Image via CrunchBase

We come today not to bury Flickr but to praise it. Any service that can competently manage 5 billion images is worthy of admiration, and the Flickr interface and platform have long been posterchildren for user-friendly and effective web applications. That’s what makes this post from the New York Observer so noteworthy:

Major, major stumble from Flickr today—a Zurich-based photoblogger says Flickr deleted his account by mistake and lost his 4,000 photos.

Mirco Wilhelm has the original files saved elsewhere, but the photos from his extensive Flickr collection had been linked to from all over the web, including the official Flickr blog. Those links will now point to deadspace. Additionally, the followers he had accumulated, tags, photo captions and copyright information have been wiped out and may not be restored.

So where did Flickr’s vaunted platform fail? What design wisdom can we derive from this object lesson? When can we expect the salient code-review article to be posted to Hacker News?

Never, because it wasn’t a design flaw or programming error that cost Mirco Wilhelm his 4000 photos. It was plain, old-fashioned user error.

Wilhelm, you see, had submitted a support ticket to Flickr some days before his mass-photo loss. He reported that another user was posting stolen photos to a Flickr account. The support ticket, naturally, included a link to Wilhelm’s own Flickr account. Unfortunately, the investigating Flickr admin simply mixed up the account IDs on the support ticket and deleted Wilhelm’s account rather than the suspect photo-thief’s. And like all suspected copyright-violating material, those photos were permanently erased from Flickr’s archive — with no way to get them back.

Wilhelm has the original photo files which he can laboriously re-upload, re-tag, re-group and re-label, but he can’t recreate the URLs associated with originals. He can’t get the community — or the link equity — of his Flickr account back. All because a single admin inadvertently transposed two identifier strings.

Human error is responsible for a third of all data loss. The Flickr admin’s mistake is entirely understandable (and increasingly likely to recur given that Flickr-parent Yahoo is laying off more employees and Flickr troubleshooters will grow more, not less, overworked). That said, this simple mistake cost Mirco Wilhelm years of work and investment in his Flickr account. You can’t put a price on that.

Backupify for Flickr could have restored much if not all of Mr. Wilhelm’s lost photos, including the upload dates, tags, descriptions and — most importantly — the original URLs of every image.

Once upon a time, hardware failure was the leading cause of data loss, with human error following closely behind. In the age of cloud storage, hardware error has been removed as a serious threat to data — but human error has grown in significance. With the convenience of cloud-based access comes the risk of that many more fallible human beings influencing (or erasing) your data. It could be an overtired admin. It could be hacker that gets ahold of your password. It could be you, simply mis-clicking your mouse with dire results.

No one can design a totally user-proof system — not even Flickr. That’s why you need a third-party backup, even in the cloud. As the Observer post notes, “Despite a growing reliance on cloud storage across industries, negligence or a rookie mistake by a new employee could irreversibly wipe out user data — be it Facebook friends, blog posts or a photographer’s oeurve.”

You can get 2 GB of Flickr backup for FREE with Backupify Personal. Set up takes less than five minutes. That’s a small price to pay to ensure your complete Flickr archive is safe.

Online storage may be safer than any single hard drive, but that’s not the same as your data being invulnerable. Just ask Mirco Wilhelm.

Hope you’ve got a good backup plan.

UPDATE: Flickr was able to eventually restore most of Mr. Wilhelm’s photos. And they’ve given him free Pro service for the next 25 years. While we’d like to believe Flickr would have gone to such extraordinary links for any customer that suffered such a loss — regardless of whether the blogosphere raised a PR stink over it — we humbly suggest that investing in a Backupify account is the safer bet than relying on the kindness of vendors.

In case you hadn’t heard, Yahoo is shutting down several of its Web properties, most notably the Delicious link-sharing service. (Yahoo and Delicious have posted a vague response to the news leak, boiling down to “Yahoo wants rid of Delicious, so we hope someone else takes it over.”) That leaves Delicious users in a quandary — what do I do with my links now?

There are a number of sites you can transfer them to and enjoy much the same (if not more) functionality that existed on Delicious itself. Many of these services can directly import from Delicious, and for those that don’t, Yahoo will reportedly allow all users to export their Delicious links before shuttering the site. (And for those of you that used Backupify to archive your Delicious links, you know you’ve got an exportable/importable backup. Thus ends the shameless plug.)

Below are the four best options for your post-Delicious link life.

Pinboard.in is absolutely the best Delicious alternative, except that it isn’t free. You’ll pay a one-time fee of about $9 to signup (the cost escalates as Pinboard gets more users, so act now). For $25 a year, Pinboard will save a copy of the actual page you bookmarked a link to — forever — and will allow full text search against those snapshots. Here’s Pinboard’s own comparison of itself and Delicious. Pinboard isn’t quite the link discovery tool that Delicious was, but it’s probably the best pure link archiver out there.

Diigo is your preferred Delicious successor if link discovery and group linksharing is your main goal. Diigo has all the browser plugins and sharing widgets you could ask for, and it allows you to highlight and annotate any link (or page cache) you save, throw it in a public or private group, or label it with a public or private tag. The base level is free; if you want to do lots of highlighting or screencapping, you’ll need to spend $20 – $40 per year.

Licorize organizes your saved links into projects, so if Delicious was a Get Things Done tool for you, Licorize is the way to go. You can’t use traditional tagging, but that’s a feature, not a bug, as Licorize funnels every saved link into a kind of action item. Only save the links that actually lead to progress. Licorize is free, but if you want to share your projects with others, you’ll need to spend $5 per month.

Evernote is arguably the most popular notetaking — and content-saving — service available today. If archiving all your favorite links and page caches and camera snapshots and audio notes and anything else any web-connected device can capture  in one place is your primary goal, Evernote is the service for you. It lacks, however, any of Delicious’ public-facing tag tracking or global trending options. Evernote is your notebook of the web, so don’t use it if sharing is your main goal. Basic Evernote is free but your uploads are limited; you’ll pay $5 a month to take the brakes off and let others edit your notebooks.

Honorable mentions

Google Bookmarks – While strangely unconnected to other Google products, you can build Delicious-like functionality with lists — but it takes work. Still, it’s free, at least until Google abandons it like it did Wave.

Zootool - Similar to Delicious, but focused heavily on image sharing.

Connotea – An academic reference tool similar to Delicious, but with almost no public lists or tag sharing.

Faves.com - A more basic Delicious clone, which uses crude topic aggregation rather than conventional tagging to organize links.

Historio.us - A Delicious clone predicated on full-test search of links, with little to no group or tagging functionality.

Just remember, any online service can shut down at any time, and not all of them will be as professional about it as Yahoo. You may not get notice, and you very likely won’t get an export option. Protect the time, energy and data you’ve invested with a third-party backup like Backupify.

Because the smart ones always have a backup plan.

For Google Me to succeed, Facebook doesn’t have to lose. Google Me just has to offer something Facebook doesn’t — open standards. Wired sums it up thusly:

If Google’s legions of engineers really have been told to prioritize “Google Me,” or whatever this putative product might be called, its only chance is to give most of it away in the form of open standards, setting the stage for a multitude of interlinking social networks rather than the current system of “one network to rule them all.”

On the surface, this is great news. Google is notoriously more open than Facebook, and will almost certainly offer far more data portability than Facebook ever thought about. But there also a downside to Google stepping onto Facebook’s turf.

First, Google is prone to rather heinous privacy gaffes, much of which are a function of its scale. When Buzz exposed your Gmail contacts, that was a huge issue, in large part because Gmail is the product for power webmail users. Similarly, when Google started mapping Wi-Fi hotspots in a manner that involved packet-sniffing, it gathered a huge swath of data because the Wi-Fi maps were intended to cover the entire continental US. If Google Me has a privacy stumble, odds are lots of data will be laid out for the world to see because Google never gathers a small data set. And there’s plenty of history to suggest that will happen.

Moreover, Google is a prime hacker target, and not just because it’s feuding with Microsoft, Yahoo and Apple. Google has national enemies, most notably China. Being based on open standards will make Google Me easier to hack, and will also mean that far more third-party sites will be able to download your data and then carelessly expose it — just like with Facebook.

Like most folks, we’ll celebrate the day Google Me launches and we get a real, viable Facebook alternative. But we’ll also take some serious privacy precautions — and we’ll have a backup plan.

Related articles by Zemanta

• A Google rival to Facebook? Let’s hope so (cnn.com)
• ‘Google Me’ New Facebook Competitor (outsidethebeltway.com)
• Is Google Prepping a New Run at Facebook? (cbsnews.com)
• “Google Me” social network rumors fueled: it’s real, heavily staffed (fastcompany.com)

It’s a pretty well established point of law that you don’t own your employee e-mail account, your employer does. Thus anything included in your employee e-mail account is fair game for human resources or your employer’s lawyers — you have no defensible expectation of privacy on corporate systems. But what about personal e-mail accounts you access on corporate computers on the corporate network via the corporate Internet connection?

Well, personal e-mail is still private, even at work. Loving Care health services found this out when they copied employee Marina Stengart’s Yahoo mail correspondence to her lawyer made on a company laptop. The Supreme Court of New Jersey ruled that such copying was a violation of Stengart’s legal expectation of privacy.

This was only true, however, because Loving Care’s security policy doesn’t explicitly lay claim to private communications made over corporate systems. Thus, even your password-protected Gmail account could be fair game if you sent any Gmail messages using corporate equipment. The New Jersey Supreme Court said that snooping on employee Yahoo Mail messages sent from work PCs was illegal only because the cited company security policy laid claim solely to “all matters on the company’s media systems and services at any time.” Loving Care didn’t own Stengart’s Yahoo Mail account, so the policy didn’t apply — even if Stengart sent messages from a company PC over the company network in the company office during business hours.

The default position is that your privacy is protected so long as the company doesn’t warn you it’s allowed to snoop. If Loving Care’s policy had read “we reserve right to copy and access any personal communications sent by any means from company assets,” Stengart would have been hosed. Your e-mail privacy extends only so far as your employer’s e-mail policy is vague.

Bear that in mind the next time you send snarky comments about your boss to friends from a work PC. Gmail may be relatively secure, but no password is immune from a court order.

Related articles by Zemanta

• feature: The cloud and the future of the Fourth Amendment (arstechnica.com)
• NJ Court Upholds Privacy of Personal Emails At Work (yro.slashdot.org)
• Government Backs Down in Yahoo! Email Privacy Case, Avoids Court Ruling on Important Digital Civil Liberties Issue (domainmacher.com)
• Internet giants and civil liberties groups unite in opposition to warrantless police searches (trueslant.com)

Image by Oversocialized via Flickr

Google makes money by being your information middleman and serving ads along the way. The search page is the universal home page, and so long as that search page is Google’s, they’ll continue to print obscene gobs of money. So what does this have to do with the “slow motion train wreck” that is Google Buzz, the new FriendFeed clone that Google tacked onto Gmail mere days ago? It’s all in how you connect the dots.

Facebook is doing a bang-up job displacing Google as the universal home page, to the point that Facebook basically owns social page referrals, which is to say links that were recommended by friends, rather than search engines. If social referrals start to rival basic search in volume, Facebook could outflank Google. This problem will probably only get worse for Google as MS Outlook becomes the “social hub” of MS Office, drawing in LinkedIn, MySpace and Facebook data for the old fogeys who don’t use Web mail. (Moreover, Facebook is moving aggressively into mobile — the same mobile space Google wants Android to own, despite a huge rival in Apple’s iPhone — which means Google could get outflanked on two fronts by Team Zuckerberg.)

The answer? Google needed its own social referral engine, and shared links on Google Reader weren’t good enough. Enter Google Buzz, a redo of the FriendFeed service that Facebook bought not so long ago.

That’s all fine and good, except for two problems. First, Google hates user configuration; it wants to pre-optimize your experience by tuning it with what it already knows about you. This usually means using your Gmail contacts to connect you to whatever new service they roll out. For Google Buzz, this meant that millions of Gmail users were now automatically publishing their contact lists to millions of other Gmail users, which could be a big deal if you had confidential sources or clients that other people really shouldn’t (ethically, legally, or socially) know about. Reporters, lawyers, doctors and philanderers were all severely put out.

This might seem an obvious flaw that should have been caught in user testing, except for our second problem: Google rushed Buzz to market without subjecting it to standard user testing.

Google users screamed about the privacy problem, and Google had a fix rolled out in days — though not fast enough to avoid some government investigations of privacy violations. (Don’t laugh; Canada takes that stuff seriously.)

Google is fighting a pitched battle with Facebook and Apple. The stakes are getting higher, which means time-to-market for new Google, Facebook, and Apple services is going to get shorter. It also means the temptation to shoehorn a new feature into an old one — like Google did by trying to springboard Buzz by sucking in 175 million Gmail users — is going to be an ever more tempting tactic. TechCrunch lists the long and sordid history of “force-feeding” users new Web services – Facebook, AOL and Yahoo all appear on the list.

To paraphrase Battlestar Galactica, all of this has happened before, and all of this will happen again. Possibly to you.

At some point, this trend is going to result in a huge data exposure or, worse, a huge data loss for a major, brand-name Web service. Don’t be surprised when it comes down the pike, and don’t be without a backup plan.

Related articles by Zemanta

• Google Tweaks Buzz After Overblown Privacy Backlash (wired.com)
• Facebook Drives 44 Percent Of Social Sharing On The Web (techcrunch.com)
• Google Buzz from a medical blogger’s perspective (casesblog.blogspot.com)
• No escaping social email: Facebook and MySpace coming to Outlook (digital.venturebeat.com)

The always insightful Joel Spolsky reminds us that it is not enough to simply back up your data:

The minimum bar for a reliable service is not that you have done a backup, but that you have done a restore. If you’re running a web service, you need to be able to show me that you can build a reasonably recent copy of the entire site, in a reasonable amount of time, on a new server or servers without ever accessing anything that was in the original data center. The bar is that you’ve done a restore.

Let’s stop asking people if they’re doing backups, and start asking if they’re doing restores.

So, does Backupify accommodate data restoration in addition to our data backups? To a point.

Some of the Web services we backup simply don’t support restoration. Twitter, for example, won’t allow timestamped uploads to its database. We can retweet everything you’ve posted, but we can’t backdate the tweets. We’re working with Twitter to change that, but ultimately it’s in their hands, not ours.

On the other hand, we can roll back your Gmail or Blogger accounts almost exactly as you left them — because Google provides a very nice upload API that accommodate thoses kind of backdated data injections. The same is true of Yahoo with Flickr.

Our performance with other services varies, but we strive to make as much restoration as is possible available to you. Where the service doesn’t accommodate true restoration, we still provide full access to all your backed up data in a handy XML format. You can see exactly what you had, even if the social network or Web app won’t let you put it precisely back in place.

It’s not perfect, but it’s far better than having no record at all. What is Facebook’s native data-restore option? Oh, right, it doesn’t have one. We can do better than that. You can, and should, too.

Related articles by Zemanta

• The Shifting Role of Storage in the Data Center (datacenterknowledge.com)
• Twitter Is Broken, But Here Is The Contributor API In The Wild (thenextweb.com)
• Theory: Danger’s Sidekick data explosion was (DUDUNT! DUDUNT!) sabotage? (mobilecrunch.com)

Tanya Forsheit at the Information Law Group blog authored a lengthy and fascinating series on the legal implications of cloud computing, but one particular passage caught my eye:

Under Rule 34 of the Federal Rules of Civil Procedure, a party may serve on any other party a request within the scope of Rule 26(b): (1) to produce and permit the requesting party or its representative to inspect, copy, test, or sample the following items in the responding party’s possession, custody, or control.  Who has control of data in the cloud?

Translated from legalese, this means that if a cloud-based data storage company is sued, it can be compelled to hand over any data in its “possession, custody, or control.” Does that mean Google can be compelled to hand over your Gmail account contents to a plaintiff, even if that plaintiff isn’t you? More simply, does Google own your Gmail data, or do you? The answer lies in how your EULA is worded.

Given Google’s “you ain’t got no privacy” stance and Yahoo’s willingness to sell your data to Uncle Sam, there’s a pretty clear legal policy from two of the bigger players in the game. If they own your data, that means they can decide when and how you access it. Right now, that access is fairly unfettered, but they could change it at their whim, at which point it will be you suing your own cloud-based data store for access to your own information. Personally, I’d have a third-party backup (or three) on standby, just in case.

Related articles by Zemanta

• Bank Sends Email to Wrong Gmail User, Sues Google For His Identity (mashable.com)
• Industry Expert Panel Discussion on Data Privacy (cloudave.com)
• Top 5 Google Account Security Tips (techburgh.com)
• How Far Should Google Go To Protect User Privacy In Lawsuits? (techdirt.com)
• Outed Blogger Plans To Sue Google; Skank Model Mess Gets Messier (techdirt.com)

The above is the price list that Yahoo charges the government for turning over user records. Not only does Yahoo have no problem turning over the contents of any user account — including all stored e-mail — upon request, but the company views abrogating your privacy as a profit center. Not a great profit center, mind you, as the complete contents of your account is only worth $40, but it’s clear the company is ready and willing to sell you out. Cheap.

Anyone else feeling the need to keep a copy of your online data with a third party, just in case the original cloud provider decides the government has more of a right to your data than you do, or that you’ll have to fork over $40 to know what Uncle Sam knows about you?

Me too.

Related articles by Zemanta

• Google Unmasks Anonymous Blogger (blogoscoped.com)
• Mozilla Guy to Users: Leave Google, Switch to Bing (mashable.com)
• Search engine privacy (wired.com)
• Is Privacy in The Cloud an Illusion? (elasticvapor.com)