The Skies Are Clouding Up Even More

In the Newsletter issue of a year ago I wrote about the need to be concerned with using "the cloud" - digital servers in cyberspace - for important computer data. (see "Cloudy Skies?" by Harrison Eiteljorg, II, CSA Newsletter, January, 20010; XXII,3.) A recent experience shows this concern to be more real and more pernicious that even I, cynic at heart, believed when I wrote the prior item.

I ended that previous item with a comment that I would surely use cloud computing at some point in time - not thinking that I had been doing so since buying my first iPhone. Apple synchronizes the data on the iPhone with data on my MAC (actually 4 MACs, two personal MACs and two of CSA's), making sure that all calendar events and all entries in my address book remain the same, whether I am using one of the MACs or the iPhone (and since writing the earlier item, my wife's iPhone). This system should, if any system can, be foolproof. I am using Apple computers, Apple iPhones, and Apple programs; thus, the company controls the operating systems, the software, the connection processes, the synchronization processes, the servers, and the server software.

For the sake of this discussion I must point out that Apple's address book and calendar applications allow names and related data (in the address-book program) and events (in the calendar program) to be grouped so that I can choose to see, at a given moment, only those people in my address book that I have called archaeologists or only those I have labeled as family. Similarly, calendar events can be grouped so the home-related or work-related events can be shown alone (or in a specific color on the calendar).

In November I added about 20 or 30 people to my address book, almost all of them in a new group for the purpose of sending holiday greetings. At the same time I added some people already in the address book to the new group I has created. After a couple of days of doing this at home on my laptop I realized that neither the people I had added nor the new group I had created for them were showing up on my office computer or on any of the other computers/phones; of course, the people already in the address book were not added to the new group since the new group did not exist on the other computers/phones. Once the problem was realized, I tried a variety of things to get the synchronizing process via Apple's cloud server to work properly. (Note that at no time did the added names or group disappear from my laptop.) My son had experienced something similar with the calendar, and he suggested possible cures, which I also tried. Finally, in exasperation, I contacted Apple and attempted to repair the damage via their chat service.

After trying all the normal avenues, the Apple support person sent me a link to a web page that described a process for exporting all my names and addresses to a neutral format, eliminating all those names and addresses from my computer, and then importing the names and addresses from that neutral format. The hitch? Of course, there was a hitch. All the groups (I had 16 groups into which names and addresses had been put, with many individuals being in multiple groups) disappear in this process. That was an unacceptable outcome; so I manually transferred data from the laptop to a home computer, synchronized that computer to the cloud data, removed the contacts from the laptop, re-synchronized the laptop, and then continued as before, with automatic synchronization. Of course, I cannot really continue as before. Now I know I dare not trust the synchronizing process implicitly and must, as a result, regularly check to be sure synchronization is occurring.

Consider what this means. Here is a company that controls every part of the process: all operating systems, programs, and connecting procedures. But when there is a glitch, it has no proper data recovery system. Note also that this is not a once-in-a-million event. It happens often enough that they have a web page detailing the (inadequate) repair scheme. They do not have a way to force the system to reform around a proper original data source in such a case; nor do they seem to have any responsibility to me, as subscriber, for the loss of time. (In the end, I was able to get everything back to the way it should have been after spending somewhere between six and ten hours on the problem.)

If this is what happens when one company, controlling all parts of the system, operates a cloud-based procedure, . . . . I should, in fairness, add the important caveat that synchronizing data across multiple machines is a very different thing from allowing someone to work with software and data hosted on a machine somewhere in cyberspace. People have been doing that for many years without knowing it. My point is simply to reiterate the critical issue: if your data are too important to be lost, they should be in your control, even if they are also - ALSO, not exclusively - in the clouds. It seems that, sooner or later, those clouds will generate rain, rain on the parade of the trusting user.

-- Harrison Eiteljorg, II

About this document:

All articles in the CSA Newsletter are reviewed by the staff. All are published with no intention of future change(s) and are maintained at the CSA website. Changes (other than corrections of typos or similar errors) will rarely be made after publication. If any such change is made, it will be made so as to permit both the original text and the change to be determined.

Comments concerning articles are welcome, and comments, questions, concerns, and author responses will be published in separate commentary pages, as noted on the Newsletter home page.