[innonate]

Nice post :) Yeah, I think the biggest misconception about the photo space, seeing how many of us cropped up at the same time, is that you can spend enough time on product rather than the hard tech stuff to make a dent in the world. We're 3 years in and just now starting to release the product stuff we're excited about because the tech challenges are so real.

Anyway, you all did an amazing job and I'm glad you shared this post. Never saw it back when you first posted it.

[Veratyr]

Thanks for doing Picturelife! Saw it in one of the comment threads here and have been using it since.

Love the custom S3 backing especially and the speed you guys move at (Lightroom support and Android updates came out so fast) as well as your amazing support (don't know how Amy puts up with me)!

Just recommended you to a friend in fact :)

The only thing I'd change at the moment would be to make a Cloud backend API so that customers can plug in Dropbox, Google Drive, Box.net etc. with something simple like a URL, separating storage from display etc.

Oh and it'd nice to have the option to pay you $2 or $3 since I'm using S3 :)

[jmathai]

  The only thing I'd change at the moment would be to make 
  a Cloud backend API so that customers can plug in 
  Dropbox, Google Drive, Box.net etc. with something simple 
  like a URL, separating storage from display etc.

For what it's worth, that's specifically what our differentiation was with OpenPhoto/Trovebox (we support about 7 storage services along with being open source[1]). We've moved on from the consumer space because there didn't seem to be a large enough market for that specifically; separating data storage from application logic.

You can see our Kickstarter from 3 years back[2].

I asked in another post how offering custom S3 buckets was working out for PictureLife. Curious if someone else found a market for that.

[1] https://github.com/photo/frontend

[2] https://www.kickstarter.com/projects/jmathai/openphoto-a-pho...

[innonate]

Thanks for the feedback. Firstly, Amy is very awesome indeed.

And as for the recommendation accept payment for the S3 service, it's something we've considered – likely a choose your own dollar amount thing. In the meantime, feel free to send any money you want to [e-mail] :)

As for using other storage options – at this point we don't feel like we can provide as high quality a service using other storage systems. S3 works well because our architecture is oriented around it, and housed within the same network, allowing us to process, serve, and analyze photos efficiently. Using Dropbox, for instance, would introduce huge amounts of latency and instability in this process.

[jmathai]

I've got a PictureLife account. I should get around to checking it out again.

To be honest, the same reason I started OpenPhoto/Trovebox is the same reason I don't trust a 3rd party service with my data. I noticed you guys supported personal S3 buckets. I'm curious how that worked out for you as that was one of our major differentiators (we supported upwards of 7 third party storage services, incl. storage at University of Southern California). We didn't find a market for it in the consumer space.

If you're in the bay area I'd be more than happy to grab a coffee.

[innonate]

Would love to hang next time I come out to SF area.

Custom S3 buckets are great for us because they don't cost us a lot to support, not many people ever want them (relative to the general public), but it allows us to do something awesome for the people who do want them.

I'm curious about your experience about supporting other storage options as well – we only do S3 because we know we can give a really high quality experience with them and it fits in nicely with all our processing steps. Complexity aside, I'd be concerned that something like Dropbox, for instance, woudl have high latency with their API and then make Picturelife seem slow.

[jmathai]

The way we dealt with that was for storage systems like Dropbox we store originals there but cache all of the thumbnails in an S3 bucket. It also didn't make sense to store thumbs in a user's Dropbox account.

All in all we treated every storage system independently but it was easy for us because that was the design from the beginning.

Even in our case the vast majority of consumers use storage we provided. We even support migration between storage services and that got some good use but nothing of any significant scale.

Drop me a note when you're in the Bay Area. @jmathai on Twitter or contact info in profile.