Who is Yourself?

The cover of the current issue of the Communications of the ACM is about End-User Development. Which prompted me to reflect about the history of "Do-It-Yourself" IT. Because, time was, when "doing-it-yourself" referred to End-Users.

The first wave of DIYIT that I recall was when "rocket scientists" hired into trading groups on Wall Street became disenchanted with bureaucratic and unresponsive IT departments which dealt mostly with mainframes -- so they went out and bought Sun workstations in order to do it themselves.

Sun, of course, went up-market, built machines more like mainframes, and started selling to IT departments instead of people.

So the second wave was when professionals (investment bankers, accountants, department managers) became disenchanted with bureaucratic and unresponsive IT deparments which dealt mostly with mainframes and Unix machines -- so they went out and bought "IBM-compatible PCs" running DOS in order to do it themselves.

But somehow, IT took over the management of PCs running DOS (well, Windows, actually). So, now, there is a third wave of "Do-It-Yourself" IT. In this one, however, the "yourself" in question is not the person trying to do "real work", but the IT professional herself. It would seem that IT professionals are becoming disenchanted with bureaucratic and unresponsive vendors -- so they are going out and downloading open source software in order to do it themselves.

The question I have to ask is: how come IT needs to be involved? In the previous two DIY revolutions, the whole point was for "real people" to bypass IT and get the job done. Is it that non-IT users today are less technically sophisticated than they were 15 years ago? Or is it that commercial software (SunOS and DOS) was easier to use and administer back then than open source software is now? Or is there some other explanation? Why shouldn't this wave of DIYIT leave IT out, too?

This whole line of reasoning was kicked off by associating "End-User" with "Yourself" in "Do-It-Yourself" IT. I.e., who is "Yourself"? The other train of thought to come out of this idea was -- assuming that IT should somehow be involved -- who in IT is that someone?

The first time I attended a panel discussion about open source software, all the vendors on the panel made the same argument. The audience was IT management, and the argument was that the staff (SAs, DBAs, developers) were sneaking Linux (in those days, "open source" meant "Linux") in to the organization unbeknownst to them. So, they (the managers) had better get started buying more Linux services from these vendors. It seemed to me that the implication was that open source software was causing IT management to lose control of their departments. Just as PCs and workstations before them caused IT management to lose control of computing budgets. The solution back then was for the vendors and IT management to work together to wrest control of the computing technology away from "end-users" and back into the hands of IT management. And at this panel, it appeared that the strategy for open-source vendors hadn't changed: work with IT management to prevent "regular people" -- be they users or SAs -- from "doing-it-themselves".

That means that the "yourself" in "do-it-yourself-IT" doesn't include IT management. That means that I'm not one of "you", I'm one of "them". Darn.

Maybe this wave of "do-it-yourself-IT" is going to be the one where IT managers get to be one of the "yourselves". I sure hope so.

But I keep coming back to the following (and my final) thought (and if you know someone who works in IT management, you can ask them this question):

Who gets to decide which software can be downloaded off the web and installed on computers connected to the corporate network (servers or desktops)?

Are those people the "yourself" in "do-it-yourself"? Does that mean the people who don't get to make that decision can't "do-it-themselves"?

Or, asked another way -- before we can truly have "do-it-yourself-IT", don't we need to have networks where people can be allowed/trusted to download and install software on their own recognizance? Because the infrastructure is sufficiently resilient to tolerate mistakes?

If one of the functions of IT is to prevent people from installing software that they think is (or might be) appropriate to do their jobs or to be more productive, how can anyone "do-it-themselves"?