Q. What's the difference between a database and a website?

Let's start with a couple general definitions:

• A database is a structured collection of specific content selected to be included. They're optimized for information retrieval (such as by search).
• A website is a collection of pages on a single domain (something.com, something.edu, etc). 

Some websites, like Amazon or Flickr, operate more like databases than "plain" websites, and that might be a good mental model. You can browse Amazon, but probably you're searching by keyword, then applying filters to get things with 4-star ratings, in the color purple, from a particular brand, in a specific price range. Amazon's website is effectively a database of stuff sold by Amazon. Searching is free, but then it costs money to actually get things.

They aren't mutually exclusive! Websites also use databases in their infrastructure to operate. However, that's not the kind of database we in libraryland are referring to. 

• A library research database is a subscription resource that collects formally published information in a system that can be searched and filtered for results. Access is limited to those in a community (verified by library card or other login). 
  • Also, if an instructor tells you to only use the databases for your research, this is the kind they mean!

If you really want to dig in to the differences and distinguishing the two, read on. However, in practical terms, this doesn't matter! It's more important to be able to distinguish the type of information presented by the site-or-database, recognizing whether it's a journal article or a blog post or a newswire feed and so on.

Examples & Discussion

[database] and [website] notes are added where different criteria favor one type or the other.

Consider The New York Times website.

• We access it through a link, and the homepage is at a simple Something.com domain (NYtimes.com). [website]
• There's a search (upper left) but it's not prominent. [website] The search does have a few filters: date, section, and type. [database]
• Many articles are presented right away, and you can browse by section to find more. [website]
• Some content is free, but it does have a lot paywalled. [website]
• We can make an account (especially if we subscribe), but we don't have to. [website]
• Content is formally created by the NYT.

Overall, it has more attributes of a website than a database.

Now let's consider EBSCOhost's Academic Search Complete.

• We access it through a specific link on the library website.
  • The link is complex: it starts with the Lone Star proxy login, which is attached to another url for Academic Search Complete itself. (https://lscsproxy.lonestar.edu/login?url=http://search.ebscohost.com/login.asp?profile=ehost&defaultdb=a9h)
• You can't access any content without signing in to an account. [database]
• This is an expensive resource! The library subscribes on your behalf. [database]
• The only thing you see on the homepage is a search box: no promoted articles, just some links relating to search. [database]
• The content (once you do a search) is a mix of journal articles, news articles, videos, and more from a variety of publishers. [database]

Overall, it has the characteristics of a database.

If we search the web for Academic Search Complete, we get to a product webpage for it. Website characteristics:

• Simple url on the ebsco.com domain.
• We have an informative page with links to more options, like other variations of the ASC database (i.e. browse-friendly).
• There is a site search, but not much filtering is available: only telling it to show news, products, or resources.
• There's no content other than the info about the database: no journal articles, nothing.
• No account needed for access to this site or these pages.

Let's get trickier.

I searched Google Scholar to get a journal article and from there, I went "upstream" to the homepage of Taylor & Francis Online.

• The homepage is dominated by a giant search bar [database], though if we scroll down, there's some "trending research" and other links [website].
• When we search, there's a robust set of filters (article type, language, date, publication, subject). [database]
• Search results include mostly articles, then there's a "databases" tab, which provides 2 "database records." 
• Access is mostly paywalled (only 86 of 2,559 results are open access). [database]
• One of the search results appeared on a Google web search, which is how I got here anyway. [website]

So what is this, website or database?

I'd say it's overall a database, just not one we have access to. Or perhaps it's a website that functions as a database. :)