Standardize the language we use across our applications
It’s in the best interests of our users to maintain some level of consistency in the language we use to label or describe various features in Tor Browser, Tor VPN and Tails. These types of strings are often referred to as “microcopy” in the design biz.
For instance:
- Aligning each app and platform to use similar strings would help reduce the cognitive load for users moving between each platform, by using language they already recognize and are familiar with.
- When a particular string has been iterated upon based on user feedback/research, each app and platform can benefit from that improvement.
- Creating a centralized resource where we can share common microcopy may also help the team work more efficiently, and if we reuse certain strings between each app, could potentially reduce the workload for localization teams too.
However, it’s unrealistic to assume we can use the exact same strings between each app in all (or even most) instances. Exceptions will always be necessary to suit the specific context of each app, for example small screen devices would benefit from shorter and more direct descriptions than desktop platforms. Or another example: it is desirable for the copy used in Tor Browser to maintain a level of consistency with Firefox, as the majority of the copy in Tor Browser is inherited from upstream. As such, the same string we use in Tor VPN may need adjusted to suit Firefox’s tone of voice or general patterns.
That said, I still think it would be useful to create a library of microcopy as a reference guide, with the objective to try and align each app and platform as much as possible (again, within reason). While the exact strings may be different, hopefully the key words used and general intent behind the strings are the same.
A Google Sheet might be a good starting place to document these strings, as we could potentially integrate the sheet with Figma and use it as a CMS of sorts for common microcopy. This could also be extended with translations in selected languages to make the production of localized mockups easier in future. Perhaps integration with Weblate itself is possible, but this would need investigation, and the technical effort of maintaining such an integration may exceed the other benefits we’d get from such a set-up.