Now that everything is moving to the cloud internet, you might think that data loss is a thing of the past. Sadly, as the past few months have taught us, this actually isn’t true; we first had the Microsoft/Danger disaster, and now we have Palm and Sprint facing a class-action lawsuit over data loss for webOS phones. All this raises the question: how safe is it to store your precious data on the internet, and do you really trust the internet?
In the Microsoft/Danger disaster, it turned out to be not as bad as everyone (including the involved parties themselves) had thought; Microsoft was still able to recover the lost data, meaning that users still got their stuff back. Still, I’m sure Sidekick owners were less than pleased.
The issue with the webOS is a bit more problematic, but at least it only affected a relatively small group of people. As it turns out, the online restore function has problems restoring a backup made by a newer version of the webOS to an older version of the webOS. It affects a small group of people, but Palm and Sprint have been hit with a class-action lawsuit nonetheless.
I won’t dive into the merits or validity of such a lawsuit (I’m sure the fine print in the appropriate contracts have some cop-out remarks about data loss), as I want to talk about how you people feel about storing your precious data on the internet (and yes, I refuse to use that other stupid marketing designation. This man is not for turning on this one).
So far, the problems have been confined to mobile phone data, which is most often not nearly as precious as data found on computers, such as photos, sensitive documents, school work, and so on. People usually still have an old-fashioned real-world address book lying around. In addition, Microsoft eventually recovered the data, and the Palm and Sprint case only affects a small amount of people.
But we’ve seen the writing on the wall. Google is hard at work making sure you store your stuff on their servers. This means that the document you make, the photos you transfer, and whatever else you create on your computer is now no longer stored locally – it’s out of your control, out of your hands. Data protection and loss prevention are no longer an individual responsibility – they’re Google’s.
Is this a future you want to live in? Surely, you say, Google won’t lose my data, They are one of the biggest technology companies in the world, highly skilled in these matters. Well, but what about Microsoft? They’re even bigger than Google, yet they still managed to mess up royally.
It is cases like the ones described above that make me very weary of any of these store-it-on-the-internet ideas. People claim this will solve all of our data problems, but all moving your stuff to the internet accomplishes is that it’s no longer your responsibility – which is a whole world apart from it no longer being an issue. The problem has not been solved – it has simply been moved.
As much as the hoity-toity technology companies of today want you to believe that the internet is the place to store your data, they are wrong. In the end, internet or no, there is still only one tried and true way to keep you data from getting lost: it involves common sense, and a 60 EUR external hard disk (or a fancy RAID setup if you’re a nerd). Mac OS X has an easy-to-use and effective (but technically crude) solution, and you can get similar solutions on Windows and Solaris (ZFS).
Data loss prevention should be in your own hands, not in those of some company. What can happen at Palm, Sprint, and Microsoft can most certainly happen at any other company. Placing your trust in them is ignorant and silly, and will come back to bite you in the rear.
Well said, Thom, I personally just couldn’t agree more. With a choice of hard drives, solid-state memory devices and the likes of CD-ROMs and DVDs to store things on, why would we want to store things with a potentially insecure third party?
The idea of remote storage is not in itself intrinsically bad, but not intrinsically one hundred per cent trustworthy either. So I use every storage device that I can – hard drives, external hard drives, USB sticks and even a second-hand Chocolate phone which just happens to have eight gigs of free storage . . . portability is a good idea, but it’s also good to know that it’s always with you and not potentially lost at a third party’s remote location.
Don’t forget about redundancy just because it’s in the cloud. Put your important data on more than one service.
Don’t worry – I don’t just store on my own physical media, I use others via the Internet too. That was my point originally – because I really don’t trust physical media much more than I trust the “cloud” – I take a very old-fashioned “belt and braces” attitude towards most things in life . . . including the remote storage of some files I consider most essential.
But the caveat against depending on such services was illustrated by changes in Yahoo’s storage facilities . . . as they brought in things like Flickr and removed the existing user storage for pictures, for example. Documents and pictures had to be removed, and where are they now? On solid media.
Actually I find it hard just to sit down and do these things through being so busy and lacking bum-on-seat time – and I use services like Google, for example, but I don’t like the idea that they may decide to use my data against my will some day so I keep a relatively small amount of stuff in http://www.startforce.com/, although there are others that I use (this one is becoming rather full now, in fact). Startforce has continued to become more useful but it only has 1Gb available. I have mainly documents and pictures there – and only things that I consider absolutely essential.
I found Startforce after seeing a link by Eugenia (http://www.osnews.com/story/18099/Another_10_Web_Operating_Systems_…) and the previous article by the same author . . . this is not purely for storage, and there are probably better services available. Maybe someone should write a new article on this subject because we need to be aware of services like this? Perhaps grouping individual service providers according to their available storage?
Could be interesting.
No amount of home based backups are safe either. Think fire , flood, theft,etc. I mean old photographs fade , become degraded, but when people said you could back it up digitally , people were scared of that process too. My Father lost years of family records because he backed up everything to one hard drive, and when the surge of power from mobo killed both hard drives…..
When you look at how many things are fouled up in software daily, this doesn’t seem silly at all. It just sounds like the next thing in a long line of things that cuase people alarm. Namely it’s new, and people don’t like new when old works just fine.
Edited 2009-12-09 23:53 UTC
I have a seperate disk with a backup of everything I have and copy the rest over the internet with duplicity (deja-dup gnome-interface) to somewhere else.
The nearest thing to 100% safe I could find.
With duplicity it’s also encyrpted, so I don’t have to worry about that either.
Their is also Duplicaty for people who need Windows-support.
With my data in my home; I control it. With my data in Google’s home; google controls it. As someone said in another discussion;
“Anything you post on facebook is being put on someone else’s computer and they can do whatever they want with it.”
I’d also suggest that keeping your data at home does not stop you from keeping off-site backups. Make your home storage securely accessible over the internet. Make sure you have backups. Make sure one backup copy is stored outside your home. Think of the family photo library; a copy at home for viewing, a backup and a copy at grandma’s which she can view or you can restore your own from. No reason at all to trust one’s data to a third party corporation with more reason to misuse and release your information than to protect it.
You are not going to release your important information in the wild unencrypted, aren’t you?
As long as it’s encrypted and you also have a physical copy in a secure place, you have the best of both worlds. I would bet that it’s safer in Google’s datacenters than in your own pocket.
I’m all for backing up to the internet, just very uncomfortable about it being the *only* location for my data. Services such as carbonite.com are a great idea, I think. “Cloud” backups which are encypted locally so theres no privacy issues.
Personally, I have a colo which I download to, and then download from. This I then backup locally if I deem it important enough. This works well for me because I can access my data from anywhere as well as having the convenience of having it locally.
I’d never feel comfortable having my data only residing on a companies servers though for reasons Sidekick have demonstrated. And what if the company goes bust?
Before a company goes bust, it will be legally obligated to find any possible way to remain alive and return a profit (assuming corp). At the end, this may even mean shareholders demanding the wholesale information retailing. Ain’t corporate late wonderful.
That’s illegal, and depending on the country, it could be a federal crime. In the end, you could make zillions.
For a contradictory point of view, look at it this way – Google backs up the data stored on their servers. Most users *don’t* backup the data on their personal machines, or if they do, it’s to DVDs or an external hard disk sitting next to the machine. Safe from a hardware failure, but useless in the event of a disaster (a fire, say) that destroys originals and backups alike.
So for the average user, Google is a much safer place to store their data. Most users don’t have redundant off-site backups, distributed across the world. Google does.
Bingo. There are lots of reasons to store data locally, but safety is definitely not one of them. Unless you have remote and redundant backups, you are way better off
storing your data in the “cloud” as long as the company has the right infrastructure. Yes shit happens, but this is about as safe as you’re going to get without spending millions yourself.
Or, I could store it both places for double the fun. There is no reason to store backups solely in the cloud. The cloud is simply a convenient offsite backup provider.
How about a User Data bill of rights, something along the lines of:
Any user generated data will be exportable in one of two forms:
1) An opaque blob that will allow the restoration of user generated content for the period of 6 months.
2) An documented data file and an API that would allow community written tools to allow restoration of user generated contents.
So users that contributed to a site, amazon reviews, pictures on flickr, comments on blogs, documents/spreatsheets, contacts, and related information could be backed up as often as the user was comfortable with. In either case the backups would be signed so that the provider could be sure that a user restores what they backed up.
Seems like that would significantly lower the risk of losing important data that’s stored in some random cloud.
Agreed, and preferably the latter – any company hosting a person’s data should provide a means for that person to fetch that data in a useful form (i.e not cryptic binary blobs, nor screen-scraping web pages).
Thom, I agree with everything you’ve said, and that’s not usualy the case
. Well put.
One thing we really need to keep in mind is the next step after this. First they store your content, next they sift and use it for their own purposes, and finally they decide what you can and cannot have. I really am afraid we’re looking at the first, somewhat undisguised, attempts at making Big Brother a reality. We really need to think about where this is leading, and the fact that we really don’t have any measures in place to protect anyone’s rights here at least not worldwide. This is one giant gray area, and it’s all too easy for that one line too many to be crossed.
In this whole situation, I’m utterly astonished with how little I see in the media about the problems of transferring the control of the data to the storage owners — as mentioned by darknexus. People seem to think:
“Oh, Google is so coool! yes, I will give them my data and they will never do anything bad with it – they promise they won’t, so that’s like… a sure thing!”
Arrgh, people, did no one ever break a promise he gave you?!!
And the worst thing is that as much as I’m afraid of that, I do also use gmail… I try to avoid it as much as I can, but it’s so well executed… mommy, I’ll quit gmailing!.. umm, tomorrow, ok?..
The very least I can do is try to disable Google Analytics & Ad-Words — those robots silently stalking me on nearly every step I make in the thick mist of “the cloud”… <shudder>
By the same token, it’s been quite amusing to see people up in arms about all of the terrible things that Google might potentially do at some unspecified point in the future (those hypothetical bastards!).
Yet the same people ignore concrete, valid reasons to criticize Google for things they’re doing today. E.g., in order to upload a video to youTube, you must agree to give Google unrestricted, irrevocable rights (in perpetuity) to that video – which also means they can use it for commercial purposes without providing any compensation.
The cloud is good. It’s the stupid amateurs that don’t know how to use it whats the problem. Cloud is cool. I no longer need /dev/null I now just point to /dev/cloud this is a technical innovation that can not be matched.
i will never trust my data with any cloud or such service, least of all google. anyone that would are simply insane and needs help. i wont use software that locks my data in snotty formats either – life is simply too short to deal with such crap
Guess I am “insane” then… Google handles my email, phone calls, instant messaging, searches, most documents, and my online payments. How many times have I lost data? Zero. How much has it cost me? I have actually made money… my AdSense revenue is higher than the 3% I am charged to take payments via Google Checkout.
I know all you security folks hate the idea of letting Google sift through your personal data. However, anything that consolidates my data, makes it redundant, available on any computer with internet access, while improving the accuracy of searching for information gets a bigs thumbs up from me. Then again, I don’t have anything to hide… I would probably read 100% of my email to my mother, or even my girlfriend’s mother.
Permanent Fatal Error … it is hardly about “I have nothing to hide” philosophy [and it’s a childish way of thinking too]. It is rather about: “will they use my data to track me / make money on me / spy me / sell it / use targeted marketing on me … or … “will someone else gain an access to the data stored on their servers?”. Just use your imagination. Don’t be that straightforwardly naive.
I hope you’re using Tor and you encrypt all your messages (SMSs, mails, chats, etc), because otherwise you are already being tracked and served.
Could I grab your bank information and credit card numbers.. I’ll just hold on to them for safe keeping and you can let me know if you loose your copy. It’s ok, you have nothing to hide so why value any privacy?
How can you live like that?! Or… where *do* you keep your dirt?
He probably lives in a small town, where everybody knows everything about each other. Nothing wrong with that; not everyone has a corpse rotting in their cupboard.
In my experience, the people worth talking to have some form of dirt somewhere. Those who claim not to can be interesting for a time, just to watch them claim how they don’t. And then there are those fascinating folks who really don’t seem to have any, which make them either interesting due to the profoundly boring lives they are willing to put up with, or fascinating because they are so good at hiding their dirt. Then there are folks like the current Dalai Lama who get into politics and leave there dirt out in the open where almost no one notices it.
Not to mention that privacy is basic principle of modern democracy. My dull ridiculous life is still my life. I don’t like to share my browser history, DNS traffic, e-mails, the music I listen, my online shopping, IRC conversations, etc. without my explicit consent. Simply because it is my life and my data.
We might as well get rid of such ridiculous laws that prevent private parties from opening normal snail mail. And so on.
Basically the same BS argument used in war against terror and whatnot.
Edited 2009-12-10 17:12 UTC
Never mind that every time you access the internet all of that information is being logged in multiple places? I don’t buy it.
I assume you mean the concept of “the only people who care about privacy are those who have something to hide”? Just because I have nothing to hide doesn’t mean that I want the government to listen to all of my phone conversations and reading my email. Quite the opposite in fact… the government should respect my privacy and rights at all times.
I do, in fact, live in a small town. Like most people, I have skeletons in my closet, but I learned years ago that keeping secrets online is just short of impossible, so I just don’t put things online that I don’t want people to see.
As for Google using my information to target me for marketing… that’s fine with me. As I said in my original post, the more Google knows about me, the better it can deliver the search results and services that I will be interested in. Bottom line is, the pros of Google knowing a lot about me outweigh the cons.
There are a ton of open source encryption programs out there. No one says you have to let them sift through your files. You need to start thinking out side the box.
Even though the data stored on the “cloud” server farm is your’s. It technically is the property of the “cloud” service.
Also… What happens to the data if the cloud service goes under?
I say I prefer to stay a traditionalist.
I trust in the cloud, the problem I see is that it has been bad implemented, all data in one server? that’s stupid, I’d make sure that my cloud provider has the information replicated in more than one place to avoid data lost.
Months ago my work Laptop was stolen with all my .doc files, source code and mail information, where do I get that now?, I won’t backup all my information daily, it takes time and is silly, source is safe with SVN, but mail? .doc files I made in the fly?
Since then I use gmail as my account and my .doc and .xls files are in MS Office Live, I feel safer. But of course It won’t replace my own backups.
Put it this way – is it worse implemented than the usual alternative, where someone’s data isn’t just on one server, but on one physical disk on their laptop? The worst cloud provider is at least no worse than what for most people is the status quo…
L.A.R.T. request? This is one for the FAIL blog.
You are silly. Yes, you, a developer, a knowledgeable person that should know how to make a backup and do it.
I have some stuff stored on the internet where it’s convenient, but I don’t trust my most critical files to the internet.
BTW, you should look up the definitions for weary, wary, and leery. I’m extremely weary of people mixing them up.
I see the problem in this way: You can use the metro to move from one place to other one in your city: it is cheaper, maybe it is safer and it is faster than your own car… but… you want to have YOUR car to go wherever you want, to do whatever you do, etc.
Maybe my data can be stored safer in some place in the internet, but I want to have my data with me; I want to be responsible of my own things and if I lose my information, it will be my fault and no one else. Ok, having my backups stored in the Internet is a very good idea, but having all my stuff just there, sounds silly to me.
Other thing: where I live, having all my stuff stored in the Internet is not an option: I have a 128kbps Internet connection and I pay maybe more than you all pay for your a-lot-of Mbps connection.
Edited 2009-12-10 03:09 UTC
Very bad and innacurate example. Cars are obsolete, they are not the safest way of getting to the places [most people die in CAR ACCIDENT crushed between steering wheel and the rest of the vehicle], while metro is one of the safest way to get where you want. One definiteliy doesn’t have to have a car. It’s stupid ego and unconsciousness that makes him want it.
Cloud computing on the other hand is not safe, while local storage IS.
It isn’t.
Care to elaborate?
Not every part of the globe is doomed with earthquakes, floods and alikes.
When you cut that out, the only thing that’s left is HW failure or human failure.
Offline backup [USB keys, cards, tapes] is the cure for HW failure. Then we have human failure … and that’s a whole different story, because you CAN actually make yourself being more careful.
Fire is enough. More destructive than earthquakes and floods. Available at your location too!
All you need is to keep backup of your data, which you want to preserve. But keeping working copy on internet is just convinient. You can work from multiple computers without copying and syncing and all this trouble with usb-flash.
“Data Liberation Front”. This is what we need.
From a Mac OS X perspective, how do I backup GMail?
Mail.app through IMAP, from what I get, is not the perfect solution due to how it thinks of labels as folders (either you get multiple copies of the same e-mail -which I guess would make thinks not easy to restore-, or you only backup the All Mail folder and loose all labels -not a valid option-). That, besides getting a bit on the way as a backup tool.
Googling I found about http://www.gmail-backup.com for Linux and http://gmailkeeper.com for Windows. Nothing comprehensive (i.e. includes labels) for Mac OS X.
You get locked in to Gmail once you start using tags, that’s the lesson here. Although I’m sure you can replicate that locally with a suitable client — as yet unknown to me…
http://www.dataliberation.org/google/gmail
Will it transfer the labels used in Gmail to (for instance) Thunderbird’s labeling system??
Labels in GMail show up as folders in any IMAP client, allowing you to retrieve your messages as easily as e-mail from any IMAP server. The only downside is that messages with multiple labels will appear in multiple folders, causing multiple copies to be downloaded.
Thanks! It would be more interesting if the Gmail labels were transformed into Thunderbird labels, and more space-efficient too.
That’s a backend issue. The IMAP server translates labels into folders. I don’t know if the IMAP protocol itself supports labels/tags.
I work with a lot of people who are worried about the potential digital dark age- either from file formats no longer becoming readable, or data simply disappearing due to disk failures… Already it’s hard to read data taken on punch cards or magnetic tape, but books from ancient history still exist.
So, Google might be the solution, as long as you trust Google with all your data (to have, and to not lose). Or you could spend the money and have offsite storage yourself.
Perhaps we judge the durability of works from ancient history based upon what has survived, and that of modern works based upon what might be lost.
The vast majority of ancient works are lost to us. At best, we know of them by a reference in something that survived. Most have simply been lost.
Personally, I’d consider the loss of the great majority of today’s blogs to be a plus. Still, I suppose we should preserve even the tripe of today, so that future generations will at least have a fair warning as to what not to repeat. Not that they’re likely to listen…
Edited 2009-12-10 05:17 UTC
Well, Sturgeon’s Law (“99% of everything is cr4p”) notwithstanding, I’m not looking at this from a perspective of “oh well, it happened in the past” but more of an “how can we prevent this from happening to us now”. Future archaeologists will probably both thank us, even if they laugh at Geocities. Even on a shorter term it’d be nice to be sure my CD-Rs from 2001 are still readable, or the consolidated DVD-R from 2006.
Bergman’s corollary: The percentage of published works which are not crap decreases as publishing becomes more and more accessible.
Maybe we should concentrate on the non-crap. If Wikipedia and Twitter were both drowning in the river, and you could only save one…
That said, some of the most fascinating insights we have into 1st century Roman life come from the voluminous wall scrall in Pompeii. I suppose one man’s crap is another man’s fertilizer.
Oh please, be it the Wikipedia. I couldn’t live without it. I hope the rest of social services die forever, for the good of humanity.
What do you mean? I farted earlier today while working on a Django app and it smelled really bad. I wished, at that moment, that I had had the foresight to register a Twitter account so that I could share it with my followers. But by the time I had half-registered with them, the aura of the experience had faded.
Edited 2009-12-10 17:12 UTC
But think. That twit could have saved a generation of future archeologists trying to understand the absurd life of mankind in the 2000s.
Thereby contributing to unemployment in their era. The intricate interaction of human flatulence, and the flapping of butterflies’ wings, upon the tapestry of time is difficult to fathom.
Edited 2009-12-10 17:45 UTC
Most of that is lost because of religion (the destruction of The Library of Alexandria) or lack of interest in copying the originals (which ruled out a lot of thrash).
You are totally missing why online storage is such a great thing because you are looking at it as an individual rather than from a business perspective.
There is one sure rule that forms the foundation of a successful career: “Always have somebody else to blame.”
You see if I’m Joe the middle management IT guy, and I store the company data in-house and something catastrophic happens to it, I’m going to get canned. But if it’s been out-sourced to an online company, I’m in the clear. It’s their fault. In fact, I would recommend to the boss that we sue their pants off.
I wonder if I can earn commission on revenue from litigation…
That is so NOT true. If I’m a CTO of a company, and the online service that I chose loses my company’s data, and it impacts my company’s ability to operate, I’m GONE. Fired. You couldn’t just blame it on the online service, and wash your hands. It doesn’t work that way in reality.
The even more relevant question is: how to ensure that your personal data is destroyed properly in the Cloud?
With conventional storage schemes you can write zeros to data, encrypt your data, smash your drive with a hammer. But with the Cloud, the only thing you have is the reassurance from the brother.
Edited 2009-12-10 06:31 UTC
The whole idea of storing my files on a remote server, with high capacity, various protective measures and all the things that I can’t afford at home makes it quite compelling and reasonable. There is just simply no way I can make something even half as durable and secure as the Google servers are.
However, there’s still several reasons why I don’t store my files out there somewhere; because I have no idea and no assurances of the files actually being gone when I delete them, it would takes ages to download or upload all the stuff I wish to backup, and most importantly, I don’t trust people I don’t know; there is ALWAYS bound to be someone working there that likes to sifle through my files, go through my personal photos and so on. The only way I could prevent that would be to encrypt all my files beforehand, but then I wouldn’t be able to use the files until I unencrypt them again.
I just like to think that giving full control of and access to all of my digital belongings, without me being able to control him/her/them, is too risky.
Exactly.
Thom, each day in IT work you are confronted with user deleting documents, losing PSTs, nuking share drive folder three structures.
How do you think we survive? We force user to use shared drive and then backup regularly to be able to restore the data if the issue arises.
Seems familiar? It is, it is saving on the Internet, only that YOU own the server.
Now, unless you do not want to start a serve in your house, put it on open DNS and be the IT manager of yourself, even backuping data has it’s limits ( what will you do? Burn a DVD each time you modify your Word doc?).
Google is in fact offering to do that for you on their servers and I assure you that no backup system is flawless in IT ( Have ever have to say to a mother that her child will not come home tonight? I had to say to a researcher that his mailbox was nuked and lost. Close enough).
If we are talking backup, the internet is a solution that existed only for companies brought down to single usage.
If we are talking ( as it seems to me) of the rights on data and thus the privacy issues that it all involves, well, we should have started from the day they offered us a Gmail, now it is a bit late, don’t you think ?
Dont be daft, a proper backup means that you have at least 3 copies of your data in different places (drives/media) It’s a foolproof protocol even used by Nasa on spaceflights. It will work for the home user as good as a company. Try that storing your data on a cloud. You have no idea what it will be used for and how it will be stored. In Fact you loose control. Simple. If you are to lazy to be bothered with backing up dont look for an easy way out. There never is.
Come on Thom, I understand your point of view, but tell me…where do you keep your money, in a bank account or behind the brick at home?
that is different, just think about it
Well, if that doesn’t win the award for ‘ridiculous analogy of the decade’, there must be some ridiculous analogies out there!
How would Thom get wages without a bank? How would he pay his direct debits?
People in this day and age are practically _required_ to use a bank. Cloud storage is an option.
On this topic– What about Dropbox and similar? You always keep a local copy as well as the server copy so if either dies, you have your data back. And on the issue of privacy, there’s nothing to stop you from using AXCrypt or similar to encrypt on a file level.
Edited 2009-12-10 11:47 UTC
I find it a valid analogy.
You don’t need the bank to store the money for you. You can get your money at any time and safely store it behind a brick.
Required to have a bank account to get your money, but you can run away with your money at any time if you want, with a lot of risk.
And cloud storage, if you think a little bit, is just an extension to what we have already being using for ages. If you send documents in a mail, they are available to both your mail provider and your destinatary mail provider. Unless you encrypt all your traffic, you have been giving away your data to the public since you clicked the send mail button.
I use it, not regularly, but when needed. I don’t implicitly trust it, but I don’t really have any sensitive information.
This is ridiculous. Simply the amount of “personal” (in the lack of a better word) data transferred in the intertubs is overwhelmingly bigger than anything transferred or stored in conventional terms.
And by the way, in civilized countries normal letters are not open to your mail provider.
So? Google doesn’t index your mails to make a profile of you? I’m sick of people here crying out loud about privacy when none of them use Tor and move in the interwebs using their own name, their real IP, and send mails with their documents and share important information without even thinking, don’t encrypt their MSN chats, etcetera. I recognise the importance about privacy, but privacy doesn’t come from the hands of corporations and laws.
Ohh, I see. I used the word “mail”. I should have used the word “email” instead. I’m sorry, I was still talking, ehm sorry, writing about electronic mail and attached documents, ehm sorry, bits of information.
Not only that, but even a cheap data center probably is about a billion times safer then any other way you could personally store data. Unless you have your data on raid-10 machines in a climate controlled room with redundant power and security guards.
Er, right. So occasionally some big name companies lose some data. Thank god for personal storage: because no one has ever lost data which has been saved on their computer’s hard disk. </sarcasm>
“or a fancy RAID setup if you’re a nerd”
RAID is not a backup-system.
If a file get’s deleted from the RAID system it will remove it from all the disks.
What you want a seperate copy, a snapshot of what you had before. Preferable several.
it is a backup if he uses it as a backup. one copy of data in one place, one copy in another. one copy happens to use RAID. RAID backup.
I heard it also helps exterminate bugs.
The fact is no one cares about you as much as you do. If you want your property to be safe, you have to take care of it yourself.
No one cares? And this is a fact?
He said ‘as much as you do’. He was saying that you need to look after number one.
Which most people would agree with,
I tend to believe Google cares much more about my data than myself.
But I bet say Stevie Jobs cares more about his data than Google. The point being that there are many people using the Internet. I can’t believe I had to write that.
Edited 2009-12-10 17:24 UTC
I can’t believe you can’t read between lines and I also can’t believe I actually have to write this. And besides, who f–king cares about Steve Jobs. Google and other ad based companies “need” to care about your data, because it’s gold for them, probably much more that for you. Your data is invaluable to them. Delete from Gmail and you are only hiding, Google still can access it (they have said it).
I can’t believe you are so foolish.
I should have said “no one cares as much about you as you do”, that way the people who only read half sentences wouldn’t be mislead
RAID is a backup solution against hardware failures. They happen, but most data loss is about user mistakes.
You should have a backup for important data from the past so you can get back in time when you or someone does something stupid with your data. And make sure you backups at several points in time in case you or someone does something stupid with your backup data.
Now what support can you trust to keep your data? Answer: none. All supports have failures. The cloud, the hard drive, and the DVD all have failures and will always have failures. Simply put, you can’t have your data 100% safe. What you can do is mitigate the risk.
So let’s say the cloud has a 0.01% failure rate (let’s say that 1 user looses all of his data every year for 10000 users using the service). And let’s say your hard drive also has 0.01% failure rate. Now if you copy your data both on the cloud and on the hard drive, your data loss risk is greatly mitigated: 0.01% * 0.01% = 0.0001%. Now if you copy your data to DVD and the DVD also have 0.01% failure rate, you are down to 0.000001%.
Bottom line: when you copy your data somewhere, you reduce your risk of loosing your data dramatically. The cloud is not the perfect, but it still reduces your risks.
The privacy issue is a completely different topic.
I think the article is greatly oversimplifying things. I have bunch of stuff stored in “the cloud” and think the convenience far outweighs any potential risk. The trick is to rate the value of each piece of your data vs. convenience.
I have some data is which is vitally important and which I take good care of. I also have tons and tons of data which, honestly, has close to zero value. I have a bunch of stuff on google docs for example and find the service amazingly handy, but if I lost all those docs tomorrow it would have close to zero effect on any important aspect of my life. The cloud (hate that term) is perfect for that kind of data, data where the convenience of being able to access it from anywhere (and easily share it with people) outweighs any of the negative aspects.
I remember a Google outage some months ago.
People lost data, and a lot of them had their accounts cancelled too.
So, if it happens when you need the data/service they are providing – where is your backup ?
Even in a personal view you must have a “B-plan”. And in the case of your death, your family should know how to recover your outsourced things – money, data or physical objects. And don’t put the spare key of the box inside the box ;-).
Still, using the Internet to store data is not a bad idea. But do not make it the only copy you have.
“affects a small amount of people. ”
Sorry, but I don’t buy that one. I’m not trusting any of the companies with my personal data. Only replicable content goes/stays online.
This whole “cloud computing” [client-server] way of using internet is just simply BAD. I prefer to be angry on myself when something bad happens, than to blame others. Besides – I can have strong crypto, multiple backups … yes, I am better in saving my own data than some silly companies that only cares for the money.
My data is invaluable for me. And one more thing – computing – as a whole thing – is PERSONAL for me. Not social, not massive and not collaborative. I do my work, I send and get things, but that’s just about it.
“Only wimps use tape backup: _real_ men just upload their important stuff on ftp, and let the rest of the world mirror it
”
How can I be the first to post this?
I wouldn’t call Time Machine technically crude. Technically simple, more like. It’s pretty much the only backup software in existence that achieves the goals of incremental, partial and full-system restore *while at the same time* allowing you to browse the disk as if it’s a normal hard drive. You may consider storing backup data as plain old files and folders to be “crude”–I consider it the best implementation of backup that I’ve ever used.
This is why Opera’s Unite is not as retarded as *some* tend to believe. Having control over your data has more advantages than just privacy
Yeah, right … :
“Although Opera Unite claims to “directly link people’s personal computers together,” to use it you need an account on Opera’s servers, and all of your exchanges pass through Opera’s servers first. That’s an effective way to get around technical difficulties like NAT firewalls, but more important, it makes Opera the intermediary in your social interactions — not Facebook, not MySpace, but Opera.”
source: http://www.infoworld.com/d/developer-world/opera-unite-real-plan-pu…
And I double that.
Using Opera proxies is optional…
It’s connectivity options are kind of limited (only clear text protocols like http, ftp, cifs). It’s dirt cheap though and has the minimum; two drive bays and RAID1 support. RAID; not just for nerds anymore.
I think the real value of the cloud is synchronization and redundancy. To store your data all in one place is foolish, and that includes the cloud.
However using services like Mozy and Carbonite for backing up data you also have at home is very smart, if you do not have access to remote back up otherwise. Saying storing your data in the cloud is foolish is short-sighted and foolish.
The average home user doesn’t back up their data. If they do, it’s to an external disk. What happens when there’s a break in and everything’s stolen? What happens when there’s a flood? Fire? The cloud, in conjunction with automated services like Carbonite could save them.
Sure, you could burn DVD’s and/or move the external disk to work or a relative’s every once and awhile, but who actually does that, besides a handful of geeks?
The cloud brings somewhat of a co-location/disaster recovery site to the average user, and that’s a good thing. Saying otherwise is quite ignorant.
“In the end, internet or no, there is still only one tried and true way to keep you data from getting lost: it involves common sense, and a 60 EUR external hard disk (or a fancy RAID setup if you’re a nerd). Mac OS X has an easy-to-use and effective (but technically crude) solution, and you can get similar solutions on Windows and Solaris (ZFS).”
Yeah, tried and true, except when you have a burglar, fire, or flood. Then you’re screwed, just like everyone who didn’t back up at all.
It could happen to anyone, even Francis Ford Coppola.
http://news.bbc.co.uk/2/hi/entertainment/7017433.stm
Edited 2009-12-10 15:19 UTC
There are three general categories of data for me. Immediately disposable, moderately disposable and finally not so disposable. Use trash can to store the former, CDs and local RAID to store the middle, and etch the latter onto my ass cheeks in binary so I can use my time machine should all else fail.
All right, if I can get a free backup provider on the cloud, internet or whatever, why not use it??
I do my own backups on tape, dvd and pendrives, but some disaster can happen on home and maybe I would not be able to rescue my own data…
So? what I would do?? Just zip the data, use pgp, gpg or some stuff like these, upload to the cloud and we are done!!!
I think you need to understand the difference between “backing up” and “storing”.
Storing your data on the internet: bad.
Backing up to the internet: couldn’t hurt.
You do know that Google has been doing this longer than Microsoft or Palm and Sprint and they don’t seem to have issues. Maybe it is how THEY are doing it!
When cloud providers guarantee 5×9’s of reliability, and have the kinds of support contracts that vendors like IBM and others provide, I’ll consider it. Until then, the cloud is more like a casino than a datacenter: You take your chances and, perhaps, you reap some reward. But for large organizations, there’s simply too much risk at present to justify committing without support. I means, seriously, has anyone ever tried getting somebody at Google on the phone? They don’t answer the phone. And this is acceptable when you have millions of dollars in transactions riding on their performance?
Just for reference, that’s about 5 minutes of downtime per year. There are other significant issues, of course. But a five nines requirement seems a bit excessive for most applications. And, of course, the truly relevant number is “How reliable they can be” – “How reliable your local Datacenter can be”.
Edited 2009-12-10 17:55 UTC
5×9’s isn’t excessive if you’re an enterprise and your entire business is riding on customer guarantees. Google and other companies like it don’t get that. It’s simply going to take some time before we reach that point. Not saying that it can’t happen. But it’s gonna require having adequate failover support, and nobody is investing in that kind of capability for Internet computing — which remains at “sort of good enough”. That may work for a lot of small businesses, though.
But if you’re just a regular business, like most, it is very excessive. My customers would balk at the cost of the last couple of 9’s, at least. Because limited down time for those services doesn’t cost them that much.
Google is trying to actively court business. But as I said before, that’s a tough sell unless you have (at the very minimum) phone support.
http://www.google.com/enterprise/enterprise_solutions/index.html
Google is saying “we’re ready for business”. But a lot of companies that I talk to are saying, “show us that you can support our businesses before we sign.” Likewise, there are very serious privacy concerns with having Google store your data on servers side-by-side with your competitors, and possibly mining your data for their own purposes. I realize that somebody has to pay the freight — and there is no free lunch — but do you realize the liability when Google’s servers get hacked? I’m sure if you read their agreement very carefully, you will see broad liability exemptions for Google. As with anything else, caveat emptor.
The problems with your data on the cloud are not technical problems. The problem is largely legal and regulatory and those will be worked out in time.
Just like we have regulation on sprinklers or fire alarms, we’ll probably end up with regulation on data warehousing. Standards on the kind of backups… locations…
Alternatively, maybe regulation will not come, and you’ll just end up with really good SLAs and what not. You’ll probably have to pay for those.
Give it some time…
Edited 2009-12-10 17:47 UTC
I’ll never, EVER surrender my personal data to the so-called cloud! I make a regular backup of my computers and if something goes wrong, I’m in control of when, where and how to restore it. And if something goes REALLY wrong, (corrupt backup when trying to restore, bad disk, etc.) I only have myself to blame, and not some company out there who’s going to “sorry for the inconvenience” their way out of the whole thing and cover their rear ends hiding behind clauses from their terms of service. Microsoft can crash their Windows Live or Danger servers all they want; I’m not going to be affected the least bit.
I don’t really use any Google product other than YouTube, and that’s because I haven’t found a degooglized alternative. GMail is for collecting the spam that is generated when you register in forums and the like, not for real email. My personal email is POP3 from my ISP saved to my computer and backed up by myself in Outlook .pst files. I don’t use Google Docs, or Google Wave, or Google Voice, or Google Maps… or whatever is Google’s next excuse to track your each and every move. Of course, no Android either! Even if Google were truly the savior of the digital era (which I don’t think they are) and their intentions when collecting all of your data were good, I don’t want to even imagine the armageddon that will ensue the day some hacker (inside job or from outside) hacks into Google’s servers and grabs that data from users for not-so-good purposes.
Quite good point.
In another news: “Security researchers have spotted the Zeus botnet running an unauthorized command and control center on Amazon’s EC2 cloud computing infrastructure“.
You didn’t have to wait long.
It hardly requires any cracking. What about the day that Google gets bought, or forms some partnership, and changes its slogan from “Don’t Be Evil” to “OK, I guess we’re powerful enough to be evil now”.
The problem is if you are actually storing the data on internet as the only copy of the files.
It is another thing if you use internet as an alternative backup area, then i see no reason to rule it out as it is only a copy, maybe one of many copies, of the data.
When people hear that they can store all their data on the “Internet” or “the cloud” their eyes glaze over and all they can think of is “Woooow… data everywhere!”
The simple fact is your data is being stored somewhere, on some server or servers. The golden rule of data reliability still applies: The more copies of your data there are, the more resistant it is to loss and/or corruption. Keeping this in mind, backing up or storing your data on other servers via the Internet is just adding to the reliability of your data. It’s not foolproof, it doesn’t guarantee your data will never be lost or corrupted, it just decreases the chances of that ever happening.
So, just use common sense!
I don’t agree.
First the gold rule of data reliability is probably the integrity of data, which is again undermined in the Cloud schemes.