One of the prime concerns of The Computer, and indeed of all citizens of high clearance, is how to ensure one gets accurate and unbiased feedback from clones of lower clearance. It may come as a shock to some, but joint research conducted by R&D, IntSec, and CPU has revealed that under certain circumstances, as many as eighty-seven percent of all citizens of clearance GREEN or below will supply the answers they believe a questioner of INDIGO or higher clearance wants to hear, instead of accurate and truthful information. (Naturally, the citizens shown to do so during the research were convicted of treason, but R&D assures us they have weighted the test so as not to be biased by that.)
As I'm sure all readers of this report know, having accurate information is vital to the successful creation and execution of any plan devoted to the improvement of Alpha Complex. (I myself blame inadequate information as to the state of the power lines in the wake of the Toothpaste_Disaster for the 3000% price spike that brought such profits to the now-defunct WhirledRON_Energy Service Firm -- had its operatives had information, as I did, that those lines simply could not support such rapid reversals of current flow, they wouldn't have bounced that electricity back and forth so many times.) So how to ensure accuracy in our data-gathering, preferably requiring a relatively small number of executions for treason?
Many programs have been attempted, some with more success than others. F-PFS/A/I/DP is somewhere in the middle ranks -- more effective than the 'Tell Us The Truth or We Blow Your Brains Out' initiative, but less so than a carefully graded suite of Truth Serums. Its major guiding principle is that time grants a certain distance to all things. In the heat of the moment, a citizen might be tempted to glorify himself, or cover up his mistakes, or try to appease a superior, all generally in the hopes of gaining a reward or avoiding punishment. F-PFS/A/I/DP is regularly scheduled to occur three weekcycles after whatever event triggered the need for an accurate assessment. This is generally after any standard rewards and punishments for the incident have been levied, and so the expectation of either is lessened. (Of course, the occurrence of such is still fully at the relevant official's discretion.) It is also far enough along that long-term results of the incident which may not have been obvious at first, such as morale repercussions or radiation damage, are clearer and can be assessed more accurately.
The exact details of the program are of interest only to the mid-level CPU functionaries assigned to carry it out, really. What is of most concern to citizens of our rank is that at the end of the process, a half-meter-tall stack of forms, reports, and assessments has been generated, which is assured to have at least 89.37% accurate information -- a 57.02% increase over immediate 'scene-of-the-action' reports in most cases. (All numbers courteously supplied by Artand-R-SON-2, of CPU Accounting Services.) These details are of great use to our Long-Term_Projections.
Some skeptics have criticized the program on several counts. Most notably, they have pointed out that citizens who have died or been convicted of treason in the intervening three weekcycles are not factored into the report at all, skewing the results severely in cases where, for instance, slow-acting effects kill important participants before they can be interviewed for the followup. Also criticized is the fact that, when used, F-PFS/A/I/DP is the last scheduled followup for an incident, thus ignoring longer-term effects that may not be fully evident after a mere three weekcycles. While there is some validity to these claims, the only way CPU seems to have seen to prevent them is to schedule even more followup assessments over the short and long term, a plan which would hopelessly waste valuable Computer resources and generate overwhelming mountains of paperwork. (Fairly typical of CPU and the pitfalls of bureaucracy, some would say -- I certainly can't deny that more competitive Service Firms typically do better when freed of CPU's tangling regulations.)
Ultimately, F-PFS/A/I/DP is a flawed but still useful tool in Friend Computer's arsenal, and quite worth using.
Bureaucracy has its pitfalls, but the lack of bureaucracy is far, far worse. The more competitive Service Firms certainly do more when they ignore CPU regulations, but as for better...over the course of a weekcycle or two, yes, but that's taking the short (and treasonous) view. In the longer span, groups that ignore required bureaucracy begin to suffer from information shortfall and manifold inefficiencies in workflow pattern. Research has shown a cumulative and exponential degradation in ability among such groups, and recovery from this decay takes long enough (typically on the order of months to years) that the short-term productivity boost is simply not worth it.
Added to this is the fact that ignoring the required bureaucracy immediately begins harming the productivity and decision-making ability of all groups other than the shirking organization. While some in the "more competitive" Service Groups might treasonously welcome that, it is not a productive attitude for the Complex as a whole.
Much of the information in this report on the Toothpaste_Disaster would be utterly unavailable and unknowable had the relevant paperwork not been filed and regulations followed. Your negative attitude is a sad example of the bias which, apparently, even reaches the echelons of the High Programmers.
For once, I must agree with my colleague Err-U. Failure to follow proper procedure leads to all sorts of problems. An inadequate paper trail makes it difficult for IntSec and Troubleshooter teams to track down misappropriation of resources and other forms of treasonous malfeasance. Furthermore, logistics get snarled up as JustInTime distribution ceases to be on time, other Service Firms find that resources allocated to them are no longer available, and citizens find themselves mandated to be in more than one place at the same time, a problem that has not yet been satisfactorily resolved by R&D.