Partager :

Vidéo Jan 24, 2025

Software Internationalization – with Adam Asnes

Welcome to three questions with MotionPoint. I'm your host, Dominic Ditherdy. And today, I have the absolute pleasure of speaking with Adam Asness, CEO and founder of Lingoport.

Lingoport helps companies and organizations internationalize, localize, and test software so it works well in every target language and local format.

Adam, how are you doing today?

Good. Good. That's a very nice introduction. Merci.

Absolutely. Well deserved. For for a veteran in the industry, LingoPort's been around, you know, for over twenty years, and it's it's a pleasure to speak to speak with you today, Adam. It really is.

Thank you. So pleasure as well as mine, and and nice to meet you, Dominic.

So Thank you, Adam.

Looking forward to this.

Let’s do it then. Let’s dive right in. So we’re gonna talk about internationalization, of course, and the customer experience.

So Okay.

How would you describe internationalization and its impact to the customer experience?

Well, I I'm gonna say, you know, there there's the obvious things that that, you know, people will prefer to, interact within software in whatever language is is most comfortable for them. And I think that there are, some people who get lost on this, initially because, let's say, you know, they they take took a trip to Paris and, you know, as soon as they tried to speak French, people switched to English because they didn't, you know, like, because they were being accommodating.

There’s lots of experiences like that.

But, and that’s speaking with a US centrist centricity.

The fact that people might speak English doesn't mean it's their preference. So for instance, I work with a lot of people around the world, and it's very easy to notice within their correspondence or screen sharing that even though they speak English really well, they are choosing on their computers and in most of the software they're used they to to work in their native language, the language that they're most comfortable with. So, whereas there are markets where you might get away with English as, the language, you might not be the preference, and that speaks volumes to the user experience.

And then there are parts of the world where you're just not really competing if you're in English, even with, technically oriented products because they can sort of sometimes get away with a little bit more. But there's so much more to internationalization in the user experience than just the language, although that's the most obvious thing. Mhmm. You know, think of numerical formats, here in the US written as one comma zero zero and many other places is one point zero zero zero as the number one thousand.

In some markets, the number grouping is going to be different. So, you know, we have, blocks of three digits together, others with blocks of two. Right. So, you know, if if you're talking about things like numbers, you really wanna have that right in the the user's mind, or do they trust your application?

There's so many what we call locale specific formats from addresses to even how plurals are handled. I I might even bring up plurals again because it's something very non obvious to the average developer or even marketing individual that, pluralization is is a big deal in some markets.

Gender of, objects, all kinds of things come into play. And even for a US company, if their customers are multinational companies, they're gonna have internal data that is multinational.

And so if you’re not parsing an address format, right, or a phone number Mhmm.

Or any number of things like a currency, it's gonna look bad. So I I think that, you know, it used to be a little harder to talk companies into the business case of internationalization.

It’s now a lot easier.

You still see applications that have zero internationalization, and, you know, we have this whole services division that that goes and fixes that. But more often than not, it's that there are internationalization bugs that are just simply sitting around for years and years because nobody, is going to take the time to fix them, especially if they can't actually locate them in the code. It's not always obvious.

So, I I joke that, you know, IT and N issues go to Jira to die, basically a bug tracking system because, you know, you look at it and you say, well, is this a priority fit a thing? It's it is if you're in this other country and if you're a localization professional, it's an important bug. It's really a bug, not an issue.

But to, many others, they may not judge it that way and not wanna go on to the sleuthing k, aspect of why is this date or time in a format that we don't want. You know what I mean? Like, finding that is not always obvious.

So Yeah. It kind of reminds me of of back in the day in high school when we were studying my middle school, we were saying, like, significant figures. Right? And and and and every in every language and in every culture and and region, you know, there there's formatting and and ways that are easier for you to recognize and understand. This was made and built for me, and when it doesn't feel that way, not only are you having people focus more on their frontal lobe and making conscientious, like, rational thinking, you know, instead of letting it just be, you know, a little bit unnoticed. Right? Sometimes you think about the word or the saying, like, you know, if nobody says anything then there's no problem, right?

Right, right, right, right, and you you rather have a customer experience that feels familiar where you’re not really having to struggle, you know, is this day in the right format, you know, is this is in dollars, but I, you know, I want to buy, you know, using the euro and, you know, creating that experience where there’s parity.

Right? And you're not really or what's the call? Like, you know, maybe feeling like you're underwhelming or under serving other markets. n'est-ce pas ?

Mhmm. Mhmm. Mhmm. So let let's, remember that business case really rules any globalization project, whether it's localization, localizing a a website, documentation, or ultimately software. And and when I talk about internationalization, I I I really am in the software domain.

Right.

But, there's always it always comes back to the business case and not, you know, way before you get to the technical aspects of it. Like, people have to believe in the business case.

And, you know, what what I'm sort of fond of of of giving is an example is that, you know, a a product designer will give a lot of thought into the color blue and the shade of the color blue and, the layout of the UI, and they'll prototype that, you know, in most organizations these days, you might have the designer building wireframes and say, Figma and and really focusing on what is that user experience going to be on. Yet that same organization has, nothing to catch an internationalization issue until, like, maybe they find it after localization. So really late in the process. Yeah.

And, use systems that really automate that effort from internationalization to localization to QA and really establish processes that they actually have already for design, deploy, you know, development, QA, and, deployment. Like like, those are really well established capabilities in an organization. And yet, if you look at the business case, back to the business case, one of a company's most significant ways to grow their revenue is to just by making things available in in a customer's language and to you know, that brings on other things like distributors or, you know, Salesforce, other investments that kind of need the best product possible because it doesn't you know, in in the realm of product development cost, it doesn't really cost that much more to do it right for global markets. Right? So it's it's one of the better investments you can make in your product development to back up your global organization and to grow your company. So it always comes back to the business case.

Absolutely. No. And and we we see that a lot in different types of, formats in the translation and localization industry in general. Like, for example, you you you you'll spend hours designing a landing page, and then you translate that website, and you didn't account for the word growth that tends to happen when translating from English to another language, and now there's misalignments and broken templates Yeah.

And it’s it’s it’s about building applications and landing pages that render an experience that’s readily available to be translated or to be offered into another language, and what happens is especially here in the US is, you don’t really have translation and localization or internationalization as a core competency, and it’s really not something that marketers, you know, developers tend to be thinking first.

They always optimize for their flagship experience. Most of the time, that is in English, and, it does trickle down to a a big fault and opportunity later on. So that's brilliant. I really like how you articulated that, Adam. And I think I think it'd be great if we could kinda move into that next question here. It's what are some of the complexities associated to you know internationalization especially internationalizing software applications?

What do you say?

So so again one of the chief complexities is to it is is as as high level as making sure all the all the stakeholders understand the business case and Mhmm. The complexity of what's going on, which I'll I'll go into. So for instance, that starts with, both your senior management, kind of trickles into your product management, your localization management. Those should be the really, the the the banner, they should be leading, carrying the banner, off of this, because it's their job, but also to impact the developers, the designers, the developers, and the testers so that you have really, like, a a a a complete view in the organization.

So it's it's planning, it's product management, it's design, it's development, and QA. It's not just a developer problem.

Although it's you know, when it comes to building everything, the developers have to do the right thing. You know?

It's just another form of coding quality, but until you have that message of why you need to look deeper, it's the same thing with security. Like, people took security really lightly, and then, you know, people were getting in the newspapers for bad reasons for security breaches, and they had to build security into the process of how they write software.

It's the same thing only instead of the stick being hit with the stick, first with a security breach, you have, you know, sort of sort of the carrot ahead. You know, if you internationalize well, you got a good product. The stick is that you don't wanna go back and have to fix stuff. I mean, it's just like no developer wants to do that. n'est-ce pas ?

Yeah.

So, that’s what I’d say, you know, first.

And then in terms of complexity, it's really easy to have I t n issues creep into software development. So, you mentioned things like, string length, you know, that that suddenly the words are, too broad for, you know, let's say in a product, it it's the button. And I'll talk about that in the next question.

Like like, I had this great conversation yesterday with a customer about this.

But that that's one obvious aspect. But, you know, a lot of times, it's as simple as, you know, word order. You know, did did did the developer bake in word order? You know, because literally developers are taught to build a message based on various variables.

So a simplified example is I have five, a quantity, red, a description, apples, a, you know, the thing with a parenthesis for the s at the end if there's more than one. Right? So all of that is learned in programming one zero one Mhmm. To build the string in parts because it's efficient coding, but that doesn't work for localization.

So people have to unlearn and think in a different way, because the, number, format may be different. The, descriptor, the red of the apples might come after the apples, and the pluralization of the apples might be a different thing. And the apples might have a gender.

Right? So you have to make it so that the translator can be successful and give you a good translation.

Otherwise, you get word salad. And, again, you you it's it's very cute for for many Americans to understand, when they go to a Chinese restaurant and the English translation is kind of funny. I'm I'm picking on one type of restaurant. It's cute, and you accept it in the Chinese restaurant. That's okay. Right?

Yeah.

You wouldn't accept it in in in a bag that's in in an applications that's managing your money. I mean, there's no freaking way. You'd be like, what is this? Do I trust this thing? You know? Is this You're absolutely right. n'est-ce pas ?

Yeah. So yeah. I get that. Yeah. If I were to surmise that, it's it's like Yeah. Well, forget about the complexities are technical and not going back and fixing it. That that that's that's hard enough.

The biggest complexity is, fundamentally, you know, thinking and getting the organization to think in a way that’s inclusive of, you know, avoiding these pitfalls in the future Right.

Which from, if you get it right at the beginning, we'll render very, like, easy to do, easy to scale internationally, solution, not only for for, you know, customers in other languages, but for for your own internal operations to not have to go back and fix things. If you get it, you have it in your mind to to build it in a in a way that's internationalized from the beginning, you'll be better off moving forward. So it's it's almost like a, you know, shift in the paradigm thinking of being very isolated into English only English only. And I'll get them picking like you were just isolating the US market and and actually being very, you know, practicing diversity and inclusivity and in and inclusion at the product development level.

Right? And that Right. That comes from the CEO, and it trickles down. That's that's very interesting.

Alright. I get I guess I'll I'll move on to the next one, which is in your experience, and I think that we we were getting into this already. Right? But how should internationalization be baked into every sprint and and release for an efficient and fast rollout of Broadcom?

Mhmm. So, I’ll give some examples.

I’m gonna say that, yes, it should be baked in because the the easiest way to fix an ITN issue is to fix it when it’s created, not when you found it after localization.

Right? Or or even in testing. I mean, testing is better than not finding it, but but that means going back and and fixing. So Yeah.

Like, our solutions will either check things as the developer goes along. That's a little bit more intrusive, within their development environments.

It will automatically check things every time they commit, which is a developer word for save.

Yep.

And, furthermore, it will absolutely check automatically during a pull request, which is when a developer submits their work to be included in the product and reviewed by their peers.

Mhmm.

Right? And so, they look at those pull request reports and say, oh, there’s ITINN issues, ITINN being short reports and say, oh, there’s ITN issues, ITN being short for internationalization.

And here exactly is where the issue is. So now instead of a sleuthing effort, it's like a quick fix, generally, to to do something about it. And so you don't build this backlog of issues that nobody wants to deal with later. So that's the first thing. And we have integrated testing, stuff and, as well on the QA side, there's there's quite a bit to help testers be able to test without having to speak a target language.

And, then, at the design phase so, like, I was talking with a customer during a webinar. I was interviewing her just yesterday, so timely.

And she was talking about how she was using one of our products to, automatically, just using machine translation, work on at the Figma design stage. Mhmm. So and what they would do, again, using our stuff, is see in the actual language, what is the longest possible string that's going to be here? And so the designers, when they create their wireframes, they see it.

It's not like it's not like they find out later that in German, it it's this long and, you know, in French, it's only this long. Like like, they see it right up front. And so they can design around that, which is, you know, very powerful. And and given how inexpensive a machine translation is now to to use it, you know, you're talking, you know, a few dollars per millions of characters.

And it's not that hard to to do training. And now we have a chat g b t, turbo, which is also good if you're willing to put in the work and open up to using it, which is another question. But, but I I'm not saying that's your final translation, but as a design tool, that's very powerful because you get an immediate feedback, not something, you know, three days or three weeks later. You you get it right away.

So that’s that’s an example of integrating not just internationalization checking, which we’ve been doing for years, but this this kind of loop of machine translation.

Or, you know, you can use pseudo localization also, but I I like the machine translation impact the developer. For the developer's seat, this is what this will look like.

Yeah.

Also, not just not just the designer is very powerful because you have developers adding error messages all the time, you know, like like the designer didn't think of or last minute changes. We also have all kinds of string management there, as well. So it's ITN, the localization automation, which we do a lot of.

Mhmm.

It's it's sometimes as simple as, hey. Somebody some group has their own version of a JSON format that they're using to pass to localization. There's no way that the translator can work with that. So we'll we'll transform that for them. So there's there's all these things that get labeled as internationalization that actually kinda creeps out beyond traditional internationalization.

But it's really fascinating how you can use automation and systems to really make this super, you know, small but mighty. That's how she described it. Localization team that that has all this power and all this impact on the organization.

Yeah. I don't know why I'm thinking about, like, the the effect of a droplet in a steady, you know, pool of water, and it it calls it all those ripples, you know. Yeah. Yeah. You get it right at the beginning. It it in most emotion point, we tell we tend to call that in context translation where a translator Right.

Can actually see how that translation is gonna render.

Because if you're translating into German, there's a certain amount of word growth. Right? There's following a word in German for everything we just said in this conversation because it's and, and being able to to manipulate the type of translation you would actually publish by understanding how it would render on a user interface.

Right.

It is a massive benefit into what we first talked about, which is the customer experience. And you don't want to have, like, you know, a broken templates, you know, unit units of measurements that are that are erroneous, you know, currency exchange rates that are not Right. You know, making you up as profitable as you should or as, you know, showing that the parity that, you know, every different market deserves to to be shown, from your product. Uh-huh.

So so it's it's it's it's very interesting talking to you about this because, you know, we're more we're more on the local website translation and localization side, but that is not as scalable or as effective if internationalization isn't properly set up at the beginning. And that's Yeah. That that was a that was a great conversation. I have a bonus question for you, Adam.

Sure.

Explain explain to our viewers why I ATN stands for internationalization.

Sure. That's, it it first of all, that's like history in the nineties or something. L ten n is also for localization and then, you know, g eleven n, you have t nine n. I I mean Yeah. Yeah. Yeah.

I I I'm happy to say I eighteen n is is a pretty common term. You you can go on, you know, Google Analytics and see, you you know, like, it's almost as much searches on I eighteen n as there is internationalization, which is kind of interesting. It's just a bizarre fact that, I have to be aware of.

So it’s I eighteen lev letters in the middle, n at the end.

So it’s called a a numeronym.

Numeronym.

I actually do not know that that’s actually a word.

Yeah. Yeah. Yeah. You know, there's a famous famous, venture capital firm, Andreessen Horowitz, and I forget. But but they actually borrowed that, and they call themselves the, a, and I forgot the amount of letters that is in between, and z. That that's what they go by professionally. Yeah.

So they they borrowed it from an I eighteen n problem they had many years ago, which is When I first ten seven.

When when I first started in the industry, I started seeing these these these hacker, and we have to use these hashtags and oh, like, but what does I eighteen end? Like, why does that stand out?

Right. What is this?

Eighteen letters in the word internationalization between the I and the n. That’s a lot of letters, you know?

Well well, when you have to write that when you have to write the word internationalization over and over again, you you really start to appreciate I t n. You know?

It's it's almost like a finger twister. So, I t n is nice and easy. So that that's funny.

But anyway said numer numeronym?

Numeronym.

That's a new word. We need we need to have a new a numeronym for the word number. I'm messing myself up over there.

Globalization trivia. There you go. Love it. You know, may maybe you'll you'll you'll win a contest at a bar with that. I I I doubt it. Maybe at LocWorld. No LocWorld should have a localization, trivia.

That’s true. We we got we got we got a we got a send it as a jeopardy question too.

I think I’m doing There you go.

Alright. Well, that does it for three questions with MotionPoint. Adam, thank you so much for having, for for being here, and until next time. Thank you very much. We'll see you again.