As a software developer I like to work with everything that is related to software Localization known as L10n. Besides being a developer working defining the architecture that will be adopted in a given project and doing the hard “FUN” work writing the code, I’m also a translator if you don’t know it yet.

One thing I've been trying to do recently is to be able to use localized strings that are present in an external assembly [ DLL ] using the ResourceManager object.
I have localized strings in resource [ .resx ] files that are specific for each locale I support. I place these .resx files in a separate class library project to maintain things organized.

So, suppose the namespace of this class library is MyProject.L10n and the .resx file name is Localization.resx. This gives me access to a class named Localization within the code. I also have Localization.pt.resx. I support English and Portuguese locales in my project for now. This naming pattern allows me to have in the future a file called Localization.es-ES.resx for Castilian Spanish (as written and spoken in Spain) and another one called Localization.es-AR.resx for Argentine Spanish. During runtime the .NET framework will select the correct .resx file to extract the localized string from based on the current culture the user has set while browsing my website.

After adding a reference to this class library, I'm able to use this code in my ASP.NET MVC project in a Razor view:

MyProject.L10n.Localization.LocalizedString;

This works as expected, but it's not what I need, though. As you see the localized string key [ LocalizedString ] is hard coded. I want to be able to use the method GetString from the ResourceManager object so that I can write code like this:

ResourceManager.GetString(item.DynamicLocalizedStringValue);

The problem and the catchy here is that in order to use the resource manager the way I want, I have to point it to the external assembly this way:

grid.Column(
columnName: "Type",
header: Localization.Type,format: (item) => new ResourceManager("MyProject.L10n.Localization", typeof(Localization).Assembly).GetString(item.Type.ToString()))

This part does the tricky: typeof(Localization).Assembly

In the code block above I’m using WebGrid that is a new helper that comes with ASP.NET MVC 3. It simplifies the task of rendering tabular data. When I do item.Type.ToString() I’m actually getting different values for each row of my grid and I pass this dynamic value to ResourceManager that in return gives me the translated/localized version of a give string key.

Going even further I’ve implemented a Razor’s Helper method in a file called Helpers.cshtml and placed such file inside the App_Code folder. This is the helper’s code:

@using System.Resources
@using MyProject.L10n

@helper GetLocalizedString(string stringValue)
{
    ResourceManager rm = new ResourceManager("MyProject.L10n.Localization", typeof (Localization).Assembly);

    @rm.GetString(stringValue);
}

Now it’s just a matter of calling the helper this way in whatever place/view I need it:

grid.Column(
columnName: "Type",
header: Localization.Type,
format: (item) => @Helpers.GetLocalizedString(item.Type.ToString()))

The above code is way more clear than the one I showed your before…

Hope this post helps shed some light in this subject since the only thing that should be done is to get a reference to the assembly that holds the Localization class and pass it to the ResourceManger’s constructor.

Just out of curiosity I wanted to know how many words I have already translated from English to Portuguese in respect to Translating ScottGu's Blog to Portuguese.

This appeared to be a great chance to play with Google Translator Toolkit (GTT) API since I use GTT to translate Scott Guthrie’s posts.

GTT gives me the number of words it finds in the source document. I could count them one by one but that’d be a tedious task. Don’t you think? That’s what computers are for.

Google Translator Toolkit is a pretty good tool because it helps translators translate better and more quickly through one shared, innovative translation technology. It uses machine translation when possible and still allows human intervention.

When a document is uploaded for translation, GTT pretranslate the doc with a combination of previous translated docs by human translation (translation memories), machine translation, etc. Great technology put to work here. That’s why I’ve chosen it.

Given the above, the translated word count I’m interested won’t be an exact figure but it does show a realistic figure about my work as a translator. So let’s find this magic number using GTT API.

Basically what one needs to write a GTT client app is very well described at Google Translator Toolkit Data API v1.0 Developer's Guide. Pay special attention to the Getting Started section as it teaches you how to set up the Google client library. Refer to this: Getting Started with the Google Data Java Client Library.

Now I’m using a Mac and so I played with GTT API with Eclipse for Mac OS if you mind. The programming language is Java and the following code creates a console application.

This is the code I wrote to satisfy my curiosity:

import com.google.gdata.client.gtt.*;
import com.google.gdata.client.gtt.DocumentQuery;
import com.google.gdata.data.gtt.*;
import com.google.gdata.util.*;

import java.io.IOException;
import java.net.URL;

/**
* @author Leniel Macaferi
* 12-11-2010
*/
public class GttClient
{

    static final String DOCUMENTS_FEED_URI = "http://translate.google.com/toolkit/feeds/documents";

    public static void main(String[] args) throws IOException, ServiceException
    {
        try
        {
            GttService myService = new GttService("GoogleTranslatorToolkitClientApp");

            // Your Google username and password go here...
            myService.setUserCredentials("YourUserName", "YourPassword");

            URL feedUrl = new URL(DOCUMENTS_FEED_URI);

            DocumentQuery query = new DocumentQuery(feedUrl);

            // Send the query to the server.
            DocumentFeed resultFeed = myService.getFeed(query, DocumentFeed.class);

            printResults(resultFeed);
        }
        catch (AuthenticationException e)
        {
            // TODO Auto-generated catch block
            e.printStackTrace();
        }
    }

    /**
     * Iterates the document feed and prints some information to the console screen.
     * @param resultFeed
     */
    private static void printResults(DocumentFeed resultFeed)
    {
        System.out.println("...done, there are " + resultFeed.getEntries().size()
                + " documents matching the query in your inbox.\n");

        int i = 1;
        int totalWords = 0;

        for (DocumentEntry entry : resultFeed.getEntries())
        {
            System.out.println(String.valueOf(i++) +  ") "
                    + "id = " + entry.getId().substring(DOCUMENTS_FEED_URI.length() + 1)
                    + ", title = '" + entry.getTitle().getPlainText() + "'"
                    + ", number of words = '" + entry.getNumberOfSourceWords().getValue() + "'");

            totalWords += entry.getNumberOfSourceWords().getValue();
        }

        // Here's where I satisfy my curiosity... :D
        System.out.println("Total words translated so far = " + totalWords);
    }
}

As you see the code is straightforward.

Make sure to replace the strings YourUserName and YourPassword to match your GTT login information.

When I ran the code, this was the output I got:

1) id = 00001vipkz2ce0w, title = 'a-few-quick-asp-net-mvc-3-installation-notes.aspx', number of words = '482'
2) id = 0000082e1f41udc, title = 'add-reference-dialog-improvements-vs-2010-and-net-4-0-series.aspx', number of words = '399'
3) id = 0000206w1a7tog0, title = 'announcing-entity-framework-code-first-ctp5-release.aspx', number of words = '2165'
4) id = 00001qlrxh63jls, title = 'announcing-nupack-asp-net-mvc-3-beta-and-webmatrix-beta-2.aspx', number of words = '1500'
5) id = 00001zm1lv1meio, title = 'announcing-silverlight-5.aspx', number of words = '784'
6) id = 00001vgl6hzbwg0, title = 'announcing-the-asp-net-mvc-3-release-candidate.aspx', number of words = '1999'
7) id = 00001rtmq1go4cg, title = 'asp-net-4-seo-improvements-vs-2010-and-net-4-0-series.aspx', number of words = '1113'
8) id = 00000snuf95gd8g, title = 'asp-net-mvc-2-model-validation.aspx', number of words = '2925'
9) id = 00000joo5ihwykg, title = 'asp-net-mvc-2-release-candidate-2-now-available.aspx', number of words = '549'
10) id = 00000si5lunjbi8, title = 'asp-net-mvc-2-released.aspx', number of words = '524'
11) id = 00000f8wakzalts, title = 'asp-net-mvc-2-strongly-typed-html-helpers.aspx', number of words = '705'
12) id = 00000en7g3nkvls, title = 'asp-net-mvc-2.aspx', number of words = '619'
13) id = 00001so4spobfnk, title = 'asp-net-mvc-3-layouts.aspx', number of words = '1817'
14) id = 00001snnr32adc0, title = 'asp-net-mvc-3-new-model-directive-support-in-razor.aspx', number of words = '792'
15) id = 00001vldns8skqo, title = 'asp-net-mvc-3-server-side-comments-with-razor.aspx', number of words = '653'
16) id = 00001r2yti5z4sg, title = 'automating-deployment-with-microsoft-web-deploy.aspx', number of words = '3784'
17) id = 00000m0if2iqmtc, title = 'built-in-charting-controls-vs-2010-and-net-4-series.aspx', number of words = '637'
18) id = 000010zzw2fq1a8, title = 'cleaner-html-markup-with-asp-net-4-web-forms-client-ids-vs-2010-a', number of words = '1725'
19) id = 00001ih7gktv6kg, title = 'code-first-development-with-entity-framework-4.aspx', number of words = '5365'
20) id = 00001r7ki4lh05c, title = 'debugging-tips-with-visual-studio-2010.aspx', number of words = '1692'
21) id = 0000151qr16itj4, title = 'download-and-share-visual-studio-color-schemes.aspx', number of words = '311'
22) id = 00001ibpvmetdkw, title = 'entity-framework-4-code-first-custom-database-schema-mapping.aspx', number of words = '1930'
23) id = 00001f6hlprohz4, title = 'introducing-asp-net-mvc-3-preview-1.aspx', number of words = '2833'
24) id = 00001so0gu8f3ls, title = 'introducing-razor.aspx', number of words = '3312'
25) id = 00000ug44uvcow0, title = 'javascript-intellisense-improvements-with-vs-2010.aspx', number of words = '930'
26) id = 00000jtavmijvgg, title = 'jquery-1-4-1-intellisense-with-visual-studio.aspx', number of words = '179'
27) id = 00001q9k1j435s0, title = 'jquery-templates-data-link-and-globalization-accepted-as-official', number of words = '828'
28) id = 00000b4mu45b4e8, title = 'microsoft-ajax-cdn-now-with-ssl-support.aspx', number of words = '309'
29) id = 00000ao1rdg9dds, title = 'my-presentations-in-europe-december-2009.aspx', number of words = '1684'
30) id = 000010q75bisw74, title = 'new-lt-gt-syntax-for-html-encoding-output-in-asp-net-4-and-asp-ne', number of words = '977'
31) id = 00000lubanp79c0, title = 'no-intellisense-with-vs-2010-rc-and-how-to-fix-it.aspx', number of words = '449'
32) id = 00000tqxxxv1h4w, title = 'optional-parameters-and-named-arguments-in-c-4-and-a-cool-scenari', number of words = '841'
33) id = 000010ydwjqjzeo, title = 'pinning-projects-and-solutions-with-visual-studio-2010.aspx', number of words = '667'
34) id = 00001wjhyhw29ds, title = 'search-engine-optimization-seo-toolkit.aspx', number of words = '711'
35) id = 000007lw738nbwg, title = 'searching-and-navigating-code-in-vs-2010-vs-2010-and-net-4-0-seri', number of words = '1305'
36) id = 00000e1cqi0ta0w, title = 'silverlight-4-demos-from-my-pdc-keynote-now-available.aspx', number of words = '613'
37) id = 00000743lq060hs, title = 'url-routing-with-asp-net-4-web-forms-vs-2010-and-net-4-0-series.a', number of words = '1045'
38) id = 000020npa0inls0, title = 'using-ef-code-first-with-an-existing-database.aspx', number of words = '2520'
39) id = 00001vo9oowkpvk, title = 'Using-Server-Side-Comments-with-ASP.NET-2.0-.aspx', number of words = '460'
40) id = 00000eafvti4sn4, title = 'visual-studio-2010-and-net-4-0-update.aspx', number of words = '391'
41) id = 000017d48tqvpc0, title = 'visual-studio-2010-productivity-power-tool-extensions.aspx', number of words = '682'
42) id = 000008fncz9zpc0, title = 'vs-2010-and-net-4-0-beta-2.aspx', number of words = '590'
43) id = 000007m1ubf8zcw, title = 'vs-2010-code-intellisense-improvements-vs-2010-and-net-4-0-series', number of words = '734'
44) id = 00000jxh5t9ebr4, title = 'vs-2010-net-4-release-candidate.aspx', number of words = '748'
45) id = 00001r3ulw3aygw, title = 'vs-2010-web-deployment.aspx', number of words = '1352'
46) id = 000008sxc7ta800, title = 'wpf-4-vs-2010-and-net-4-0-series.aspx', number of words = '3171'

Total words translated so far = 59801

If I was to charge for those translations and considering that each word translated costs $ 0.07 (market price), this fast math gives how much I’d have accrued so far:

59801 x 0.07 = $ 4,186.07

$ 4,186.07 / 46 = $ 91.00 average per doc translated

That’s a lot of money, but as you know already I do not charge a thing to translate ScottGu’s blog. That’s something I do to help others and to keep myself up to date.

Hope you liked the curiosity that made me write this post, the reasoning regarding the math and of course this simple java console application.

Notes
It’s important to mention that I started using GTT on Oct 17, 2009, that is, more than a year after I started translating Scott’s posts. According to my records, I have translated in fact 77 posts so far since April 2008. Those remaining 31 posts ( 77 - 46 = 31 ) didn’t figure in the math above. :(

GTT cold have folders just like Google Docs so that one could organize their translations by client or whatever.

I tried to get only 100% translate completed documents but GTT doesn’t give me this info. It’s true even if I mark the translation as complete. Although GTT shows 100% complete in its UI, when I read the value of entry.getPercentComplete() it gives me not 100% but what is described at Word count and translation completion. So I had to consider every document even those that I still need to finish translating.

Download
You can download the sample app with the necessary libraries at:

https://sites.google.com/site/leniel/blog/GoogleTranslatorToolkitClientApp.zip

As a form of gratitude to Babylon I decided to write this post. A post is worth a thousand words almost literally in this case.

When it comes to computer translation, don’t think twice. Take a look at Babylon. Babylon is a Windows client application that resides in you taskbar. It offers translation from/to a varied set of language pairs.

If you are not a native English speaker as is my case and if you work in the software world, chances are that you had to resort to any online dictionary to find the meaning of some word. English reigns when it comes to programming languages and any other sort of computer related stuff. With Babylon the translation from/to English to a varied set of languages is just @ a click of the mouse. This amazing software product has helped me a lot since I started the computer engineering course in 2003 and I think it'll continue to play a big role in my English learning process during the years to come.

Nowadays it's practically impossible to work in the software industry without a grasp of English grammar, so that if you want to know a software product that can help you, read on.

Being a Brazilian speaking Portuguese and an avid user of Babylon since 2002 I can tell you that Babylon played and plays a big part in my English language learning. Despite having a 4.5 years course of English background, it’s always not enough. Learning is an infinite process.

Much of my contributions to ProZ comes from Babylon. At the same time I help others I learn a little bit more of English each day.
I just can’t express in words how much Babylon has helped me during these past years. At least this is an attempt.

Below is a screenshot of Babylon 8 with word auto-completion (auto-suggestion):

When I get to a word that I don’t know yet I just select it with "Ctrl + Right Mouse Button" or copy/paste it inside Babylon’s text box to get instant translation that comes from a number of online dictionaries including Wikipedia, which is the world’s largest encyclopedia/source of information.

Babylon has its native dictionaries/glossaries but the community also develop custom ones making them free to use. Such dictionaries/glossaries are in most cases specialized ones, that is, they refer to specific fields as computer networks, software, electronics, etc. In contrast with free content, there is also premium content which you pay for.

Behind the curtains Babylon uses built-in browser functionalities as is the case of translated data returned in the form of a webpage and history navigation.

Besides translation to and from any language, full web page translation, full document translation (Word, PDF, Text), integration into Microsoft office spellers, leading dictionary packs - Oxford, Britannica, Merriam-Webster, etc, Babylon also does currency, measurements and time conversions. With that all, Babylon can be called a killer app.

I’d like to remember when exactly I started using Babylon and in which version it was, but that is a difficult task… Sometimes you get a little bit nostalgic.

Babylon has been evolving and in each new version new features get added and existing ones are refined. Version 8 in my humble opinion is fantastic!

Watch this video demo to see Babylon in action.

With Babylon there’s no more language barriers when it comes to read/written words. Take it with you and you can rest assured that you’ll have that word translated as fast as a mouse click.

I just created an account at ProZ. ProZ is the translation workplace providing translation service and translation jobs for freelance translators and translation agencies. It's is also home to the world's largest community of professional translators.

Take a look at my Translator profile at ProZ.

If you ever need a translation work be it from English to Portuguese or from Portuguese to English, please, feel free to contact me at

I really like this kind of job, that is, to translate any type of material related to my area of expertise: software, technology and computer related stuff.

I translate ScottGu's blog from English to Portuguese - see Translating ScottGu's Blog to Portuguese post. ScottGu's blog is one of the most accessed Technology blogs in the world.

I'm looking forward to work with you.

Joe Stagner from Microsoft wrote a post in which he calls people to volunteer for translating Scott Guthrie's Blog posts originally written in English to Portuguese. I immediately contacted Joe emphasizing that I would like to help. Since then I've translated three posts and I really expect to translate many more.

You can access the new blog created especially for the Portuguese community at http://weblogs.asp.net/scottguportuguese

Other great news for the Brazilian community is that the Developer Division, more precisely the Community Localization Team from Microsoft has released the Brazilian Localization of Visual Studio 2008 and .NET FX 3.5. More details can be found on this post at Somasegar's WebLog.

Again, I volunteered to participate in these community localization projects assisting with translating into my native language: Portuguese.

I think it's a good idea to participate in such projects. You can learn a lot more. As Joe stated in the title of his post "Do you translate ScottGu's blog in your head?"; my answer is YES, I do. I always do that. Besides, it's worth it to help others that still don't know the English language.