(4) Files

• In ordinary language "file" refers to, e.g., a Word document, an MP3, a digital photograph, etc.
• To a computer a file is simply an abstraction for some arbitrary chunk of information
• A file has (at least) a name, a size (in bytes), permissions (who can read, write, or execute it), and a type
  • Most basic type distinction: text vs. binary
  • Filename suffixes (e.g. .doc, .mp3, .jpg) are only hints - changing them does not change a file's type

(5) Filesystems and Directory Structure

• There's an indirect relationship between files and storage media
• This relationship is managed by the file system
• The interface the filesystem presents to users is the directory structure
• Directory structure is (mainly) hierarchical: directories can contain files and other directories
• Different ways of visualizing directory structure
  • Folders in a graphical user interface
  • Nested lists in command line interface
• Directory structure is increasingly de-emphasized in user interfaces (example: iPhone)

(6) Graphical View of Directory Structure

(7) Textual View of Directory Structure

(9) File Paths

• A path is a sequence of symbols for uniquely describing a file or directory's location in a directory structure
• A path consists of a sequence of directory names, optionally followed by a file name if the path is to a file
• The elements in the sequence are separated by a delimiter character (/ on Mac and Unix, \ on Windows)
• Example file path on Linux or Mac:

  /home/dilanm/Desktop/notes.txt

• Example file path on Windows:

  C:\Documents and Settings\Dilan Mahendran\Desktop\notes.txt

(10) Absolute vs. Relative File Paths

• File paths may be absolute or relative
• An absolute path starts at the "root directory" and always refers to the same file or directory
  • Root directory on Mac, Unix, and Linux is /
  • Root directory on Windows is (for example) C:\
• A relative path is interpreted relative to some current location
• Relative paths may use some special characters
  • . refers to the current directory (the directory we're in)
  • .. refers to the parent directory (the directory above the one we're in)

(11) Absolute vs. Relative File Paths

• Assuming we're in the directory named 'Desktop', the following paths are all equivalent:
  • Absolute:

    /home/dilanm/Desktop/notes.txt

  • Relative:

    notes.txt

    ./notes.txt

    ../Desktop/notes.txt

(12) Why Relative Paths?

• Directory structure varies from computer to computer
• Absolute paths name every directory from the root
• An absolute path that points to a file on one computer may not point to anything on another computer
• A relative path usually only names directories beside or below around the current directory
• Using relative paths, files in or below a certain directory can refer to each other without making assumptions about the directory structure of the whole computer
• Thus relative paths are more portable
• This is important for web sites, which are usually moved around to different computers

(14) Web Servers

• "Web server" is used to mean two things:
  1. Software that receives requests from web browsers and sends back the files that compose a web page
  2. A computer running such software
• A web server is not a special kind of computer
  • If you have a Mac, you have a web server [http://docs.info.apple.com/article.html?path=Mac/10.6/en/8236.html] (Apache)
• In particular, (computers running) web servers have file systems and directory structures just like any other computer
• Note that Unix is far more common among computers being used as web servers than it is among computers being used for personal computing
  • We typically Unix (and Mac) -style paths in this course

(15) URL Paths

• A web URL has two main parts: the domain and the path

  http://[domain]/[path]

• For example:

  http://courses.ischool.berkeley.edu/i253/f11/

  • The domain is "courses.ischool.berkeley.edu"
  • The path is "/i253/f11/index.shtml"

(16) URL Paths vs. File Paths

• URL paths might be the same as file paths, but usually they aren't
• Web server software is responsible for translating URL paths into file paths
• Some URL paths are not translated into file paths at all, but are interpreted in other ways (e.g. looking up something in a database)
• Example of a URL path and its corresponding file path:
  • URL path:

    /i253/f11/index.shtml

  • (Absolute) file path:

    /courses/i253/public_html/f11/index.shtml

• Like file paths, URL paths can be absolute or relative

(19) Resource Identification

• Early hypertext systems usually only identified documents within the scope of a single system (e.g. HyperCard)
• The web uses URIs (Uniform Resource Identifers) as global identifiers
  • http://bcnm.berkeley.edu/
  • mailto:joehall@ischool.berkeley.edu
  • tel:+1-510-684-2179
  • urn:isbn:0451450523

Global naming leads to global network effects... the value of an identifier increases the more it is used consistently

Architecture of the World Wide Web, Volume One [http://www.w3.org/TR/webarch/]

(20) URIs & Resources

• A URI identifies a "resource."
• But what is a resource?
• A resource is anything identified by a URI.
• Resources are a way of conceptually dividing up some information space

(21) URIs & Resources

• URIs do not merely identify; they can be used to access or interact with a resource
• The details of how to do this are determined by the URI’s scheme, e.g. the "http" part of http://bcnm.berkeley.edu/
• The usual method of interacting with a resource is by dereferencing the URI

(22) URI Schemes

URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]

http://courses.ischool.berkeley.edu/i253/f11/geolocation#(4)

• URIs in their general case are very simple
  • the scheme identifies how resources are identified
  • the identification may be hierarchical or non-hierarchical
• Many URI schemes are hierarchical
  • it is then possible to use relative URIs such as in a href="../"
  • the slash character is not just a character, in URIs it has semantics
• Query components specify additional information
  • it is non-hierarchical information further identifying the resource
  • in most cases, it can be regarded as input to the resource

(23) Resources & Representations

• When a URI is dereferenced by a web browser, it receives some data
• That data is called a representation of the resource identified by the URI
• Resources are abstract, representations are concrete
• Resources may have various representations, and they may vary with time

(24) 1 Resource, 2 Representations

(25) 2 Resources, 1 Representation

(27) DNS & HTTP

The two basic protocols which every Web browser must implement are DNS [Internet Architecture; Domain Name System (DNS) (1)] access and HTTP [Hypertext Transfer Protocol (HTTP) (1)]. However, most operating systems provide an API for DNS access, so the browser can use this service locally and only has to implement HTTP. TCP [Internet Architecture; Transmission Control Protocol (TCP) (1)] (which is required as the foundation for HTTP) is usually provided by the operating system.

(28) The Web's Protocol

provided by CacheLogic Inc. [http://www.cachelogic.com/]