Drew Houston's (Dropbox) YC Application | Hacker News Reader

Drew Houston's (Dropbox) YC Application

files.dropbox.com

152 pointsgiu26 comments

I liked this part >>

# Why would your project be hard for someone else to duplicate? This idea requires executing well in several somewhat orthogonal directions, and missteps in any torpedo the entire product.

For example, there's an academic/theoretical component: designing the protocol and app to behave consistently/recoverably when any power or ethernet cord in the chain could pop out at any time. There's a gross Win32 integration piece (ditto for a Mac port). There's a mostly Linux/Unix-oriented operations/sysadmin and scalability piece. Then there's the web design and UX piece to make things simple and sexy. Most of these hats are pretty different, and if executing in all these directions was easy, a good product/service would already exist.

With Google's GDrive clearly listed as a potential competitor, which had to be mentioned several times in the application, I didn't find the application as "strong" as I would have expected.

Of course, I love + use DropBox, and GDrive has never really happened (not in the dropbox-like incarnation people expected) but at the time the application was written it seemed as though things were very stacked against DB.

I always think that for a space like backup and sharing of data/files a startup always had much chance of success than GDrive or other biggies.

Reasons -

1. Its important to iterate extremely fast and keep on improving your solution which a big company like Google cant really do any more.

2. This space does not require integration with other consumer apps offered by big companies like emails/calendars/social-network. Therefore there isnt much to win for big companies to win by focusing enough on competing in the space. Its more of a distraction really.

3. Backups of data is very important feature for which people are willing to pay. Hence consumer is also very demanding. Must easier for a startup to keep on listening to consumer than a big corporate.

Another similar startup which is doing extremely well in similar space compared to biggies is http://www.druva.com/

There was a version of this with PG's comment that was even more interesting to read. Edit: found here http://news.ycombinator.com/item?id=801503

The URL of this submission is the same as the one of the submission you linked; it's strange that my submission made it through the dupe filter.

Thanks for the link, crocowhile; pg's comment there is really that more interesting to read

I think the dupe filter only checks for stories within X number of days. This was past the limit => new thread.

The URL is not the same. In the other submission Dropbox was still called getdrobox.com

Source of this submission is pg's following comment: http://news.ycombinator.com/item?id=1938348

To quote it: If there's one thing applicants don't get, it's the value of being concise in the answers. Not brief, concise.

> What are people forced to do now because what you plan to make doesn't exist yet?

> Email themselves attachments.

For me, that was the killer feature that got me hooked 2+ years ago.

Today we use Dropbox for so much more. It's become the ultimate networked drive that takes advantage of your local hard disk.

The one feature I wish it had was encryption.

I understand the wish for encryption, but once you've put your encrypted dropbox in the cloud, it's there forever. If anyone is going to break the encryption it's going to be:

1) Someone with massive computing power (e.g. cloud based)

2) Someone with a long time to work on it

Who's to say DropBox wouldn't setup background processes to brute force encryption keys? Who's to say your encryption keys might not be trivial to brute force 10 years from now?

And if you use password protected documents you can't change your encryption key because they can always look at a previous version where the key was different.

More to the point, if you don't trust them to store your data, why would you trust their encryption? If they were malicious (and I'm not claiming anything) they would deliberately weaken the encryption 'accidentally' and you would likely never know.

If the dropbox client encrypted all files using a shared secret known only to myself, then my files would be protected by an additional layer of security where I manage the keys, that makes a lot of difference.

Obviously there is no such thing as absolute security, but that's not a reason not to take protective measures.

How would the dropbox client encrypt the files if only you knew the shared secret? You have to give the client your shared secret, and then we're back to where we began.

No because the client is on my machine. See how truecrypt works if you want to understand this.

In the real world, DB's servers are more likely to be breached than file-level encryption, by orders of magnitude. So encryption would be a very welcome addition.

Unfortunately it has a very real cost for DB: data redundancy could no longer be leveraged to save storage (because 2 identical files would appear different to DB). I suspect at their level of scale that would move a few cells in their CFO's spreadsheet.

You can create a TrueCrypt volume or Mac encrypted disk image in your Dropbox folder: http://wiki.dropbox.com/TipsAndTricks/IncreasePrivacyAndSafe...

This reduces the usability of syncing and sharing, which are Dropbox's biggest selling points.

Tarsnap is an alternative if you just want encrypted backup: http://www.tarsnap.com/

Sure, I actually use both Truecrypt and Tarsnap for other things. They are both excellent tools, but Dropbox with client-side encryption they are not.

It's awesome how honest he was in who his competitors were and how similar they were to his product (including the fear that Google would eventually overtake) and yet all through the application he seemed very confident that his product was going to succeed and surpass what the others attempted to do.

I remember showing my friends the first presentation video when they started giving out private Beta invites. I was so stoked to have something like it that my friends could also use with me. I'm very glad everything worked out for him and the Dropbox crew!

OT: The Windows machine I'm using at the moment has the/a Norton... security suite installed (firewall, anti-virus, et al.).

Upon navigating to the linked page, a pop-up appeared warning that the site is "unsafe".

Which causes me some concern that the "mainstream" campaign against Dropbox and others may already be gearing up.

I wish my employer would get Dropbox. Not being able to host it in-house is probably a dealbreaker though.

Providing ideas which could be patented on the application form seems quite foolhardy (for the applicant) .. is that a trick question?

EDIT: Actually, I suppose it serves two goals -> could show potential ingenuity, but also reveals how 'trusting/guarded' the applicant is.

In the US at least, you can still patent an idea for some amount of time after it's been made public.

My favorite part:

# How long will it take before you have a prototype? A beta? A version you can charge for? Prototype - done in Feb. Version I can charge for: 8 weeks maybe? (ed: hahaha)

Just another example of how optimistic developers can be (I know I'm guilty of the same).

For those of us who haven't been tracking Dropbox closely, how long did it actually take?

"How long have the founders known one another and how did you meet?"

I wonder how many single founders make a terrible joke answer to that question and if some sort of filtering could be applied. Drew handled it well, I would have just put n/a

seems like 1 million is the number everyone goes with...so if someone sees a very promising startup at demo day, would be a good number to offer.

But with that 1 million acquirers will want the team/product for X years to ensure they get something of value (the product creation ability / pitching skill of the founders).

Reply on news.ycombinator.com