/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Google Translate Adds Crowdsourcing Features To Improve Its Translations

Over the years, Google Translate has gotten significantly better at giving its users (relatively) legible translations fo most commonly used languages.  It's still far off from being perfect, though …

TechCrunch Frederic Lardinois

Context & Ripple Effects

Google has been signaling a quality push on Translate for a while: a September 2013 interview laid out ambitious machine-translation goals, and this week's crowdsourcing features are the first concrete mechanism attached to them — letting users flag bad translations and submit corrections that feed back into the system.

The pickup was broad for a feature announcement: Inside Search, 9to5Google, The Next Web, SlashGear and WebProNews all ran it on or about July 26, 2014, reflecting how central Translate had become to Google's consumer surface even while the description concedes output remains far from perfect for most commonly used languages.

First-order effects

  • Translate users gain a direct channel to fix mistranslations in place, turning every correction into training signal rather than dead-end feedback.
  • Google gets a free, continuously refreshed corpus of human-verified alternatives for exactly the phrases its statistical models get wrong.

Second-order effects

  • Rival translators (Microsoft's Bing Translator, Yandex.Translate) face pressure to match the correction loop or cede the compounding-data advantage, since each user fix widens Google's quality gap at zero marginal cost.
  • The feature deepens Translate's lock-in as default browser and Android translation layer: better output drives more queries, which generate more corrections.

Third-order effects

  • If the pattern holds, machine translation consolidates around services that own both the query stream and the human-correction loop — a classic liquidity network effect where scale begets accuracy begets scale.
  • Crowdsourced correction becomes a recognized dataset category in its own right, raising questions about contributor rights and how public-data permission boundaries apply to user-submitted language fixes.

The trend: Machine translation is shifting from closed statistical models toward human-in-the-loop systems where user corrections compound into a proprietary data moat.