a:link, a:visited 
{ 
  text-decoration: none;
  background-color: green;
  border: 1px solid blue;
  color: black;
  display: block;
  width: 200px;
  text-align: center;
  margin-bottom: 1px;
}

For id selector (the # should be put in front of the id):

#mainPoint 
{
   font-size: 24px;
   font-weight: bold;
   background-color: red;
   opacity: .7;
 }

<span id="mainPoint"> content </span>

For class selector (the . should be put in front of the class)

.highlight {
   font-size: 20px;
   font-weight: bold;
   font-style: italic;
   background-color: green;
   opacity: .6;
 }

<p class="highlight">Lorem ipsum </p>

:link (to add style to an universal link)
:visited (to add style to a visited link)
:hover (to add style to element which we mouse over it)
:nth-child (to add style to nth child of the element)
:lang (to define language of certain element)

1. The No Free Lunch theorem applies very heavily in supervised machine learning as there is a large number of learning algorithms that can be utilised. The choice on which one to use depends on the size and structure of the datasets.
2. Decision tree alhorithms are one of the examples. This are fast and easy to train. Hence, it remains a favourite in machine learning. An example would be yes/no answers for an input such as presence of fever and whether or not it indicates pneumonia. More data can be incorporated into the decision tree to make a more complex algorithm.

1. The best source would be the journalist themselves. However, since we cannot be in two places at once, we have to rely on other people as sources.
2. Sources should be transparent and reported. They should not remain anonymous so that the credibility of a reported would not be jeopardize.
3. Sources can be anonymous if real harm could come on to them when the news article is published. However, we need to weighed it carefully because if the case is called onto court, a journalist isn't protected by law.
4. We can gather information though interview, crowd-sourcing, documents and conference.