This episode will introduce you to the world of BiDi localization and all its challenges. Gilad Almosnino worked at Microsoft for 20 years and he’s here to challenge the status quo.
In this interview, you can learn about Gilad’s mission to improve the standards for BiDi users. But we have a lot more for you:
- Why you need a robust and inclusive design
- The problems of mirroring UI
- Going against the rules to make Windows better
- Why MS, Google, FB, and Apple gave up BiDi users
- What’s wrong with Unicode’s BiDi algorithm
- Why customers don’t have the power to change things
- Human Design System
- How to… bribe people?!
This is episode 36 of my social practice, also known as The Localization Podcast 🙂 #localization and #translation insight delivered to you by the power of voice, this time with Gilad Almosnino.
⏲️ Timestamps:
0:00 – Intro & Gilad’s background in Microsoft
07:47 – Design issues, lack of robustness and inclusivity
15:01 – When you first noticed issues with BiDi?
24:05 – UI mirroring
30:08 – Who should be fixing BiDi issues
33:16 – Why MS, FB, Google and Apple gave up? But Spotify and Netflix rock
47:15 – The issues with the text input
56:30 – Hebrew vs Arabic, collaborating together
1:06:10 – Human Design System
1:13:58 – Gilad as a paragliding instructor
1:22:39 – What is wrong with the localization industry
Andrej Zito
Let’s go. Welcome to the podcast first of all.
Gilad Almosnino
Right. Thank you. Great to be here.
Andrej Zito
How are you doing?
Gilad Almosnino
I’m doing great. Not too bad.
Andrej Zito
You have a special time in in the month, right?
Gilad Almosnino
Yeah, yeah. We’ll get back into that. Probably later.
Andrej Zito
Maybe we can talk about it later. So Where exactly are you right now?
Gilad Almosnino
I’m in a small town in North Israel. Not too far from Haifa. Haifa is the big city. A small town called Kiryat Tiv’on about 15,000 people.
Andrej Zito
Is it your hometown?
Gilad Almosnino
Yeah, same town. I, I was born there actually live in the same house. I’m 42. Now. So back to my childhood home. Yeah.
Andrej Zito
I remember when we were doing the recording for the first time you were a little bit mysterious about your age.
Gilad Almosnino
Really?
Andrej Zito
Yeah. I think I was just listening to the recording. And before we started doing this, and I think you were saying like, Okay, if you tell me about the by the issues, when you first notice them, then we will guess how old you are?
Gilad Almosnino
Yeah, exactly. Because I think you were you were asking about, you know, when was the first time you saw BiDi issues? And I said, Wow, you’re gonna find out very quickly. I, I when I first saw computer, it was running DOS, so
Andrej Zito
Yeah, okay. But you didn’t get into localization in Israel, right?
Gilad Almosnino
No, no, I actually moved to the states when I was young. And and I think when I was around 20, a friend of mine said, You know, I was going through college, and a friend of mine said, Hey, you know, Microsoft is looking for Hebrew native speakers to test out the latest version of Windows that’s coming up, which was Windows ME. And he said, You know, it might be a fun gig, they don’t pay a lot, you know, it was like 15 bucks an hour or something. But you know, you might enjoy it. And I had plans, like I thought, you know, Hey, I’ll just work at this gig for about six months or so. save enough money to go to India. And, and, you know, and that’s, that was my expectation, ended up being about 20 years in the company. You know.
Andrej Zito
Why did you want to go to India?
Gilad Almosnino
I don’t know, I just, you know, it was one of those things where maybe maybe it was spiritually connected to to India or something, just, you know, something there was, you know, the adventure.
Andrej Zito
Was it supposed to be your first time to visit India? or?
Gilad Almosnino
Yeah, no, I haven’t been I actually haven’t been there at all. So I never got to go to India. Never got to it. But it was just one of those things when you’re I don’t know, when I was 20. And I wanted to be adventurous. I said, I go to India or something. Right. So that’s, that was the motivation, there was nothing behind it. But yeah, I ended up taking a rather long trip at Microsoft, which was just as entertaining. I think.
Andrej Zito
What what made you stay at Microsoft for so long? Like, how do you remember your first work experience with them when you didn’t think about the job very seriously.
Gilad Almosnino
I always thought about the job very seriously. But I think I don’t know what the drive behind the job was. I initially started as quality assurance, right. And I think very quickly, people realized, you know, they offered me a full time position within less than a year because my my work was was very, very good. BiDi tends to be a lot more buggier. But it also requires that the person looking at it has a very deep understanding of what’s going on, you know, but I think very quickly, they realized that, you know, you can’t, you can’t have to just do in quality assurance, because the issue I would find a lot of the issues were design changes. And so having design changes come in A very late stage is a very big problem. And so very quickly, you know, I kept that quality assurance title for quite a while, maybe like 12 or 13 years. But my job was nowhere close to quality assurance. I was basically free to do whatever I wanted, in terms of impact on the product, and vision. And so the title really wasn’t anything that drove me ever. The the team, the title, it was never an issue, right. Like it was never something that I felt bound by. So my I was always trailblazing, if you will.
Andrej Zito
When you have this freedom, was it related only to Hebrew? Or was it about internationalization? Vidal languages in general?
Gilad Almosnino
I think the fact that I was so good at what I did for Hebrew and Arabic and Vida in particular allowed me to venture to other areas as well, you know, sometimes the issues come out with bidat languages a lot more clear than they would on other languages. And so when you’re able to take a issue found on bidang, say, while but if you think about some of the other locales, they’re going to get impacted by this, then quickly, you start not just thinking about vida, but just thinking about the larger picture of internationalization. Some, some of the issues or design changes, some of the issues are technical issues, but a lot of I think they’re nationalization work is being able to project a vision forward for a product. And that takes very special people. Right? It’s not, not everybody can do that. And I think I was very good at it. And I enjoyed it. And I had the total freedom. And, you know, some managers love my freedom, and some managers didn’t know how to, they didn’t know what to do with it. Because I’ll give you an example. For many years, my impact was with, you know, design changes and bugs and making sure that we’re shipping something that’s relevant to the market the BiDi market or internationalization in general. But take a manager that’s being pressured to have their team produce automation. I mean, it doesn’t make a lot of sense for somebody like me to sit there and write automation, when I’m making significant changes to the product design changes, you know, and so for me, it was always a struggle to do the bare minimum, to keep the managers happy with the formal title, and then make sure that I’m driving innovation across the product, and challenging myself. You know, I’m always here for that challenge.
Andrej Zito
When you talk about fixing or improving the experience for bida, and how it affects the other languages, are we talking about, let’s say, fixing the root cause, which would have an impact on the source English product?
Gilad Almosnino
Yeah, yeah. Even sometimes you’re able to some of the best design changes, basically, reduce bottlenecks, you know, you want to produce a robust enough design that can handle all these different variations of locales, and localization or whatever. But if the designers are robust enough, the user experience is great. Right? Both for the user, both for the developer developing on that platform, it’s it’s a win win situation. So I guess, being somebody like me, my, my, my, I guess my my best qualities is being able to reduce these bottlenecks and make the design or at least the people working on the design, think that they have to produce something that’s robust enough, so that down the road, or they have something that’s sustainable, that works that makes sense, and not some patch.
Andrej Zito
When we’re talking about robustness. This may be a stupid question. But do we always want to have just one source and one design for all the languages? Or have you come across any examples where you might have a different source for BiDi languages?
Gilad Almosnino
Well, I don’t think so. But But I’ll give you an example. Okay. So one of my first ideas when I came to Microsoft as a full time employee, and during my interview, I said, I have an idea. You know, this was 20 years ago, this is like 2000, I think, or 1999. And I said, we need to have gender based localization. So the UI strings are based basically translated for the user with female form and mass masculine feminine form, okay, but not not on the same UI. So basically, if you’re a male, you can select Hebrew male version, or if you’re a female you can select if you choose Hebrew v version. And I think that solves a huge problem. Because you know, the machines today, even even today, they refer in some languages, you just can’t work around the gender issue when it comes to localization. And so the, the translation often refers to a masculine form, that’s not very inclusive. And so I think today with the technology of machine translation, we can actually take the Hebrew source and do the the analysis and application of, you know, masculine to feminine form, rather easily in rather cost effective in terms of resources. But you know, what’s stopping us from doing that. So like the resource loader on Windows won’t allow you to have more than two versions of more than one version of Heber. So you’re never going to be able to put this other version in there. So that’s one example where the design isn’t robust enough, you’re not able to do two versions of Hebrew, two versions of Arabic. By the way, when when, when we when I brought up the idea, one of the, one of the locales, I thought it was great was the Japanese locale. Because the Japanese, the Japanese market thought, Wow, we can create a version of Japanese for children. You know, because they’re, they’re having a hard time reading the UI, because they’re not at that level yet. And we, we could use a second version of Japanese so that if you’re, you know, seven to nine years old, you get a version that you can read. And so that just goes to show that, you know, one thing that came from, you know, thinking about gender, inclusiveness for Hebrew and Arabic, quickly turns into age, inclusiveness for Japanese. And you just can’t do that today with Windows, for example, because they’re not able to, to have two versions of the same language loaded.
Andrej Zito
So what happened to that idea?
Gilad Almosnino
The funny story was that came to my, I think it was like the director, level manager that hired me into Microsoft, he was a great guy, I’m still in touch with him today. During reading, I think he’s a VP of internationalization at Facebook. And he is actually the guy that invented m UI, the resource loading technology that allows the localization to be separate from the code. And, and I came to him and I said, You’re and I have this great idea for gender based localization. And this is 1999. You know, people are still using the same profile at home, they have one computer, it’s not like today, your phone and your Facebook and whatever, any. And he laughed, and he said, Well, I mean, my wife is German, he said, me, my wife, we use the same profile. And so you want the machine to refer to me and mess in feminine form. I don’t think that’s a good idea. And and I laughed, and I said, he was a great guy, it was all a good conversation and said, you know, Bjorn, you don’t understand me, you’re not a woman. And and he looked at me, and she’s like, you want to talk to my skip level, then skip level was was Laurie Brunel. And I had a chat with her. In general, I think they, they thought it was good idea. But the I think at the time, they just couldn’t allocate the resources to have the resource loader modified to have the idea come come through. So I think this, you know, plenty of years later, same issue for Windows, at least, maybe for Facebook, you know, where there were, that’s not an issue, they can have another version of language. But so that idea. Today, I’m actually pushing that idea as part of my work as the chairman of the Hebrew Hebrew support committee in computer systems, that the standard Institute of Israel, we are looking to push that idea forward in, in one way or another? Where would you push this? Wow. Ideally, we’ll start with a local standard. And then hopefully roll it up to W three, C or Unicode. I’m not sure what the challenges there, what what kind of challenges we’re facing when we come when we approach those standardization bodies. So it’s not my top priority as the chairman. But it’s certainly a work item that I’d love to see happen. It has impact, by the way, in a lot of things. It’s not just, you know, being able to localize for Hebrew and Arabic in a more inclusive way. It also I think, has impact on other things like machine translation that learns off existing resources or texts, prediction and spell checking, all kind of feed off existing strings and existing language out there. So the moment the minute you make the language more inclusive, the better those technologies are going to be in terms of accuracy.
Andrej Zito
So let’s go back to when you were young before you joined Microsoft. And let’s let’s, let’s repeat the question that you already spoiled what we discussed last time. So, as a user before you started working in the industry, have you noticed any issues like that? Or is it only after you joined Microsoft?
Gilad Almosnino
I think because of the years that they started, you know, DOS didn’t have a UI, so that wasn’t an issue. And then just interoperable interoperability those days just didn’t exist, you know, Unicode came into play many years later. So I didn’t really pay attention to those issues, because software wasn’t even available in most cases, and not in languages like Hebrew and Arabic, let alone input experiments of any kind. And so I think it was only windows 9598, when Microsoft finally released a Hebrew in Arabic version of Windows. And I might have gotten a glimpse of that. But just to make kind of put things in perspective, windows, I think, 98 came two and a half years. I mean, the Hebrew version of Windows 98, came probably two years after the original release, just to show you how difficult the release process was that those those years. And so I wasn’t exposed to those things very much before I started working for Microsoft. And but the one thing about, you know, going into Microsoft and starting to work on Hebrew and Arabic version almost immediately, because it’s a pre release, build. I mean, the issues are so clear, right? Like you’re seeing UI mirroring bugs, and all text layout bugs, and those were probably the two biggest bug categories. Those are very easy to spot for the train.
Andrej Zito
So yeah, let’s talk about the issues in a bit more detail. And maybe you can also compare it to the situation right now.
Gilad Almosnino
Right, um,
Andrej Zito
Let’s start with software,
Gilad Almosnino
Software. IN windows in particular, I mean, any any software. But I think the two biggest challenges for anybody going into this market is UI layout, UI design, because the Hebrew and Arabic markets require that the UI layout is reposition. So it flows from right to left. If so, and that’s a significant amount of resources that have to be invested in making sure that that happens correctly. It’s not a switch, you flip and everything just magically happens. And that, you know, and we’ll talk about later about probably, how engineering systems and design systems today just just don’t do that well. And so for a developer, it’s not about switching something and making sure that the UI layout is is, is working correctly, you know, there’s a lot of thought that goes into it. Imagine, it’s not just the just the UI, it’s the experience, right, you have a lot of things that you have to take into consideration when you reposition the UI to flow from right to left. because not everything flows purely from right to left. And take for example, immediate control. If you have a multimedia player, the play button, the back and forward buttons next track forward track buttons are actually left to right oriented. What happened is the markets received, you know, take in cassette players and CD players from you know, that were just manufactured in probably in China. And that’s what they receive. So they’re we’re used to seeing the play button flow from left to right. They’re used to having the the track control flow from left to right. And they also want to see the progress bar on something like that flow from left to right. And so if a developer approaches a media player that’s built for English and just says mirror this, all sudden, everything flows from right to left, it might look okay. But it might just the user is not going to like it. And so there’s a lot of thought that has to go into what makes sense there in terms of mirroring the UI. There’s also a lot of back end work that has to do with content. So imagine if you have like a swipe control, for flipping between pages on a book, okay, so usually if you’re flipping a page on a book, like a PDF reader, where the pages are spread horizontally, then you flip from right to left to go to the next page on English, right. And so for Hebrew, you have to go if the If you think that in Hebrew, you’d have to go from left to right to go forward. And so if you want to circle back to robust design, it’s not about hard coding the direction because you never know what kind of book you’re going to open. So imagine having a UI, that’s English, and you open a Hebrew book. And now you want to be smart enough to understand what the content is, and then make your UX interaction adjust appropriately. That’s robust design. That’s perfect. That’s natural. I think I actually have a patent pending application on that. Particular I do, I did apply for a patent for it. You know, we had a exactly that we had a PDF reader that had the function of being able to swipe to flip the page on Windows. And we wanted, I wanted the design to be robust enough to allow both directions based on the content. But it’s really hard to understand what the content of a document is today in terms of language, like, how do you? How do you know what what the language of the document is? And then adjust the layout appropriately? And so do you have a good solution for that? How would you do it?
Andrej Zito
I think when I was working at Moravia, so my early days, long time ago, yeah, I know that one of my team members, and localization engineer, once we were doing a delivery for Microsoft, by the way, because I started working for Microsoft. Windows, by the way, as well. Yeah. And he he swept due to file management. inconsistency, he swapped Chinese simplified and Chinese traditional rules, of course, the files. So of course, it was it was a big mistake. And then we were asking our r&d department to come up with some way to check what language the files are. And you’re basically trying to locate the unique characters that are only applicable to certain language, but I’m not sure about Hebrew. Do you have any specific characters or something that’s only used for Hebrew?
Gilad Almosnino
Yeah, you could probably look at the Unicode range and say this is it, but that that isn’t going to be 100%? bulletproof, right?
Andrej Zito
Because sometimes you don’t have those characters.
Gilad Almosnino
You might you know, if Yeah, you might have, you know, you might have a mixed document with English and Hebrew, and you never know. But the user knows when you open the document, what what language the document is in, right? Like, if you just stare at it for a second, you know what language is.
Andrej Zito
But when you sit mixed English and Hebrew, how would it look like? Like there’s only a piece of English in the otherwise completely Hebrew text, or some parts of it could be English and Hebrew?
Gilad Almosnino
I mean, imagine a technical document where you have a Hebrew technical document with a lot of English, you know, how do you what what, what percentage of of measurement? Do you have to make sure it’s a Hebrew or English document? Right? It’s a very tricky question. More so on content on the internet, for example, my solution was very, very simple. It wasn’t about that. I didn’t want to go scan the document at all, I thought the user knew exactly what document they’re looking at. And guess what? The initial swipe direction is the direction. So the minute the user swipes, you know which direction the document has to flow. And so that I have a I have a patent pending application, I think, I don’t know if we went through or they cancelled it. But that was my basically, patent pending moment, if you will, to make sure designers are robust enough. And, yeah, so that’s about robust designing user experience and UI, UI layout and directionality and UI layout. The second biggest issue with BiDi, in particular, is text interaction yet
Andrej Zito
Before we before we get into that, I’m wondering, when we’re talking about the design, you were saying that some of the controls they need to stay in place. So it’s not like revert or flip everything globally. So how are these exceptions handled? like can you design software, you can set some controls at as, okay, stay like this all the time and flip.
Gilad Almosnino
I haven’t seen a design system to date. Like I recently looked at Adobe, Adobe has a design product, I forget what it’s called. In there was internationalization features like pseudo localization, which was really interesting. But there’s nothing about mirroring UI a lot of the knowledge about which controls get mirrored is out there. Like I wrote a very robust A set of guidelines on how to mirror certain UI elements. And but the design systems don’t take that into account because they’re just, they just don’t look at it in particular. So it’s not like you’re able to take your design in any design system, just flip it in for the design system to be robust enough and smart enough to say, you know, if you have an on off button that’s horizontal, you should probably mirror it. You know, the design system isn’t smart enough to say this particular control has to be mirrored, it’s also the design system isn’t smart enough to say, if you’re mirroring, media control, don’t mirror the play button in the back and forward track controls, there’s nothing out there that actually does that. Not a design system, not an engineering system. There’s like, like, it’s missing the, like the artificial intelligence to, to help developers and designers avoid the mistakes. Right. It’s a blank canvas. And so I think that’s one of the biggest steps that the industry has to do with internationalization in general, not just by the by, that’s great. But imagine if you had a design system that’s robust enough to help the designers and developers actually jump into a localization project, and not be in a position where they’re exposed to making some of the inherent errors, because they’re just, they’re not aware, and the design system didn’t, you know, prevent them from doing that. So we just don’t have that today, that tool isn’t available in the industry. And then again, the internationalization knowledge isn’t being in many companies, and at least the big ones isn’t, isn’t being spread effectively enough to make sure that we don’t do these mistakes. If I look at Facebook, or Google or Microsoft or Apple today, they all have mistakes, that they ship with bugs. And so I think it’s just a reflection of their engineering system in design system that is unable to help them produce. Experience, that’s a lot more, a lot better. With with lower overhead.
Andrej Zito
I still don’t understand it. And maybe you can explain it by telling us what how the mirroring technology works. But what I still don’t understand is, let’s say, Okay, I, I’m a developer, I create something. I don’t know anything about pie. So I’ll just flip everything because I read it on the internet. And then I set it for QA. And the QA person, like you tells me, okay, this is good, you flip this, but these individual controls here need to be as they are in English. So how would he do it? How can he set it?
Gilad Almosnino
The you know, I don’t come from a developer background. But my understanding of what developers do is they flip a switch, basically, that says beer, everything. The issue with the switch is that everything that’s inherent there, all the other elements, beer with that. So what happens is a developer and that’s why it’s so like this, what people don’t want to go into the markets because a developer has to sit there, do the initial mirroring, which basically mirrors everything. And then all the elements that have to get unmetered you actually have to manually go through them and Amir them. And so the two issues there are one that technology is kind of like a bucket of cold water, right? You just like just sit there and like you, you death somebody with it, right? And then you go in and you have to do all this tedious work. And hope that you did it correctly. because not a lot of companies have BiDi experts. Right? And going the like one of the easiest things for most languages is you know, go just Oh, go ask the native speaker. But with the BiDi languages, you can’t do that. Because you really have to have somebody who understands user experience to help you understand how to mirror or Unmarred things. Right. So it’s not as easy as asking as a Hey, Hebrew speaker, you know, if you have one in the office, can you tell me if this is mirrored correctly, you get three types of people. The first one is going to tell you it’s mirrored correctly. The second one is going to tell you it’s mirrored incorrectly. And the third one’s the worst. You know what the third one is? He’s the one that hates the mirror UI. He’s like, why do you guys even go through this painstaking process of doing this right? You should just use English. And so.
Andrej Zito
I thought it the third guy was is bad but this is what you need to fix in the design phase.
Gilad Almosnino
No, no It’s the the other the worst guy than the one that hates takes the mirrored UI and just says, Yeah, you should just not do that.
Andrej Zito
But was there, was there- so far, I got the impression that it’s all up to the poor developer to fix this. But was there ever any initiative to move this work to the vendors? Who are the ones working on localization? Was there anything like this possible in lock studio?
Gilad Almosnino
I think maybe in the past, they had some power to do that, I hope they’ve stopped doing that. Because, you know, just having a robust system means that you’re doing this through the code and not through the localization layer, because that’s buggy. I mean, imagine, you know, the localizer, you know, does that incorrectly and then you’re trying to figure out whether the code is introducing the issue or the localizers, introducing the issue. And that was a lot of the work for localizers. At Microsoft to understand whether or not the UI mirroring issue was introduced from the localization process or from the code. And so when the system got really robust, like, probably 90 plus percent of the mirroring issues were code issues, and the localization team was just focused on, you know, localizing strings. I think the better approach also is to have the feature team that’s working on a particular UI, on the UI mirror, because they have to also own the experience, right? They have to have this end to end design, develop localized experience. You know, during my time at Microsoft, at least at the end of my tenure, as an internationalization expert, we were doing design walkthroughs, that basically looked at the design phase of things. And I would take things take mock ups in PowerPoint, and give them back as mirrored so that the design team understands exactly what has to happen. developer has a very good blueprint. What work they have to do. And so that was Microsoft, I don’t think they do that today. By the way, I think the the reports I published our, our indicator that that isn’t happening anymore. But when it when it was working, it was working very, very well, because it did a few things. It helped educate people inside the company, they didn’t have to become expert, but they had access to an expert, they were learning. And then, you know, as time went by two, three years into it, they were able to do a lot of these things on their own, and only asked me the really complicated questions, right. And so I think when you’re able to integrate internationalization into the process, that’s a very basic step. So if the entire design process has an internationalization expert there, you’re going to save a lot of time and resources, you’re going to make the release more predictable, you’re going to have your bug counts lower end that works really, really well. And more. So if you have a tool set in engineering system that prevents people from doing what we call this low hanging mistakes, or the low hanging fruit, the stupid mistakes. And that doesn’t exist today. That just no one has something that works that well.
Andrej Zito
Why do you think the things went south?
Gilad Almosnino
For for what for Microsoft, just in general?
Andrej Zito
For Microsoft, or in general.
Gilad Almosnino
I think just in general, you can look at a few companies and say they decided to go internationally. And they did very, very well. So I’ll give you an example, Spotify recently, not recently, but about a year ago, when I came in to Israel. And what a delight, they did an awesome job. You know what I’m not about criticizing companies, I want them to kind of think about what they do, and I want to empower them, I want to motivate them, I want them to do things better. So you know, you can look at my reports and say, you know, he’s, he’s maybe sour or something. But it’s really not the case. I love Microsoft. We love people there. And I want them to do better, right? I want to motivate them, I want to empower them. And if you read my reports, you can see what the expected behavior should be. So they can just take those reports and fix everything, right. Like, I’ll give you the example of Spotify coming into the market. You know, here’s, I think one of the things that companies do is when they decide to go into a market, they’re intentionally going, they’re being intentional, is so important. And Spotify, nailed it because they were intentional about coming to zero. They performed UI mirroring, with very few issues. They did the text handling inside the application with very few issues. And it’s just a delight to get something like that. I thought they did a wonderful job because they were intentional. You know, they are very focused on coming into the market and doing something very, very And I think Microsoft and Google and Facebook and Apple, were also very intentional at one point going into these markets, the issue with somebody like Microsoft or Google, you know, they’ve been in the market for 20 years or so. And so they’re no longer intentional, as far as they’re concerned, they have this checkmark, you know, we shipped to this market, we’ve done the work to support it, but they’re no longer intentional. And you can see that with the quality of the product, the robustness of the design, you know, as their design changed, the fundamental technologies kind of stayed behind in terms of internationalization in vitae. So I think being intentional is such a, such a fundamental quality to have when you go into a new market, if you want to be successful. And, and Spotify does that very well. And most recently, I think, also, Netflix did a very good job. And that’s what we see the last few years are this, you know, the Netflix chart of new users. And, you know, it’s a, it’s a beautiful chart, you know, it’s, it’s, it’s look at all the growth, and all the happy people you have, because you’re intentional about going into a market with local content and great experience. And so you have to be intentional.
Andrej Zito
I totally get it that when a company is introducing or entering a new market, they want to have the best experience for the users to make the first impression great. But in the case of Microsoft, was there no feedback from the user base? Or does it need to reflect in the in the number of users or the number of sales?
Gilad Almosnino
Yeah, if you look at Microsoft, as an example, I’m sure other companies are struggling with the same issues, like a lot of these companies has have feedback mechanisms. Okay. So some of the feedback mechanisms are just telemetry, right? Nobody’s not a user behind them. But we just get a telemetry reading from a machine that, you know, says machine is happy machine is not happy drivers, working drivers not working things like that, right. That’s one thing. But when it comes to user feedback on design, you know, it’s, it’s, it’s, it’s a multi dimensional problem. So take a market like Israel, right. It’s a very demanding market, in terms of the, you know, the resources that need to be allocated to get into the market, because of the UI mirroring and tech support. And but Microsoft, for example, the insiders program, has a huge bucket of feedback, huge bucket of feedback, you know, Israel is 9 million people on a good day, when everybody switches on their computers, even even if they decided to have the, I don’t know, a crowdsourcing sponsored event with everybody sending feedback, you get 9 million. How does that compare to the Chinese market? Right? Where you’re dealing with a few billion people. And so and then you might have some Program Manager sitting there saying, Hey, you know, you have the small market complaining about issue compared to the Chinese market comparing a complaining about an issue. I think it comes to the point where they’re not able to scale the feeder program to be locally relevant. And that makes people very, very frustrated. Because they feel like they’re getting a, a, a reduced experience. They feel like a second class citizen, then they don’t feel included. You know, it’s all about inclusiveness, and being able to include inclusively add user feedback, regardless of market. So that they feel like somebody’s tailoring the, the product for them. And so I think with big companies like Microsoft that the feedback from smaller markets, is it just gets drowned, right. And so, and then you have the language barrier and other things, right. So that’s 1111 aspect, one dimension of the user feedback issue, when it comes to international markets in general. The second thing is like, I think that a lot of these companies have unconscious cultural bias when it comes to international feedback. And I say that in a way of, they need to be cognizant of that issue. So I think some people will have an easier time prioritizing a piece of feedback that comes from the market that they’re in at that moment. So if you’re sitting in Palo Alto, and you get a piece of feedback from, you know, from the United States, compared to a piece of feedback from, let’s say, Dubai, you’re probably going to treat the piece of feedback from Palo Alto, United States. More than you would the piece of feedback from far away. And so.
Andrej Zito
Is there no country office in Israel that would be pushing this?
Gilad Almosnino
So for all the multinationals, Israel is a high tech nation, right? It’s a startup nation. There are plenty of r&d centers here. But the heat the Israeli r&d Center, for example, for Microsoft doesn’t deal with the human version of Windows at all. That that isn’t there, they do cybersecurity, they do artificial intelligence. That’s, you know, that’s the that’s why this market is so strong in terms of startups. That’s what they do. And so it’s not about that, because in Microsoft isn’t, for example, with the news. I’ve talked to the folks that the windows insiders program, and they said, you know, you really have to make a local version of the insiders program, not not to say that you have to develop features that are specific to the market, but you want to have people running the feedback program, in their native language, inside the market. Like one of the things that is inclusive about the windows insiders program is it’s, you know, they’re, they’re never going to be able to have the podcast that they do the Live podcast, on a schedule that meets everybody’s needs, right? Like, you’re going to do it like 12:11pm 11am pacific time, you know, somebody in Israel is going to be sitting there at 1am in the morning going, I don’t want to listen to this. And then they’re, they’re not even going to listen to the local feedback, right, then you’re, you’re back at this huge bucket where you’re competing for attention. And that’s just depressing, right. And so I think there’s a lot of things the multi dimensional issue is with with that, right, being inclusive of people is that the other thing is that I’ve noticed with certain companies, is that, let’s say you had some UI clipping on the English version of a feature that you’re releasing, right? You’d never released that with that kind of bug, like, let’s say it was a major clipping, right? You’d go and fix them? Well, here’s where that unconscious cultural bias comes in for international design and localization, the feature team will say, you know, let’s see if the customer complains about it. Mm hmm. And then your quality bar, you have two standards, basically, you have the English Standard quality bar, which is the top of the notch, and then you have this, oh, we’ll wait to see if they complain about it. And then that, you know, that the user doesn’t want to go that route, they’re not going to complain about something that’s broken, they’re not your quality assurance team, they’re having a horrible experience, you’re going to reach out, they might even not even use your product down the road, you know, and so I don’t think people that have this unconscious cultural bias, are cognizant of that, of that approach. Right? You just, they do it unconsciously. That’s, that’s why I’m not angry about it. I think it’s an issue that just has to be aired. And that that’s something that I had to when I was working at Microsoft, it was one of the things that I you know, I fight for bugs, I I’d fight for bugs till the day like that thing was kicked out the door, right? It would drive people crazy. But what I was actually fighting wasn’t the technology wasn’t the it wasn’t that it was the device, right? Like, I don’t want my customers to complain about this, this has to go out the door, just like it would for English without this clipping or without this horrible luck. And so you could see that today.
Andrej Zito
You were talking about Spotify, and how well, well of a job they did. But he still mentioned that you notice some some small issues. Can you just I’m just curious, what what are those?
Gilad Almosnino
Just certain texts let you know when you have? Well, first of all, they did UI mirroring very, very well, with no issues, which is almost unheard of. I mean, if even if you go to settings and stuff, everything is just is done very, very well. And so kudos for that. The issue with any application that displays content, and you never know what the content is going to be is that you have to be smart about displaying that content correctly. And so there’s top technology, most of the major platforms today will detect whether or not you’re displaying content in BiDi or not. And then make certain adjustments like layout and alignment so that the content flows from the so that the text flows correctly. And so Spotify has just minor issues where sometimes the text isn’t aligned correctly. When it comes to like showing. I think it was like the artist bio, I think or something like that. They’re Welcome to reach out all pointed out, more than happy to but that’s it. Really minor because it’s a very small fix, right. And so those are the issues where developers even if they nail you on mirroring, you know, bite is very unique, because now you have to have text handling, where you don’t have to do that for other languages. You know, the text sometimes has to flow from right to left, sometimes have to flow from left to right. And that’s what why we talk about robust design. You know, even on English UI, you might have to display Hebrew text from right to left, laid out in line from right to left. And so having a robust design that’s able to handle that correctly, is, is is priceless. Because it creates a great user experience. And by the way, today, they don’t nobody does that very well, by the way today. So the technologies that are doing that kind of adjustment, are very, very old.
Andrej Zito
So it doesn’t mean that when I search search for Hebrew artists, on Spotify, using my English UI, it will show right to left.
Gilad Almosnino
I don’t know, it might, I don’t know how this is, again, this is about a rich internationalization experience. Like, imagine looking at you’re in Vancouver, imagine, like, googling, sorry, not googling but searching for Israeli artists, and then having the bio come up, you know, what’s the expectation? Like, if Spotify is like, done things very well, they know that you want to see things in English. Are those true the English bio of that artists? I’m in Israel in my UI languages, Hebrew, they might show me the Hebrew version of the of the Bible, even if I googled wrote if even if I sorry, search for, like the Rolling Stones on Spotify in Hebrew. Maybe they have a machine translated bio, that’s in Hebrew. And they can display that, right? Because it’s more about, like, you don’t know if the user is multilingual. Right? And so you want to be able to tailor the experience to the best of your ability to make sure that stays in the right language. So I don’t know how they do it. But, you know, it wouldn’t make a lot of sense for them to have you read a Hebrew bio if you don’t know Hebrew.
Andrej Zito
Right, right. Yeah. Makes sense.
Gilad Almosnino
Right.
Andrej Zito
And so the second thing, when it comes to the software, beta issues that you wanted to talk about was text input.
Gilad Almosnino
Well, text, text user experience in general, whether it’s displaying text, we just talked about being able to align content automatically. You know, that’s, that’s a technology that’s 20 years old, you know how they do that, by the way, in most tech stacks, whether it’s Google, Microsoft, Facebook, whatever, they, they try and sample the first few characters of a string. And then if it’s, if it’s a BiDi string, like Hebrew or Arabic, then they’ll apply the right properties for the alignment and layout. Okay. And so that can go horribly wrong. Because you could have a English string that starts with a Hebrew word, like if I said, Haifa, and Tel Aviv, our city names in Hebrew, and Haifa, and Tel Aviv were spelled in Hebrew. And that’s the beginning of the sentence, then the the rendering engine would think that’s a Hebrew string, and align it from right to left, then you get a reading order issue, like the strings go all over the place. And that’s where the introduced errors. And that’s why, you know, one of the things for me and standards to Israel’s to push these technology companies that platform companies to start really investing in technologies to make this experience a lot better, because we’re seeing what we’re seeing today is 20 years old, that’s 20 year old solution. Imagine how many technology leaps we’ve done. To make that experience better. And imagine if you’re able to understand have the machine be smart enough, the platform smart enough to understand language, and detect language that has impact for other locales, not just BiDi, and imagine what you can do with that. I mean, so you really want to get to the position where you understand the string, if you’re displaying a string, what is what is the what is the language that that dream is written in? So that you can make all kinds of other great experiences. And so for us, that’s a huge thing, right? Because, like when it gets wrong, and it often becomes like, it’s when it’s often wrong. The experience is horrible. Like imagine having a full sentence aligned incorrectly. And every time you try to get to the beginning of the line, it’s basically all over the place, right? Because it’s it’s not, it’s not even. And so that’s a very poor experience. And then on top of that, when it comes to the text input experience in text selection experience, all of that is today on every major platform influenced by directionality of the string. So, I’d love to maybe we can integrate into this podcast, a demo video demo of how bad text selection is on Hebrew and Arabic under certain situations. So people can visualize it, you really have situations that are very, they’re not corner cases, where the user will try and select text. And you know, in English, when you select text from left to right, right, left, whatever, it’s a, it’s a continuous motion, there’s no breaking points, right? There’s no spaces that are unselected in the middle. Well, in certain situations, when you have like a mixed bag with strings in Hebrew and English, and you start selecting, at one point, the selection becomes unlinked, you actually have a big chunk of text that isn’t selected in the middle. And the selection process is kind of flipped around and reversed, and then goes in the other direction, and then flips back to the original direction. And that has to do with the fact that all of these, all of these technologies are based on the Unicode bida algorithm. That’s a Unicode Consortium of the big standardization body, and the Vita algorithm there. It’s a very smart algorithm, but it’s 20 years old. And one of the issues with it, is that it’s not backed by usability studies, right. And so one of my biggest pushes in the standard Institute of Israel right now is for every standard that we push, and one of them is text interaction, is to have robust usability studies, so that we can back up our, our research, within standardization efforts with real user data. And not just expert assumptions, right, because I’m an expert. But I’m sure that I’m going to learn some new things about how people use computers, right. And this is one of those prime things where I have to have the data, so that I can go to Unicode my standard and say, not only is the standard, you know, pass rigorous standardization required requirements in Israel. And it has a bunch of subject matter experts behind it. It also has here, here are the usability results, right? Like, here they are, we actually, one of the other things I did is bring into the committee, excellent experts in I have a PhD candidate that is doing a PhD on UI mirroring, and how that impacts the ability of a user to understand the UI, Yulia Goldberg, that name. And she’s also going to be responsible for helping us kind of shape the usability studies around our standardization efforts for UI layout, and text interaction. And so we’re very, very happy to have we have another expert. We’re very happy to have these experts come in, we have another couple of experts coming in that are leaders in Israel for data visualization. There’s no standard today that indicates whether or not you know, a chart should flow from right to left or left to right. I’ve been in the business for 20 years, I can’t tell you the answer for that, you know, and here are these people that are in the market. And they’re experts. But it’s very important for me to have them also backup their their standardization efforts with usability studies. So for data visualization, for example, we have some crowdsourcing that was done for user research. And we found that the guy that did their research, basically found that, you know, 70% of people prefer charts from left to right, and 30% of you know, Midas speakers prefer charts from right to left. And so how do you create a standard that’s robust enough to accommodate both of those? You know, how do you how do you make sure that you don’t Trump 30% of your users? Right? Trump will trample Sorry, my English is my second language, right. And so I think one of the one of the initial ideas that we’re going to bring into the usability studies is to mesh user experience with an UI layout for data with accessibility. And what if we had a indicator on the chart that would indicate to the user whether or not they should read the chart from left to right or right to left, and so that that might be a possible solution where we integrate the best of all worlds, to make reading charts, more more, easier and more coherent.
Andrej Zito
So would the chart be just in one direction with an indication of how to read it?
Gilad Almosnino
Maybe it’s a chart with one direction that with an indicator like a little arrow that says, you know, you should read this from left to right or whatever, maybe if you wanted the design to be robust, and, and people wanted to flip that, then you could click on that little arrow, and it would flip it to right to left, and then they would remember your preference, you know, but still, but still kind of, you know, indicate that, you know, a chart reads from from a certain direction. And those are exactly the things are worth investing in, in terms of usability. And this is where companies that, you know, produce these solutions aren’t, aren’t able to think about these things, because they’re not intentional about coming into this market. And they’re not aware of this issue. Right. So we’re, I guess, we’re very focused. And we are very intentional about making sure that we spread the standardization, so that an impacts design on the major platforms, right, and that’s my other job, being able to reach out and take these standardization efforts and not keep them local, make sure that they go to Unicode, make sure that they go to W three C. And, and so that we have a very robust set of standards that’s actually internationally recognized.
Andrej Zito
Is Hebrew,very different from Arabic When it comes to all these issues, or?
Gilad Almosnino
No, no, just just to give people a sample example of First of all, there’s I think, 420 million BiDi speakers in the world. That’s that’s quite a large number. And I don’t think people I don’t think most software developers understand that there are differences between Hebrew and Arabic, not a big deal. I think percent symbol positioning varies. And so there’s, there’s very little things that you know, there’s very small things. Not not enough to say that this market is difficult, because you have to tailor to every different locale, right? It’s, it’s very few things that are different. And I think if you have a robust enough platform, most of these things are handled for you. In terms of UI layout, there’s absolutely no differences. No differences in terms of percent symbol positioning, and the Arabic question mark is flipped. But if you have a really robust platform that’s handled for you. And also Arabic uses native digits, which are, you know, some platforms support that better than others.
Andrej Zito
Did you think about collaborating with someone from the Arabic countries?
Gilad Almosnino
Oh, yeah. So you know, last time we talked was what a few weeks ago, and now that we’re talking, there’s already two countries that signed a peace agreement with Israel. So the Middle East is already a better place. And if there’s anybody in the UAE or Bahrain that’s watching this, welcome to reach out, we’d love to chat. We’ve already put in a request through government, there’s a government office for regional corporate cooperation, we’ve already put a request to reach out to our counterparts in UAE. Frankly, I don’t need that office to do that. If I knew who the people were, we could just start working. I don’t need this official that that’s never been my Mo. I don’t do the official. But I do have connections, for example, in Egypt, and we do have people from Jordan helping us with these really standards. And when I was working at Microsoft, we work with a variety of people from the Middle East. There’s no reason that Israeli standards exclude Arabic, from from their standardization effort. And we actually were bringing in Arabic speakers to help us make sure that we’re on par and take into account Arabic requirements. Because again, we want to be intentional about you know, Israel, one third of the population are Arabic speakers. We want to be intentional about providing them an inclusive experience a great experience. And so that’s why we’re you know, we’re making our standards more robust so that they include regional considerations and we enter other And the other motivation behind that is, I want these really standards to reflect regional requirements, so that when I go to Unicode, I’ve already done that work. And so they’re not going to, I don’t know, if Unicode or anybody else pushes back at me and says, Have you made? Did you make sure your neighbors are included? Now? I’ll say yes. Right. And so that’s my, my motivation. That’s why I’m intentional about doing these things. So that they’re very, very inclusive.
Andrej Zito
How can people participate in the usability studies? How are you planning to do it?
Gilad Almosnino
Yeah, we’re gonna ship, we don’t have a solid plan on the usability studies yet. I think designing them is a very, very fine art and profession. And that’s what we have, you know, I have a limited limited experience with usability. I did market research at Microsoft, but I didn’t do usability and design research. I think it takes a very professional set of people to go in and make sure that that happens correctly. Because you want to make sure that you are doing usability in a way that really looks at things from all different angles, and not just the angle that you’re trying to prove, right. And so I’m very good about that. And I have some very opinionated people on my committee, that think that things should go in a certain direction. And I say, I want those same people to be the devil’s advocate, right? Like, you want to look at things differently. And so I think it’s going to be a set of usability studies that are done in a professional environment where we bring people in. We were fortunate in Israel to have both Hebrew speakers, Arabic speakers, and even Farsi speakers. So we’re able to do this very inclusive, look at usability studies. And I hope that we’re also able to crowdsourcing, crowdsourcing through the internet, so that anybody that wants to participate, we’re going to be able to to include them. So we have a larger sample size, it’s about creating that large enough, you know, the critical mass of that you have enough input. And so when the time comes, if you’re you can follow the committee’s work on LinkedIn. We have a page there that I update, or you can follow me, and I’ll make sure to announce that we’re doing that kind of usability study. And we’d love to have people, you know, participate all kinds of people, not just, you know, that’s the other thing. How do you get the right demographics, not the computer geeks? How do you get the How do you get the teachers? How do you get the information workers? How do you get the government employees? How do you get the people who work with the things day to day to participate in user research? That’s a very big challenge.
Andrej Zito
Anything else that you want to cover when it comes to BiDi?
Gilad Almosnino
No, I think I think we’ve covered most of it. Yeah. I can tell you that I’m very passionate about BiDi. And I’m, I’m very good about asking people about their experience. And one of the things that drives me is to produce better experiences for people. I’ll give you an example. My my oldest daughter goes to elementary school, and they’re, you know, they have the elementary school has an office 365 subscription. And they wanted to do the annual reports, you know, and they generated like a Word template, somebody worked on the Word template very hard. It was a very, you know, templates can be tricky, right. But when you have multiple people editing, a template, then becomes even more tricky. And when you have multiple people editing a document in bidar. And the directionality of the text is so unintuitive in terms of the errors that the user introduced, that becomes a nightmare. And my, my daughter school actually had to pay an additional, they actually had to hire an external company to provide a office 365 based solution for like five grand, so they’re able to produce a so they’re able to produce these reports. And that just goes to show you that being intentional about going into market, you know, maybe 20 years ago, this was a, you know, acceptable experience, but today it’s not. And so you have to be intentional, and you have to keep being intentional about supporting the market.
Andrej Zito
What was the problem with the templates, don’t get it is like the text input or just the way you get the templates?
Gilad Almosnino
So when you’re typing in word for example, or even Google Docs, there’s a little control that says alignment, right? Like you can align the text in any direction. But in binary, it’s more complicated because you also have to adjust the UI, the text layout, so that the order of the words is correct. So you could have left align right to left text. That’s not a problem you could have. So if you wanted to have right, right aligned English sentences, you can have that and they’ll read correctly. But in Beit if you align the text and get the layout incorrectly, then you’re going to have all kinds of characters just jump around like the the coma, the periods that the exclamation marks. And so people will sit there for hours tried to adjust the alignment and not the reading order, which is a different control. And that just goes to show you that the solution doesn’t work anymore, because it’s not intuitive. And the users just aren’t aware. And it’s not new. It’s been there for 20 years. Right. And so that’s that’s a really big issue, because it just goes to show that nobody’s intentional about having a good experience.
Andrej Zito
So besides BiDi clearly excited about what are you curious about?
Gilad Almosnino
I’m curious about a lot of things. Most recently, I’m into the Human Design System, which is a system that allows a person to understand how their personality works. What are the mechanics of the personality? There is a few types in this. There’s five types of personalities in the system. And a very detailed map that helps you understand what are the unique mechanics that you have in you? And how, how can you live in a more calibrated and precise way, so that you feel that you’re that you’re feeling good that you’re correct, that you’re living your life the way you’re supposed to? on many levels? Right. And so being able to look at that, that’s one of the things where I found in the last few months, you know, personal development? You know, I thought for many years, I was an alien. Seriously. Yeah, people looked at me and said, you know, what, what’s going on here? What’s up, guys? Oh, you know, how do you very unique, right,
Andrej Zito
How did it manifest to be an alien? Like, did you not fit in anywhere?
Gilad Almosnino
I had a, you know, living in the United States for 20 plus years, when I always had a calling to come back to Israel. And so people around me would ask me why, you know, you have a great job here, you’re enjoying yourself, like is good here. You know, it’s great. Well, I go back to Israel, and always had a vision of going back to my house here in Israel, though, you know, I returned three years ago. And turns out my design is very sensitive to its location, and its community. And it if I’m not in the right place, if I’m not eating the right food, if I’m not around the people that I feel, are supposed to be around me that I’m not in the right place. Right. And so when I found that out, that made sense. The other thing about my design is the only consistent thing about me is inconsistency. Right? And I think we’ve talked about that, like, how do you handle that on the professional level? Right? And so today, when I’m, you know, I do a lot of consulting work. How do you handle that on the consulting level, because you have to deliver on a certain date, right? And so today, I have a relief, when I don’t promise, I try not to promise to deliver on a certain date, not because I don’t want to because now there’s other things that are driving me. And so I try to give myself a larger, you know, timeframe to deliver on things so that I don’t feel pressured, and I don’t feel miserable, trying to accommodate for that, you know, that deadline. And so that’s the Human Design System. And that’s been probably the biggest pivot that I’ve done most recently in recent months.
Andrej Zito
I can imagine that when you were working at Microsoft, it was pretty much deadline driven, right?
Gilad Almosnino
No, I was true enough to my design while working at Microsoft. I have to sample things I have to be very having the need to sample things. I have to be very I need a variety of things around me. So when the manager or met one of my manager you know, one of the things where they would let me do what I need to do, was really letting me live my design, and everybody around me thrive. Because, you know, one day I’m working on tech stacks. Other days, I’m working on UI layout. Another day, I’m working on user or cows or whatever. And another day, I’m talking about design robustness, right? So every day is a different thing. So it’s not that I’m shooting all over the place, I’m sampling and giving a reflection of a reflector, by the way, 1.3% of the human population, the rarest type of design there is. And so I’m reflecting to them, what I think is division forward, not necessarily working on it. But allowing them to see things forward so that they’re able to execute on it. And so when you have all these things together, I’m in heaven. Right. And so deadlines, yeah, deadlines came in went, some of my managers knew how to leverage by design, to amplify. And that’s the best thing you can do with somebody like me, and other ones wanted to condition me, which was the worst thing you can do. Because I, anybody who’s been conditioned is going to be miserable. But anybody who’s living their design is going to be very centered and very, very happy. And so I’ve had one manager, that that just knew that I don’t think she was aware of Human Design, but she was very spiritual. And so working with her was great, because she very quickly understood how my mechanics work and use that as the best tool you could write, she used to send me out to understand the health of the organization, which is the, the highest level of what you can do with a reflector, like, go and talk to everybody, because everybody can talk to you and not feel threatened. And help me understand what the health of the organization is. Right? So I can make the proper adjustment. So everybody’s happy. You know, and so that manager, by the way, was a new Aurora, which now, you know, she left Microsoft to become a coach, which is the, you know, knowing her, like, amazing how people that are centered and correct about their lives, can find the path that is very correct for them very, you know, integrated with their whole view of life. There’s, I mean, I’m so happy that she went that path and not, you know, didn’t stay in technology company, you know. So that’s one of the things I’m into,
Andrej Zito
But did you know about it before? Or is it the recent thing?
Gilad Almosnino
No, no, no. I, after my divorce, I met a my wonderful partner. And she started the, you know, she’s into these kind of things. And she signed up for the course. And, you know, one of the things to, to check was, you know, if you have a partner, what kind of type are they in for, you know, we dated for a year before that. And she looked at me and she said, you know, in my statistics of knowing people, you’re just, you’re different. You’re not, you know, your something about you your ad in a cat, put my finger on it. And then, you know, they she said she opened the map. It was like, oh, now it all makes sense, right? And so today, actually, I started, just today I was in my first class, there’s a six month course I think it’s six months, where I’m practicing living by design. And so very excited to have that happen. Because I get to sample different people. I get to experience my design, I get to understand how other designs work. And it’s it’s a very interesting thing for me to happen. The other thing I do is i’m a paragliding instructor. I’ve been flying for 13 years. So I, on my free time, when I’m not advocating for the Best Buy to experience in the world and more inclusive design and more. Being more intentional about going into international markets, is I jump off mountains and fly for miles and miles and miles or kilometers. We’re in the national, international business. It’s extreme sports, I’m saying. I find it rather relaxing.
Andrej Zito
13 years. So you started in us in the US?
Gilad Almosnino
Yeah, I flew off. I used to fly a lot off Tiger mountain in Seattle. That’s where I learned. And so it’s one of those things where you know, one thing about paragliding isn’t the I’m not an adrenaline. I’m not an adrenaline junkie. It’s very relaxing for me, but it kind of makes sense. sense for me to have a sport like that a wind powered sport, because with wind power sport, just like the ocean, you never get the same day ever. Like you never get a wave that repeats itself. You know, I, you’ll never get an identical wave to repeat itself, right? Whatever you fly that day or whatever you surf that day or whatever, is never going to come back, that moment will never come back and never be seen, you’ll, you’ll develop some patterns, right? Like the air is more rowdy, or the air is more calm. But you’ll never get the same experience over and over again. Because every every day is different. And having being a design that needs to sample a lot of things. This is ultimate, this is never going to be boring for me, because it’s always different. And I’m always sampling it a little bit differently. And so it’s been one of those things where I’m able to take that, you know, being an instructor and having flying alone is one thing, being instructing people to fly off a mountain is a very different thing. And one of the things I learned from paragliding is how to manage risk, risk versus reward. And one of the other things I did is being able to, I was able to take that into my work at Microsoft. How do you how do you motivate people to take a risk? write some code has to be get fixed, it’s a risky thing. How do you change user experience? How do you position yourself so that they’re, I’m, I’m motivating them to jump off the mountain, essentially, now I have to motivate them and have them look at it in a way that they feel that their reward is worth the risk. And so a funny story about that is, you know, I told you, I used to fight for my bugs, you know, when you have a milestone based system, at a certain level of them, when you reach the end of the milestones, you know that they stop fixing bugs, because they want to stabilize everything in so only the toughest bugs get fixed or whatever and some things you have to live with. And most people would back away, they say, Oh, the bug bar is here. Now, I’m not going to talk about this bug anymore. I you know, that’s one of the things that made me great and great at Microsoft, it also got me into a lot of trouble. Because I was I’m so motivated about being able to produce a great experience of when the bump bar came. And I knew there was a one line code fix that can fix a bug, regardless of how minor or bad it was, I’d go and fight for it. And I’d go through the bug bars. And I’d go through, you know, two days before we’re chipping. I’m emailing the VP of engineering saying, Hey, thanks for surprising us windows 10 was supposed to come out with the new, you know, there’s a way to, to clip text, right, like you have the three dots, you know, when you have a string clip, there’s string clipping, it’s called. So there’s a way to do that with three dots. Well, they wanted to have a surprise, and they decided to do like a fade effect, designed to fade effect was great and developer went and did that fade effect. And he completely screwed it up for bida. So the leading edge of strings faded away, we can’t even read it. And you know, they’re like, oh, we’ll fix it down the road. It’s not so bad. You know, it’s, it’s back to that unconscious cultural bias that we’re talking about. And, and I said no. You know, and they said, Who are you? Right? Who are you? I said, Well, let’s, let’s play right? Like, let’s let’s put together I’ve gotten better over the years with it said, Why don’t we put together an email director level email, so they can say yes or no. Right? And if the director said, No, I’d go to the VP level and say, This is unacceptable. We’ve never shipped something like this. And I don’t care what the feature team says. We’re not doing this. And the VP level would agree. And then I was joking. I was I was laughing so hard when they because the VP I had a bug that was escalated to the VP level. And the poor pm that was available for the feature was partying on a boat because we’re at the end of the release. And he gets a phone call. And they’re like come back in we have to fix this. And he’s like, Where’d that come from? Right. And so being able to take those risks, being able to motivate people to take those risks is something that I brought from extreme sports from paragliding. And and so that my, my, actually my secret weapon there was I had to motivate people sometimes and say, You know what, do me a favor, just go and fix this. I’ll take you out for a flight and you know, they’re either And so that was my secret weapon and nobody knew about was, how do you? I don’t know what you call it’s not it’s not how it’s not blackmail, right? It’s it’s the opposite. It’s like how do you incentivize
Andrej Zito
Bribing people?
Gilad Almosnino
Bribe. There you go, how do you bribe your developers to do stuff? You?
Andrej Zito
How would you motivate me because I share with you that I have fear of heights. So I think the experience would be great if I overcame the initial fear. So how would you motivate me?
Gilad Almosnino
There’s, there’s a few things you have to consider. First of all, the flight itself, if you’re flying level is very slow, you’re doing about 1520 miles an hour. So it’s kind of like riding a bicycle very quickly. There is no motor. And if you do the flight at the right hour, there’s no like, turbulence. Right. So the flight, you know, for somebody like you, when I take people up, make sure to feel whether or not like, one of the questions I asked people is, you know, if they say they’re, they’re, they’re afraid of heights, you know, only like three to 5% of the world population is actually afraid of heights where they step up on the ladder and become when you have visually visual problems, right? Like they, they start seeing things, right. And if you don’t have that issue, then you you’re not really scared of heights. You know, even I, when I stand, you know, if you go up to the Space Needle in Seattle, and look down below, you become scared, right? So it’s more of a fear of falling, and not the fear of heights. And if you have just a fear of falling, well, that’s, that’s okay, you can overcome that. And so, once you realize that, and if you have a good instructor than you, you’re able to have a conflict, like, I would not do things to you. Like, if you don’t like rollercoasters, if you’re not the type that gets on a roller coaster, because they like it, I’ll have a very calm level flight with you. So that you’re able to have a great experience. And you’re able to feel that you’ve achieved something if you won’t do it again, ever. But you’ll have that sense of accomplishment. And that’s back to the risk versus reward, your reward versus the risk, if you have the right person with you, is substantially the rewards substantially better or bigger. And you’re able to walk from that experiencing from that experience, and then challenge yourself to do other things. And so that that would be my, you know, motivational talk to you. Right? Or anybody anybody really.
Andrej Zito
Going, going back to localization, my interest question, what do you think is wrong with our industry?
Gilad Almosnino
I think, actually, no, I don’t think localization is looked at as the redheaded stepchild of everything of the development process. And that’s a that’s a, it’s a multi dimensional problem problem. But I don’t think you have the robust enough experience, whether it’s design, you know, you don’t have that robust end to end engineering systems today, to make localization inclusive enough, so that you’re able to produce software that goes internationally without a huge overhead. It comes it’s it’s from the level of the research is an inclusive in international markets, the design systems are robust and inclusive of designing, you know, design considerations for different markets. So basically, the developer experiences aren’t robust and inclusive enough of international markets. And what happens is, so every one of those pivot points, is left to struggle with the short fallings of their design or design systems, right their systems in and so how do you scale internationalization so that it doesn’t become this mouth to ear kind of, you know, tribal knowledge, right? Like me, and you know a lot about this. And if we sat together, we could talk about it for hours. But imagine the poor developer that has to sit there and figure things out, or the designer that’s trying to figure things out. They’re still relying on this tribal knowledge. And the technology has been developed to have a robust enough end to end system that allows a designer to simple design system and have some kind of artificial intelligence to say, hey, you’re going to mirror this for this locale. Here is what you should do or here’s how we’ve done this. For you, and here’s why, you know, so that you’re able to have a system that’s smart enough to not only do things correctly so that people that don’t have the knowledge are able to come into these locales with very little friction, but also to educate them as you do this kind of work, right? And today that’s, you know, relying on people in the tribe, the internationalisation tribe, localization tribe, and, and that has to transition into a technology based solution. So if we’re able to scale to that we’re able to not be seen as a text, you know, we’re perceived many times as a text. How many of our localization colleagues sit there and say, Oh, you know, the design team, finally, let me sit on the design sessions. But they said, Don’t say anything. What, I’ve heard that multiple times, from different companies like the design team let you sit in, but they won’t let you comment. So that you don’t disrupt their design process. That’s not being very inclusive. You know, and so that’s an issue. That’s a real that’s a culture issue. That’s a technology issue. How do we stop becoming a text? You know, and more so in big platform companies and companies are in it for the long haul? companies that are no longer intentional companies that have people transition from role to role? How do you make sure that tribal knowledge transition into significant improvement in design and engineering systems so that people don’t make mistakes? I think that’s our biggest challenge. That and the unconscious cultural bias, that has to do with, you know, the rd process in many companies when it comes to international users. And I think that’s, that’s, that’s a huge, that’s a huge iceberg. So those are the main things that concern me about the industry, in general. Because you’re seeing more and more big companies shrink their localization teams, because the technology is there to do machine translation or whatever. And, and if they think the localization team is just about translation, then they stopped innovating. You know, that’s when the innovation stopped. Because they’re no longer think about the international user, they just think, oh, we’re just going to relabel everything should work. You know, and that, that doesn’t, that that’s what innovation stopped. And that’s a big concern.
Andrej Zito
I think this would be a great final words from you. But just in case, you want to end the more positively, what would be your final words to the industry, the people listening?
Gilad Almosnino
Don’t be afraid to be a trailblazer. Don’t be afraid to challenge the status quo. Don’t sit and think that the current standard in design systems and engineering systems, art aren’t something that you can change. Don’t be afraid to do it. I challenge them all the time. My next challenge is to challenge the Unicode vida algorithm that hasn’t changed in 20 years. And that’s, that’s a huge task. But when you talk to some people, they say, you know, we comply with standard. And I say the standard isn’t good enough. And I think that’s what I want. I want all your trailblazers, if you need your support, if you need my help, if you need your motivation, if you if you want somebody to reflect with you, on what you can do, to challenge to be forward looking to make sure that our industry is robust, in full of value for years to come. You know, don’t be afraid, find somebody like me find somebody like Andre that’s sitting in front of me and do it. Do it. You know, I’ll be more than happy to.
Andrej Zito
Correct. It’s perfect. I have nothing more to add. Thank you. Thank you very much. Gilad, looks like we finally made it.
Gilad Almosnino
Yeah. Thanks for having me. It’s always a pleasure. And thanks for having these podcasts that mean a lot.
Andrej Zito
Yeah, no problem. Thank you. Awesome. Have a great day. Bye.
Gilad Almosnino
All right, Cheers. Bye.
