Thesis
Aside from having an absolute blast writing blog posts, whether articles, recap posts or just my own soapbox musings on happenings around the JB space, I’d like to continue taking a more active role in our translation department. Currently we do not always have systems in place to moderate translation quality, and subsequently some of our translation is going slowly or the quality is lacking. Furthermore, we have no incentive structure to reward excellent translators, and lastly, some of our documentation is in need of streamlining on the English end to ensure high quality translation/tone on the foreign-language end.
Abstract
Currently translators are paid 10 cents per word, regardless of how long they have worked for us, their output, or their quality. We vet translators once when they join, and after that we don’t. Unfortunately this has opened up some opportunities for exploitation, and some translators have been resorting to using translation bots, which offer a substandard product. The goal is to create an accountability structure to prevent this from happening, and also an incentive structure to create a team of dedicated, high-quality translators. I’ve already started working on this with Zotico, who oversees translator payouts and assignments, and who I would work with to implement this system within the current translation structure.
Motivation
As someone who once worked in translations many moons ago, I feel strongly that not all translations are created equally. I also believe our voice should remain consistent across languages and cultures, and we can only achieve that with a solid team of translators under good direction, solid accountability constraints, and who are rewarded for going above and beyond consistently. I also think we should amp up translation, particularly translation that deals with JB culture/marketing efforts, so that we can spread the JB culture across many languages.
Basic structure and Work Needs
My proposal is to finalize and deploy an accountability structure by securing a team of fluent, high-quality language raters of various languages, and having them evaluate translator samples on an ongoing basis. These evaluation opportunities will be randomly generated. The ratings received from these evaluations will determine the quality of a translator, with certain bonus-benchmarks in place as a reward for good work. For instance, for each time a translator hits 10,000 words translated with consistently good marks from the rater team, we may give them a $100 bonus (the precise number of words to bonus ratio is up for discussion!). Also, translators who are clearly using bots/google translate, which is an ongoing problem, will be banned from participating in translation. The accountability-team will also be reimbursed $25 per session per person, with evaluations taking place ongoingly (at least once per language) within a 3-month span. The total cost for these evaluators will be less than $500 over a 3-month span. Zotico will continue his critical role of tracking these payments, disbursing them and assigning translation projects. My role will be reserved to this accountability system and evaluating translation content with the team of raters, managing the team of raters and setting the pay structures based on the raters’ feedback.
Additionally, I will do a pass over existing English documentation with a technical writing focus in mind, making sure to double back to original authors where certain statements may be unclear for translators. This is a pretty important need as some of our clauses and grammatical cases in English don’t exist in other languages, or vice versa other languages have some that we don’t, and therefore the existing writing given to translators should rely as little as possible on complex syntax and sentence structure, within reason. Evaluating these documents and streamlining their meaning for translators is an ongoing need as documentation is produced, and off the bat as the well of existing documentation is processed.
Risks
Paying translators more in bonuses means that we are increasing our expenditures. I do believe the return on investment in this case is totally worth it, since we will be able to have peace of mind that our translations are in fact accurate and accessible in other languages. Another risk is that by paying me, we are also undertaking another ongoing expenditure. It is also worth noting that Zotico will have some additional work if this proposal passes. As such he may also request an increase in his payout, which again is an increased burden on the treasury.
Cost
I would request an increase of $2000 per funding cycle, to $4000 total, to do this work, and roughly $500 or less of funding to Zotico’s translator funds in every 90-day period for raters. This includes carefully vetting first-time translators, finding and coordinating with raters who also need to be vetted, organizing rating cycles, and ensuring that the translated work makes it where it belongs in a high quality. I’ve already started compiling a rater database and will continue fleshing out the system over the coming days/weeks as we source high-quality raters. Additionally this includes going over in detail existing documentation, coordinating with original authors to ensure clarity, and streamlining it ahead of giving it out to the translator team.