Build 30 Workstation Computers

Several years ago I worked for a company that experienced a sudden need for thirty personal computers. My boss asked me to "build thirty workstations" for Monday (it was Friday afternoon).

I'd built home computers before as a hobbyist, so I was comfortable with the idea of assembling the machines. And in my enthusiasm for completing my job I neglected to make even a rough estimate of the time required for the tasks involved versus the schedule. So I spent Friday evening considering required specs and identifying vendors with the suitable parts.

An optimistic early Saturday start on the shopping turned into a late evening of despair. No one vendor had all the parts (or sufficient quantities), so far more driving and inquiring was involved than I had expected.

Sunday morning found me in the office with a mountain of parts in packaging. I quickly regretted my choice of low-end compact case. Fitting the parts inside the cramped case was a hassle with each machine. I didn't experience any efficiencies from the repetition. The tools I used to assemble the machines were a problem, it was all branded vendor swag from a conference a few weeks earlier.

I made a trip to Fry's after a few hours and bought better tools, but it was still an RSI-inducing experience. The resulting computers were serviceable, but quite loud. I was concerned about cooling, but didn't want to bother with actually measuring the cooling required, so I went overboard and put several powerful fans in each machine. As the optical drives spun-up while I installed Windows and Office (from separate media), the noise level was deafening. It should have occurred to me earlier, but it wasn't until Sunday afternoon that I lifted my head from my task and realized that I should have disregarded my employer's literal instructions and simply fulfilled the intent of his request.

I could have gone to Costco on Friday afternoon, picked up a pallet of machines with pre-installed OS and productivity software, and spent the weekend doing something fun. And it would have cost several hundred dollars less per machine. And the machines wouldn't have been so intolerably loud. I've long since left that organization, but for all I know there are still employees working with the home-made boxes, loud as hair-dryers.

Dave Says

This is a great example of how rushed planning and picking parts without too much consideration can result in a computer that isn't what you wanted it to be. Spending the additional time to check the reviews and benchmarks really does pay off in the end.