'Cloud' is safe from 'black hole'

Where are our files most secure? Are we better off keeping them on a single hard disk, a technology prone to failures and other mischance, or are they more likely to survive if stored on multiple servers around the planet?

For better or for worse, the second answer is the obvious choice. Store your files on multiple servers, and they're tucked away in what's commonly called the "cloud." If one data center goes down, you're backed up by another, your files always available.

Now this fits interestingly with another trend, the use of programs that run over the Web, programs like Google Docs or the powerful office suite from Zoho (www.zoho.com). We need to realize that we are in a transitional moment for human knowledge. You might consider it a kind of informational "black hole."

Consider: We have books from the earliest days of printing and manuscripts from before that, preserved and available for scrutiny at university libraries and in some cases on the Net.

But what about all the information we've been committing to hard disks and other unreliable storage media, such as CDs? Books, especially volumes printed on low-acid paper, can survive for centuries, but I have audio tapes from the 1960s in reel-to-reel format that are unplayable not only because it's hard to locate the proper equipment, but because the tape itself is so delicate. Our VCR tapes, many preserving priceless family moments, are equally fragile, just as early color photography tends to lose its vibrancy in short order.

We need harder forms of storage, in other words, and while the industry figures out how to achieve it, we're living through an era that may not translate well in the future.

The problem with today's information is that it needs continual refurbishment. Its guardians need to move it from one format to another to preserve it, making us all librarians -- information custodians -- of our private collections. The sheer volume of data being generated makes this a challenging process, but if we don't do it, future researchers won't be able to make sense out of CDs or diskettes that have long become unreadable.

All of this is why moving computer functions onto the Net is a stop-gap for information loss.

Widely distributed data are less likely to be compromised by a single, catastrophic problem. And with storage becoming less and less of an issue (in fact, our computing infrastructure seems to double storage capacity in roughly the same way that Moore's law predicts a doubling of computer power every 18 months or so), the limits on how much we can put into the "cloud" give way to the still-thorny problem of keeping the data secure.

Into this environment comes Google's new browser, called Chrome.

Most of the reviews I've seen compare the browser to Internet Explorer or Firefox, or other browser entries like Opera and Safari. On that score, Chrome does modestly well. On many Web pages it's quite fast, and its re-positioning of tabs at the top of the screen is sensible, with a protective wall between what is happening in one tab and what is happening in all the others, to avoid any one page crashing the browser.

I also like the ingenious Omnibox, which combines the search box and the address bar. But these useful features tell only part of the story. Chrome, while currently in beta-testing mode and not yet available for any operating systems but Windows, is more than a highly intelligent re-thinking of what a browser should look like. It also rejiggers the basic concept of what a browser is for.

Here's how: In its long-term quest to strip away Microsoft's hold on office software, such as word processors and spreadsheets, Google has been developing Web applications that have been, by comparison, underpowered, slow, and accessible only through Web surfing software. The solution is to re-make the browser itself by providing it with the tools needed to make Web programs behave like desktop programs. That means tuning up the key tools, which is why Google has completely made over a technology called JavaScript.

Google's so-called V8 project is aiming at a massive performance increase. You won't see that improvement yet -- most sites aren't designed to take advantage of these changes, and again, we're dealing with beta software. But watch Chrome for a few months, and you'll begin to see its power emerging in more responsive and powerful Web programs.

Google is hardly the only company exploring this turf. Microsoft is deeply involved, and so is Adobe. All sense the power of the "cloud." And if we ever do move away from desktop software for many daily tasks, using some local storage but leaving most of our information on the Internet, then we are looking at a possible fix for the information black hole problem. It's plausible to suggest that within 10 years, the data loss we are experiencing through our spinning-platter hard disks and delaminating CDs may be supplanted by robust and redundant storage in an online setting. No one would doubt such methods are sufficient.

We still need rock-solid hardware storage at the local level, failure-proof devices we can plug into machines at home and in the office to save the items we choose to keep off the network storage cluster. But the move toward Web-based software slows the trend of data loss, ensuring that the bulk of what we do now can be adequately preserved and read by future historians.

The great infrastructure project of the digital age will be the preservation of future history, the kind of gift a wise civilization would want to pass along to its children.