Basic Examples (1) 

Generate a random (probably-fictitious) metal subgenre:

In[1]:=

Out[1]=

Scope (3) 

Here are five random (probably-fictitious) metal subgenres:

In[2]:=

Out[2]=

It may be easier to read as a Column:

In[3]:=

Out[3]=

Using UpTo provides a different way to get five random (probably-fictitious) metal subgenres:

In[4]:=

Out[4]=

Here is an array of random (probably-fictitious) metal subgenres:

In[5]:=

Out[5]=

The nesting is easier to visualize using Grid with the Frame→All option:

In[6]:=

Out[6]=

Possible Issues (2) 

RandomMetalPseudoSubgenre[n] returns unevaluated when n is any data type not mentioned in "Usage":

In[7]:=

Out[7]=

In[8]:=

Out[8]=

In[9]:=

Out[9]=

In[10]:=

Out[10]=

RandomMetalPseudoSubgenre uses RandomInteger and RandomSample under the hood to create as much randomness as possible. For small samples, this is generally enough to avoid duplicates:

In[11]:=

In[12]:=

Out[12]=

Even so, duplicates may still exist in larger examples:

In[13]:=

In[14]:=

Out[14]=

In[15]:=

Out[16]=

Neat Examples (6) 

While all of the examples are neat, some stand out more than others. For example, sometimes, you get really long subgenre names:

In[17]:=

Out[17]=

In[18]:=

Out[18]=

And sometimes, the subgenre names are really short:

In[19]:=

Out[19]=

In[20]:=

Out[20]=

On average, however, the lengths of the subgenre names tend to be more average. Experimental evidence suggests that the Mean and Median of the subgenre name lengths are both around 50:

In[21]:=

Out[21]=

In[22]:=

Out[22]=

To make metal seem more academic, you can test the Mean/Median hypothesis by using StringLength on longer and longer lists of subgenres. This yields a pair of datasets (where x is the length of the subgenre list and y is the Mean/Median of the corresponding list's string lengths) which seem to support the hypothesis.

Here is a plot of the corresponding Mean data:

In[23]:=

Out[1]=

And here is one for the Median data:

In[24]:=

Out[3]=

The trend can be summarized by taking the Mean of the means and the Median of the medians. Again, the results are very close to 50:

In[25]:=

Out[25]=

In[26]:=

Out[26]=