Rule number one: never trust the cloud computing
It was August 27 of 2007 (my birthday) when columnist John C. Dvorak warned against the dangers of things such as “cloud computing”, “software as service”, “Web 3.0″ and the many abbreviations that in these years are trying to catch the attention of the public and sell as new what is the most old fashioned computing architecture ever existed.
Then Dvorak used as an example the crash of the servers for the authentication of Microsoft Windows with the Windows Genuine Advantage (WGA) tool, a crash that could have had catastrophic consequences on a huge amount of systems. Despite the time passed, unfortunately, those potential consequences continue to hang over as a sword of Damocles on all the naive people which uncritically (and often enthusiastically) embrace the overwhelming majority of the web-based services designed as replacements for the corresponding applications to run locally.
In these months and years, in fact, the situation have done anything but worsen, and the increase of the user base for the manifold galaxy of cloud computing has highlighted the absolute unreliability of the Internet, as it has been designed in the Sixties, as the base platform on which permanently host applications, data and more generally the digital life of users, firms and organizations of any kind.
Internet is unreliable, transient, impermanent by definition, and the clearer evidence of such established fact comes just from the behaviour of a corporation, Google, that has made networking and the worldwide network the foundation of its enormous advertising business and its own existence. A series of events piled in a few days is useful as a manifest alarm bell on the excess of hype that by cloud computing could cause a new crash in the IT, explosive now more than ever when the economy is in full recession.
The abuses and the inadequacies of Google
Mountain View is spending many of its energies on the convergence between remote applications and the digital ecosystem of users and companies, and just in the days when the Microsoft servers crashed CEO Eric Schmidt was evangelizing the crowd of Seoul Digital Forum pleasantly talking about the new digital world of cloud computing, or “Web 3.0″ as suggested by the audience, of an infrastructural model “very different from the mainframe era, very different from the current PC industry“.
In the model conceived by Schmidt the applications run on any device, by a thin client or inside a web browser, are fast, secure, reliable, resolve problems and facilitate the life to users and firms. In the real interconnected world however, here and now, a simple extra punctuation mark, a slash (”/”) at the wrong place is enough for Google to end identifying as potentially harmful any single result of any search of its search engine.
The accident, occurred on January 31, lasted only an hour and the situation came back to normal after the cause of the problem was identified in a simple human error. And what if theoretically such an error, always possible because employees aren’t machines and sooner or later they make a mistake, would have conditioned not the web search but the productivity suite Google Docs, making temporarily unreachable the projects of companies that already use it as a job tool for their business? What if the information stored on remote servers would be impossible to recover if not through the failed Google services, how the supposed advantages of a computing “in the cloud” would end if it wasn’t even capable of guaranteeing the most trivial access functionalities as any off-line word processor?
But even more dangerous than Google’s unreliability are its tendency to abuse users’ data and the inadequacy of support for who became victim of a fraud made possible by the security flaws of a popular social networking service. It gives the heebie-jeebies to know that Mountain View delete posts from music blogs hosted on the Blogger platform, following an RIAA’s report on the alleged distribution of copyrighted tracks and careless about the fact that nothing illegal is happening indeed and the blogger at issue (Ryan Spaulding) is behaving exactly as advised by the press release sent to him by the labels.
To whom like Mark Ghosh put then an excessive trust in Google services happened something even worse, he was left alone by a non-existent support service after having suffered a phishing attack on Orkut: the portal, one of the many designed to promote personal relationships old and new, is affected by a huge amount of security problems, is the favoured target of an endless series of worms, malware, crackers and cyber-criminals willing to exploit at their own profit users’ personal data.
Ghosh, who says to be accountable for an Orkut community counting up to 25,000 users, tells the horrible story about the seizure of his account after the attack, the useless attempts to ask Google for help and the fact to have essentially been alone with himself facing the issue. “Are we all under the false hope that someone in these big companies actually cares about the people that use their products?” Ghosh asks himself, “Is the online world doomed to failure in circumstances or are we willing to make a stand only when it affects us?“.
Where the users’ data go
Google is the IT company that more than any other one, in these years, publicised Internet as the solution to any problem and the answer to any need, and if not even the corporation that shaped the net in its likeness is able to guarantee that minimum level of trustworthiness required to take care of sensible information, personal and professional data then it means that the entire system doesn’t work: the user loses control on his own data, returns back to the dependency already existing tens of years ago between client and mainframe and in exchange he obtains zero guarantees, a mean (when it goes well) support and the tangible risk of an abuse or the selling off of the aforementioned information to the best bidder.
It already happened, it’s happening again now that the Facebook founder has announced the intention to exploit the more than 150 millions members of one of the biggest social networking services for marketing purposes, by creating the largest database for marketing research of the world. By offering to the interested companies the chance to lead targeted researches on a loyal and easily sortable user base, Mark Zuckerberg has finally found the way to monetise Facebook by selling off privacy and the likings of who thought it was only a question of “friends” old and new, photos, common interests and other nice things.
Not even a blind, at this point, could ignore what awful hazard is innate in handing over the keys of one’s own digital world to a company whose only desire is to make business and profits. The intriguing cloud computing meme hides the largest hoax of IT from the times of speculative bubble of the last Nineties, a phenomenon that pretends to bring the technology clock back of many decades (if the net does it all then the PC upgrade rush has no meaning at all…) and that is destined to disappoint all the promises of reliability, speed, simplicity and security done up to now.
Web 2.0 is dead, and even cloud computing doesn’t feel very well. The last nail on the coffin of the hopes that “in the cloud” it feel much better than off-line is the simple psychologic mechanism of survival, that livens or should liven any user, executive or CEO aware of the fact that if the on-line services ecosystem provider goes bankrupt, or decides to delete the account for whatever reason (maybe for a human error like happened to Google), data, contents and business disappear forever. Or end up being on a post-mortem auction on eBay.
Related posts
- PC gaming: a platform to rule them all
- Why The Fool blocks Google Safe-Browsing
- Parallel computing and GPGPU, the super-PC genesis between universal libraries and proprietary platforms
- Browser war, Firefox at 20%. Google Chrome what?
- Google Chrome starts the Third Great Browser War
Comments
5 Responses to “Rule number one: never trust the cloud computing”
Leave a Reply
-
· Subscribe to Sir Arthur's Den ·
-
· © Some Rights Reserved ·
The exclusive contents of Sir Arthur's Den are distributed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License, and must be ascribed to Alfonso Maruccia as the sole owner of the related copyrights.
Refreshing counterpoint on the Cloudwashing of all things IT. The fragility of the network (Internet) and the protocol (opportunistic IP) makes today’s reality of the cloud something significantly less than the promise.
Browser
Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 1.1.4322; InfoPath.1; .NET CLR 2.0.50727)
Beside the warnings I’ve expressed in the post, I think that the network as an appliance platform offers some interesting stuff, but I hate hype and particularly the marketing hype of who tries to sell sh*t describing it as gold…
Browser
Mozilla/5.0 (Windows; U; Windows NT 5.1; it; rv:1.8.1.16) Gecko/20080702 Firefox/2.0.0.16
http://tinyurl.com/bcpj4l
Oh, that’s funny
Browser
Mozilla/5.0 (Windows; U; Windows NT 5.1; it; rv:1.8.1.16) Gecko/20080702 Firefox/2.0.0.16
Royal Pingdom does a great job of tracking outages in the cloud. They’re way more common than most would think. And check out my blog post on Google Apps SLA. Google’s SLA is written to allow 21 hours of downtime a day!
Browser
Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 1.1.4322; InfoPath.1; .NET CLR 2.0.50727)
Harsh. I like it.
The very idea that people would store their information on an external service is baffling.
It’s hard enough for me to trust the hardware and software I use, letalone trust hardware and software and people — all of which I am not even *allowed* to examine.
Browser
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.10) Gecko/2009051209 Mandriva/1.9.0.10-1pclos2009 (2009) Firefox/3.0.10