I immediately checked to see if Garrison has a Twitter account. While my search was not exhaustive, I didn’t find a likely candidate. (Call me crazy, but I’m pretty sure that the user claiming to be “garrisonkeillor” is an impostor).

This makes sense. Twitter gets a lot of abuse from people who don’t actually use it, and fair enough: when you first hear the thumbnail description of the service, and if you’re still getting used to the idea of blogs, Facebook, and the rest of what the kids are calling “web 2.0″ and “social media,” it’s easy to assume that such a format is good for nothing but the sharing of trivial information.

I’ll be the first to admit that Twitter users share a lot of trivial information (I’m certainly guilty), but the great thing about this and other services is that the recipients don’t have to keep listening. As soon as someone’s feed—or “stream,” as the whole combined Twitter/Facebook/etc. output is often described—bores you, you can opt out.

My friend Brad wrote about this recently, comparing the relative intrusiveness of unwelcome/unneeded information sent by Twitter to that sent by email:

If you are like me, you have probably heard a number of friends complain bitterly about Twitter (and, to a lesser extent, Facebook status updates) by saying something like: “Why do you think I care what you had for lunch?” It’s a fair enough question if you discount the opt-in nature of most social media. That is, if your analogy is “Why would I want an email about what you had for lunch?”

But that’s a false analogy — I’m not emailing you, and if I were, I would definitely not email you my lunch menu. It would be rude. But, there may be some people who might find it interesting that I am eating at a particular restaurant, or eating a particular dish, or just that I’m having lunch. The transaction cost of letting them know is near zero, and the burden on others’ attention is near zero too.

Besides, to anyone who actually uses Twitter, it quickly becomes clear that the people who lampoon the medium by saying “I don’t care what you had for lunch” are telling us more about the limits of their own imaginations than they are about the limits of this mode of communication.

This morning, I experienced an unexpected upside to the opt-in nature of “the stream.”

Most mornings these days, I try to get in a little work on my novel (tentative title: Learning to Lose). I’ve decided that two of the supporting characters—a married couple—are struggling to conceive, and I wanted to know what sort of books, gadgets, medications, and related items they might have lying around. My first stop was the Wikipedia entry on infertility, and I did find it to be a good start.

Then, on a whim, I posted the following question on Twitter, which—because I’ve set up my account this way—then updated my status on Facebook:

For my novel, can anyone tell me some things a couple undergoing infertility txmts might have lying around? (Hold the ribaldry, please.)

In response, I got the following advice. Even the non-writers among you will easily be able to see how much more usefully, idiosyncratically human these details are, as opposed to the more schematic view offered by something like a Wikipedia entry.

Commenter 1: pregnancy tests, sex toys, porn, alarm clock (so you can get up early enough to get ur daily ultrasounds done before work), under eye concealer, knitting/books/etc for dr’s ofc waiting rooms, lots of rx meds including gross stuff like vaginal suppositories, bedside calendar & pen for recording “activities”. and no, i’m not in this process. i just happen to be in the car w/a doctor & parent who’s gone thru this. she sez feel free to contact her for more info if needed.

Commenter 2: Baby name book, syringes, what to expect when your expecting. they are already living as if they are expecting a baby.

Commenter 3: prescription drugs with names like “medroxyprogesterone-135.” Mine were actually compounded specially by a local pharmacy. Many women use a cream or gel for this purpose, but I couldn’t sleep with the constant sensation of having peed myself. Let us know if you have other questions.

Commenter 4: Thermometor! Sutton, never having “been there” I did egg donation for a cousin in 1990..let me recollect..(sometimer’s disease strikes in your 50′s)..I’ll get back to you on this one.

And so on. The comments are still rolling in, and, from the sound of some of them, it seems clear that I’m gaining insight from people with personal knowledge of the subject. This last fact leads to two points I wanted to make about the opt-in nature of “the stream” and how well it worked for me in this situation.

1. For one thing, I had no idea that I knew anyone with personal knowledge of fertility therapies. But since my tweet/status update was going to be seen by, potentially, hundreds of people, I figured it was worth a shot.
2. Even if I had known which of my friends has personal experience with this issue, I might have been reluctant to approach them with questions about such a potentially emotionally fraught issue—especially since, having been contacted directly, they might have felt obligated to help me out, even if they didn’t really feel up to it. But because tweets and status updates aren’t “to” anyone, I could just put the request out there and see who felt like responding. No one was put on the spot, and no one shared anything more than they wanted to.

Some final thoughts: I’m a Garrison Keillor fan, for the most part, so I hope this doesn’t sound like an attack on him. I think he and many other people misunderstand Twitter, but, on the other hand, if users who willingly refer to their posts as “tweets” can’t have a sense of humor about the whole concept, they are taking themselves way too seriously.

And while, I haven’t had lunch yet, I had eggs for breakfast.

As I related in my last Mozy-related post, I started using the on-line backup service in late 2007, and all went reasonably well at first. The initial backup took well over a month, but — once it was complete — it was a great load off my mind to know that I had 30 gigabytes worth of photos, music, personal and work-related documents, and applications and settings saved remotely. My data was protected against both machine failure and — because my data was spread out all over Mozy’s servers — theft, fire, or tsunamis.

Then my hard drive failed. I had Computer ER, a computer-repair shop here in Missoula, replace it with a new one, and looked ahead to restoring my files quickly and easily from my Mozy backup.

There are several methods of restoring from a Mozy backup, but the company’s tech support told me that the ideal method for my situation was what is called a “web restore” (i.e., downloads, as opposed to using the Mozy application on the desktop, which didn’t seem able to find all of the files I had specified for backup).

I signed onto the Mozy web site and began my web restore. I selected the files that I wanted restored (which turned out to be far from all of them) and submitted my request.

In a few hours, I received an email from Mozy indicating that my requested files were now available for download. Following the link, I found a web page with about seven downloadable files, each named something like “download 1” and each containing about 1 GB of data. I also learned that these downloads would only be available for seven days, at which point I would need to start the web restore process over from scratch.

Over the next few days, I downloaded all of the seven files, which turned into “disk image” files on my desktop. These had names like “restore_2009_02_27_17_23_572237.dmg”; when I expanded these files, they generated “mounted disks” with names like “345802.2.dmg.”

As I explored the various mounted disks, I discovered that fully five of them contained a folder named Documents, none of which were complete.

• One version of Documents contained the subfolder “writing” but nothing else; another version of Documents contained a version of “writing” and several other subfolders, etc.
• My Documents subfolder “Clients” was available in several versions, none of which contained all of my original Clients subfolders.
• In turn, one version of my Clients folder would contain the subfolder for a given client but only some of the completed projects that should have been in that folder; the rest were scattered across several other versions of the Clients folder, in turn contained within several versions of “writing,” within several versions of Documents, etc., ad nauseam (and I do mean nausea).

What all of this meant was that I could not easily drag and drop the contents of the restored files back into the proper locations on my hard drive. If I dragged Documents from one mounted disk onto the hard drive, the next version of Documents I dragged in would erase the first one. I couldn’t even drag and drop the first level of subfolders, since, for example, I had several versions of the subfolder “writing,” each containing different files.

After mucking about in disbelief for a while, I determined that the only way to reassemble the files and hierarchies I had before would be to go folder by folder, starting at the lowest level of subfolder and working backwards to the main folder, combining files as I went. For my Documents folder alone, this would have meant having five finder windows open at once in order to be able to compare the contents of the various subfolders, so as to ensure I wasn’t leaving anything out.

At a rough estimate, this would have taken me days and days. I quickly gave up on this method. Too hard to monitor, too hard to be sure I was being complete, too much time wasted. Instead, I pulled all of the versions of the contents of my various restored fragments of my Documents folders into specially named folders on my hard drive. It’s all there, and I should be able to find everything by using my laptop’s search function, but I can’t just navigate through folders to an old project like I could before.*

To say this is not the outcome I’d been expecting from a restore would be an understatement.

Thinking maybe I’d missed something, I checked the Mozy web site. Nope, they do advertise their service as being “easy to use,” and — as far as I can find — there is no fine print reading “unless you actually need to restore anything” or “but of course your file hierarchy will be left in smoking ruins.” The site also presents a list of snarky alternatives to Mozy (as in, only an idiot wouldn’t use Mozy), such as “pay $200/year for an online backup service that uses old, mediocre software.” In fact, after this experience, it is difficult to imagine software more mediocre than Mozy’s, at least when it comes to the actual restore process.

My upgraded hard drive has approximately four times as much space as my old one. Looking ahead to the kind of shambles that would result from this sort of restore of that much more information, I quickly decided that there must be a better way and decided to cancel my Mozy account. This put me on the radar of the company’s Customer Retention department; a representative contacted me, writing:

”We would love the chance to work with you to resolve any issues you have and to make sure your experience with Mozy is positive. If you like, I can escalate your issue and get any problem or concern is resolved quickly.”

In response, I explained the problems I’d been having in essentially the same terms as above, and — saying that I felt that Mozy was advertising more than it was capable of delivering — I reiterated my desire to cancel my account and requested all of my money back ($5 a month for about a year, so around $60).

I take it as a final verdict on Mozy that the next contact from Customer Retention was not an explanation of how I’d gotten something wrong, how the restore process is really easy if you just take these additional steps, etc.

Nope. The next contact simply advised that my account had been cancelled and a whole $5 had been refunded to my account, and expressed the “hope that we can do business in the future.”

Not likely.

_____
*I also discovered that I had apparently not saved all the files necessary to restore my calendar; that was my fault, not Mozy’s, but it’s a good word to the wise that even backing up requires some technical knowledge to do well.

I’ve been using Mozy since the fall of 2007, when I stumbled across it while researching backup options. I was about five months into working as a full-time freelancer, and I was getting more and more worried by the fact that not only my photos, music and personal files but also the work I was doing for clients were all unprotected.

At first, I was going to get an external hard drive and use a synchronizing program, but the pessimist in me was quickly swayed by the arguments for remote backup systems, such as Mozy. An external hard drive is a great way to protect against data loss from machine problems (disintegrating hard drive, file corruption, etc.), but what about fire or burglary? In one of these worst-case scenarios, if the external hard drive is in the same location as your computer, the backed-up data might be gone, too.

I was also swayed by the price difference. Setting up an external hard drive and a synchronizing program would probably cost at least $100. Mozy costs $5 per month and protects against anything from accidental deletion of a single document to total loss in a tsunami (admittedly uncommon in Montana, but you can never be too careful).

I also thought that an online backup solution like Mozy just seemed elegant. Like Gmail, Flickr, Google Docs, and so many other similar services, Mozy felt like it was letting me tap into the distributed “survivability” that was the original purpose of the entire internet — which, after all, was originally designed to support the continuity of the U.S. government in the event of, say, a nuclear exchange.

So, in November 2007, I signed up for Mozy’s Mac beta version and started backing up.

The initial upload of the data that I wanted backed up took about a month or two. That’s nothing against Mozy — I ultimately selected close to 30 gigabytes of data for backup, and I was using a cellular modem in an area with poor tower coverage, resulting in low bandwidth — but the amount of data and the connection speed weren’t the only factors slowing things down.

The additional factor was the frequent errors that caused the service to disconnect itself, which meant that — when I wasn’t keeping a close eye on the upload process — I would sometimes lose a day or two in what was already an immensely long process. When I complained to Mozy Support (available to “home” users only by email and — during business hours — web chat), I was told that these interruptions were “known issues” for which there was no fix but to uninstall and reinstall Mozy on my computer.

And these interruptions didn’t go away once I had completed my initial upload. I had to go through the reinstall process about once every two months for the year and a half that I was using Mozy. This was understandable as long as I was using a “beta” product (the software industry’s term for a product that is still being tested and improved), but the product came out of beta while I was using it, with no noticeable reduction in the problems I’d been experiencing.

Frequent reinstalls of the Mozy application wouldn’t have been such a big deal, except that — each time — I also had to re-select the files I wanted backed up. When uninstalling Mozy, there is an option to save your configuration, but I learned early on that doing so also “saved” whatever problem I was having. It seemed that the only way to successfully eliminate the errors I was experiencing was to utterly wipe the application, its log, and its configuration settings from my machine, and to start from scratch after reinstalling, including reselecting which files I wanted backed up.

And having to reselect what files I was backing up introduced the risk that I might overlook something I had selected in a previous configuration, which, in turn, would signal to Mozy that I no longer wanted that particular file backed up.

As an aside, another potential problem with Mozy for some people is that, if you (1) back up a given file and then (2) accidentally erase that file, Mozy only keeps the backed up version of the file for 30 days, after which it interprets that file’s absence from your machine as evidence that you want it to become absent from the backup as well. If you think about it for a moment, this is a pretty strange policy, since it means that — if you don’t spot an accidental deletion within 30 days — you’ll never be able to restore that file. It also means that you can’t store something on Mozy long term, say 5,000 photos you don’t have room for on your machine while you are waiting to upgrade your hard drive, or something like that.

Still, for all these limitations, Mozy seemed like good value for the price. Then, earlier this month, I needed to restore everything, and I learned the hard way just how little that $5 per month was really buying me.

Next, Part 2: Mozy is great okay… until you need to restore. And isn’t that kind of the point?