While it's true that Atlas got stuck with a pretty tough gig, you have to wonder where the world would be without him. There he sits on the floor of the cosmos, like a dropped marble parked deep beneath a heavy dresser — unseen, yet holding up the world. Infrastructure such as Atlas isn't sexy, and it's interesting only when viewed through the geekiest frame of reference. Nevertheless, its importance could be likened to the importance of tracks to a train.
This sobering reality applies at all scales and in all contexts. Consider a sound designer or composer in a project studio. Remove all word clocks, sound-library organization, project tracking, AC grounding schemes, templates, and studio furniture, and what's left? Not a functional setup, that's for sure. Viewed another way, think about a company that grows from 30 to 300 employees without adding dedicated IT staff. An awful lot of time would be wasted with everyone sitting around because their computer systems broke down, and no techs were available to repair them.
These are extreme examples, but they make the point that infrastructure is of critical importance in enabling a person to perform the work at hand, and it is therefore an easily justifiable area for investment. The problem, of course, is that everyone hates spending money on infrastructure. How many of us set aside significant funds for high-quality cabling, acoustic treatment, and ergonomic studio furniture? Not many, I'll bet. A similar issue arises in corporations. How often have you seen companies try to save money by trimming support staff?
It's not difficult to understand why infrastructure is such a late finisher in budget races, in spite of its huge importance. Infrastructure doesn't make money; there's no profit in it, only cost. Anyone trying to run and grow a business is going to spend money where it can make money. But there is a flaw in this logic that is large enough to make even Atlas stumble: although it is true that infrastructure does not make money, failure to invest in infrastructure can potentially cost more than investing in it.
Examples of this are easy to spot. If you lack a CD duplicator, you can't make money duplicating the CDs that you've mixed and mastered for a client. That duping job will have to be sent elsewhere. After you lose out on that business a few times, you don't need to use quantum mechanics to deduce that shelling out a couple of thousand dollars on a relatively low-volume duplicator could bring in jobs that would pay off that investment in a matter of months.
More difficult to see are the incremental costs. Work goes much more slowly on a sound-design job when the library sounds are scattered across different media, with no central documentation or easy auditioning method other than loading the source media. You can work far more efficiently with a well-organized sound library, accessed from a decent database. Without efficient systems, if you are working on a project for a fixed price, then you are effectively lowering your hourly income. And if you are working at an hourly rate, you are making your price less competitive. Either way, it's going to mean less money in the long run.
I have often made infrastructure improvements in my studio by attaching them to a certain event or project. A prime example is when I have had to move my studio. I have made infrastructure improvements as part of setting up in the new location. Another good hook often accompanies new projects. The very nature of a project may suggest or impose specific infrastructure demands, such as needing to have a sound-library system to do a sound-design job.
Like most people, I have glaring gaps in my infrastructure that are crying out to be filled, but I must contend with time and money limitations. Nonetheless, I have never regretted making a needed and well-researched infrastructure acquisition or improvement, even when it has been time-consuming and expensive. I've always found it handy to have a good Atlas around.