Dropbox Hits 25 Millions Users, 200 Million Files Per Day | Hacker News Reader

Dropbox Hits 25 Millions Users, 200 Million Files Per Day

techcrunch.com

199 pointsssclafani74 comments

Thanks to Dropbox my photo sharing site has the world's best uploader. (1) Since it's an ordinary folder, all photo management software (iPhoto, Aperture, Lightroom, etc.) can export to it without installing a plugin. (2) People can queue up photos to upload while traveling even if they're currently without Internet access. (3) Mobile clients are already built, and people can use the same client either to upload to my site or to send to their desktop for post-processing.

If you're running a site that people upload files to, I highly recommend integrating with Dropbox. They have an API, but I just did it by letting people share a new folder with box@ourdoings.com and pulling photos from that folder.

Sounds great for people that already have dropbox, for others that might not work so great. Most savvy computer users understand dropbox, but sometimes (less savvy users) just can't get it to work.

I wouldn't make it the only upload method, but I highly recommend it at least as an option. And with 25 million users, there's a good chance a lot of your new users already have Dropbox, especially within the early adopter crowd.

I know it will probably end up being the first $1 billion+ exit for YC but I'm really hoping it's the first IPO. It'd really put an end to the myth that YC startups are "dipshit companies" once and for all (not that Heroku didn't do a decent job of that).

Your comment makes me wonder if Dropbox has an exit strategy. Or if they're going to do like Digg and sit there until the entire industry moves on. With anything cloud-related being big busines right now, Dropbox should quit winners and either sell-out or go public.

"Your comment makes me wonder if Dropbox has an exit strategy. Or if they're going to do like Digg and sit there until the entire industry moves on."

Those are not the only two alternatives.

Especially not when your company has a solid plan for making money. Digg never did, Dropbox does.

I don't believe Dropbox has an exit strategy, at least not for the time being. From what Drew Houston has to say about the company[1], the big vision is to give everyone universal access to their files, regardless of where they are or what platform they're on [2]. That's something that they seem willing to commit to for the long haul.

[1]http://youtu.be/eaJKPCdimLA - actual quote being (21:16 mins): 'I'm completely dedicated to Dropbox ... I mean we're starting with files but there's all kinds of other directions we can go ... we're a long long way away from you being able to sit down in front of any computer, device, and have access to all your stuff.'

[2]The analogy that Drew used was Athena, which apparently is this infrastructure at MIT that gives you access to your stuff no matter which workstation you're on, on campus.

In the context of GP comment ("sit until the industry moves on") I think Dropbox has one potential problem -- that people would start realizing that their data is their data and paying more attention to the privacy aspect of the "storing in the cloud" services. They sure are convenient, but the loss of tight control of one's private data is an issue.

It all goes in cycles, and if something happens that prompts companies and individuals to lock down their files, then Dropbox and alike will go from "universal file access" service to a trivial file sharing app. Still useful, but not as grand.

That's a good point. I'd add that such cycles tend to happen within geographical boundaries (e.g.: this year, something happens in France that makes people in Europe lock down; the next year it's the US, etc) and so it's unlikely that Dropbox would grind to a complete halt, seeing as they've become increasingly international.

> It'd really put an end to the myth that YC startups are "dipshit companies" once and for all.

Of course it won't. Dropbox will just be labeled a fluke and the same rhetoric will continue. The underlying issue is not the lack of 1B exits, but rather a form of animosity or envy depending on how you view it.

Nobody says that.

Yes, they do:

http://www.avc.com/a_vc/2010/07/lead-investors-dipshit-compa...

http://venturebeat.com/2010/07/29/angelconf-ron-conway-micha...

1. No, he says other VCs think that the funding landscape in general creates dipshit companies. Not that YC companies are dipshit companies.

2. It's Arrington talking. ;)

Dropbox is the best startup to rise in the last few years. In terms of startups I use that have actually changed the way I use computers in a huge way, it is right up there with Facebook. It is one of the services that I simply could not imagine living without, now that I know it exists.

If you haven't used it, or are still on the free plan, you really should try it out and buy at least 50gb. Stick all your most important files in there, and forget about that annoying thing called "backups".

>Stick all your most important files in there, and forget about that annoying thing called "backups".

Good luck with that, when dropbox has its first inevitable data loss incident.

If you don't have physical control over the hardware, it doesn't count as a proper backup.

The neat thing about Dropbox is that your files get automagically mirrored to all your computers with Dropbox installed. So unless the client blows up too, odds are good that a data loss incident on their servers wouldn't be catastrophic.

This isn't an enterprisey backup solution by any means, as I don't believe previous versions are stored locally. But it's a whole lot of really convenient redundancy.

Does it automagically sync file deletions and corruptions, too?

It keeps file revisions server-side. So you'd have to have one heck of a failure to find out:

1) Have all your nodes syncing at the same time and connected. 2) Introduce file corruption and deletion. 3) Have all your sync'd machines get the deletion. 4) Have Dropbox's file revision history go wonky on their side.

> 1) Have all your nodes syncing at the same time and connected.

That's not necessary. All you need is your nodes to connect and sync at a time before you notice a problem, or if after you notice a problem but accidentally let it connect (consider a non-technical user here).

> 2) Introduce file corruption and deletion.

If this didn't happen, you wouldn't need backups. By discussing backups, we're already assuming this might happen. Unless you consider backups to protect only against theft or fire. Accidental deletion and corruption are also major factors.

> 3) Have all your sync'd machines get the deletion.

That's what the system is designed to do, so that's a given.

> 4) Have Dropbox's file revision history go wonky on their side.

There well be one error that leads to both file revision history going wonky and the introduction of file corruption or deletion.

I'm not having a go at Dropbox, it works as expected. But backups need to be independent, not heavily integrated. Otherwise what you get is some kind of pseudo-backup that won't cut it in particular, relatively common failure modes.

It certainly isn't "one heck of a failure". It's one failure.

The clients hold copies of old revisions for three days. That's a healthy margin to find the corruption in case of the dropbox servers failing terribly.

> The neat thing about Dropbox is that your files get automagically mirrored to all your computers with Dropbox installed.

Is that correct? My understanding is that all your computers are connected to the same remote server on which your files reside, so that they all have access to the latest version. That does nothing to prevent loss of access / loss of data if the remote server goes down.

That's the beauty of Dropbox. All your computers are connected to the server, but automatically sync whenever there's a change.

What that means is, you work with the files as if they were completely local to your computer, since they are. Dropbox monitors for changes, and makes sure to update all the other copies of these files. But you always get the speed (and convenience) of working with local files.

That also means that what the parent says is right - you have extra copies of your files lying around on any computer that is connected to Dropbox. These are full physical copies of the latest version of each file.

Of course, if there is some bug that makes Dropbox think all your files were deleted or something, that change could propagate to all your other computers at once and delete all your backups. I agree that it's not very likely.

> Of course, if there is some bug that makes Dropbox think all your files were deleted or something, that change could propagate to all your other computers at once and delete all your backups. I agree that it's not very likely.

Not very likely, and if you've something like OSX's Time Machine it'd be trivial to go back.

Thanks for clarifying. That really does seem like a more robust system than what I had in mind.

Although I agree that it's not a real backup, there are people out there who have nothing else except for their laptop harddrive.

Dropbox isn't a professional backup for business critical stuff, but it's a massive step up for people who's previous backup strategy was "email myself the data every so often"

I care less about the backups because I have Backblaze covering my whole Dropbox folder anyway, but I can forget about that annoying thing called "USB sticks" - for me the real benefit is file transfer and access across devices, not backup.

I should mention that I do physical backups every few months. But like other comments here said, unless there is a server data loss, and the client decides to delete everything from your other computers, only then is everything really gone. I also have laptops who sync once every few days or so. If something ever happens, my laptop will almost certainly have an almost-up-to-date version.

That's what I've done. And it's amazing having access to all the files from my iPad, iPhone and generally, for those few times I need it, from everywhere through the website. It's made my life a little simpler, that's for sure.

Proving that dreams sometimes do come true, the Dropbox Y Combinator application from 2007: http://files.dropbox.com/u/2/app.html

Google never got around to releasing the "G Drive". None of the other competitors for Dropbox ever caught on, either.

It'd be really interesting to see how many of their users are paying subscribers. Most of the heavy Dropbox users I know do have pro accounts, and for me, it's the best spent SaaS money I've every spent!

Wow, the most impressive thing here is the exponential growth: 2 million users in late 2009, 4 million mid 2010, 25 million now.

I imagine word of mouth has been a big factor for Dropbox. Who here has not at some point sung the holy gospel of Dropbox to someone else? Especially study groups.

Amongst layman's or techies (coders, etc.) Most "normal" people don't seem to know what DropBox is outside of "introduction" via a "techie." Very anecdotal..it would be interesting to note paying vs free users, which could be (very lightly) related to layman / techie usage (VERY lightly).

Their affiliate program apparently has helped immensely with their growth. http://www.quora.com/How-effective-is-Dropboxs-referral-prog... "As of Apr 2010, the referral program with 2 sided incentive permanently increased signups by 60%

September 2008: 100,000 registered users January 2010 (15 mos later): 4,000,000 Mostly from word-of-mouth and viral: 35% of daily signups from referral program 20% from shared folders, other viral features Sustained 15-20%+ month-over-month growth since launch"

Looks like O(n^2) growth to me. If we take month 0 as being in late 2009, then it looks like at month n they have about 19x^2/162-5x/6+2 million subscribers.

This is true of any function if you evaluate it at only 3 points.

That was the point.

Sigh My point was that calling their growth exponential based on three points is silly, because with only three points you can fit a wide variety of curves.

Has dropbox taken money since the $6M in 2008?

Their financials must be incredible. If it costs $2/yr to support a free customer, they must already have 500k paying customers to cover the cost. They could just as easily have 1M paying customers and be wildly profitable.

My guess is they are keeping their financials a massive secret, because if they became public it would invite competition from copycats with a budget for TV ads.

They do have something like 40 employees, so they're spending on the order of $10M/year on salaries + benefits + office + etc.

That said, I get the impression (from how PG talks about them, the fact that they seem to have supported this massive growth in users & employees without more funding, etc) that they are doing very well, and improving quickly.

I don't know - the Backblaze guys did several blog posts on the economics of online file hosting/backups and it came out very expensive in terms of hardware. Plus the custom software and management. Dropbox accounts are generally small, so maybe that's making it easier. I'm not saying they aren't doing fine, just that online backups are low margin (Dropbox is special, they aren't regular 'online backup' providers, I realize; and they operate at a premium price point; but still...

The backup services are a different beast. Huge competition, AdWords cost is $5 per click minimum, frequently as high as $15-20. This translates into an insane customer acquisition cost, $50+ is not unheard of, meaning in turn they need people stay with them for 2-3 years just to turn some profit.

The online backup looks simple, and technically it is. It's the business side that is tough.

Plus, Dropbox isn't doing hardware (they use Amazon's). I'd imagine they are S3's largest customer.

They may be S3's biggest customer, but Netflix is easily AWS' biggest customer overall.

I don't think a Dropbox competitor that tried TV ads would do well. TV ads fall into two broad categories. First there are those that are trying to build up the brand. For instance, when Charmin runs an ad showing bears cleaning up after implicitly answering an age old question, and touts how Charmin leaves less pieces stuck to your ass than the other leading brands, they are not trying to get you to leap off the couch and rush down to buy Charmin.

Rather, they are trying to get you to remember Charmin and its superior ability to keep your ass free of toilet paper pieces when you are next in the market picking up toilet paper.

The second category of ads are those that want to get you to actually buy as soon as possible. They want you to call ("Operators are standing by!") or go to their website right then.

Let's consider the first category. Do those kind of ads work for a Dropbox competitor? I don't think so. They might make people aware of the category Dropbox is in, and some people will remember subsequently when they are online and maybe go looking for the product. Most likely they will then do a search, and if Dropbox has played its cards right, it is going to have sponsored ads in the search results, hitting on the same selling points that the competitors TV ads were hitting. Dropbox will end up capturing much of the benefits of the competitor's ads raising awareness of the category.

As far as the competitor goes, they will have spent a lot of money on TV ads, and have no clue whatsoever if the ads actually brought them any business. When we were doing TV ads for a software-based service, this was a nightmare. At one point we were doing things like calculating signal propagation from the FCC data on antenna height and power for stations that ran our ads, in order to figure out out zip codes might have seen have them, and then trying to look at statistical anomalies in purchases from those zips compared to zips in similar regions that would not have been able to see the ad, and attributing those to the effects of the ads.

Now let's consider the second category--the ads that try to get you to buy right now. If you've got a big budget, you can run ads at times that you know well in advance. You can have operators standing by. You can track call center response as the ad shows. If you are are really big, you can have a control center that looks like something out of NASA (see the first season of "Pitchmen", and watch how Billy Mays and Anthony Sullivan would monitor the results as a new infomercial rolled out). But unless you've got a really big budget, more likely you are buying your ad time via a system where it isn't determined when your ads will run until fairly late, depending on what other people are bidding for the time. It's a lot like the way search ads are sold--you and the other advertisers have a budget with an agency, and they are deciding how to divvy out their inventory of TV slots.

It's hard to have operators standing by for this, especially since a lot of these slots will be well out of prime time. So you are relying on getting people off the couch to your web site. Dream on.

Personally, I'd be happy if I saw competitors doing TV ads.

The notion that TV ads fall into two broad categories, is a false dichotomy.

There are other reasons for TV ads. One is to inform consumers about a product or service that they didn't know existed. Viagra and Swiffer come to mind. There is a difference between why Viagra bought ads (and now no longer does), and Charmin ads.

I'd be willing to bet that Dropbox runs an ad in the 2012 superbowl. It only costs $3M, and they are a new product category that has mass appeal. They need to lock down their brand as the goto place for the service. Much like swiffer is now the brand for disposable mops, dropbox need to be the default brand for "your files available on all your computers and devices".

Dropbox it's ok, but what I'd really love is a software which does something a bit different: to arrange all the unused space in the hard disks of a LAN as a big virtual folder, with redundancy and all.

Every time I think of the Outlook-and-Word users I have at the office with 500Gb hard disks...

Maybe I should send an unsolicited job application to Dropbox for a (sexy) data scientist position. 25 million users is pretty crazy.

go for it :) jobs+hn@dropbox.com

Do you even need UI designers especially that there is basically just no UI/Interaction?

have you ever seen the web interface?

Yes, but I honestly believe it suffers neglect and a lot can be done to improve it. I have never seen a posting for a UI Designer hence wondering.

Also the major USP is that Dropbox behaves just like a folder on the OS and not a seperate entity.

I was just pointing out that there definitely is a UX piece and plenty of room for designer contributions.

Not to mention the installers and the rarely-used menus and preference windows. Just because you don't see them very often doesn't mean they wouldn't benefit from thoughtful design.

If you have the chops, you probably should. I'm pretty sure that until recently they had a position listed for a data analyst type.

A hockey stick made of money! Thanks Dropbox for setting such a kickass example of a great technology product and company.

Dropbox worked great for me for quite a while. Then my Keypass password.kdb file got corrupted. This file contained all the passwords to my personal / freelance and office projects. Impossible to retrieve the data. Major PITA. So now I just use it for funny pictures and music but I don't trust it with really important things anymore.

I'm pretty sure dropbox does file versioning, so you should be able to revert to an earlier (uncorrupted) version of your file. Tell me if that works for you, b/c I use dropbox+truecrypt for hosting my super important docs...

Can't seem to find a mention of file versioning anywhere. When I login into Dropbox I find the file via search, clicking on it just downloads it.

When I try to open the file it says The following error occured while opening the database: Hash test failed. The key is wrong or the file is damaged.

I'm not the first one with this problem: http://forums.dropbox.com/topic.php?id=26321#post-163634 and http://superuser.com/questions/187885/keepass-lost-password-...

To get a previous version of a file, login to Dropbox.com, browse to the file you'd like to get a previous version of, then hover over the file and click the blue arrow on the far right that'll appear beside the file change date. This menu lets you, among other things, restore old versions of the file. Select the "Previous Versions" button to see all of the versions of the file from the past 30 days on any free or paid account, or if you have the packrat upgrade enabled, then you can see unlimited previous versions. Hope that helps!

Looks like the directions are here: http://www.dropbox.com/help/11

It does seem like it only keeps 30 days worth of versioning though...

It keeps 30 days for free accounts. That's unlimited for Pro IIRC (although you have to manage according to free space)

Edit: apparently policy changed at some point and versioning does not count in your quota. Also, Pro accounts need the Pack-Rat extra to enable unlimited versioning.

I wish I found out about this within 30 days of my file getting corrupted. Strangely enough I posted my issue on the Dropbox forums and a Dropbox admin didn't mention this, he just said it's a problem on Keepass' end.

I use dropbox on multiple computers. one of the computers is running jungledisk to backup whole dropbox content. The problem is both services work on top of amazon cloud but I will eventually resolve this issue too.

Jungledisk lets you choose where you want your data hosted. You can choose Rackspace over Amazon IIRC.

Versioning is only done on files that have changed, and a key file wouldn't change very often.

This is definitely a service that I have no problem paying for. Whenever people say to me that people will not pay for online services and you have to use ads, I use Dropbox and Netflix as an example now that Netflix's streaming service has really taken over.

Dropbox - the result of building something people actually want.

I am massive fan of Dropbox and like to see it doing well. I have talked about its virtues on many Quora posts. I have ussually recommended using a Truecrypt volume within the Dropbox folder to guarantee encryption.

Now reading findings like this and the authentication issues exposed in the last few weeks I'm getting quite worried about the security of the whole solution. Going to have to change a lot of my posts to say do not store anything sensitive without Truecrypt: http://paranoia.dubfire.net/2011/04/how-dropbox-sacrifices-u...

Dropbox had helped me accomplished the impossible.

Convincing a girly teenage girl to have a habit in backing up her computer regularly.

No other service could do that, not even Time Machine.

Amazing... One thing I would like to point that even though people claim that systems such as Dropbox are "simple" it is really not so simple to implement. Just ensuring that Windows piece does not mess up your computer (eat all your CPU, bandwidth etc.) is really hard. There are a lot of not-so-sexy work behind the curtains.

BTW, we just started private beta for cloudHQ for Dropbox (http://www.cloudHQ.net/dropbox). Basically this service provides synchronization of Google Docs and Dropbox files and backup of Google Docs to Dropbox. And we have a cool thing that you can edit Dropbox files directly with Google docs (you need to install our Google Chrome extension). We need some beta testers ....

im wondering how many of these users are paying customers, since subscription is dropbox's only source of income as far as i know.

Awesome numbers. I wish I could participate in design/implementation of a system like that.

Anyone know if that is 25 million active users or 25 million registrations?

Registrations, of course.

Reply on news.ycombinator.com