AI summary by TwinMind

Will YouTube Ever Run Out Of Video IDs?

Tom Scott

5m 10s

YouTube Video ID System Structure

  • Every YouTube video has unique 11-character ID in URL
  • Uses base 64 encoding system instead of base 10 or base 16
  • Characters include: 0-9, A-Z, a-z, hyphen, underscore (64 total)
    • Standard base 64 uses slash and plus, but YouTube uses hyphen/underscore for URL compatibility
  • IDs are random numbers, not sequential counters

Number Base Systems Comparison

  • Humans count in base 10 (0-9)
  • Computers use base 2 (binary) - difficult for humans to read, gets too long
  • Base 16 (hexadecimal) commonly used: 0-9, A-F - efficient but not easily human-readable
  • Base 64: 2^6, easy for computers, relatively human-readable
  • Base 65,536 exists (Sam Hughes creation) - includes characters from every language, described as "ridiculous and unnecessary"

Why YouTube Uses Random IDs vs Sequential Counting

  • Sequential counting problems:
    • Requires synchronization between servers handling uploads
    • Need to assign number blocks to each server
    • Extensive tracking required to prevent duplicates
  • Security vulnerabilities of incremental counters:
    • Easy enumeration - users can guess other video IDs
    • Breaks unlisted video privacy (videos not publicly listed but shareable via link)
    • Reveals business metrics to competitors (exact customer count)
    • Enables bulk data downloads
    • Exposes personal details (Florida website example mentioned)
  • YouTube solution: Generate random number, check if taken, use if available

YouTube Upload Statistics

  • 400 hours of video uploaded every minute (last released stats)
  • Millions of videos total on platform

Base 64 Capacity Calculations

  • 1 character: 64 possible IDs
  • 2 characters: 64 × 64 = 4,096 IDs
  • 3 characters: 64³ = 262,144 (quarter million+)
  • 4 characters: 16+ million (enough for London population twice)
  • 7 characters: 4 quadrillion
  • 11 characters (YouTube's system): 73,786,976,294,838,206,464 possible videos
    • Enough for every human on Earth to upload one video per minute for ~18,000 years

Dictionary Filtering

  • YouTube likely filters out actual words from IDs
  • Particularly blocks inappropriate/rude words
  • Filtered words represent tiny minority of total possible combinations
  • Negligible impact on overall capacity

System Scalability

  • Technically possible to run out of URLs
  • Practically impossible given current usage
  • Simple solution if needed: add one more character to ID length
  • YouTube "planned ahead" with 11-character system

TwinMind

Get summaries like this for your own meetings

TwinMind takes notes in the background during meetings, lectures, and calls, then writes the summary for you.

Free plan to get started

No bot joins your calls

Phone, desktop & Chrome apps

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome