English to French IT Translator Blog

Wordfast Pro 4 Review: A Huge Step Backward

Although CAT tools love making us upset in various ways, you have to admit they’re generally getting better with time. Take Trados, it’s far less buggy than it used to be, much faster at processing files and its plugins add tons of useful features.

So when news broke that Wordfast Pro 4 was out, I was curious to see where improvements were made. How disappointed was I to find out that the software had completely changed, in a terrible, terrible way.

It’s painfully unresponsive

When you open a piece of software and see a bunch of Java threads show up in your task manager, you know you’re in for a bad day. 2002 all over again. You know, those web apps that took forever to load with the coffee cup icon – and barely faster after that? It’s exactly the same thing. Really, Wordfast Pro 4 actually uses a browser plugin to render its interface, so the combination is the same, and the experience is very comparable. Slow as hell.

I understand they’re trying to push their online version of the tool, but it doesn’t excuse the unresponsiveness. Some web-based solutions like Memcloud Cloud have shown that’s it’s possible to have a fast and clean-looking app online, and not something that reminds me of the cheap and buggy shareware of the early 00’s. The thing just feels heavy. It lags all the time, again and again.

File processing is even slower

Now, what about processing speed? I ran a small experiment on a 20k project I was working on. The original files were in Wordfast’s txlf format, but I’ve been using them in Trados thanks to the xliff filter.

So, numbers.

File opening is where Wordfast seemed to suffer the least. It took it 27 seconds to chain the files and present me with the result. The same process took only 12 seconds in Trados. So that’s more than twice the time for WFP4, but it’s nothing compared to some of the other features.

The next thing I tried to do was to run a file analysis with an empty TM. Shouldn’t take very long, right? Trados needed 6 seconds to do that. Wordfast 6… minutes! OK, not quite, 5 minutes and 54 seconds (yes, I timed it). That’s 59 times slower than SDL’s product.

Even such a thing as the “Find” feature seems to take ages. Still in the 20k word project mentioned above, I compared how fast Trados and WFP4 could find a string located towards the end of the chained file, with the cursor placed in the first segment. No regular expressions or anything complex, just a good old plain text search. The result was almost instant with Trados (as in too fast to be timed), while Wordfast needed a full 6 seconds.

The interface is a disaster

Besides the general slowness of its functionalities,  Wordfast also suffers from a catastrophic interface. Every single action is tedious and far more complicated than it should be. You want to switch from a project to another? You’ll need to close the former one first (which does… nothing at all besides wasting your time). Need to run an analysis on a file? Make sure it’s closed first. If it’s not, you’ll need to close the error message, click the Wordfast tab, go back to the editor interface, close the file, then go back to the project interface and finally be able to run your analysis. Trados does that automatically for you and even offers to reopen files afterwards.

Time to export your finished project! You’ll need to manually choose an export path. Each and every time. Can’t just the software offer some default option, like the path you’ve used the previous time?

Oh, and the ribbon, wonderful. If you fold it, you’ll still see the tabs. But if you click on one of them, the ribbon won’t unfold automatically. What’s the point? Why do you need two steps for that? Even that good old Office 2007 is able to handle that for you.

You may say all these are fairly minor inconveniences, which is true. The problem is that they add up to the point anything you try to do ends up in frustration. You keep clicking around, hoping that you didn’t forget to do something the software expects you to – because it won’t do it for you, and it won’t let you know until the very last moment.

The editor is a pain to work with. It apparently needs to load something for each of your actions. For example, let’s say you want to get back to a previous string to make a correction. You try to scroll, but the editor won’t show any text while you do it, so you just hope to get lucky and find the piece of text in question by trial and error. When you’re done scrolling, you have to wait a couple of seconds until the text actually loads and displays (this is a desktop app, just keep the text in memory!). Then, when you click the cell you want to edit, wait again! The editor will need a moment to load whatever it needs until it lets you edit anything.

Didn’t someone test the software and realized how absurdly user-unfriendly it was?

Is this even a finished product?

Wordfast Pro 4 feels very amateurish overall. The interface is not only difficult to use, but also filled with buttons that do nothing or that are enabled when they shouldn’t be. Just an example: when you click an online TM in the Project TM tab, the Export button is enabled. You can’t export an online TM, but it makes it look so. But if you press the button, a window opens and offers you to select a TM… from an empty list, because you can’t export online TMs. Don’t you think an error message would be less confusing? Or even better, can’t you just disable the button in the first place?

Again, that’s just one example among others. You’ll often find yourself puzzled by such dead ends.

Above all, it has lost its direction

I used to recommend Wordfast to beginners for its relative simplicity. Up to version 3, it had a simple, clean and reactive interface, and all the basic features you would expect from a CAT tool. No fancy productivity hacks or project management features (that we translators don’t need anyway), but more than enough to work efficiently on most projects.

Whether that was intentional, I don’t know, but Wordfast had its place on the market as an easy-to-learn tool.

In contrast, Wordfast Pro 4 doesn’t seem to have a clear positioning. It tries to mimic both rival desktop and cloud solutions, but fails spectacularly by only offering the worst of both worlds. It doesn’t have the features and processing speed of Trados, nor the simplicity and reactivity of Memcloud. Instead of focusing on its strengths, it is vainly trying to chase competitors, years later. And still, I don’t remember Trados Studio 2009 being that bad.

Conclusion

I’m baffled by how much of a step backward Wordfast Pro 4 is compared to its predecessor. I don’t see much to save there. Some of the interface flaws can be fixed fairly easily, but it will still be slow and awkward. And in terms of features and processing speed, it’s so far behind its rivals that I just can’t see it getting its head out of the water anytime soon.

By trying to mimic the competition, Wordfast lost its soul, the speed and ease of use that made it an interesting entry point into the CAT world. And it gained nothing in the process. It’s become a slow, buggy, incoherent software headed nowhere.

I have stopped accepting projects involving WFP4. In theory, you can only use Wordfast to check TMs/TBs and edit .txlf files in Trados or MemoQ, but some of the custom fields won’t get updated, which apparently causes issues with some of Wordfast features. Yet another area where WFP3 was better, and a fatal flaw for me. Goodbye, Wordfast.

By Anthony Teixeira | January 18, 2017 | Uncategorized |

Internationalizing Ikinari Maou for LocJAM Japan

Introducing Ikinari Maou

Simply put, Ikinari Maou is a puzzle-solving game disguised as an 8-bit RPG.

It was created during the Tyrano Game Fes JAPAN 2016, a jam where the goal was to create short games with TyranoBuilder (or using the TyranoScript language), a visual novel creation tool.

The game won the 2nd prize, and was praised for its originality, the quality of the challenge and its clever use of TyranoScript’s capabilities.

As a parodic RPG, the game ticked all the boxes for the LocJAM’s needs: reasonably short, diverse (lines from different characters, menus, system messages) and using a vocabulary typical of video games.

The Localization Tool

For the purposes of the contest, Shikemoku-MK, the creator of TyranoBuilder, kindly wrote a localization tool that reads scenario files and extract localizable strings, with the possibility of previewing and generating the game in the target language.

Technical Challenges

Most of the technical obstacles we met during the internationalization process were due to the nature of tool. TyranoBuilder is originally designed for novel games. You will typically have the scenario read in a windows, with the occasional selection to make through textual buttons.

Ikinari Maou, however, isn’t your typical novel game. As mentioned earlier, it could be described an RPG/puzzle-solving game, albeit a very scripted one. And to keep that old school RPG feel, the developers had to push the tool to its limits and use its features in creative ways. Long story short, we also had to find ways to accommodate the specifics of the game to make it localization-friendly.

  • In terms of volume, the main task was to convert graphical buttons in text buttons to make them localizable without the need to produce new image files. Remember, the contest is aimed at translators, many of whom aren’t familiar with advanced image editing tools, so everything had to be plain text.So we went over the script files and converted image buttons (“button”) to text buttons (“glink”), switching the image attribute of the former to the text attribute of the latter. We also tweaked TyranoScript’s CSS files to render these links a little more nicely, as options were a bit limited inside TyranoBuilderThis part went fairly smoothly as text and image links essentially work the same way… for the most part.
  • On the name entry screen, switching the graphical confirmation button to a text version had the game crash and it took us a while to figure out why. As it turns out, graphical buttons and text ones work a little differently in the background. The way TyranoScript was designed, clicking on a text button would actually kill the name entry field before any extra code can be applied, effectively making it impossible to save the name entered (and making the script crash by calling an element that has been deleted). The solution here was to dig a bit deeper and make a couple of edits to TyranoScript’s JavaScript files, the core of the tool that converts script (scenario) files into the HTML your browser displays.
  • This time we had to prepare not only the game but also the localization tool, and so we had to debugging on both fronts. One issue that appeared is that the tool originally didn’t extract strings under the “ptext” tag, which allows you to place text anywhere on the screen (outside the space reserved to the “story”) and was used in a couple of places in the game’s scenario files.
  • Another challenge was to find a good format to make strings easy to translate in a broad range of software. We decided to go for a tab-delimited .CSV file using UCS-2 Little Endian encoding – a format that can easily be opened in spreadsheet editing software and most computer-assisted translation (CAT) tools. It is flexible, universal and allows for instant implementation of translated strings.
  • Finally, we had to adjust the original text at a number of places, because of the limitations of our localization tool and TyranoScript. For example, variables were not localizable, so we had to remove a number of things, such as the default character name. We also had to make sure there would not be strings concatenated on a single line. Otherwise the options were to either automatically add a space every time, which would have made punctuation non-compliant in certain cases, or ask translators to add line-breaks manually every time. Obviously the latter option would have been a little tedious, so we just opted to move the text around to avoid issues.
  • As for the font, we used Atari Small, which we carried over from the very first edition of the LocJAM. It covered all of our needs and matched the game’s feel.
By Anthony Teixeira | January 3, 2017 | Case studies, Game localization, Localization, LocJAM |

LocJAM Japan Presentation – Kyoto Study Group (December 2016)

Here are the slides for the presentation I gave for the Kyoto Study Group organized during LocJAM Japan. You will find a quick introduction to LocJAM Japan (what it is, reasons to participate, etc.), basic information about the tools we are using this time (TyranoBuilder-based games, and Tyrano Translator to localize them) and an overview of the game localization process applied to Ikinari Maou, the game we are offering for localization.
The idea is to have a grasp of how a typical game localization project works. Although we’re taking Ikinari Maou as an example, the whole process can be applied almost identically to most games, including analog ones.
Speaking of which, the presentation is based on the one I gave in March for LocJAM3, with a full text version available here.
By Anthony Teixeira | December 17, 2016 | Case studies, Game localization, Localization, LocJAM |

LocJAM Japan Unofficial FAQ and Notes

This year I had the chance to be closely involved with LocJAM Japan‘s (Japanese to English game localization contest) organization. Anticipating some of the questions and comments that are likely to arise, I decided to put together a quick, unofficial FAQ. I will do my best to update it during the contest, but once again please note that this is a personal initiative and I can’t guarantee all questions will be answered.

Known issue: Under certain circumstances, the text can overflow a bit in the main message window

The frame width is set to a slightly too small value in the .css file of the game. Depending on the text used, it might overflow a bit as on the picture below shared by Thomas Bruckert.

15369016_1204549326299024_3192057755960753697_oSince the issue is on the game side, it is absolutely fine if it happens with your translation. More generally, minor display issues won’t be taken into consideration by the jury. If you are still worried, you may use non-breakable spaces to force a line return – for example, that would be between “big” and “threat” in the example above.

Known issue (typo): “ファイラ” >  “ファイア”

There is a small typo in the list of “じゅもん”: “ファイラ” should read  “ファイア”, as spelled in other parts of the game. The jury has been informed and won’t penalize the brave translators who tried to translate this exotic name.

typo

Known issue: unlocalizable brackets

15304544_10154122201018193_5287771767486086972_oThe Japanese brackets on the screen above can’t be localized – this one is on us, sincere apologies! Technically the issue can be fixed within the .csv file, but it won’t be taken into account by the jury.

The game won’t start when I click on “Preview localized game”

This is a known issue with the first package we shared. To fix it, you can either:

 

  • Download the latest package on locjam.org
  • Open transrator.csv in Excel and save it as it is

So what is this CSV file? How do I open it?

The translatable file is using the CSV (tab-delimited) format, with UCS-2 Little Endian encoding. It needs to be strictly preserved for the translation to be correctly updated.

You can use spreadsheet software (Excel, OpenOffice Calc, etc.), text editors (I personally recommend Notepad++) or your favorite CAT tool to open and translate it. Specific instructions for Trados/MemoQ, and Excel are available here.

My text isn’t aligned properly, am I going to be penalized for it?

The goal of the tool is to give you a preview of your localized game, rather than to generate a final product. Accordingly, you will not be penalized for cosmetic issues you have no control over, such as text alignment. You should however try to avoid overflows and text encoding issues.

transrator.csv, seriously?

Yes, the translatable file is named “transrator.csv”, not “translator.csv”. This is an innocent mistake from the tool’s developer. He is not an English native speaker (nor a language professional) and unfortunately didn’t have time to fix this small typo. In no way is this meant to make fun of our profession/non-native speakers.

I want to add spaces to my translation but I can only display one at a time

If necessary, you may use HTML entities in your translation (for example, “ ” for a non-breakable space)

What should I do with lines starting with ScenarioName?

Do not edit “ScenarioName    XXXX.ks” lines, or your translation won’t be read properly. If you’re using a CAT tool, it could be a good idea to lock these segments at the beginning of the project

I translated the file, but the old strings are displayed when I run the game

– If the translated strings are not updated after you edited the file:
1. Make sure the file was saved at the right location with the original name and encoding (UCS-2 Little Endian). Also make sure you didn’t add tabulations as plain text (if necessary you can add extra spaces with “ ”)
2. Delete the following:
– “maou/trans.json” file
– “trans/IkinariMaou” folder
Restart the app. Select the project again (maoh/index.html), and the tool should read everything from zero again.

I played the original version of Ikinari Maou and the graphics/text/layout are different

Yes they are. The game was edited in different ways and for different reasons to make it more easily localizable. For LocJAM Japan, the version shared on locjam.org is the only one that matters. But you are very welcome to try the game in its original format and enjoy the better visuals.

Some text is missing from the translatable file/I can’t seem to find a string of the translatable file in the game itself

While we did our best to include only relevant strings without forgetting any, we’re not excluding the possibility  we missed or added some by mistake. Please feel free to report such issues, but don’t worry, the official file shared on the LocJAM website is the only one that counts. You do not need to fix it and we will not edit the official file during the contest.

By Anthony Teixeira | December 8, 2016 | Game localization, LocJAM |

Ikinari Maou Walkthrough: How to Beat the Game (LocJAM JP)

1) After the introduction, don’t answer “こうさんしますか？” and click on どうぐ > エーテル > ゆうしゃ instead
2) Click on じゅもん > チェンジ(MP30) > まおう（１ぴき）
3) Click on どうぐ > エーテル > まおう
4) Click on じゅもん > タイムリープ（MP100）
5) Click on じゅもん >エーテル > (White arrow) > ゆうしゃ（１り）
6) Don’t answer “こうさんしますか？” and click on どうぐ > エーテル > ゆうしゃ instead
7) Click on じゅもん > チェンジ(MP30) > まおう（１ぴき）
8) Click on じゅもん >エーテル > (White arrow) > ゆうしゃ（１り）
9) Let the magic work!

By Anthony Teixeira | December 8, 2016 | Game localization, LocJAM |

[LocJAM Japan] How to Translate Ikinari Maou in Trados, MemoQ, Excel, LibreOffice

Ikinari Maou is the game offered for translation for LocJAM Japan, a Japanese to English game localization contest. This article offers pointers to translate its localizable strings in different CAT and non-CAT tools, and may be updated as questions arise. Please also note that this is an unofficial guide, with no guarantees whatsoever.

The localizable file was successfully tested in both SDL Trados Studio and MemoQ. You can create a project as you normally would, with the following settings:

SDL Trados Studio (2014 or later)

To translate in Trados, please use the following options for the .csv file type in your project options:

How to localize Ikinari Maou (LocJAM Japan game) in Trados

MemoQ (tested on version 2015)

When importing the file in MemoQ, please use the following configuration:
Localization of Ikinari Maou in MemoQ

In both cases, your CAT tool should be able to export the .csv file in the right format and encoding, without any further modification.

Excel

If you are working with Excel or similar spreadsheet software, you will be able to open/save the .csv file directly, but you may have trouble exporting files in the right format if you add certain characters (quotation marks, etc.). In that case, I would recommend the use of the free code and text editor Notepad++. Once installed, copy the source and target columns from Excel (or similar software), open the original (untranslated) file in Notepad++, select all the text (Ctrl + A) and replace it with the content of your spreadsheet content. Save the file, and everything should work smoothly.

Of course, you can translate directly in Notepad++, but you will most likely find Excel more comfortable to do so.

LibreOffice

A member of the IGDA LocSIG Group on Facebook, Anish Krishnamurthy, kindly shared the following settings to open the file in LibreOffice:

15317817_1546447495382355_7852266833573856023_n

By Anthony Teixeira | December 8, 2016 | Game localization, LocJAM |

Game Localization Link Roundup – October 2016

October was another exciting month in the game localization industry. This time, I’m sharing a few fascinating interviews with industry insiders as well as insightful articles about the localization business.

During the past month, we also made tremendous progress with LocJAM Japan’s preparation. We now have a fully functional tool and an internationalized game ready to be shared with the world. We were lucky to find a game of the perfect size for our needs AND incredibly fun to play. Even if you’re not joining the contest, look forward to playing it, next month in Japanese or in English when the winning entries are announced!
The importance of Localization Quality Assurance

Localizing a Unity Indie Game: The Hidden Costs

Sega, Ubisoft, Deep Silver, Gameforge: What’s New in Game Localization?

We Work With an Army of Translators, Says GameHouse Translation Manager

Interview: Localising DRAGON QUEST VII: Fragments of the Forgotten Past –
An interview with Oli Chance, who had kindly answered our questions about Ni No Kuni a few years ago

Japanese visual novel Steins;Gate 0 gets EU release date, new trailer – The game’s localization lead discusses the complexity of adapting the story to the west

8 Top Internationalization Changes in iOS 10

By Anthony Teixeira | November 21, 2016 | Game localization, Localization |

Game Localization Link Roundup – September 2016

After a relatively calm summer, September was richer in game localization-related content. Besides the Shin Megami Tensei IV: Apocalypse story, which ended with an official apology from Atlus, a few interesting interviews with veterans of the industry popped up, and the trend continued into the first few weeks of October.

We are also preparing to officially announce our plans for LocJAM Japan, so you can expect a busy end of year for the localization community!

Interview: Brian Gray on localizing Gotta Protectors

SMTIV: Faux Pas-calypse – Or how a string accidentally left untranslated can result in apologies from the developer. The issue itself is far from a first, but these things usually get swept under the rug and it was interesting to see a developer officially touch base with the gaming community regarding this

A Look At Untranslated Text In Video Game Localizations – A small compilation written after the above-mentioned story

An Interview With Mastiff: The Passion Behind Localization

Sword Art Online: Hollow Fragment – Death By Localization

Video: How Capcom localizes games like Monster Hunter – Based on Andrew Alfonso’s (Capcom) presentation at the GDC 2016. In this video, he explains how Capcom’s localization directors address different challenges with the localization of Monster Hunter 4 Ultimate as an example. A very complete presentation and a rare chance to hear the story directly from in-house people.

By Anthony Teixeira | October 24, 2016 | Game localization |

Why Localization Quality Matters For App Sales

Improve app sales with quality localization services

The importance of localization is now well documented. You can find dozens of case studies out there, with stories of sales increased by a 2 to 3-digit percentage, sometimes simply by localizing app descriptions. This promising reward doesn’t come without effort, though.

Although localization can be a very powerful sales enabler, quality is essential to make the most out of it. Indeed, localization-related issues can ruin your globalization efforts, and even have a negative impact sales for your app’s native language.

Reviews influence downloads

Reviews are a decisive factor when people decide whether to download an app or not. A couple of bad reviews could be enough to cut your download numbers in half, and lack of/unproper localization is one of the most common sources of negative reviews.

Nobody likes an application that looks like it was processed through an automated translation engine. By releasing low quality localized versions of your app, you could cause ratings to dive and lose the trust of potential customers – even in the original language version. In that respect, localization can be a double-edged sword if not handled properly.

ASO: you need the right localized keywords to be found

Working on App Store Optimization (ASO) is one of the most important strategies to improve your sales. So when you try to reach new markets, it makes sense to localize every aspect of your app, from description to keywords, to ensure it shows up in search results – ideally ahead of competitors.

When you do it, it is vital to use the right keywords in each language. If you use low quality localization services or if your provider lacks market knowledge, the odds are high that you will lose out to rival solutions, because you won’t be targeting the right keywords.

You need to work with someone who perfectly understands the target market and is familiar with your industry. A synonym won’t do here, you need the one keyword that most people interested in your app are going to look for.

How to make the most out of localization?

– Hire a professional, specialized translator. The providers you hire shouldn’t just be native speakers of the target language. They need to perfectly understand your product, the market and what users will be expecting. Remember, a “good enough” translation won’t cut it. Only an accurate and perfectly localized product will allow you to truly optimize downloads and purchases. Look for experienced translators who clearly list applications as one of their main specializations. The more targeted the better. Ensure they have a good understanding of app localization challenges and concepts such as ASO, and check their track record.

– Test localized versions. It doesn’t have to be super expensive. Even if you have a good translation for your app, different types of issues can happen during implementation and harm your efforts. Only release your localized app when it is truly ready to achieve better ratings and the improved sales that go with. If you are short on resources or budget, start with just a couple of languages and expand progressively.

By Anthony Teixeira | October 11, 2016 | Game localization, Localization |

Is Localization Testing Necessary And How To Reduce Its Costs

Localization testing - How to reduce costs

Localizing your games for different languages and markets can be an extremely effective way to boost your game sales. Besides reassuring your potential audience, it is an important part of App Store Optimization (ASO) and can help your games get featured on mobile game stores.

Very promising on paper, but localization isn’t free, neither is localization testing. It can be tempting to skip the testing part once you have all the localized assets in your hands. After all, you spent enough time picking up reliable language service providers, so nothing can go that wrong, maybe a few cosmetic issues at most… right?

Proper localization testing IS necessary, no matter what

Compared to the console market, localization requirements for mobile game stores are pretty loose – and that’s a mild way to put it. It may sound like a blessing to developers, but it is a double-edged sword that doesn’t do gamers any good, as they are often left with games that look unfinished or impossible to decipher.

More languages = More sales? True, but only if you get it right. The only thing a bad localization is going to give you is a long string of bad reviews from upset users, potentially dragging down your overall sales – including in the game’s original language. You should strive for quality in every single aspect of your game.

Even if you hire a strong localization team, there are still lots of ways things can go wrong. Lack of context can lead to inaccurate translations that will harm user experience or make you the next Internet meme. Implementation issues, such as text overflows, encoding problems, missing strings and the like will make your app look buggy and unfinished. There will always be something, the same way games always have bugs before functional testing.

It doesn’t matter how much and in how many languages you localize: whatever you do, try to do it right. Any half-hearted effort could ruin an otherwise great game and prevent it from selling as well as it deserves.

Budget limitations are obviously the pain point here. However, you can get most of your localization testing needs covered at a reasonable cost.

How to keep costs down without affecting quality

There are lots of ways you can optimize the QA process and avoid overspending. Here are a few general tips to help you start out:

– You don’t need to do everything at once. If you don’t have the resources to properly localize your app, reduce the number of localized languages or start by translating store descriptions and texts essential to the understanding of the game. Players will be more tolerant toward a solid game written in a foreign language than a game that looks unfinished and buggy. Start with the languages you consider the most important, then add more as sales pick up.

– You don’t need to outsource everything. Having garbled characters or overflowing texts in main menus, tutorials, etc., things every player is going to see, is simply unacceptable. Yet, these issues are very easy to detect, even without any knowledge of target languages. There is absolutely no excuse not to do any testing at all.

Your budget may be tight, but you can probably find a couple of hours to do implementation localization testing for your game. Reduce the scope for your external resources and save on QA costs, as your testers will be able to focus on the text itself, rather than spending time reporting issues you could have found yourself. You should also consider pseudo-localization to address some of these issues before any actual translation is implemented.

– There are different levels of testing, choose what works best for you. If you really don’t have the budget to do extensive testing, you should at least aim at an acceptable quality standard, something people won’t be complaining about. You ideally want to have a perfectly polished product, but in general players will tolerate a small number of minor issues. At a minimum level, test implementation yourself and ask your translator or another professional to check the game in context (to make sure everything makes sense) and have a final look at spelling/grammar.

It will help to have a well-defined testing plan and cheat codes so testers go through your game quickly. If you manage to optimize that part of the process, you can probably keep testing costs at 30-50% of the translation fee. Once again, you should consider more thorough testing once finances allow, but it should be enough to achieve a decent standard of quality.

– Don’t make it more complex than it needs to be. It’s really all about optimizing the process. Translators will be able to fix most issues by editing localizable files themselves. Technical issues aside, you maybe don’t need them to fill a complete report with dozens of fields, screenshots and so on. Again, the key is to have them focused on testing itself.

By Anthony Teixeira | October 5, 2016 | Game localization, Localization |