author: Marlin Ouverson
Someone1 on Google+ mentioned:
"The hype around cloud reminds me of the hype around outsourcing or programming without programmers. These are marketing concepts, not technical solutions."
To which we replied, as follows.
Yes, more hype than usual. Plus, perhaps consumers need a reminder that the model before personal computers was that applications and data resided not at the terminal but in a clean room attended by experts who took care of things like maintenance and adding features and data storage. And granting, or denying, access.
"Like building on an earthquake fault or in a flood plain, with the cloud we are choosing risks and future costs that are outside our power to control in exchange for mere convenience."
On the technical side, "the cloud" seems formed from vaporware but it is acquiring meaning nevertheless. It's the sense that the user can relax, offload the backups, don't worry about upgrades, let others deal with security and scalability, and have easy multi-machine access to the same documents and software wherever they go. That's the allure, and the warm, fuzzy word "cloud" is disarming, too: good coinage.
But that concept, in which the user unwittingly relinquishes control as well as responsibility, is a sharp turn back toward the old mainframe model and once again exposes users to the risks associated with giving others control of our lives and our personal data. Like building on an earthquake fault or in a flood plain, with the cloud we are choosing risks and future costs that are outside our power to control in exchange for perceived convenience.
Cloud-based software applications may charge unending rent for as long as you use the software. That would be like writing a check to Microsoft for every month you use MS Word or Outlook. And the host could deploy forced upgrades that are unneeded and unwanted — consider the user backlash, for example, when Facebook unilaterally changed certain settings and overhauled its entire look-and-feel. There may be no way to opt out of changes to cloud-based software.
Perhaps more important is the issue of data independence. Your own data may include songs and video you have purchased, photos you have taken and documents you have created. Before you choose to store it in the cloud, you have to consider things like: what happens if the cloud host goes out of business? can you get a backup copy of all your data? will your data be in a format that can be transfered to another cloud service or onto your hard drive and still be usable? is it encrypted both during transmission and before it is saved to the cloud's storage?
And there may be social consequences, as well as personal ones, to cloud computing. Ideally, having your music in digital format means far less manufacturing and transportation costs and, hopefully, huge savings in energy consumption when you consider the size of the music-buying market. This is a green-leaning technology. But having your data in the cloud means it gets re-downloaded whenever you access it from a new machine or when the copy on your device differs from the version in the cloud — changes may be saved to the cloud repeatedly, and then downloaded to every device from which you access it. Again and again. Add the costs of massive server arrays, cooling plants, secure buildings and storage devices at the "server farms," and it's easy to see that, over time, cloud-based computing could even end up more costly and more detrimental to the environment than the older methods of pressing and shipping physical copies.
If "the cloud" is mostly made of smoke and mirrors, what actually backs up the dream with real services? Those features largely are server configurations and massive disk arrays. That about sums it up, though the implementation can be so intricate and complex that it makes the concept seem nebulous, hard to grasp. We like the parts we understand and the rest is… cloudy. There is no doubting the convenience, in many cases, but with our personal accumulations of data growing steadily, it is well worth taking the time to understand the tradeoffs and how to mitigate the potential costs and losses that may be the fallout of today's cloud and the rush to profit that underlies the commercial forces advocating it.
1. Thanks to Dirk Harms-Merbitz for the initial remark.
Copyright © 2011, 2012 by Marlin Ouverson. All rights reserved.
Original Facebook "Note" posted
8 September 2011
Expanded version presented here 15 September 2012.