In 1994, John Seely Brown and Paul Duguid wrote that the three advantages for paper-based newspapers were:

a) the determination of what is news as only a certain amount of information will fit within the paper limits;
b) the layout of the news will assign each news item a level of importance and suggest links between different items; and,
c) uniformity of information as the news are the same for every reader.

I guess these are all still valid but are these still advantages?

With so many (and so good) Internet search engines available, is the amount of information still a problem?

If I don't know the newspaper editors, why should I trust them to decide what is important to me?

If my experience and preferences are different from the person next to me on the train, why should I read the same news as she does?

For those of us not yet fully equipped with mobile e-tools, it is great to be able to get a paper version of the newspaper and take it to read. A newspaper is also a great topic to start a conversation (how many times have you started a conversation with someone because we wanted to know more about what they were reading, or because you wanted to discuss an item both of you have read?) and to learn about someone's preferences and style (you, no doubt, will form an opinion about a person depending on what newspaper he is reading).

The interesting thing is that all reasons for liking printed copies of newspapers, when used in public situations, are exponentially better on an online context. Blogs, wikis, content syndication and other web 2.0 tools are opening new possibilities, creating new working habits, and forever changing the face of news publication.

Reference to Brown and Duguid's paper:
Brown, J.S. and Duguid, P., Borderline Issues: Social and Material Aspects of Design, Human-Computer Interaction, 9 (1), pp. 3-36 (http://www.sociallifeofinformation.com/Borderline_Issues.htm)