WiserEarth in Different Languages

Helping to build a global and multilingual WiserEarth community

We are here to discuss and advocate ways which WiserEarth  can be more accessible to all regions and languages.

GROUP DETAILS

Created: Apr 02, 2008

Updated: Nov 22, 2009

Membership: Open

Semi-Private

Created: Apr 02, 2008
Updated: Jun 04, 2008
Viewed: 331 times

Topic: What does it mean to Internationalize?

Posts (1 - 6 of 6)

Sort by: Ascending | Descending
Login to Post a Reply.
Sm_avatar
First of all, this is what it means to "Internationalize" and "Localize" according to Wikipedia: "In computing, Internationalization and localization (also spelled internationalisation and localisation, see spelling differences) are means of adapting computer software for non-native environments, especially other nations and cultures. Internationalization is the process of designing a software application so that it can be adapted to various languages and regions without engineering changes. Localization is the process of adapting software for a specific region or language by adding locale-specific components and translating text."
http://en.wikipedia.org/wiki/Internationalization_and_localization - this wikipedia article also has "Focal points of internationalization and localization efforts" which are interesting to read.

I think it's important to open up the discussion on Internationalization to the WiserEarth community so that we could better understand its importance and share ideas.

Please share your thoughts on this group, and other ways for WiserEarth to internationalize and localize.
Sm_avatar

Mike, I don't think the core reason for the existence of the group is about internationalizing so much as making WiserEarth accessible to people who don't speak English as their first language. In this regard, I think that since the web is 'international' already, WiserEarth simply has to be able to make itself 'multilingual'. As a result, I changed the title to WiserEarth in different languages. I think the key challenge is how to make all of this wonderful data available to others around the world in other languages:

 

- does the community automate the translation?

- does the community set up different sites for each country, or region?

- does the community look for volunteer translators and do this manually?

- or a mixture of all or some of these options.

Sm_avatar
MichaelK about 1 year ago
This comment was removed by a WiserEarth editor for the following reason:
corrections
Sm_avatar

Hey Camilla, I agree that this group currently serves more to make WiserEarth multilingual. Doing that takes care of the basic essentials for making WiserEarth a more international site.  But I think that since a site that is multilingual is not necessarily accessible to every 'international' user, there also needs to be discussions centered around possibly more technical, or culture-specific topics.  Perhaps this discussion can take place in another group?

 

For example, these are some questions we should explore in the future...

 - How does difference in culture affect the user-experience of WiserEarth users outside of the U.S?

 -  Is the Area of Focus taxonomy understandable or relevant?

 - Do we need to have multiple, but interconnected "versions" of WiserEarth for different countries? (Different home page, groups list, comments section)

 - How fast does it take to load WiserEarth in a developing or third-world country?

Sm_avatar

I had a couple of discussions in the last few weeks around the language piece.

You guys raised a lot of good points.

I will start a wiki page, and puts all the questions you raised.

Let's see what we can learn from other site, and start involving people to we can maybe imagine WiserEarth in Spanish to start with.

Sm_avatar
bowo about 1 year ago

Camilla asked: does the community set up different sites for each country, or region?

 

For true connectivity and collaboration, I think it would be better if WE in different languages be hosted in a single www.wiserearth.org site. This would have the advantage of:

  • minimizing technical development cost, simplifying the process, and maintaining consistency of the user interface and experience across different language versions of WE.
  • some people are multilingual already and having content and people of all languages in a site would simplify the process of connecting content, and facilitating collaboration people/orgs/groups. WE of course need to determine English as the working language. Cross-language communication would be carried out via English (perhaps with the help of volunteer interpreters/translators).
  • funding would I guess be concentrated and minimized, as there will only be one site to maintain. This does not mean that data center should only reside in one place (the US currently), mirror sites can be set up in Europe or Asia as Peggy hinted to make access from regions far from the US faster.
  • not sure how the governance model will look like, but the need for country representatives forming the core governance body will be there. A more democratic and powerful General Assembly minus the Security Council (modeled after the UN) perhaps?

Other steps that seems necessary to accommodate diversity:

  • Working groups for each language/country should be created to lead the effort in community building, outreach, content building, translation effort and other needed efforts. Identified leaders for each working group should be made of the core WiserEarth governance body.
  • Customization should be made possible for the homepage and the hub pages (local, national and regional. Will this be implemented sometime in the future?)
  • Don't think the areas of focus should be different between languages/countries, as they are conceptualized from the beginning as the common thread running through NGOs around the world. They need to be more comprehensive for sure and the translation need to be understandable to each language community.
  • Specific issues relevant to each language/country can still be facilitated by the creation of groups.
  • Powerful filters to show only content and people in each language should be developed (think Google's search in "the web" or "only from the [country name]).

These are only the pluses, are there minuses of this approach? Would love to hear them for further discussions.

 

 

Camilla asked:

- does the community automate the translation?

- does the community look for volunteer translators and do this manually?

 

Have done some serendipitous research in this regard, with the answer being both. The translation effort must start manually, but can be automated to a certain degree once sufficient "translation database between languages" is developed by the community.

 

Discovered that in the translator community they have softwares that aid their work in translating documents. The latest development of the software enables the translation know-how of a great number of translators be aggregated into one large software-assisted human-auto-translation engine. Much like google translation, but using the database of translations by real, professional translators!

 

WiserEarth can imagine setting up such translation software/databse and integrate it into WiserPlatform somehow. Or if technically too difficult / impossible, WE can enable the community of volunteer translators to access, add and extract out translations from the software/database to then easily copy paste into WiserEarth's wikipages/wikispaces for further editing/refinements.

 

Here's a breakdown of the concept, explained further below it:

1. Each translator can develop his/her own translation memory (TM).

2. A termbase (TB) for each area of focus can be developed together by all translators.

3. These TMs and TBs can then be integrated into the translation software/database.

4. Any new translation effort can benefit from this database, where the workload for each translation can be reduced significantly.

 

Now, a more detailed explanation of each:

 

1. Each translator can develop his/her own translation memory (TM).


From a leading software in Computer Aided Translation (CAT):


A translation memory is a linguistic database that continually captures your translations as your work for future use. All previous translations are accumulated within the translation memory (in source and target language pairs called translation units) and reused so that you never have to translate the same sentence twice. The more you build up your translation memory, the faster you can translate subsequent translations, enabling you to take on more projects and increase your revenue.

 


2. A termbase (TB) for each area of focus can be developed together by all translators.

 

Again from a leading software in Computer Aided Translation (CAT):

 

Terminology is the foundation of all communication. At its most basic level it is the study and ultimately usage of words or phrases that have a particular meaning, these words or phrases are referred to as terms. Terminology is growing in importance as terms are becoming increasingly adopted by organizations to describe a company, product, service or even a unique selling point.


A termbase is a central repository, similar to a database, which allows for the systematic management of approved terms. It provides definitions and indicates when a particular term should be used. Use of a termbase alongside your existing translation environment ensures that you produce more accurate and consistent translations.

 

3. These TMs and TBs can then be integrated into the translation software/database.

 

I found three example of this where the TM of a large number of translators is connected via the web and thus accessible to all translators:

a. Lingotek's Language Search Engine (LSE). Commercial?

b. Wordfast's Very Large Translation Memories (VLTM). Partly-commercial. The client comes at a cost, the VLTM is free.

c. Across Language Server. Partly-commercial. Personal edition is free with access to some key features of the server.

a. Lingotek's Language Search Engine (LSE). Commercial?

 

Language Search Engines (LSE) work similarly to Internet Search Engines. Rather than searching the internet, however, an LSE searches TM's (Translation Memories) to find useful segments of previously translated documents. [Image]

The Lingotek Language Search Engine is unique in the world in it's capability to conduct meaning based searches against millions of previously translated segments to find the most complete and accurate resources that closely match the actual meaning of the source document. Unlike any other tool, the Lingotek LSE finds sentence fragments, phrases, whole sentences, even complete paragraphs where terms are used in the same context, and with the same meanings as the source document.

Where other tools perform character searches seeking only exact sentence matches or rely on outdated fuzzy matching technologies, they often discard, or simply can't find the data most helpful to the translators. The Lingotek Language Search Engine leverages cutting edge, Google-like, meaning based search algorithms that instantly find, and prioritize all the data that will be most useful to the translators.

While industry leading TM tools often get bogged down, or even crash when dealing with large TM files, the Lingotek Language Search Engine operates much like Google and other Internet Search Engines, and operates faster and more accurately as the amount of searchable content increases. Thus, the translation process increases in both accuracy and speed as it is continually used.

The Lingotek LSE also supports 99% of all the worlds languages and offers targeted glossaries, selected searches, spell checkers and other features that make the translation process as quick and easy as possible for a qualified linguist.

 

b. Wordfast's Very Large Translation Memories (VLTM). Partly-commercial. The client comes at a cost, the VLTM is free.

 

The VLTM project offers translators a set of Very Large Translation Memories, accessible with Wordfast through the web. VLTM use is free and anonymous. The translator works as usual, but can leverage valuable information from a very large public TM in addition to her/his local TM. The VLTM does not replace the local TM, it complements it. All languages are supported.


«Prime content for free? Too good to be true!» This is probably what you thought when first trying Google.

  • Confidentiality You only receive translation units from the VLTM. Your translations are not recorded in the VLTM (unless you specifically set up a sharing workgroup - see below). Your client's intellectual property is safe.
  • Gratuity Connection to the VLTM is free and anonymous. We do not charge money for the translation memory we serve. We don't even send advertizing. The VLTM is a pure give-give, pro bono project.
  • Goodwill VLTM users are welcome to donate translation memories to expand the available database, or to set up Wordfast to write to the VLTM when confidentiality is not at stake. Contact us if you wish to donate a TM. Donations are final and become the property of Wordfast, whatever their final destiny is in the future.

c. Across Language Server. Partly-commercial. Personal edition is free with access to some features of the server.

 

The Across Language Server contains the entire functionality for organizing, delegating, and efficiently processing translation projects.


It serves as a central platform for all corporate language resources and translation processes. It provides a uniform workspace in which all involved internal and external actors meet - from editors and project managers to service providers and freelance translators.


With its help it is possible to recycle content, control processes and integrate corresponding systems. As a result, qualitative foreign-language content is quicker available and translation costs are significantly reduced.


Among other things, the Across Language Server comprises the following:

  • crossLAN, crossWAN, and crossWeb access modes
  • Collaboration tool crossGrid
  • Translation memory crossTank
  • Terminology system crossTerm
  • Multiple-format editor crossDesk
  • Project control utility crossProject
  • Workflow control utility crossFlow
  • Quality management utility crossCheck

plus, optionally

  • Standard interface crossConnect for crossAuthor
  • Standard interface crossConnect for content systems
  • Standard interface crossConnect for software localization
  • Open crossAPI interfaces for user-specific system integration

 

4. Any new translation effort can benefit from this database, where the workload for each translation can be reduced significantly.

 

As a result of the above, and as the database of translation for sentences and terminology expands, each new translation effort takes less time to do and gains in accuracy.

 

Concluding this section, I'll restate my previous point:

 

WiserEarth can imagine setting up such translation software/databse and integrate it into WiserPlatform somehow. Or if technically too difficult / impossible, WE can enable the community of volunteer translators to access, add and extract out translations from the software/database to then easily copy paste into WiserEarth's wikipages/wikispaces for further editing/refinements.

You do not have access to post to this record
1 to 6 of 6 Posts


Contributors to this Page