http://www.amazon.com/gp/help/customer/display.html/?nodeId=...
"We also reserve the right to suspend or end the Services at any time at our discretion and without notice. "
And Google's:
> We are constantly changing and improving our Services. We may add or remove functionalities or features, and we may suspend or stop a Service altogether... Google may also stop providing Services to you, or add or create new limits to our Services at any time.
http://www.google.com/policies/terms/ (that's where a link to Terms of Service from Google Drive page at takes https://support.google.com/drive/answer/2450387?hl=en you, those general Google ToS)
For better or for worse, nearly every ToS you will see anywhere includes a provision like this.
Nothing lives forever. No company lives forever. Amazon, Microsoft, Google...may all go away some day.
And, of course, they may change their business focus and decide to go a different way. I would expect any of them to provide notice and let people get their data out...but one should always assume that a service will go away at some point.
You're talking about 1, and sure, there's not a lot you can do.
As far as 2 and 3, back in the day, when you paid for a service you might sign a contract, and they could not simply decide to cancel your service in the middle of the contract because they didn't like you, for reasons not covered by the contract.
Of course, that also assumes that they wouldn't just put "and by the way we can cancel your service at any time for any reason including just because we don't like you" in the contract, or that you would have some kind of negotiating power to demand different terms or go with a competitor with better terms. But obviously they _would_ just put that in the contract too, as they did in the ToS which is theoretically a contract too.
And there's really another thing too specific to the Amazon terms posted above. They are advertising the service as 'unlimited', but then the terms say they can shut you off if your use "substantially exceeds or differs from normal use by other users." That doesn't really sound "unlimited" exactly, does it? I don't know if putting "Ha ha not really unlimited" in the ToS would protect them from such things as this: https://www.ftc.gov/news-events/press-releases/2014/10/ftc-s...
[1] Though personally I would disagree.
IMO, I think local hosting is the primary usecase the Tor project should be advertizing.
It introduces another problem, managing the replication and testing to make sure it's working, but if the data you're trying to protect is important enough, you devote the resources.
Looked at in this light, hosting your own physical box is less useful than using two separate online services. You still have the management and testing overhead either way, but with your own physical box you also have to maintain and manage that resource too. That only makes sense if you have existing infrastructure you can plug into, and sometimes it doesn't even make sense then.
I don't think a clear cut winner emerges in all cases. Cloud vs. DIY is situationally dependent and requires a great deal of deliberation. I think the pendulum will swing between cloud and DIY for a long time... With proponents of each making awesome cases to use either one.
The cost of using multiple solutions should be negligible compared to the cost of losing the information. If it isn't, then store it wherever, and if you lose it, so what.
A competent developer should not have to spend a lot of time learning an api, automating the process or testing the automation. If you don't have a competent developer, you should just use COTS. If you have a developer that complains about the time these tasks take, you don't have a developer, you have a technical handyman. In which case you can't engineer anything and should just use COTS, because that's all he'll realistically be able to handle anyway. If the developer is you and you can't afford to waste your time engineering your infrastructure, then yours is not a technology company and again should just use COTS.
Your operating profits should support the cost of your engineering, including the salary of a competent developer, this will typically dwarf your hosting costs. If they don't then you don't have a real business and need to spend more time figuring out how you're going to make money and less time on the technology.
If you're storing and using big data, and currently using a cloud provider, then your roadmap should include a plan for eventual self-hosting, as that's one of the few areas where self-hosting still makes sense, as costs can diverge very quickly. It's not big data unless building your own Backblaze storage pod is a viable option.
For all other applications, self-hosting can quickly become a boondoggle, unless you have competent systems administration, the cost of which will again dwarf your hosting costs. If you do not have competent systems administration, and you are owning and managing systems, then your business is a disaster waiting to happen. If the hard drives fill up on your home-built server, you will have downtime until you can figure it out and fix it. You will not have your hosting company's skilled customer support team, which handles the common cases that trip people up all the time, at your disposal.
Any time you touch the machine do do anything other than deployments, you run the risk of breaking something important. If your development is not competent either, then you run the risk of having your hygienic development process dirtied by, say, someone working directly on the production server. The problems caused by this are insidious and can take up time and attention that is better used pushing your business forward.
My post was a reply to:
> Can't your hosting provider terminate your contract?
It does solve that problem exactly.
I didn't propose a single box as a complete backup solution.
Next thing I know, all of my sites are 404ing, and when I call Bluehost, the tech tells me they shut down my hosting service because I was "abusing their unlimited storage." Then they pointed me to the asterisk that says unlimited storage isn't really unlimited and is whatever they feel like it should be on a case-by-case basis. Then I had to sit there on the phone with them and delete all my recently uploaded files before they would put my websites back online.
caveat emptor
-- Your hard drive
"Me too."
-- Your backup hard drive
"I also reserve the right to be stolen, dropped, burned up in a fire, or thrown out by your partner during a cleaning jag."
-- Your hard drives in general
This is why I hate "unlimited" in general. It's not unlimited, just tell me what the damn limit is.
That is to say, in the time it takes for your data to go from 1 GB to 10 GB, their cost-per-GB has gone down by >90%.
http://blog.kozubik.com/john_kozubik/2009/11/flat-rate-stora...
Why even bother putting "Unlimited" if they are just going to kick off the power users?
But I get it. Marketing.
The problem is making the marketing claim that this is "unlimited" when it's not. Call it "almost unlimited" or something similar and you're fine.
The first person who gets kicked off for this should file a false advertising suit. The marketing doesn't get to give you something which the ToS immediately takes away.
Maybe then these idiots (yes, idiots, because it takes an astounding lack of either morals or intelligence) will learn that lying to your customers is not okay.
^Conditions and Restrictions apply.
Verizon got smacked around for exactly this behavior a while ago (selling "unlimited" data, with throttling after some arbitrary cap). This is not substantially different.
Yes it is.
Again - explain to me how Verizon can be successfully sued for "unlimited" data (with throttling) and how these companies cannot with "unlimited" data (with undisclosed caps).
Explain how the two cases are substantially different, please. Just because Verizon are generally bastards and the companies above are generally not does not make false advertising okay or legal.
Just curious, but you say "successfully sued." I assume you mean they were sued, went to trial, and lost. I've actually tried looking this up, but I can't find this case. I've never really paid much attention to that, so forgive me.
It was a settlement, rather than an outright adverse ruling, but the net effect was the same - someone had to tap them on the shoulder and tell them to cut it out, they did, and had to pay out for their trouble.
If you go for "Unlimited" it has to be really unlimited, otherwise you're lying.
"Unlimited" service if honestly sold and serviced, would mean that this number does not exist. You could keep throwing data at them and they could keep storing it.
What is that number? Is it low? Is it high? Only Amazon knows, but in the meantime, they're pretending that this number does not exist.
False. Advertising.
The real problem is these companies want to eat their cake and have it too. They want to be able to advertise as unlimited, but not actually provide unlimited.
Me uploading my >1TB music/video collection for backup purposes that will only ever be touched by me? I honestly don't know, and knowing the way accounts are integrated, I'd rather not find out the hard way.
Logically, the acceptable amount they "want" you to store is obviously north of 5GB, since you get that for free by being a Prime member and you're paying for an upgrade, but I want to know where the line is, instead of playing this "it doesn't exist (but actually does and you're dead if you cross it)" charade.
It's much like companies having "unlimited vacation", where the net effect is that nobody takes any due to uncertainty.
I'm not sure i follow, how is it not?
If you're saying that the word can be used, but you know in Amazon's case that there is a secret number, then they shouldn't use it.. that seems unfair. There will always be a number to unlimited, hopefully published. It's technically impossible for it to be unlimited.
I mean, let me reverse the situation, and ask what amount of data/m or data/lifetime (data == tb, pb, whatever) would fit the word - for you?
The definition of "enough data" is going to vary person to person. I saw this and my mind went immediately to backups, but that is not a trivial amount of data (terabytes+), and I am not prepared to risk my Amazon account over it.
That's the problem, here. By not knowing what the number is, you are unable to make correct decisions, and with the alternative being account termination, that is unacceptable. Amazon has constructed a situation in which it's unclear what exactly they're selling, but since we know in advance that there must be a limit, they are advertising falsely.
The moment they kick someone out for this, ToS or not, they will have a legitimate claim of action. And I will guarantee you the actual number is going to be smaller than you think.
I definitely agree there. Though, i think their ToS could solve this by stating it in the fine print - assuming of course they really wanted to market the "Unlimited" word.
I did have a thought though, it was be interesting if a service like this offered "true" unlimited storage. How? Well, you could limit the upload/download rate, and ensure that you are always willing to expand faster than each user can upload.
Would make for an interesting PR spin for Unlimited.. but with so many companies touting Unlimited, it clearly wouldn't be worth it. Just an interesting thought.
Ahah just thought that maybe Amazon Glacier is implemented this way, and that's why it takes hours to recover a file.
That is why those things end up in contracts.
Someone is going to attempt to upload a pb of photos to amazon, and amazon is saying unlimited, but for personal usage (threshold quite possibly judged on a case-by-case basis), but doesn't include free storage of pbs of pictures.
At the end of the day, users need to be somewhat reasonable about what they expect from a $60/year service.
Now it's "your files, documents, etc." That's still a limitation by kind—the emphasis is on the "your." You can store as much data that you honestly, personally created for your own personal use, as you like.
Random terabyte-sized binaries? Not your personal data. Someone else's files? Not your personal data. Automated logs? Not your personal data.
But if you manage to create a few petabytes of vacation images? Sure, go right ahead.
What if it's all encrypted? Actually, scratch that, if I'm putting any substantial data in a random cloud service, it's going to be encrypted, period. Amazon gets incompressible, non-dedupable random bytes and nothing else.
What this all boils down to is that your definition is arbitrary and illogical, and has no support from the documentation, either friendly or legal, which leaves us right where we started:
Nobody knows how much they can put on this service, and the word "unlimited" is still false advertising.
There is no technical measure.
Instead, there is a heuristic that makes them curious about the purpose to which you are putting all your data usage. They then have the right to ask you to show them what kind of stuff you're putting in the cloud drive, at which point they can apply the actual measure—a human, ontological, qualitative measure of the "color" of your bits.
This is how the legal system works; this is how contracts work. Their provisions don't have to be interpretable by a dumb algorithm; they can be AGI-complete to solve for. This is why the court system has both police (instructed in an simple-but-false-positive-generating heuristic), and judges and/or juries (who are expected to then apply the human-deliberation-requiring "true algorithm" to screen out the false positives.)
Picture a savings account that holds "unlimited money." Does this mean that it holds money derived from tax fraud? No. Does this mean that it holds money used to create a correspondent deposit account for a wire-transfer service? No. These facts are intuitively obvious. Why doesn't it do these things? Because an account that does these things ceases to be a savings account per se. In the former case, it becomes an illegal conspiracy if they knowingly continue to serve you. In the latter case, it becomes a business account requiring strict money-service auditing. In both cases, if the bank detects suspicious activity (i.e. pushing millions of dollars around every day in said account), they have every right to ask you to what purpose you're putting their service—and to take the service away if you refuse to answer.
"Show" them? As previously mentioned, I'm encrypting my content. The nature of the data I'm story is both none of their business (I'm paying for unlimited data storage, not unlimited data storage of certain file types after all, per their advertising) and anyways it exposes them to further legal liability by way of asking that question.
If they can change the definition of "literally" in the dictionary they can probably change "unlimited" too.
IMO the term is still a useful marketing short-hand that has come to mean "more than can reasonably expected unless you are purposefully trying to break the system", though cellular data providers in particular have done their part to cloud the issue by having "unlimited" terms that are really ridiculously limited.
Based on my dealings with Amazon so far I have every reason to believe their "unlimited" will be, if anything, overly generous.
A buffet doesn't have to limit anyone's consumption to keep them at reasonable levels. Just like you can have broadband service with unlimited traffic. But storage is relatively expensive and can have quadratic cost if it's being constantly filled. It's a risky place to do actual "unlimited", and pretend "unlimited" is scummy.
If Amazon did something like that, they could call it "Practically Unlimited", go on about "you could store 10 thousand hours of high def video" or whatever so that normal users get the idea that it's plenty, and still give the actual limits to those who care.
Maybe the catch is that they want a limit that's high enough that practically nobody will hit it, but low enough that it isn't a net loss if someone does fill it up, otherwise their enemies could get a bunch of dummy accounts and fill them up. So they'd rather just maintain the right to preferential treatment.
- 1st line: your computer's HD.
- 2nd line: your SOHO NAS (very cheap these days, I have a 2 + 2 (mirror) TB NAS)
- 3rd line: remote backup
I can't think of any real-life scenario where these 3 copies get destroyed simultaneously.
Now the real problem here is upstream. In 3rd world countries like mine (Greece), we have an avg upstream of less than 100 kb/s. So if my pictures/video/pc/whatever backup is > 50 GB, uploading it anywhere is a pain.
ReadyNAS are way better especially the ones with INTEL cpus, since they can transcode video on the fly for some media players that require such a thing (e.g. you could stream video to your iPhone/iPad). But these are expensive and I don't need this kind of functionality.
If I had more than 5-10 TB of data to backup then I'd for a custom tower, 8 GB of memory, Gbit Ethernet of course and + FreeNAS (ZFS + FreeBSD network stack).
[2] The ReadyNAS got toasted through ethernet by a thunder!!!
NEVER keep important things only on a cloud service with no backups and no guarantee of service.
I think it's more to cover the case where you figure out a way to build a business that stores data in this, instead of in S3. Especially the case where that business is, itself, a primarily data storage or backup business, so you're directly living off the margin between "capital-cost storage for you" and "linear-revenue storage sold to customers." Amazon does not want to kill S3's business model, just yet.
Do you seriously expect a company to commit to not only guarantee that their service will be supported forever but on top of that, that they guarantee they won't shut you down regardless of what you store on their servers?
Never rely 100% on anything.
Amazon is not Google; I wouldn't put anything on Drive, given their history of product cancellations.
- Deciding that this isn't a lucrative service
- Deciding that it's had the desired effect on competition and served it's purpose
- Releasing another service that they decide competes with this but is better
What prevents them from redefining a core service to a non-core service on the go?