Hacker News Reader

Comment by miki123211 | Hacker News Reader

There's one more aspect to this that wasn't mentioned at all in the article.

In Japan, home computers never really made sense until it was far, far too late.

In the west, you'd buy a PC (or a home computer) to play games, edit documents or manage your business. The latter two were pretty much impossible in Japan, as the computers of that era couldn't handle the complexities of the Japanese language and character set. Gaming was all that remained, and if you only wanted gaming, you could just as well get a NES (known in Japan as Famicom), which was much better suited for the purpose.

Computers eventually caught up, but some of the cultural impact remained, still making them less popular than in the west.

This is one of the reasons why Japanese were so good at consumer electronics, they just needed that electronics a lot more than we did, and the devices needed a lot more features, as "just plug it into a computer to do the complicated stuff" wasn't really an option there.

View on HN

NEC sold a Kanji board for their Z80-based PC8801 mk II (released 1983). The original 1979 PC8801 didn’t have an official Kanji board from NEC, but one was available for it from a third party vendor. With a Kanji board, you could do Japanese word processing. I believe the same was true of many other 8-bit vendors.

Inevitably the greater complexity of Kanji required more advanced hardware, so Kanji-capable machines usable for business and education were initially more expensive than games-only machines that lacked it. But in 1990 IBM released DOS/V, which demonstrated that standard PC hardware had become powerful enough to support Kanji without needing any dedicated circuitry. And even before that, I believe already by the late 1980s many machines (such as NEC PC-9800s) were coming with Kanji support as a standard feature rather than an optional add-on card.

That powerful IBM system was unique to other systems being brought into Japan, in that IBM started up that business just before the rule that required a Japanese partner. So it didn't really get shared like other technologies that lead to most other Japanese tech giants.

DOS/V was just PC-DOS with some additional device drivers. IBM licensed it to Microsoft who in turn licensed it to other vendors such as Toshiba, Sharp, AST, Compaq, Dell and Fujitsu.

The biggest barrier to its adoption was vendors such as NEC (with their PC9800 series) and the AX consortium (Oki, Casio, Canon, Sanyo, Sharp, Hitachi, Mitsubishi) who had already invested in proprietary hardware for Kanji support and were threatened by the support of Japanese text on global standard hardware. (That said, some AX vendors decided to embrace DOS/V anyway, such as Sharp.)

Another barrier was that DOS/V had different APIs from those previous solutions so you needed new software that could support it.

There was this entire Japanese subculture of microcomputers/home computers/PCs. Machines like NEC PCs you mentioned, proprietary models from Fujitsu and others, the many incarnations of MSX (which I think might have started out as a Microsoft initiative?), and of course the nowadays legendary Sharp X68000, that were marketed and sold in Japan but not necessarily well known - or even available - outside of east and south-east Asia (though I think MSX also got some traction in South America).

The MSX standard was originally developed in Japan, with the cooperation of Microsoft in the US-its OS was a port of MS-DOS to 8-bit Z80 systems, Tim Paterson the original developer of MS-DOS did the port-he’d left Microsoft by this point to found his own company, so he did it at his new company under contract to Microsoft. MSX systems were also quite popular in Europe-Phillips was also a major manufacturer of them. Like other 8-bit machines, you could get an add-on board or cartridge to add Kanji support. First generation MSX machines, some higher-end models had Kanji built-in, lower-end models you needed to buy the expansion. Later generations of Japanese MSX machines, Kanji support was standard even on the entry models. Machines were also differentiated by how many Kanji they supported (JIS1 provided the ~3000 most commonly used Kanji, JIS2 added another ~1000 which were less commonly used).

There were other Japanese machines which found a foothold in some overseas markets. IBM Japan developed a modified version of the PCjr, the PC JX, for the Japanese market. IBM ended up also selling the JX to schools in Australia and New Zealand (minus the Kanji circuitry), and many bought them (I remember using one at school when I was 9 or 10.)

Similarly, Fujitsu’s cloned IBM mainframes, the FACOM machines, were popular with Australian businesses and governments in the 1980s and first half of the 1990s. They were also sold in other markets, but in other markets were generally rebadged (or even manufactured under license) by vendors such as Amdahl in the US and Siemens in Europe, whereas in Australia they were sold directly by Fujitsu and generally closer to what was sold in Japan.

The article seems to under represent how advanced and widespread i-mode cell phones were in Japan - over 60% of the population used it in the early 2000s.

I stayed with friends in Japan in 2001 who noted all their friends used email on their handsets rather than have a PC. They also could watch certain broadcast TV channels on their phone which was popular on the train.

The only people I knew with mobile email in my home country were some execs with blackberries.

I don't know exactly the reasons but Japanese software is basically embarrassing. I was talking to a good friend of mine last month who is a very good photographer about how you basically don't have GPS or a ton of other features in the main cameras from Japan--so both of us increasingly just use iPhones unless we really need to use big bodies and lenses.

Sure, some of it is that iPhones (Pixels) do a good enough job for a lot of us. But it's also that the gap has closed so much and a lot of it is about software.

Go to events in Japan and a lot of the design of posters and so forth just looks seriously bad to US (and presumably European) eyes.

And even in the large systems space, when I was an IT industry analyst, there was just a lot of quirky Japanese tech stuff that was out of step with the world as a whole.

I don't have a coherent theory for it all but Japan just fell out of alignment with mainstream patterns especially in the 90s or so.

> Go to events in Japan and a lot of the design of posters and so forth just looks seriously bad to US (and presumably European) eyes.

As an European living in Japan, I can confirm. Japanese advertisement and presentations look like a competition in who can cram the most text in epilepsy-inducing colors in a single page/poster. Bonus points for cramming both horizontal and vertical text in the same page.

I don't know if they really like it, or if they just follow what everybody else does.

The research I have seen seems to show that Japanese brains simply prefer information density over style and simplicity.

https://www.youtube.com/watch?v=vi8pyS076a8

"Japanese brains" as in there is neurological difference, or is it because that is the environment they have been raised in? (nature vs nurture)

Nurture. That's why the Asians are better in Math. Their language allows to shift from one operation into another at light speed.

Do you have a source that can back this up? Otherwise this looks like linguistic determinism and Orientalism

Ah yes, the myth that East Asian countries are better at math because their numbers are just one syllable, so they can calculate faster.

Leaving aside the fact that math is not only basic arithmetic, the reality is that they dedicate much more time in school to this kind of problems. But as usual, skills that are not used often atrophy over time, and in my experience the average Japanese adult is not much better than the average Western adult.

I will say this though --

Japanese companies and organizations (and people) love diagrams to present to the public. Whether how to navigate a website, book a ticket, check-in to a hotel, etc. the number of process diagrams in daily life (as a perceived effective way of communicating information) is unseen outside Japan.

Also, the amount of highly detailed posters in public places. For example, the departures board of a train line -- far more information than probably the average US rider knows how to comprehend even. Or is interested in.

https://en.m.wikipedia.org/wiki/File:Stem-and-leaf_time_tabl...

https://www.reddit.com/r/MapPorn/comments/rqaoj3/the_crazy_t...

You have to wonder whether this generates a tolerance or expectation that detail is expected to be understood.

That train time table actually looks more readable than a typical bus stop time table from Europe https://formstark.se/wp-content/uploads/2015/09/Otraf_buss_2...

How do American public transit agencies present the same information?

From Chicago: https://www.reddit.com/r/chicago/comments/1bxiy5s/i_highligh...

Only distributed as a PDF: https://www.transitchicago.com/redline/

Overall train map: https://www.transitchicago.com/assets/1/6/ctamap_Lsystem.png

Overall train+bus maps: https://www.transitchicago.com/maps/system/ + https://www.transitchicago.com/maps/downtown/

Sooo yeah, I agree the Japanese one is much nicer. I don't recall ever even seeing a schedule at the stations, just "train arriving in X minutes". The train map and per-line map are at every station, and the train+bus map is pretty common too.

I notice the Chicago map has multiple stations with the same name -- is this not a problem when using Google Maps and other navigation apps? How would the app know which "Pulaski" I want to go to?

Those are the names of the street the station is at, if you follow the map north/south you'll see all the Pulaski stations are on the same street. In an app you'd use the address for where you want to go and it may or may not include the train. Though the apps can also understand "Blue line Pulaski" if it really is your destination rather than a stop on the way.

That first board is just a list of departure times by hour, with express trains highlighted. For weekdays and weekends. Not exactly a huge amount of information.

I'm more impressed by their maps of where each station exit leads to, or where on the platform to wait based on your seat number. Those can be tricky to figure out as there are multiple wait points based on the train route and number of carriages. Shinkansen even take into account what end of the carriage you're closest to.

Here's some more examples of exactly that complexity:

https://imgur.com/a/Qzy46WV

https://imgur.com/a/7UtQnMF

Both are standard-looking maps, honestly - stem and leaf is so obvious I have problems imagining what other form you'd expect, and I have never seen japanese public transport timetable sign before now. It's just standard over here (Poland, UK).

The format of metro map is actually based on London metro, the only major difference is the amount of lines.

I had a Fuji point-and-shoot with GPS built-in nearly 15 years ago. The problem with GPS built-in to hardware is that it takes like 30-60 seconds of standing still to get a lock on the satellites, it doesn't work indoors, and so on. Phones work around this by downloading ephemeris data and triangulating using the cell towers and nearly Wi-Fi and Bluetooth devices, things that a camera can't do without incorporating a whole phone with SIM card into the camera and burning battery.

That's a bit of a red herring. Why do cameras integrate so poorly with phones, which could provide this information, backup and upload images etc. Same is true of professional video cameras, which are used for much longer periods of time in a single place, turned on etc. It's a mindset thing with camera manufacturers, running enormously outdated software with terrible UX on underpowered processors.

Battery life. A good camera will have a standby time of months and will do heavy image processing for days on one charge. And that's a key feature. It's also one of the reasons most people quickly turn off features like GPS and WIFI in their cameras. They're relatively useless and a massive battery drain.

But yes, it could still be made better by waking up the radios fast and having e.g. reasonably smart tethering based on an open de facto standard spec that would be only woken up when the camera is switched "on" (which is mostly a key lock in many cameras, not a boot up).

This isn't true at all of cinema cameras - which run off v-mount, b-mount etc type batteries and are expected to burn through them, and still face the same processor and feature limitations. There isn't (to my knowledge) a single cinema camera that supports bluetooth headphones for example, bizarrely.

Phones handle that this way, but it's not necessarily a good thing if that's the only way they can get the ephemeris data. I used to navigate in Japanese cities using not my phone, but my tablet. Worked fine. Then I changed to another tablet. Didn't work at all, could not. The only time I could get a GPS lock was if I could connect to a local wi-fi, which is not something easily done. Its GPS software clearly didn't even try to get the ephemeris date via the GPS signal, it relied 100% on network download. Which made it utterly useless, in the place where I needed it most - another country, non-EU at that, and my tablet doesn't even have a SIM slot, and if it did, I would have had to buy/get a SIM card (as would I for a phone, something I normally wouldn't need).

In short - downloading ephemeris data from the internet is fine, but if it also cripples the whole thing by not using the embedded GPS ephemeris data.. horrible.

> you basically don't have GPS or a ton of other features in the main cameras from Japan

I'm just a happy amateur photographer, but this fact has been annoying the sanity out of me for over a decade now. Mobile phones are taking increasingly large bites out of the camera manufacturers' market share, but companies like Fuji/Canon/Nikon/Olympus, for some unknown reason, are unable to adapt at all. Sure, there is still and will always be a market for larger glass and more pixels, lower noise levels, ergonomics etc. - but why in 2024 GPS, Wi-Fi, uploading to Instagram and similar functions are not standard in ALL cameras is beyond my understanding.

It's like watching the car industry in the early 2000's where manufacturers would pride themselves over finally putting a USB jack in their car.

I increasingly don't travel with a dedicated camera. In fact, I suspect in the next year or two I'll purchase a new phone sooner than I otherwise would to have a backup camera when traveling especially given a backup phone has other uses.

I have a couple of good cameras and glass but they're getting old and I really can't imagine upgrading at this point.

Adobe (sorry, subscription haters) made up for a lot of defects in the manufacturers' PC software for downloading etc. But I totally get frustrated by the lack of GPS metadata at this point because I hate entering that sort of thing which geo data makes irrelevant in many cases.

> Sure, there is still and will always be a market for larger glass and more pixels, lower noise levels, ergonomics etc. - but why in 2024 GPS, Wi-Fi, uploading to Instagram and similar functions are not standard in ALL cameras is beyond my understanding.

Specifically on uploading to social media and having apps that depend on external APIs — maybe it's because putting Android on such devices and having to constantly keep the social media apps updated "dates" the camera and forces consumers into an upgrade cycle because of software bloat.

I've noticed that for some reason, cheap Android phones tend to objectively slow down over the years even without software or app upgrades — as if the cheap Mediatek chipsets and RAM are designed to eventually degrade. Something similar happened to some of the wi-fi connected printers I have owned, which slowly become sluggish and unusable even after factory resets.

On the other hand, I can still use my 16 year old full-frame Nikon D700 DSLR purchased in 2008 very, very well. Even back then, it supported a GPS attachment, though it's utter trash compared to in-phone AGPS these days.

Even assuming your theory is true (which I kind of doubt), you could be confusing the symptom for the cause.

Why didn't software work for Japanese business purposes? Keep in mind the 80s and 90s were the peak of the Japan economy, and if they actually wanted a software system that could handle Japanese requirements, they had the resources to build it.

Where's the Japanese counter-part of Lotus 1-2-3 and Visicalc? Why didn't any Japanese firm make this software, or if they did, why didn't it become popular?

Certainly Kanji makes things a bit more complicated, but IIRC I've seen systems that mainly dealt with Hiragana and Katakana. It's a limited character set. Nobody says you need to use ASCII.

We could dig a big further and ask why computer were able to handle alphabet characters before hiragana for instance.

Or why no EU company came up with an OS that could fight Mac OS or Windows.

At the end of the day I think the capital and investment aspects, market protections and political situations account for way more than sheer cultural or technical aspects.

> Or why no EU company came up with an OS that could fight Mac OS or Windows.

They sort of did. For a while Symbian had two thirds of the smartphone OS market. And the real winner in the global OS marketplace today is Linux (I'm including Android here, though you could debate that), which started out life as European. But I do agree about investment opportunities.

> We could dig a big further and ask why computer were able to handle alphabet characters before hiragana for instance.

JIS X 0201 was released in 1969 and widely adopted on mainframe systems in the 1970s. It supports a modified form of Katakana (“half-width”) as well as the Latin alphabet in a 7 or 8 bit encoding. 8-bit encodings supported both, for 7-bit you could switch between Katakana and Latin using control characters. So, by the 1970s you could already do Japanese language computing if you limited yourself to katakana and romaji.

Kanji took longer because there are too many characters for a 7/8-bit code and their visual complexity required more advanced display hardware, but already in 1971 IBM was selling their “IBM Kanji System” which contained specialised software, printers and keypunches to support Kanji on IBM S/360 mainframes. In 1979 IBM added support for 3270 terminals (with additional Kanji circuitry) to enable Kanji in interactive as opposed to batch mode punched card computing.

Early systems without kanji support generally used katakana rather than hiragana, because if writing everything in katakana looks weird to Japanese speakers, doing the same thing in hiragana looks even weirder. Also, the “blockier” shape of katakana makes it more legible than hiragana on low resolution displays.

> Or why no EU company came up with an OS that could fight Mac OS or Windows.

In the 1990s, Oracle licensed RISC OS from Acorn in the UK, and sold it in the US as Oracle NCOS (Network Computer Operating System). But the whole “network computer” thing was never as successful as its backers had hoped.

Hiragana (or rather Katakana) alone could never be enough to allow businesses and government institutions to use computers for their daily tasks. These scripts are not just different ways of writing the same things, but rather certain classes of words or certain words in certain contexts tend to use certain scripts. I doubt a fairly conservative society like Japan's would have switched to Katakana for legal documents just to accomodate computers.

> Or why no EU company came up with an OS that could fight Mac OS or Windows.

Because if you're making something in the US, you have immediate access to a market of 330m[1] people, and they all speak a single language, follow roughly similar laws and regulations, and have relatively similar needs (e.g. software made for schools in Arizona is pretty likely to work well for schools in Kansas, because things like the A-F grade system stay the same).

This rich market gives you plenty of opportunities to grow quickly. Once you grow large enough, you have pretty easy access to other large and rich markets that also speak English, and then you can get into the hard stuff like internationalization.

If you start out as an EU company, you only have access to the population of your own country, and you're not large enough to handle internationalization or the 30 slightly different systems of laws, regulations, agreements with retailers and different store chains etc. It's a lot harder for you to grow at first, because the potential pool of customers able to use your product is much smaller.

You have markets like China and India, which are technically better, but at least historically, they used to be poor enough that the US was a much better option in practice.

I don't think this is the only reason for US domination, but it's definitely one of the reasons.

[edit] [1[] the population was different back then, but the difference in scale was similar

Well, what is now the EU market wasn't fully integrated into the "single market" until 1993, well after Unix (from which MacOS is obviously derived) and Windows OS's were starting to be developed. As others have said, trying to target all those different countries and languages is not that straightforward, but it was significantly harder 30 years ago.

> home computers never really made sense until it was far, far too late.

That's not really the case. There were a lot of PCs in the 1980s and Japan was a strong leader in the early laptop market as well. All these PCs and many hand-held devices for sale then handled the character set just fine.

I suspect the two of you simply have a different definition of "make sense".

Yeah, my family had Toshida laptops when I was a kid, Japan used to be a leader in the laptop market.

I don't follow your argument. PCs weren't popular because they couldn't handle the Japanese language encoding, but somehow, consumer electronics were popular because they could handle the complex tasks with the language?

JIS C 6226, the encoding for the Japanese language, was made in the 70s. While later than the US, I would not call it late.

Early consumer electronics don't have substantial text entry.

In the 1980s, a major goal of AI research was to make CJK text entry effective and convenient (OCR or alternatives). It didn't pan out in time.

When the necessary tech was reading, emoji was huge in Japan first because it helped mitigate the text entry problem.

>In the west, you'd buy a PC (or a home computer) to play games, edit documents or manage your business. The latter two were pretty much impossible in Japan, as the computers of that era couldn't handle the complexities of the Japanese language and character set. Gaming was all that remained, and if you only wanted gaming, you could just as well get a NES (known in Japan as Famicom), which was much better suited for the purpose.

Some, much complex, JRPGs such as Falcom's titles (the Ys series), began on Japanese computers (though they did end up getting NES ports anyway).

There's also the proliferation of smut on Japanese computers, since you can't have that kind of stuff of consoles. This also eventually trickles down to some "underground" computer magazines in Japan having that stuff on their pages and coverdiscs - the stuff you would typically find on their weekly gossip magazines.

> In Japan, home computers never really made sense until it was far, far too late

I see your argument and I raise you China and India.

Reply on news.ycombinator.com