Defensive design

Validation checks that data is sensible before a program uses it. A username can be required (presence), limited to 3–12 characters (length) and restricted to letters, digits and underscores (format). A person's age can be checked to be from 11 to 16 (range). Python's input() gives text, so check that it contains digits (type) before converting it. ERL's age example assumes age is entered as an integer.

INPUT username
valid = username != ""
IF username.length < 3 OR username.length > 12 THEN
    valid = false
ENDIF
formatValid = true
IF username.length > 0 THEN
    FOR i = 0 TO username.length - 1
        code = ASC(username.substring(i, 1))
        IF NOT ((code >= 48 AND code <= 57) OR (code >= 65 AND code <= 90) OR (code >= 97 AND code <= 122) OR code == 95) THEN
            formatValid = false
        ENDIF
    NEXT i
ENDIF
IF NOT formatValid THEN
    valid = false
ENDIF
INPUT age
IF age < 11 OR age > 16 THEN
    valid = false
ENDIF

username = input("Username: ")
valid = username != "" and 3 <= len(username) <= 12
valid = valid and all(c.isascii() and (c.isalnum() or c == "_") for c in username)
age_text = input("Age: ")
if not age_text.isascii() or not age_text.isdecimal():
    valid = False
else:
    age = int(age_text)
    valid = valid and 11 <= age <= 16

The ERL age input is an integer; Python checks the text before casting.

Authentication checks that a user is who they claim to be, usually when they sign in. This simple example compares a username and password with demonstration values:

INPUT username
INPUT password
IF username == "learner" AND password == "demo-only" THEN
    OUTPUT "Signed in"
ELSE
    OUTPUT "Try again"
ENDIF

username = input("Username: ")
password = input("Password: ")
if username == "learner" and password == "demo-only":
    print("Signed in")
else:
    print("Try again")

This is only a classroom example, not a real sign-in system. Real systems never store plain-text passwords; they store salted password hashes and use trusted authentication methods. Validation checks data; authentication checks identity.

Maintainability means that a program is easy to understand, test and change. Use meaningful names such as age instead of x, consistent indentation, and comments to explain non-obvious decisions. Split repeated tasks into subprograms with clear purposes, and validate input in one place. These habits help someone find and fix an error without accidentally changing a different part of the program.