Given over 6 million people are still without power right now, and may be without power still in a week, that might be more of a concern than it'd otherwise be.
Also classy is the "It's been real" signoff at the end. Kind of a stain on the startup team to end so abruptly and with so little respect for their users.
I'm not sure this will be a popular sentiment here, but I think we might want some sort of consumer protection laws dealing with things like this. Have a sort of mandatory cooldown period for (paid?) web services after a notification of cessation (or maybe a major restructuring?) of services. Somewhat like an eviction notice.
Of course people will argue that regulatory action is unnecessary: if sudden terminations are an issue, people will naturally flock to services which guarantee due notice through terms of service.
I'm so liberal I'm practically a socialist, but I don't see how you could have this sort of regulation without it having a chilling effect on new services. Generally when an event like this happens, the company behind the service is dissolved, so who is even on the hook for keeping the data available?
This is a very real practical problem. I basically never use any SaaS app that isn't at least as established as, say, Basecamp and if enough people think like me you've got a horrible chicken and egg problem. I'm not sure how you fix it, but I don't think it is via laws.
Caveat emptor. Also never trust that a single copy of anything won't vanish like a fart in the wind one day. If stuff is important to you, keep your own copies.
The reason why we choose a small number of days to download your content is simply because when you used the app, the photos remained on your phone.
We're monitoring who's downloading their content and who's not. If we see that there are a large number of users who still haven't downloaded their content, we'll adjust that date.
When you go out of business, save each user's data separately and send them an access URL. The user can pay Amazon to retrieve it if they want it back.
It is infinitely better than your data just suddenly being deleted before you had a chance to get it. At least you get to decide if retrieving it is worth the cost. Also, you can "pay" for Glacier using time/patience because you can download 1gb/mo for free.
All that said I don't think Glacier is set up to deal with a situation where one party inserts a ton of data owned by many other parties, so this is all very theoretical, but IMO it would be great to have something like this.
If my data isn't worth $0.12/GB to me I can't see being too worked up either way.
After reading the pricing and FAQ a bit more, it looks like there will have to be a 3rd party in the middle. There doesn't seem to be a "reader pays" option in Glacier.
There. I registered data-morgue.com. Now, if I can just get a free week…
I would disagree. What reasons are needed? It's a startup: we know they ran out of money. The rest is probably personal or not worth delving into. Apologetic tone? Save your PR spin. Eight days to migrate? Well, the users probably all have copies of the photos anyway. I liked the note. Short and to the point.
Agreed, though I'd add one other thing: one could argue the issue with Batch failing was the idea itself, or the execution. Perhaps the crappy "shutting down" message is a symptom of poor execution/management.
They were acquihired and will be working on other things, probably worth including that. Maybe something about the product, why it failed to catch on and what they think users should use in its place.
Will not surprise me at all that many users will miss this announcement. Turn off new accounts today, but give them 6 months and remind those that haven't signed in monthly.
And when Batch was acq-hired by Airbnb in July they said (TechCrunch):
"An Airbnb spokesperson told me that if plans change in the future, Dailybooth and Batch users will be given ample notice before those services are closed."
I wish there were a service startups could use when they're setting up new products which will provide a 60-90-180-etc. day transfer period for users if the startup ever shuts down, runs out of money, etc. Maybe integrated with more mundane disaster preparedness for the startup, too -- a datacenter ready for them if theirs is flooded, etc.
It would almost make sense for AWS or someone else already providing the infrastructure to do this for a few percent premium on every bill.
I like the concept but it wouldn't work for the very reason people don't save for a nice funeral.
Source code in escrow works because a large customer builds it into the contract; meaning you only get paid if you agree to the terms.
Consumer startups work by removing payment friction for their users, which inverts the dynamic. Plus, most consumers aren't as tech-savvy as HN so it wouldn't enter into consideration when they use an application.
Yeah, it would have to be somewhat automatic and widespread, sort of like credit card warranty/travel/etc. insurance.
It might make sense for b2b apps instead of consumer. b2b apps still aren't usually sophisticated enough to do source code escrow (and what does that really mean for SaaS apps), but a business would still be hurt if someone like Salesforce disappeared (or even a smaller product like a CRM gmail extension or whatever, if your workflow is built around it.)
Also, developers are customers of sites with APIs, so some kind of promise that an API will remain available, non-throttled, etc. would be good incentive to adopt it.
The app filled a niche that no one really needed. It was well designed and worked well, the winners in the photos space are providing more than a sharing mechanism though. They are either attaching it to the rest of your social media (Facebook) or providing new ways to take pictures (Instagram) which even there has beat hipstamatic because they have built social around that and developed their own network culture.
It also looks like a considerable number of people on HN— myself included— had never heard of it. I mean, it appears the link to their site had never been submitted before.
I don't use a lot of new consumer web apps for this reason (I did use Batch just to test, but didn't start using it heavily because I was afraid of something like this happening).
One way of reassuring me in using a product like this would be to have something in the app's settings menu that allows me to point to a different production server, and a commitment in the legal Terms of Service that in the event that the product is shut down, that the server-side code will be open-sourced.
53 comments
[ 4.3 ms ] story [ 119 ms ] threadhttp://uncrunched.com/2011/10/27/batch-may-be-the-perfect-mo...
Hint to startups: don't try to solve problems that are already solved.
Of course people will argue that regulatory action is unnecessary: if sudden terminations are an issue, people will naturally flock to services which guarantee due notice through terms of service.
I would love to see someone make that argument and try to keep a straight face at the same time.
This is a very real practical problem. I basically never use any SaaS app that isn't at least as established as, say, Basecamp and if enough people think like me you've got a horrible chicken and egg problem. I'm not sure how you fix it, but I don't think it is via laws.
Disclaimer: I'm the lead developer on OpenPhoto which does just this.
http://theopenphotoproject.org
We're monitoring who's downloading their content and who's not. If we see that there are a large number of users who still haven't downloaded their content, we'll adjust that date.
When you go out of business, save each user's data separately and send them an access URL. The user can pay Amazon to retrieve it if they want it back.
All that said I don't think Glacier is set up to deal with a situation where one party inserts a ton of data owned by many other parties, so this is all very theoretical, but IMO it would be great to have something like this.
After reading the pricing and FAQ a bit more, it looks like there will have to be a 3rd party in the middle. There doesn't seem to be a "reader pays" option in Glacier.
There. I registered data-morgue.com. Now, if I can just get a free week…
Joke.
Anyway, you would do anything with it. It's generic enough.
* no reasons/details
* no apologetic tone
* short timeframe to move your data
Suck it users!
Although it's not a very good page for their users, either.
In a situation like this, if you can't say anything professional, then don't say anything at all.
Will not surprise me at all that many users will miss this announcement. Turn off new accounts today, but give them 6 months and remind those that haven't signed in monthly.
"An Airbnb spokesperson told me that if plans change in the future, Dailybooth and Batch users will be given ample notice before those services are closed."
Hilarious.
It would almost make sense for AWS or someone else already providing the infrastructure to do this for a few percent premium on every bill.
In the enterprise space, a lot of software contracts include source code escrow, etc.
Source code in escrow works because a large customer builds it into the contract; meaning you only get paid if you agree to the terms.
Consumer startups work by removing payment friction for their users, which inverts the dynamic. Plus, most consumers aren't as tech-savvy as HN so it wouldn't enter into consideration when they use an application.
It might make sense for b2b apps instead of consumer. b2b apps still aren't usually sophisticated enough to do source code escrow (and what does that really mean for SaaS apps), but a business would still be hurt if someone like Salesforce disappeared (or even a smaller product like a CRM gmail extension or whatever, if your workflow is built around it.)
Also, developers are customers of sites with APIs, so some kind of promise that an API will remain available, non-throttled, etc. would be good incentive to adopt it.
https://twitter.com/paulg/status/130875076110323712
One way of reassuring me in using a product like this would be to have something in the app's settings menu that allows me to point to a different production server, and a commitment in the legal Terms of Service that in the event that the product is shut down, that the server-side code will be open-sourced.