Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: If I build it, will you pay? (Open photo sharing service on S3)
8 points by jmathai on Feb 6, 2011 | hide | past | favorite | 30 comments
This idea has been stirring in my head for several months but today in the car I realized the vision in full.

TLDR; If I build a photo sharing service based on Amazon S3 (and CloudFront) would you pay to use it? You'd upload photos through the service into your own Amazon S3 bucket - meaning you have 100% control over the photos. Low resolution copies (for sharing) will be stored on your CloudFront distribution.


The details

The state of photo sharing is absolutely horrible. Most people upload photos to Facebook and I won't even go into the 1001 reasons why their photo service is abysmal. Some use Picassa or Flickr, while better than Facebook they still control the destiny of your photos. Flickr's got APIs - so that's a start but it's not trivial to get all your photos and their meta data (comments, tags, etc).

What I'm envisioning is a frontend for YOUR photo collection that resides in an Amazon S3 bucket you own. A good interface to upload photos, share them, and control privacy. The whole point of this is to be open and let you retain ownership of your photos.

Lots of potential features on the roadmap but what I described is phase I.

My guess is the cost would be in the $1-2 / month range + any S3 costs you'd incur yourself. I'm hosting about 18 GB of photos and it costs me around $2 / month.


The team (me)

Who the heck is this dude and what are his qualifications? My website is at http://www.jaisenmathai.com. I was a co-founder of a photo sharing startup back in 2004. We shut down shop in 2007 but you can see it here: http://photos.jaisenmathai.com/users/jmathai/photos/tags-tavin,favorites/.

I'm pretty positive I can build the first version of this out in a couple weeks (with a basic UI).<p>If you're a graphic artist and want to provide ongoing design services then you can have an account for free :).

Wouldn't pay. I don't feel the pain from any of these photo sharing service. The storage is not the issue, I think its distribution. Imgur, flickr, picassa, facebook.. all work fine in my opinion. The magic of facebook is that I can upload and have to do nothing for distribution because all my friends and family can already see it.

If you are speaking of a tool for developers then I would like to know more details, but yes I have paid for image services that will resize and dump them into my bucket.

Sounds like you might not be the target. If you're not worried about the durability or long term management of your photos (and owning them as much as you can) - then Facebook is perfect with it's distribution.

The idea of making it available for developers is an interesting one. I hadn't considered it but if there's enough interest to build it out for some consumers then making it work for developers shouldn't be a stretch.

I would definitely pay for a service like this. It would be even more convenient if you handled billing for S3 for me while me still retaining ownership. I would pay a premium on top of S3's services. It would also be neat if you could frame this as a Mozy or Backblaze for photos only... Emphasize storage rather than community like Flickr and Facebook do.

I don't know if I'm your target demographic though:

> Hobbyist photographer

> Flickr Pro account user

> 29 years old

> Upload ~15-20 photos a month mostly because I use flickr as a showcase rather than storage

> Shoot in RAW

EDIT: Also... Please do not use a dark background. Or, if you do, please give users the option to customize the theme. Dark backgrounds are extremely irritating to me.

Yea, a Mozy for photos is a decent way of looking at it.

I haven't even got to thinking of how the site would look but I presume it'd be theme-able and let you provide your own CSS if you wanted.

I really don't want to dictate anything about your photos. I want to provide the utility of uploading/resizing etc.

I do this for myself and after 6 years and a few thousand photos I realize how valuable it is. I also know that the chances that my kids will be able to see their photos is orders of magnitude higher than if I were uploading them to Facebook.

Just want to spread the love :).

Short: I would pay.

Long: I would love to see some features that other sites haven't put in yet.

1. The ability to upload from a computer or via e-mail (from a cell phone, for example).

2. The ability to automatically repost pictures to specified Facebook galleries and Twitter.

3. Robust tagging/sorting/searching systems.

?. An offline system with a searchable database that you could synch with the cloud system would be amazing, so you could organize photos on your computer in the same way and keep everything synched up.

Edited for clarity.

Another thing that would bring over a lot of people I know is the ability to automatically import flickr/picasa photos and tags. Most people won't move from established galleries because it's such a pain to relocate all of that history -- they're locked in.

Thanks for the feedback. All of those are great features. #3 is my favorite though :). Ideally it would be open enough that adding "modules" would be easy. Not exactly sure what that means, perhaps an API with OAuth so others can build apps. Really interesting...but definitely phase II.

I feel bad saying this, but I'd also want some kind of guarantee or something that the service wouldn't shut down in a few years, since you mention that you did this before and shut it down.

Well, that's part of the beauty of this. I can't guarantee the service wouldn't be shut down but you're in the same boat with any other provider you go with.

The essence of this service is that even if I went AWOL you'd still have your photos and meta data in an easily accessible form (your S3 account and a SQL dump of the meta data).

I'd love to open source the "front end" code so you could port it simply by running it on your own server. The actual photos would never have to be migrated since....tada - you own it and it's not bound to the service I'm offering to provide.

It sounds like you're trying to sell both the service AND the software. Why not just offer to sell the software to anyone who wants it for X dollars, and for the people who don't want to host it themselves, run it for them for Y/mo?

I hadn't really given it too much thought. The intention wasn't to make money off of it. Some money to cover expenses but I'd love to not be the one managing it, having nodes of these sites all around the web.

But that's lofty - so a hosted solution is a bit easier to swallow - and I'd be happy doing that as well and breaking even (time/energy/stress, etc).

You need to eat the S3/CloudFront costs and bake it into the price. Almost nobody has an AWS account. The people that do are either using it for work/startup stuff, or are like me and get annoying $0.12 charge every month.

I wouldn't have to bake that into the price. If you signed up you would have your own AWS account and get billed for storage directly.

You don't get it. Getting billed for storage directly is not a feature 99% of the people on the Internet want.

Make it simple for your users and don't get cute with pricing models.

I do understand what you meant but wanted to clarify a few things.

1) This is really for 1% of people on the Internet (possibly less). Facebook is great for the majority but not for people who want to retain ownership and original versions of all their photos that they are confident they can get anytime they want.

2) I agree about making it simple. But I am not really interested in making it simple at the cost of the main vision of this idea - letting users "own" the photo (meaning it resides in their own amazon bucket). Unless there's a way to use FPS to do this I probably wouldn't consider "simplifying" it the way you mentioned.

This is more of a pro bono idea than one I wanted to or intended to make a lot of money on. I already have those ideas I'm working on :).

Idea is interesting. I don't think I would pay for the service, granted I'm not your target audience (I don't use flickr, photobucket, etc.).

Also, I'd go with $12-$24 a year instead of $1-$2 a month. Just looks nicer, and converts a lot better.

Unless he went with $1.20 a month and pointed out that it's only 10 cents a month (echoes of Sprint's old 10 cents a minute ad). Then not only would he have a robust feature set to compete, but also a pricing advantage over Flickr.

I wasn't (yet) considering the "marketing" of the service. But at $1-2/month I was definitely intending to bill yearly.

This sounds similar to what Jungle Disk offers, but focused on pictures. The way they approach it is that they give users 5GB per month for a $2 per month fee for storage. On average, this should cover the AWS costs and leave some margin for Jungle Disks. I think any additional storage charges after that are billed at the rate of AWS or Rackspace since they are compatible with both (I believe they are owned by Rackspace). I think this approach makes it less intimidating to consumers who have no idea what AWS is.

There are plenty of services like that - which I'm not terribly interested in copying or spinning off as a "photo" version.

I do this myself for my own photos and thought others would enjoy and pay for a service that did the say. I love that my photos are backed up on S3 and it's not someone else's Amazon account. I own it and can do whatever I want with my photos.

I agree that it's a bit intimidating. My original idea was to see how many technical people were interested - hence posting the question on HN :).

I wouldn't pay. I don't even know 1 of the 1001 reasons you think Facebook's photo service is abysmal.

Who is the target user for your service?

Here's a good answer by someone else :). http://news.ycombinator.com/item?id=2184735

As far as target, I'd presume people who want to keep control of their photos and keep them for the long term (10+ years). Who knows what's going to happen with all those Facebook photos 10 years from now.

Can you list a couple of the worst parts about Facebook's photo service?

I'm not a Facebook user, but I do know that photos are one of the main reasons why people use (love, are addicted to, etc) it.

Maybe even compare your improved solutions to the poor features of Facebook's service. Thanks, and good luck!

It's vertically integrated to be social. As such, the following are not considered important:

> Original photo size (additionally, optional sizes)

> EXIF data

> Compression

> Sharing as a link, embed, BBCode, etc. -- anything outside of Facebook can be done by copying image URL, but that URL can change without a moment's notice.

It's designed for people who take snapshots, not for photographers or artists.

Yea, that's a pretty good synopsis :).

Does Dropbox have an API that gives read/write access? Being able to drop photos into my dropbox and having them resized and uploaded automatically would be killer. Also much more user friendly than AWS for non geeks.

Unsure if Dropbox has an API to do that. The original target was geeks which is why I asked on HN.

Dropping photos into a directory and having those resized and uploaded is just a script that looks for photos in a given directory and posts them to any of the photo sharing sites with an API, no? But then you don't really own the storage of your photos (or at least aren't billed for it - which implies some level of ownership).

I would probably pay. My dream solution would back up my pictures and the let me share the ones I wanted to share. (Not a photo hobbyist, just like to take lots of pictures.)

That's precisely what this service would solve. The photos are backed up on S3. You don't really ever have to worry about switching services or anything since it's bound to your Amazon account and your S3 bucket.

can't you share photos on dropbox? and you would still have complete access to your original files.

Probably, but you don't really have a great interface for photos. Dropbox is for files and works for photos but photos have such different properties than most files.

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