A blog focused on helping bloggers and online retailers make more money online.

Yesterday, the Internet’s considerable gossip mill was alight with talk of Yahoo! layoffs and the possible closure of several Yahoo! properties. [edit: link here] My sympathies to all those employees affected.

Amoung the more notable services on the list was one-time search giant Altavista and more significantly, Delicious. Delicious, for those that don’t know, and will now never know, had the rather modest goal of allowing people to save and share bookmarks. It has been around for a while in terms of Internet time—it was founded in 2003 and later sold to Yahoo! in 2005 making it an éminence grise on the Internet block. It was one of the earlier forays into what we now know as the ‘social web’ and also one of the pioneers in the dubious use of both subdomains and  kooky ccTLD’s for website names (originally opting to brand itself with the rather lumpy del.icio.us), a legacy that lives on with bit.ly et al.

For a while it was quite popular, and on a numerical basis still is: according to Alexa it ranks around 250th in global traffic and attracts roughly .5% of the Internet’s traffic on a daily basis. No slouch. Not bad for a website that hasn’t had a major refresh in years.

But I’m not here to eulogize Delicious: it was already dead to me. On hearing the news of it’s imminent demise, I checked and discovered it has been over 18 months since I had last added anything, and probably 6 months or more since I remember visiting the site.  It’s purpose, in my case, supplanted by newer, ever more social tools like Twitter.

But I don’t speak for everybody, and with 5 million users, the surprise news was taken badly by some [apparently those who haven't yet realized that Yahoo! is the doldrums of the Internet inhabited by ghost ship websites]. The most common cry was “what about my bookmarks?” and it is this that stood out to me.

What about my bookmarks? I realize that in the fast-moving stream of the Internet, bookmarks have roughly the same shelf-life as toast, but there is still value in older content: you bookmarked it for a reason, right? So far Delicious has been quiet on what happens to all the bookmarks that users have stored and shared. And as of yet, there is no way to [easily] export your bookmarks; when Delicious turns out its lights, everything is gone in a blink.

Now, I’m sure that this unfortunate event is papered over in clickthrough Terms of Service (TOS) that nobody reads, but in an Internet where cloud computing is touted as the future, and software as a service (SaaS) is pushed as the Next Big Thing™, the issue of data ownership and data portability is a serious one to consider. Before you place your data into the cloud or entrust it to a service, make sure you know how you can get it out if you need it.

Delicious provided the framework, bandwidth and infrastructure, but the users populated it with content, the ebb and flow of interesting links that its service relied on to draw eyeballs.The data soul of Delicious belongs to its users and should be allowed to exist outside of the now-dead corporate body.

However, few corporations and EULA’s allow for this type of corporate reincarnation. The death of Delicious illustrates once again that as a user, the real value in many of these services is not their brand or the fora they create, it’s how they allow us, millions of us, to create, manage, and share our data with other users. Data that should persist.

UPDATE December 20, 2010: After a period of some confusion and much gnashing of teeth by Delicious users, it now appears that Yahoo! may not be closing Delicious, but rather looking to sell it off. Even on this news, my point still stands, ownership transfer merely perpetuates the existing encumbered data model.

As an aside, the PR bungling here is massive: alienating all your users prior to a shotgun marriage doesn’t seem like the best approach to me (most of Delicious’ front page is populated with How-to’s documenting migration away from Delicious); then again, I always imagine Yahoo! headquarters as having Benny Hill’s Yakety Sax soundtrack on permanent loop.

The idea of a ‘direct-dial’ Internet was long ago supplanted by the need for an appendix that points you in the right direction. As the web has evolved many systems and companies have risen to the role of Card Catalogue of the Internet: Gopher, Yahoo!, Altavista, and a league of others have come and gone. However, one upstart grabbed the crown in the early part of the 2000’s and has steadfastly remained king: Google. With Google’s rise the maxim ‘build it and they will come’ has been superseded by ‘PageRank and they will come.’

Google’s long and fruitful reign has triggered a Pax Romana in what was a previously a tumultuous, factious world. Google’s ‘Do No Evil’ mission statement has made it a benevolent giant. But like a lumbering Gulliver, Google can errantly become the weapon of Lilliputian skirmishes.

An entire industry, Search Engine Optimization (SEO), has sprung up parasitically around Google, feeding on their crumbs like plovers picking crocodile teeth or barnacles affixed to the belly of a whale. They exist because of the chinks in Google’s armour and the unpredictable eddies and whorls created by its army of bots as they swim through the Internet’s content soup.

SEO is one part voodoo juju, one part science, and sometimes six parts snake oil. Practitioners claim that through study and observation, the Gordian Knot of Google’s algorithm can be teased apart with the marlinspike of carefully placed keywords, formatting, link backs, and repetition. However, like Goldilocks, Google is tough to please: too hot, too cold, too hard, or too soft, Google surveys websites with a jaded and critical eye in an attempt to index content that is legitimate rather than a lowly sow’s ear of Spam stitched into a silk purse. While SEO gaming and keyword stuffing may elevate your PageRank scores, such arbitrage may just as easily dash you on the rocks of the dreaded Fifth Page.

Occasionally though, someone finds an unlocked stairwell while rattling doors in Google’s keep. Recently the New York Times ran a lengthy article on one such exploit foisted by a Brooklyn-based eyeglass retailer (I won’t give him the satisfaction of a mention here) who had risen in both infamy and PageRank (perhaps RageRank is more apt) through aggressive, if not abusive business practices. In an example of the classic perverse incentive, his garrulous customer service was a deliberate marketing ploy: complaints from shilled customers were the very thing that fueled his ascendency to Google’s PageRank summit. Not only did he not shy away from customer animus, he encouraged it. His business thrived as his search placement ride aloft on a balloon of blistering invective.

Google is not static though. It boasts of the busloads of PhD’s under its employ, and I picture them strolling through vast data centres in skinny ties and labcoats, clipboards in hand, tapping inspection glasses, squinting at gauges, adjusting dials, and turning the valves that monitor and precisely control the massive, gleaming Search Engine they have assembled. Perhaps a romantic notion: at the minimum, Google’s programmers are as busy optimizing and refining their much-vaunted search algorithm as ne’er-do-wells are at exposing knotholes and jiggering windows.

Google is notoriously tight-lipped about discussing its operations and its secret ranking recipe. However, as often happens when called out in a large publication, Google has subsequently tightened a spigot and the perpetrator’s ranking has plummeted. All this is good, the system worked, right? Except on further reflection the very fact that Google, a private, commercial company, has become The System and can exercise arbitrary life-and-death control over online businesses with virtually no oversight is terrifying. A friend of mine who operates a multi-million dollar online web business says it best: his livelihood “is semi-colon in Google’s source code away from bankruptcy.”

I am not suggesting that Google has done anything wrong here: by all accounts they have done a spectacular job making the billions of webpages online navigable. With the exception of a few missteps (from the serious Chinese censorship debacle, to clumsy privacy management with Buzz and Streetview, and the downright comical Wave) they  abide by their own Do No Evil ethos and provide a valuable service. Usually for free.

But have we become lulled by Google’s convenience? By allowing Google to become so large we now risk living in its massive shadow should it choose to blot out the Internet’s information sunshine. As a marketer and writer I chafe and struggle against the straightjacket imposed by keywords and SEO requirements. Has Google’s PageRank algorithm become a gilded cage within which we sing for its Imperial pleasure? We trust Google, perhaps too much as evidenced by recent events in Brooklyn, can we trust that Google will always act to separate the crows from the nightingales?

UPDATE Dec 6, 2010: The notorious eyewear website proprietor mentioned in this piece was arrested by Federal agents in New York on charges of cyber-stalking and interstate threats.

If only every piece of IT hardware was as useful as R2-D2.

In the Star Wars franchise, even the childhood memory ruining prequels, one of the lesser players is a small Astromech robot named R2-D2 (or Artoo, to his friends). As squat as a bar fridge, round with a domed head like a garbage can, and speaking in an odd vocabulary of beeps, boops, whistles and flashing lights, Artoo is not just Oscar to C3PO’s Felix, he’s one of the unsung heroes of the Rebellion.

The benefit of being decidedly unblessed in the looks and communication department is that Artoo bristles with a veritable Swiss Army knife worth of tools and accessories. He’s the sort of robot who could help you ‘borrow’ cable from your neighbour or show you how to jailbreak your iPhone. Because of his 31337 skills (that’s ‘elite’ for non-geeks), Artoo’s intervention is critical to the success of virtually every operation undertaken by the spacefaring protagonists throughout the trilogy (okay, grudgingly, hexilogy). So, while whiny Luke and lantern-jawed Han receive a lot of the credit and get to blow up moon-sized space stations, I like to remember Artoo working quietly in the background to open access hatches, hot-wire spaceships, and prevent death by garbage compactor. Most importantly, like a giant, 3-legged USB stick, R2-D2 is the preferred media to convey Princess Leia’s message to a Jedi Knight who retired to Phoenix, Arizona Tatooine with the plaintive missive:

Help us Obi-Wan Kenobi, you’re our only hope.

The circuitous lesson here is that if you need to store something important, store it on something reliable: preferably a robot.

There are two types of computer users—those who have lost data and those who are going to lose data. How you lose it depends on circumstances; hard drive failure (likely), computer virus, theft, fire, or leaving your laptop in the seat pocket of a Boeing 767 after landing at Denver International Airport [*whistling innocently*]. In any case, when Bad Things™ happen to your data, you quickly realize how invaluable it is and how difficult/expensive/time consuming it is to replace or recreate it (the latter fact can be attested to by anyone who saved the only copy of their thesis on a floppy disk.)

Even when you’re not busy losing data, you’re trying to share it. AffinityClick is growing quickly and we’ve got lots of computers, assorted iPhones, iPads and other gadgets, and lots of documents and data floating around. Creating shared folders on individual computers is a hassle, and while solutions like Dropbox work great, the service is not practical for sharing large files quickly.

We needed something that was network-attached for easy access from multiple computers (and iDevices), would allow us to store data with RAID¹ redundancy, act as a backup point, and share data between multiple users with some basic access controls. Oh, and preferably cost less than a decent used car (Sorry, EMC).

The Apple Time Capsule features both wireless N (in fact, it’s a full-fledged router) and Gigabit Ethernet (GigE) but offered only a single hard drive (no RAID), is difficult (impossible?) to expand, and comes along with the stiff Apple tax which accompanies shiny, white devices from Cupertino.

I personally use a D-Link DNS-323 at home. It’s a toaster-sized box that is as cheap as dirt (~$120 on sale), has a GigE plug, and offers two hard drive bays for RAID redundancy. However, as it is only a two-slice data toaster, it lacks room for future expansion.

The final contender was the Data Robotics Drobo FS (presumably for Frickin’ Sweet) which presents 4 drive bays for easy expansion, the mandatory GigE port, a sorta-kinda-RAID-like proprietary file system² that allows for data redundancy and recovery, and the promise of simple setup and configuration. All at a price that keeps the CFO from having a conniption fit.

So credit card in hand, we bought one online, along with two capacious (for now) 2TB hard drives for storage. Somewhere, in a galaxy far-far away (actually Vancouver), our own Artoo was being boxed up, ready to save our data from certain doom.

[Next: From Zero-to-Drobo in 30 Minutes (and possibly get sued by George Lucas!)]

¹ RAID provides a method of automatically distributing data across two or more hard drives to allow for data recovery in the inevitable event of hard drive failure. In this case, we are looking at RAID 1, also known as ‘drive mirroring’, which makes a duplicate copy of one hard drive onto a second one. RAID 5 is the other commonly used method which splits data and recovery parity ‘stripes’ across 3 or more hard drives.

² Drobo’s use of a non-standard, proprietary file system is one point of contention in the tech community. People who worry about such things argue this point like Biblical scholars debating the Shroud of Turin. The fact remains that despite the theological debate over EXT2 versus ZFS et al, I can’t find much evidence that this prevents Drobo from doing its job effectively. From the perspective of the guy who has to install, use, and support it, I’ll leave the debate to people who have time to debate: I just need a box that works.