Stats

Differences between revisions 4 and 5
⇤ ← Revision 4 as of 2010-06-17 14:39:12 →
Size: 1778
Editor: 162
Comment:
← Revision 5 as of 2010-10-08 14:25:10 → ⇥
Size: 2084
Editor: 247
Comment: Simplified explanation to make it more understandable for readers not familiar with translations. Added a note on OpenOffice.org
Deletions are marked like this. Additions are marked like this.
Line 6: Line 6:
These statistics represent the percentage of translation coverage in all languages for a given Ubuntu version at release time. As such, they provide an indicator of how well translated Ubuntu in a given language and release is, but they do not offer any other additional information such as how good font support or input methods for a particular language are (if needed). These statistics represent the percentage of translation coverage in all languages for a given Ubuntu version at release time. As such, they provide an indicator of how well translated Ubuntu in a given language and release is. They are not intended to offer any other additional information such as how good font support or input methods for that language are.
Line 11: Line 11:
 * Considering only the translation templates which are included in the default Ubuntu installation. In most cases, a translation template can be associated with an application.
 * Prioritising the translation templates (applications) which are most visible to users. At the moment, the method used is to blacklist all command line applications and not consider them for the calculation.
 * Prioritisation is not done at the string level (i.e. most visible messages), but at the template level (i.e. most visible applications). The current translation infrastructure does not allow for string prioritisation and 100% accurate statistics, but it has been found that doing it at the application level offers a good compromise.
 * A [[http://bazaar.launchpad.net/~ubuntu-translations-coordinators/ubuntu-translations/u-t-stats/files|series of scripts]] to process the data and produce an HTML report.		  * Considering only the applications that are included in a default Ubuntu Desktop Edition installation.
 * Ignoring the applications that do not provide visible messages to the user. That generally means not considering the translation status of command line applications for the calculation.
 * Note that this prioritisation is not done at the string level (i.e. most visible messages), but at the template level (i.e. most visible applications). The current translation infrastructure does not allow for string prioritisation and 100% accurate statistics, but it has been found that doing it at the application level offers a good compromise.
 * A [[https://code.launchpad.net/~dpm/ubuntu-translations/ul10n-stats|series of scripts]] to process the data and produce an HTML report.
Line 16: Line 16:
More technical details can be found on the [[http://bazaar.launchpad.net/~ubuntu-translations-coordinators/ubuntu-translations/u-t-stats/annotate/head:/README|code]]. More technical details can be found in the [[http://bazaar.launchpad.net/~dpm/ubuntu-translations/ul10n-stats/annotate/head%3A/README|description of the code]].

== Important notes ==

=== OpenOffice.org translation statistics not included in the final result for Maverick ===

Due to Open``Office.org no longer being translatable in Launchpad, its translation statistics data was not included in the final calculation. As Open``Office.org is a big application in terms of translatable messages, this might have had an impact in the translation coverage of some languages in the statistics.

Ubuntu Translation Statistics

These statistics represent the percentage of translation coverage in all languages for a given Ubuntu version at release time. As such, they provide an indicator of how well translated Ubuntu in a given language and release is. They are not intended to offer any other additional information such as how good font support or input methods for that language are.

The current method for calculating them is based on:

  • Raw export of translation data from the Launchpad database as the source data

  • Considering only the applications that are included in a default Ubuntu Desktop Edition installation.
  • Ignoring the applications that do not provide visible messages to the user. That generally means not considering the translation status of command line applications for the calculation.
  • Note that this prioritisation is not done at the string level (i.e. most visible messages), but at the template level (i.e. most visible applications). The current translation infrastructure does not allow for string prioritisation and 100% accurate statistics, but it has been found that doing it at the application level offers a good compromise.

  • A series of scripts to process the data and produce an HTML report.

More technical details can be found in the description of the code.

Important notes

OpenOffice.org translation statistics not included in the final result for Maverick

Due to OpenOffice.org no longer being translatable in Launchpad, its translation statistics data was not included in the final calculation. As OpenOffice.org is a big application in terms of translatable messages, this might have had an impact in the translation coverage of some languages in the statistics.

CategoryTranslations

Translations/Stats (last edited 2012-09-21 16:41:00 by dpm)