Thoughtful, detailed coverage of the Mac, iPhone, and iPad, plus the best-selling Take Control ebooks.



Pick an apple! 
Extract Directly from Time Machine

Normally you use Time Machine to restore lost data in a file like this: within the Time Machine interface, you go back to the time the file was not yet messed up, and you restore it to replace the file you have now.

You can also elect to keep both, but the restored file takes the name and place of the current one. So, if you have made changes since the backup took place that you would like to keep, they are lost, or you have to mess around a bit to merge changes, rename files, and trash the unwanted one.

As an alternative, you can browse the Time Machine backup volume directly in the Finder like any normal disk, navigate through the chronological backup hierarchy, and find the file which contains the lost content.

Once you've found it, you can open it and the current version of the file side-by-side, and copy information from Time Machine's version of the file into the current one, without losing any content you put in it since the backup was made.

Submitted by
Eolake Stobblehouse



Related Articles



Breakfast, Dinner, Launch: A New Design

Send Article to a Friend

Our launch today of the newly designed TidBITS Web site is the result of about 15 months of work in transitioning our old architecture for writing and editing articles, assembling and publishing issues, and feeding out articles and other bits online.

Adam wrote about our back-end transition for assembling issues that are sent out via email back on 11-Sep-06 in "Behind the TidBITS Curtain." That overhaul also meant a new back-end for, the site that served as our article archive.

The move today had a few components to it. First, we revamped our Web site to be more modern without being unnecessarily so. There's a little AJAX hidden here and there - the bits of JavaScript that can talk to a server after a page has loaded to update information or provide live interaction.

That was after a silent transition a few months ago when I rewrote the entire server system developed in mid-2006 from the PHP programming language to perl; you probably didn't notice when we switched from one script to another as the results and appearance were virtually identical. While the two languages have a lot in common, I've learned quite a bit about object-oriented programming (OOP) in perl, and between mid- and late-2006 had used OOP perl to build a full-blown content-management system for TidBITS. (It's called the TidBITS Publishing System or TPS; points for getting the reference before clicking here.)

That silent changeover also moved us to a modular template format that you can pull off in PHP, but isn't how most PHP code is written. I used the HTML::Template module in perl to manage this. In this approach, everything that appears on a Web page is externalized in one or more template files, allowing a beautiful separation of programming code from appearance. This will also make it dramatically easier for us to localize TidBITS for other languages.

By splitting the HTML into a separate file, this made it easier for collaborator Jeff Carlson to work on the design without having to worry about code. He had to learn a bit about the template variables and commands, but not much. We could hand files back and forth to add new features and fix bugs.

Adam will write more about the new design and the process, but I'm incredibly tired and incredibly happy. TidBITS has been longing for years to move into the 21st century. We weren't held back by a lack of ideas, but by needing the right platform, which turned out to be one we needed to build from scratch.


CrashPlan is easy, secure backup that works everywhere. Back up
to your own drives, friends, and online with unlimited storage.
With 30 days free, backing up is one resolution you can keep.
Your life is digital; back it up! <>