Monthly Archives: February 2006

Mainstreaming the mainframe

Back in the 70s, when personal computing entered the mainstream, they came about as a means of providing individuals a tiny piece of the computing power available to governments, corporations, and universities in the form of mainframes. But in contrast to the mainframe/terminal paradigm, in which the network is an inherent aspect of the model, personal computers were mostly islands. Rather than being delivered via a network, virtually all data input came via a floppy disk or the like. Networking for personal computers would not become a mainstream reality until thousands of miles of fiber optic cable later. While mainframes efficiently stored and managed all data and provided all processing power centrally on the mainframe mother ship, with dummy terminals limited to input/output, the personal computer ended up being like little pioneers, required to be completely self-sufficient in terms of processing power, data storage, and software. Even after the World Wide Web entered the mainstream in the mid 90s, it would be quite some time until broadband networks would become widely accessible and allow for offloading some of that burden to remote servers, so the little pioneers settled and became beefy workstations with massive storage capabilities and mainstream processor speeds eventually surpassing that of supercomputers of old. This evolutionary path has continued in a steady progression, to the point where personal computers are starting to look like their own little mainframes, at least in terms of the extent to which they need to be maintained—owning a personal computer today means being a mini-administrator, dealing with software updates, security issues, networking, etc. But, of course, personal computers are not mainframes. More importantly, they have long since ceased being islands. As networking becomes an increasingly central aspect of personal computing, the personal computing paradigm as we’ve now known it for the last quarter century is making less and less sense. It’s making less and less sense for files and applications to reside on a local machine, considering that more and more people work on multiple computers, and more and more people have access to broadband connectivity, which negates so many of the original reasons why personal computers came about in the first place. In my view, what instead would make sense would be for a mainstreaming of the mainframe paradigm. In some ways, that process already is underway, in the guise of the Web 2.0. web-as-platform model. Building on that, I would like to not only store these blog entries online, but all my files (some of which of course I would choose to keep private.) In addition, instead of sitting here typing this in Word, a desktop application (yup, even though it will eventually end up online, doing a lot of typing in a text area isn’t exactly an optimal user experience, so I always type everything first in Word), I would want to be able to have a web-based version of Office, in which it was provided as a service, and I never again had to worry about installing it or keeping it updated or having to attach Word docs to emails whenever I wanted to send it to someone. After having typed this, rather than clicking on a save button (which would be gone, since it would always be automatically saved), I would just publish it, without the need for any FTPing or anything like that, since it would already be online. Then, instead of sending someone an attachment, I’d just send them a link. And let’s say I wanted to collaborate with someone on writing a document – well, I would just need to give them access to it, and suddenly we’ve got a collaborative workspace, or I could open it up even more and turn the document into a Wiki. I could go on, and talk about all the advantages of moving personal files to an online space and turning desktop applications into either free online apps or subscription-based online apps (e.g. no more dealing with backups of your files, since that would be part of the subscription service), but the funny thing is that there really is nothing about these ideas that is infeasible. It’s more a question of whether or not companies that have a strong interest in retaining the status quo of personal computing, such as, oh I don’t know, Microsoft? (and Apple too), would have the boldness and vision to make their products available as an online service. The good news is that there are companies, such as, oh I don’t know, Google? (and Yahoo! too), that in many ways already are either realizing or creating the momentum for just such a shift.