Naming your TMs and Glossaries
What’s in a name?
There are two main types of databases used by Wordfast: a translation memory (TM) and a glossary (also referred to as terminology database). A TM automatically records pairs of sentences—also known as segments—during the translation process for re-use in the future. A glossary stores source and target term pairs that are specified manually by the translator during the translation process.
The names of these databases should include information that helps you easily identify them in the future, without having to open them to know what is inside. They should also be standardized and scalable in the event you want to share them in the future.
Translation Memory (TM)
Best practice is to create a separate TM for each client. However, if you have many small jobs for different clients in the same field, you might consider creating a field-based TM.
Potential information to include in the name: end-client name, agency name, field, date, TM creator, source language, target language, tool name, etc.
Here are some examples of standardized naming conventions for TMs:
- EndClient_slLV_tlLV_wfm (e.g. a US English to Canadian French TM for a company called Acme could be named: Acme_enUS_frCA_wfm)
- Agency-EndClient_FIELD_slLV_tlLV_wfm (e.g. a US English to Canadian French TM for a bank called Acme, via a translation agency called TransAgency could be named: TransAgency-Acme_BF_enUS_frCA)
- FTM_FIELD_slLV_tlLV (e.g. a TM for all banking and finance-related translations from US English to Canadian French could be named: FTM_BF_enUS_frCA)
The same principles described above can be applied to glossary names.
Here are some examples of standardized naming conventions for glossaries:
- EndClient_slLV_tlLV_wfg (e.g. a US English to Canadian French glossary for a company called Acme could be named: Acme_enUS_frCA_wfg)
- Agency-EndClient_slLV_tlLV_wfg (e.g. a US English to Canadian French glossary for a company called Acme, via a translation agency called TransAgency could be named: TransAgency-Acme_enUS_frCA_wfg)
- FG_FIELD_slLV_tlLV (e.g. a glossary for all banking and finance-related translations from US English to Canadian French could be named: FG_BF_enUS_frCA)
- You can use both client and field databases simultaneously. This allows you to leverage work done for other clients in the same field.
- The underscores are optional; the benefit of using underscores is that it stands out from other file and folder names on your computer so you can quickly identify it as a database.
- DO NOT store your TMs or glossaries on cloud sync drives like Dropbox, iCloud, etc. This can corrupt the databases. You should store all of your TMs in a dedicated "TMs" folder on your hard drive. You should store all of your glossaries in a dedicated "Glossaries" folder on your hard drive. Make sure that these folders are regularly backed up.
- Keep it simple! If you decide to use fields, use no more than 3 to 5 field-based TMs or glossaries. Below is a list containing general fields and suggested acronyms you can choose from:
- TE Tech/Engineering
- AL Art/Literary
- MED Medical
- LP Law/Patents
- SCI Science
- BF Bus/Financial
- MKTG Marketing
- OTH Other
- slLV stands for source language locale variant and tlLV stands for target language locale variant. Sometimes, a locale variant is unspecified or unnecessary. For example, there are multiple variants of English (UK, US, Canadian, etc.) giving you language codes like enGB, enUS or enCA whereas Corsican does not have a locale variant, resulting in the two-digit code co.