Not entirely fair, there are whole communities around storing 100s of TB on amazon
Customers are customers, not product managers. It is only natural to make use of a service you pay for.
Its understandable from their point of view to offer unlimited and be awesome but not expect this kind of usage that is not sustainable. So they made a mistake and are correcting it.
It's hard to see it as a deliberate strategy to pull in users and then charge them more when they are "locked in"
Do they also refund people for their time wasted assuming this was a sustainable service, or does this "correction" only work in one direction?
At the same time, I don't get, why would you encrypt your "Linux ISO's"? Let the AWS dedup do its job, don't abuse it, and everyone is happy.
> At the same time, I don't get, why would you encrypt your "Linux ISO's"? Let the AWS dedup do its job, don't abuse it, and everyone is happy.
Because if you are a self-proclaimed data hoarder, do you have the time to sort through and selectively classify your hoard to "encrypt this ISO don't encrypt that tarball" on a file-by-file basis across many terabytes?
How much would be saved by deduping anyway? If they're not deliberately making it easy/redundant, even if you got 300TB down to 100TB or such, a single order-of-magnitude reduction doesn't fundamentally change the economics of "unlimited."
Blame data hoarders, but don't blame encryption.
I only have my pictures and personal data in AWS cloud, encrypted. They way I set it up? Point rclone to relevant directories and skip the rest.
Any recommendations on the "plenty of apps" that sort your data for easy searching?
It seems that as of a few months ago, two popular (unofficial) command line clients for ACD (Amazon Cloud drive) were acd-cli[1] and rclone[2], both of which are open source. Importantly the ACD API is OAuth based, and these two programs took different approaches to managing their OAuth app credentials. acd-cli's author provided an app on GCE that managed the app credentials and performed the auth. rclone on the other hand embedded the credentials into their source, and did the oauth dance through a local server.
On April 15th someone reported an issue on acd-cli titled "Not my file"[3] in a user alleged that they had received someone else's file from using the tool. The author refered them to amazon support. The issue was updated again on May 13th with another user that had the same problem - this time with better documentation. The user reached out to security@amazon.com to report the issue.
Amazon's security team determined that their system was not at fault, but pointed out a race condition in the source for the acd-cli auth server (sharing the auth state in a global variable between requests...) and disabled the acd-cli app access to protect customers.[4]
In response to this banning, one user suggested that a workaround to get acd-cli working again would be to use the developer option for local oauth dance, and use rclone's credentials (from the public rclone source).[5] This got rclone's credentials banned as well,[6] presumably when the amazon team noticed that they were publicly available.
To top this all off, the ACD team also closed down API registration for new apps around this time (which seems to have already been a strenuous process). I suppose the moral of the story is that OAuth is hard.
[1]: https://github.com/yadayada/acd_cli [2]: https://github.com/ncw/rclone [3]: https://github.com/yadayada/acd_cli/issues/549 [4]: https://github.com/yadayada/acd_cli/pull/562#issuecomment-30... [5]: https://github.com/yadayada/acd_cli/pull/562#issuecomment-30... [6]: https://forum.rclone.org/t/rclone-has-been-banned-from-amazo...
The whole point of an unlimited tier is to attract large numbers of outsiders who don't want the cognitive burden of figuring out $/GB/month and estimating how many GB photos they'll need to store.
What we're talking about here is that they got some customers like that, but they also got a small number of customers taking them for a ride, call them 'power users' the kind of customers who (as we see elsewhere in these comments) won't stick around if the price changes.
There's nothing wrong with these power users storing huge amounts of data at subsidised price, just like there's nothing wrong with Amazon changing the pricing. They just decided to stop subsidising that behaviour and probably take a slight hit on a conversion rate somewhere.
As for your question about 'private' storage, it's a grey area. Privacy isn't absolute, especially in cases where a company is by inaction helping you breaking the law (whether you agree with the law or not). Companies work very hard to distance themselves from responsibility for their customers actions and don't want to jeopardise that by letting it get out of hand
How does this work with Google Play Music (you can upload up to 50k songs for free and listen to it "on the cloud")?
I think you are focusing on the wrong thing. Corporations don't care about the law any more than individuals do. Laws and regulations are just guidelines if you are determined enough to get your way. Look at all the Uber stories. Pretty sure people here still like Travis for his tenacity no matter what you say about his morality.
I think we often forget that humans wrote the laws we have today. They didn't come to us in stone tablets down the mountain top. At the end of the day, these laws don't matter. They are not written in stone so as to speak. We should always strive to do better. Intellectual property is a sham. I mean think about it. I think there is legitimate intellectual property, the trademark.
I think it is wrong for me to sell "Microsoft Windows" (even if I wasn't charging any money) if I had modified the software and added malware into it. But me watching a movie or reading a book without paying royalties does not hurt anyone.
Please think about it. Just because something is legal does not make it right and just because something is illegal does not make it wrong. We need to calibrate our laws based on our image and not the other way round. We write the laws. The laws don't write us.
I'm struggling to find a connection between the points that I made in my comment and the points in your reply. Suspect we have some miscommunication here... my own comment wasn't spectacularly well filtered.
I'll bite on these though;
> Laws and regulations are just guidelines if you are determined enough to get your way. Look at all the Uber stories.
Don't conflate civil or criminal law with the work of regulatory bodies, who in my experience with the FCA and OFT are very open and collaborative without any need for "tenacity".
Uber work very hard on marketing and competition, but they are allowed to succeed to regulators who WANT them to succeed despite their amoral hussle, not because of it. Regulators in my experience (the FCA and OFT specifically) are very open and collaborative. They understand that markets move on and regulations sometimes lead and sometimes follow.
> Please think about it. Just because something is legal does not make it right...
So, I'm assuming from this comment that you're quite young. Just for you information; I suspect most folks on HN are already aware of the delta between legality and morality.
I'd also recommend thinking about the subjective nature of morality, and the causes and malleable nature of it.