1. In order to build an unclassed (continuous) scale in observable, you can create a linear scale. Typing in for the domain d3.min(data) and d3.max(data), you can input the minimum and maximum values of your data set. You may also type d3.extent(data) which will the include the minimum and maximum values. Then, in order to input colors, you type the colors you would like to be displayed in parentheses following the code .range. The first color you use indicates the lower range, and the last color you use indicates the high range. So, the domain helps you to input the extent of your data, and the range allows you to decide what the visual output will be.

2. In order to implement classification in ObservableHQ, you can use the d3 library in javascript to help visualize the data in different classification methods. This will categorize the data into varying colors to represent the classes. For quantile classification, you can use the quantile scale that is in javascript. For the domain, simply input the data set. For equal interval classification, you can use the quantize scale in javascript. For domain, enter minimum and maximum values of your data range. For natural break classification, you can use scale threshold. In order to determine the number of classes shown in each classification method, you input varying colors in the range. For 5 classes, I placed five different colors.

3. In order to determine the number of classes, I used the calculation: 1 + 3.322log10(n), where n indicates the number of data point in your data set. For my data set of 64, I calculated 7 classes by rounding to the nearest whole number. The number of classes chosen should lie between 5 and 20 typically.

4. To select colors from ColorBrewer, I went to the ColorBrewer website. On this site, you can imput the number of classes you have in your data set. Then you can chose from their selection of color schemes of either single hue or multi-hue. When you select a hue, they come up with the same number of colors as you have classes. You can then copy and paste those color values into your observable notebook. To have color scales in Observable, you input the colors you chose in the .rang. Put each color in quotation marks, and separate each color with a comma.

5. You have to make sure that your values are first normalized before mapping the values to colors. You can then input colors from ColorBrewer into your range. The amount of colors you input will indicate the number of classes in your data set. The classification methods will separate those classes following that classification method's rules. For all of the different classification methods, I input 7 different colors. This allowed for 7 separate classes for each classification method. The lower values will have a lighter color, and as they grow larger, they become more darker.