFAIK you can turn off some of its many features. http://airmailapp.com/ (hint: You can use it for free if you use the beta version)

I was a huge fan of Sparrow (like you, I used it long after the updates stopped), then Airmail and Mailbox, but now I simply use the Gmail web app. I follow the zero inbox method strictly, so I only need to read/archive/reply/search emails, and it does a decent job at it (except for search where it's best in class), while being simple and non-confusing.

I was, but hadn't bought a license before the acquisition, so that part of things got a bit annoying. IIRC there are ads that take up a good amount of space.

FWIIW, Inbox has been created by the Sparrow team (among other googlers oc), so at least it is not a case where an acquihire ends with the people working on completely unrelated things.

I felt this coming given my experience using the Mailbox desktop app since it launched:

- Good experience, fairly regular releases, stability increasing with each one.

- Stagnation in releases, but app is in a fairly good working state.

- Out of the blue, big update comes in, application changes completely. For the worse, as it loses a bunch of features and is sporting a far less polished look and finish. I believe the new version is now an OS X native application as opposed to a webview.

- Frantic releases over the following weeks, killing some bugs but introducing more.

- Stagnation in releases, in its fairly broken state.

- This announcement.

From the outside, it looks like a case of an engineering team that decided it would rewrite the application from scratch in the native stack. Widely regarded as a bad idea. [1]

After numerous months spent burning money in refactoring and rebuilding features that already existed and worked, management pressure builds up, and they decide to release their "good enough" native version. After torturous weeks of back to back frantic releases to fix all of the complaints coming in, some factor or another (developer churn possibly) caused them to cease development and decide to sack the project altogether.

This is of course just speculation, a narrative I made up. That's how it looks like from the outside to me, but I'd like to hear from the developers inside, since I know they must be reading this thread.

I loved this application and I'm saddened to have to move away from it.

[1] http://c2.com/cgi/wiki?RewriteCodeFromScratch

I worked on the desktop client for several months up until the first public beta release.

No, Mailbox for Mac has been a fully native app since inception - which is how you get those nice fluid gestures.

What happened next I have no idea, since I left the company. If I have to guess, we incurred a lot of technical debts during the sprint to beta, and they were doing some major rewriting to fix them. Before it was finished, the project was axed. The new version was released nonetheless to support sunsetting.

The reason I thought the previous version was a webview was due to the not-really-native-looking settings pane (Cmd+,), which turned into a much more typical looking one post-update.

Thank you so much for the input, and thanks for helping build something I enjoyed using.

Glad I'm not the only one who thought the one supposed 'major' update did not add very many useful features and made the UI look cluttered. Also, it ran surprisingly well for a webview app, I don't see why they felt the need to rewrite it.

I just don't think they had the numbers on Carousel, and maybe not enough growth in Mailbox either. If they were getting users like mad, of course Dropbox would focus its attention on expanding them.

>I just don't think they had the numbers on Carousel

My friends experience matched what they said on their blog that people just want their photos in their Dropbox

I wish they could open source Mailbox at least... That way it might survive.

[1] "Will Mailbox be open-sourced?

Unfortunately not. We gave a lot of thought to open-sourcing the underlying system, but this is ultimately not something we will support."

[1] https://www.mailboxapp.com/faq/

"We gave a lot of thought to open-sourcing the underlying system, and none of that thought will be elaborated on here."

I'll elaborate on their behalf: "That would be hard and expensive and we have money to make."

Even if they wanted to open source their system, it is likely is that they migrated their codebase to the Dropbox platform, and it would be difficult for them extricate their code. Most larger tech company (e.g., Google, Facebook, MSFT) have this problem when they want to open source products...

God forbid anyone make any money. Where I work we feed our engineers on the appreciation of free users. That's the only way to fly. Self-entitled much?

God forbid anyone share abandoned code. Where I work we incinerate anything potentially useful when we're done with it. That way no dirty freeloaders can take advantage of it.

Let Pa be the probability that some piece of your open-sourced codebase reveals an important technique or strategy to your competitor, thus leveling a technological advantage you have,

Let Ca be the cost of that lost advantage,

Let Cb be the value of good will from the open source community

In what case is Cb bigger than (Pa * Ca)?

When Pa == 0, so like, most of the time.

It's an email app. It talks IMAP^H^H^H^Hto Gmail. It runs on iOS. How many secret, commercially important techniques do you really suppose it contains?

Suppose it writes to Dropbox using internal APIs? How many developer hours should they spend abstracting/obscuring that usage? What are they going to gain by doing that? What are they risking if that process misses something?

There' no such thing as an "internal API", assuming the API works over HTTP.

I don't even understand what this comment is trying to say. If you use HTTP to communicate between services, and those services are not publicly accessible, the use of HTTP makes it no longer a private interface?

For a site which supposedly hosting an audience of entrepreneurs and engineers -- people who understand that the value of a thing can be multi-faceted and not always obvious, or that the difficult of any job is easy to underestimate, and that to convince someone to do something you have to appeal to their incentives/concerns rather than your ideals -- the entire argument in favor of opening this app is built on pedantry and baseless assumptions.

> I don't even understand what this comment is trying to say. If you use HTTP to communicate between services, and those services are not publicly accessible, the use of HTTP makes it no longer a private interface?

Yes - if you make a HTTP call from an app, it can be trivially sniffed. Sniffing HTTP is the first thing a third party trying to discover that undocumented API would do, and you don't need source code for that at all. (This is also why you must always sanitize data coming in from a user's device, even if it's from your own app.)

You can make the argument that there would be a time cost cleaning up the internal calls that will no longer work once the servers are turned off. Sure, but: 1) there are no secrets that would give competitors a new advantage, and 2) if you don't have that time, just chuck the code over the fence and see what happens - worst is that no one uses it, which will be the case with closed code anyway.

The is just so unimaginative that I can't believe you've ever seen the inside of a large operation. How are you going to sniff traffic that's communicating on a private network? And how does trivial sniffability not extend to all unencrypted traffic, ergo nearly everything is public? That's a totally useless definition and misses the point entirely.

Your point #1 is totally unjustified: you don't know what you could learn by e.g. looking at a data structure used internally. #2 shows that you are unable to answer the question of which of Dropbox's incentives are satisfied by doing this.

Someone said consumer apps (Carousel and Mailbox in this case) couldn't be open-sourced because they use "internal" APIs.

My point was just that any API over HTTP that's used by a consumer app is not private or internal. It is a public API with unfriendly documentation.

(Note that when I say "unfriendly documentation," I'm not even talking about sniffing. Most consumer apps can be decompiled by non-experts, and then the text-based API calls would be readable.)

Undocumented API then?

That "formula" really isn't particularly difficult to work around.

First off, if you're open sourcing your codebase because you're getting out of a particular market, you have to ask whether revealing techniques and strategies to competitors in that market is really an issue. After all, if those techniques and strategies had given you a competitive advantage, you probably wouldn't be having this discussion.

Second: If you really are concerned about that, just use the GPL.

The assumption here being that the integration of the now disused product into the parent product (Mailbox into Dropbox) has no potential to reveal the internal workings of the parent product e.g. APIs or data structures. Or that those techniques would only be applicable to competitors in that market. Both are convenient for your argument, but there's no reason to think that they're correct. Moreover, there's no reason for a risk adverse company to accept those assumptions.

My point was to challenge open source cheerleaders to actually give a reason beyond their own gain for why a company should do this. Instead, we have blithe dismissals and narrowly constructed hypotheticals built on optimistic assumptions.

> My point was to challenge open source cheerleaders to actually give a reason beyond their own gain for why a company should do this. Instead, we have blithe dismissals and narrowly constructed hypotheticals built on optimistic assumptions.

I'm sorry, but what? Your whole initial argument is a narrow hypothetical "they will see our secrets" with no theory of what those secrets might actually be - what exactly do you expect in return? I gave you an answer based on your formula, and a follow-up comment afterward. Can you expand what about my answer was built on optimistic assumptions, in a way that your initial theory was not?

There are no private APIs nor secret data structures in software that you've distributed to users. It can all be decompiled and sniffed. "Oh but the competitors will see my code" is basically FUD. How many times has YC told us it's all about the execution, not the technology?

Yeah, the difference is that the "narrow hypothetical" is a concern a real person at any company would have when tasked with deciding whether something should be open sourced. It's appropriately conservative.

You, however, are asking everyone to assume that it's totally safe to reveal any/all source code.

> There are no private APIs nor secret data structures in software that you've distributed to users.

That's fine. What about the code that lives on your servers and supports the client?

When abandoning a market and an app, there is no strategic advantage in that code. This means the solution to that equation is CaCa.

I imagine Mailbox has a heavy server component, which would be non-trivial to get set up and running on your own.

If you're looking for an alternative, we built one at Nylas. :) https://github.com/nylas/sync-engine

Nylas looks awesome. Has anyone built a plugin for it that allows for "message snoozing"? That was the real ingenuity with Mailbox IMO.

Not yet! Totally something that we'd love to see, though.

That's hardly a reason to not open-source. You're not required to provide support just because you release your source code. Even a dump of their repositories would be useful for people who are interested in Mailbox.

No; but often the server components are heavily tied to some other proprietary libraries that they're not ready to open source. They could remove all proprietary code from their codebase, but that costs money - likely more than they were willing to spend on this product.

Unless open source is part of your marketing strategy (which means the effort would have a budget), it's really difficult to open source an existing commercial application.

That's a speculation on your part.. and PR spin-off on theirs.

Nobody asked them to support it, or whatever. Just dump the damn thing online so others can pick up and continue to "help fight your inbox to zero".

Clearly their intention here was profits and since they didn't see any - they kill it off. They won't do it so nobody else should try to safe email either.

I wanted to get my company (over 3,000 users) off DropBox long time ago after each update comes with new issues. But this just broke camel's back. I will make the switch happen this weekend.

> Clearly their intention here was profits and since they didn't see any - they kill it off. They won't do it so nobody else should try to safe email either.

Um, of course their intentions were profit? Dropbox is a for-profit company, nobody ever questioned their intentions as being anything else.

And it's not pure speculation; I've worked on enough internally-developed products to know it's not always feasible to open source an entire service offering after the fact. It's expensive enough to do between code scrubbing, legal reviews, etc. that you're generally only going to do it for strategic reasons (i.e. you know you can never make money doing it but want to commoditize the market space to hamper a competitor's growth, you want the community to help support your infrastructure, or your business model is open source + support).

"That's a speculation on your part"

Whereas your post is speculation on yours.

"Nobody asked them to support it"

At the same time, if they do open source it, people are going to see that it was once Dropbox, and any issues they see with it are going to be seen as the fault of Dropbox.

"Just dump the damn thing online"

Far more work than it sounds like.

"Clearly their intention here was profits"

They are a company with bills to pay and engineers to pay, right?

"They won't do it so nobody else should try to safe email either."

I don't see them going around and killing other people's email clients.

Honestly most startup backend systems are not amazingly engineered or something to be proud of, so I would guess that the code would need some considerable work before releasing it to the public if that is what they chose to do. Even after release, the project would need considerable support to get it off the ground (helping the first users, accepting contributions, establishing the project community and leadership). Simply dumping the code on Github with no instructions and leaving it to rot is not particularly useful.

Then OS the previous version that got them purchased by Dropbox.

Someone started a petition in an effort to lobby dropbox to do just that:


This is very sad news and I don't really get the decision.

The thing with Mailbox is that it was truly a great product before it switched hands to Dropbox. Once they bought it, that was the end of good functions and the product only went downhill from there.

There's a lot of potential with email clients that will help you work better, Mailbox was definitely one of those products that helped. With good clients for Mac and PC along with smartphones it would also be profitable IMHO.

The only thing I can think of is that Dropbox is headed for major firing rounds and they want to save up on resources.

I have switched to Airmail on my Mac and I am much happier now.

I'm still hoping they'll sell them to someone.

Dropbox is obviously planning to go a different direction with its core product strategy, so that makes sense to me, but those products were too good to just let die. I'm sure someone would be willing to take them off of Dropbox's hands.

That would be the rational thing to do. I don't believe it will happen though. If it was going to happen it would before they announced they're shutting it down.

I just don't get how you can spend 100m$ on something and just let it die. What did Dropbox get out of this deal?

SV is depressing sometimes.

They got the brains and patents (if applicable).

So you're saying that Mailbox was a great product for the whole month between when it launched and Dropbox acquired it?

I can tell you that behind the scenes it was far from a well architected service, and the only reason why it was able to ditch the signup waitlist and ultimately handle the needs of its userbase was entirely due to the Dropbox acquisition.

What I am saying is that Mailbox was a great product, yes.

Looking at it from the outside, it seemed that Dropbox initially just continued the initial product velocity that existed from before the purchase.

After a while it seemed that the development lost steam, features were lagging and it seemed that they lost interest.

I am sure it wasn't a very well architected service and beyond that growth was "controlled" by the invite system (which ironically only fueled the growth).

The Dropbox acquisition had a lot of potential, mainly in fueling the growth by adding people/servers and ops knowledge.

So yeah, they ditched the invite system and a lot more people were using the product but what's that worth if the end decision is to drop the product all together?

What I'm failing to understand here (and I will admit it's due to non-existent investment knowledge) is what did Dropbox get out of the 100mm paid on the product and what led the decision to drop it now.

I guess what I'm trying to say is that I don't think Mailbox was derailed by Dropbox, since it barely had a public life before Dropbox and certainly would not have survived without the acquisition, Dropbox is equally as responsible for the few good years as the original Mailbox team.

As far as what Dropbox got out of the acquisition: no idea. I know there was a lot of interest in the Mailbox team, and they were immediately given a much broader purview post-acquisition (much like the founding team of Cove when they got acquired). Ultimately I don't really think that parachuting in founders to take over parts of your company works -- three out of the four of those people no longer work for Dropbox, for instance. However, at the time Dropbox clearly craved adult supervision, and I think that was seen as the way to get it.

Oh, and if you have a product people like, that doesn't hurt either. I just found it maddening that Mailbox was built from day one not to make money, and that problem was never solved. That sort of thing can only last for so long.

I thought Carousel looked really cool, eagerly installed it on release.

Then I discovered that when Carousel is installed, the Dropbox app stops automatically uploading photos from your phone. You have to now open Carousel to do so. Broke my existing upload process, and due to iOS permissions for the new app or not launching it frequently or something the new Carousel upload process wasn't reliable.

Discovered this by noticing that weeks of pictures weren't backed up to Dropbox. Glad I didn't lose my phone.

Ended up just deleting the Carousel app rather than figuring it out. The Dropbox app started syncing reliably again.

Sounds like you had a bad experience. I enjoyed the app until recently switching to amazon primes unlimited photo uploads. The amazon photo app has a very similar feature set compared to Carousel, but seemingly worse performance on large collections.

I had the same experience. I really liked Carousel. It was a great way to organize my photos, but like you said it totally disables auto-sync.

On iOS at least, apps can't run in the background. Carousel could only upload photos when you open it (or, as a workaround, when your GPS location changes if you've given that permission).

Carousel's "Flashback -- Discover This Day in the Past Year" feature strikes me as a very clever social workaround since it gets you to open the app at least once a week.

The Dropbox iOS app has the same limitation. Presumably you opened it more to do unrelated tasks, so it makes a lot of sense to bundle them.

Google Photos has the same limitation as well on iOS. I find Carousel's syncing and Google Photos' syncing seem to be roughly on par with each other.

Crap. I absolutely adored Mailbox; it did everything I wanted, but especially:

1) The ability to snooze emails

2) A really nice UI on both OS X and Android letting me quickly swipe emails to archive or snooze them.

3) A unified inbox that showed multiple email accounts as a single inbox.

Does anyone have any suggestions for a mail client that meets these requirements? I'm quite happy to pay.

Edit: Google Inbox doesn't have a unified inbox, which drives me nuts, and nothing else I know of has message snoozing. ANY suggestions welcomed.

Outlook. Seriously. It supports multiple accounts I believe (I only use my work exchange account so not positive), and it has scheduling of messages. The scheduling hasn't synced very well to the desktop version in my experience, but a lot of people have called it the best gmail client on iOS.

Well, I guess it's worth a shot.

Although my primary OS for dealing with email is OS X and I see the OS X app is $130? Acceptable for a good Mailbox replacement, but a bit pricy for a gamble. :(

And "The scheduling hasn't synced very well to the desktop version in my experience" worries me a lot.

He means the Outlook iOS/Android app, which was a 3rd party app Microsoft bought (Accompli) and further upgraded by the people at Sunrise Calendar (also bought out) to make the app that it is today. Outlook for OSX, though, is still the classic Microsoft Office for OSX app, so it doesn't have any of those upgrades. The iOS app is my daily driver, and it works splendidly.

>nothing else I know of has message snoozing. ANY suggestions welcomed.

The new Outlook mobile apps support scheduling for later. I don't think it's as nice as Mailbox was (with multiple choices for when to schedule) but it's the closest thing I know of.

Airmail doesn't support snoozing, but it does have a unified inbox and nice UI. IMO it's a bit overloaded with power and features, but it's worth consideration.

Another commenter pointed out that you can try/use the full version for free in their beta program.

I loved it too. Cloud Magic on Android comes close.

Sad, Drew had said it's not going anywhere when the acquisition was announced.


Executive promises at the time of an acquisition mean absolutely nothing, their main goal is to reduce break-off risk (both of users and employees) due to change of control.

It's far more likely that he was overly optimistic about Mailbox's chances of success at Dropbox. Founders are notoriously delusional about this kind of thing.

Dropbox wasn't worried about maintaining the small existing Mailbox userbase, they were betting they could grow it by millions of users. When it didn't work, it messed up their plans (and promises).

Never attribute to malice that which is adequately explained by stupidity.

> Never attribute to malice that which is adequately explained by stupidity.

As a CEO I'm not sure which of those two explanations I'd prefer.

Stupidity because people who act in malice are dicks. You can choose to be nice, but it's inevitable you'll be stupid at one point or another.

To be fair, that was 3 years ago. A lot can change in that amount of time.

Yes, they could do a spin-out.

Is this Dropbox cleaning up their balance sheet in an attempt to IPO?

There's a pretty loud rumor floating around, heck it was reported on CNBC today, that Dropbox can't come close to IPOing anywhere near its $10 Billion dollar valuation. Though Fidelity did mark them up slightly at the end of November!

And for better or worse, they are going to be closely compared to Box, which has been a pretty big disaster since IPOing. For the record, BOX is valued at 1.65 Billion currently. And it has an awful lot of short interest current( a measure of the ratio of shares sold short vs the total outstanding shares).

On the other hand, I thought Dropbox was squashing Box like a bug in terms of users. Not sure if that translates to revenue though. Wouldn't investors want the most profitable company to IPO?

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact