The Third Age

I think that we are entering (have entered?) a third age of computing. The technologies coming on-line (XML, RIA, Web 2.0, and such) are giving us the capabilities to create new types of applications.

The first age is what many folks think of as 'the mainframe age'. Some folks will call it 'the bad old days', and other folks will wax nostalgic over it. I look at the ages in terms of data and how that data was organized, who organized it, and who could access it.

In the first age, data was structured and centralized. (Just like mainframe systems like it to be.) The typical applications were business applications: general ledger, payroll, inventory, and billing systems. The technology used was mainframes and COBOL, but then, that was all we had. (I am conveniently ignoring FORTRAN, but while FORTRAN applications were for science folks the data was still structured and not really shared.) The system owners organized the data and the users had to comply with the imposed structure. Data was not shared, at least not very much. And data was limited to the business problem at hand. The 'killer app' for the first age was the 'accounting set' of the aforementioned typical business applications.

The second age is 'the PC revolution'. Data was non-structured and distributed - everyone had their own computer and stored data as they saw fit. The main applications were word processors and spreadsheets, running on PCs with either MS-DOS or Windows. (I'm ignoring the pre-PC microcomputers such as Apple II and Commodore PET. I'm also ignoring minicomputers and Unix.) The primary users in the second age were businesses, as in the first age. The data did not follow the first-age flow of 'input-process-output' but was composed and then distributed, usually on paper but sometimes on floppy disk or BBS. There were many originators of data but each originator had few recipients. The killer app was the spreadsheet, but the word processor helped.

The third age changes the rules again. The data is not the rigidly structured datasets of the first age or the non-structured documents of the second age; data is semi-structured with markup tags. Data is stored on central servers but accessible from anywhere (well, anywhere on the internet).  The technologies include web servers, web services, AJAX, XML, and other 'new stuff'. Typical applications include LiveJournal, MySpace, FaceBook, LinkedIn, and DOPPLR. I don't know that we've seen the 'killer app' for the third age (yet).

The big difference in the third age applications is the user set. The third age is driven by customers, not businesses. Individuals enter the data (journal entries for LiveJournal, contacts for LinkedIn, and travel plans for DOPPLR) and share it with others; the sites do not create the data and provide it to users. In third-age applications, the users who enter data are also the users who share the data and retrieve the data. (Not always in proportion to how much they create.)  The site provides the API and the storage area; the users provide the content and organize it themselves.

I don't see the applications of the third age replacing earlier applications. Instead, I see the new applications opening up new functions and features. Just as the second age expanded the computing universe (we still have accounting running on mainframes, don't we?), the third age apps (OK, if you insist 'web 2.0 apps') will create new capabilities that do not replace our beloved word processing and spreadsheet applications.

There may be some overlap, for example collaborative document composition, and I predict that some applications will move from the PC to the web because the data sharing model is a better fit. (Project management, for example. MS-Project will die as a Windows application and move to the web. The data sharing needs of team members will force this change.)

The lesson here is to embrace the new technology and more importantly, embrace the new capabilities for sharing data. Not all applications have to use them, but some will, and we will all be more effective with the change.