CASE STUDY

Protocol Labs & Transifex

Protocol Labs Leverages Crowdsourcing With Minimum Admin Process Overhead

Protocol Labs is an open-source research, development, and deployment laboratory. Their projects include IPFS, Filecoin, libp2p, IPLD, Multiformats, and many more. Protocol Labs aims to make human existence orders of magnitude better through computing advancements.

Business Needs

IPFS uses JS, GO, and Rust technology stack. They have several graphical user interface tools like browser extensions, web dashboards, and desktop applications to make it easy for people to interact with the IPFS stack.

IPFS depends on crowdsourcing, which means that community volunteers do all the translations. Over some time, the prolific translators are promoted to reviewers and collaborators. Collaborators are trusted people from Protocol Labs, or the community, while reviewers are promoted based on their long-term commitment and the number of contributions in recent years.

This required better organization management and shared translation memory to localize short explainers and labels for these GUI applications and tools.

protocol-labs-transifex-localization2
“Translation memory across projects is the killer feature. It ensures language does not diverge too much across projects.”
Marcin Rataj
Solution

With Transifex, Protocol Labs has improved the entire translation process, allowing for better communication and management of the translators’ community. Here are some of the essential Transifex features that helped Protocol Labs achieve their goals in supporting the utility websites and GUI tools with leveraging community contributors while reducing the overhead associated with admin and process for short explainers and labels with their translators:

Better content and people organization management
The ability to have multiple projects under a single account, the translator teams being able to work on multiple projects, as well as to re-use existing translators to newly added projects.

Shared Translation Memory
As Marcin Rataj notes, “Translation memory across projects is the killer feature. It ensures language does not diverge too much across projects.”

GitHub Automation
Automatic sync of the source files from GitHub is set up. The translators are informed via Transifex email notifications whenever new strings to be translated are added. These notifications go out within 24 hours after a pull request is merged on GitHub. With source content updates every few weeks, this workflow is critical for the team.

The IPFS community members ask to join in a specific language, and approvals are manually handled by the team. Transifex is used to respond to any issues coming from the community, providing clarifications about any specific label’s meaning or even User Interface localization. The team carefully names the translation keys so that the translators are given enough context simply by reading the key name within the Transifex interface. Protocol Labs team link to https://github.com/ipfs/i18n and https://www.transifex.com/ipfs/public/ from each project’s README.

While translators use the Transifex interface to fill-in issues, the IPFS team handles the translation syncing before each release. This is done across all the projects, and updates happen quite frequently. IPFS uses JSON files with ICU format translations to make it future proof. The process is quite simple.

  • Automated sync from GitHub to Transifex takes place once a day.
  • Each release is preceded by the project maintainer, ensuring a fresh sync to transfer the
    translations from Transifex to GitHub.
Results

While IPFS does not actively track any key performance indicator, they have seen a steady growth of requested languages. There is also a surge in the number of contributors. This shows significant success concerning the localization projects IPFS is undertaking. On top of this, the improved user and developer onboarding is an additional indicator of a good return on investment of IPFS’s localization efforts.

Crowdsourcing and localization automations have helped IPFS become more accessible to non-English speakers, helping them create a better internet for tomorrow. It also ensures a certain level of similarity of the language across the projects while keeping track of contributions of the translators. With much better content management and easier organization of people, IPFS can now maintain and manage a number of projects with a single structure for organization.

Other Case Studies
hubspot_transifex_localization
HubSpot Leverages the API for Flexible Continuous Localization Workflows.
READ MORE
wishtrip-trasifex-case-study
WishTrip Achieves a 30% Increase in Lead Conversion Rate from Localized Content.
READ MORE
vwo-localization-transifex3
VWO Localizes Landing Pages to Increase Conversion Rates for the Non-English Prospects.
READ MORE
FacebookgithubGoogle+Fill 88Twitter