Figure 1.

Total accesses on non-school days. Banned accesses are plotted on this graph, although they are just barely visible at the graph's bottom line, since total accesses overwhelms them. For a close-up of banned accesses only, see figure 2.

Figure 2.

Compare with figure 1. This graph plots banned accesses only, magnified at 100 times the scale of figure 1 for visibility. Bans in the sex category make up the majority of total bans, and generally closely parallels the line representing total bans. Bans in the other categories are minimal. There is a single peak for the "universal" category at 7 AM which probably represents one person's determined effort to bypass the censoring proxy server by accessing www.anonymizer.com.

Figure 3.

Banned accesses are plotted on this graph, although they are just barely visible at the graph's bottom line. For a close-up of banned accesses only, see figure 4. This graph is much more bell-shaped than the non-school day graph, because accesses during the hours school is in session completely dominate all other accesses.

Figure 4.

Compare with figure 3. A close-up of banned accesses on school days, magnified 350 times for visibility. This graph is less bell-shaped than figure 3, indicating that on these days, users in the evening (adults) are more likely to be banned than users during the school day.

Figure 5.

Total accesses on non-school days, with images excluded. Banned accesses are plotted on this graph, although they are just barely visible at the graph's bottom line, since total accesses overwhelms them. This graph is plotted on the same scale as figure 1. For a close-up of banned accesses only, see figure 6.

Figure 6.

Compare with figure 5. This graph plots banned accesses only, on non-school day with images excluded, magnified 100 times for visibility. Bans in the sex category make up the majority of total bans, and generally closely parallels the line representing total bans. Bans in the other categories are minimal. Compare with figure 2.

Figure 7.

Total accesses on school days, with images excluded. Banned accesses are plotted on this graph, although they are not visible at the graph's bottom line. For a close-up of banned accesses only, see figure 8. This graph is much more bell-shaped than the non-school day graph, because accesses during the hours school is in session completely dominate all other accesses. Compare with figure 3.

Figure 8.

Compare with figure 7. Banned accesses on school days, with images excluded, magnified 350 times for visibility. This graph is less bell-shaped than figure 3, indicating that on these days, users in the evening (adults) are more likely to be banned than users during the school day. Compare with figure 4.

Figure 9.

This graph shows the percentage of accesses banned at any given hour, comparing school days and non-school days. Note that the lowest points on the graph are between 7 AM and 3 PM, school hours. After school (when most teachers make their lesson plans) it jumps up slightly, and then evening hours between 10 PM and 11PM seems to be the time when dial-up users feel free to be less careful about the sites they access. On non-school days, the late night peak shifts later, roughly 11PM - 2AM, possibly because users need not wake up so early the next morning.

Figure 10.

This is Figure 9, with images excluded. Percentages are overall somewhat higher (remember, viewing unbanned sites generates many images, while only a relatively few images are banned). Anywhere this graph drops or only increases slightly from Figure 9 indicates that many of the bans resulted from banner ads being banned. The most noticeable shift upwards is in the 10PM peak, which indicates that there, most bans were not banner ads.

continue: Category Definitions

back to Table of Contents

This document last updated on Thursday September 07 2000.