Tokyo commuting in tough times

Many Japanese companies have implemented telework or flex time during the COVID-19 crisis so that employees can avoid Tokyo rush hour. My office has shorter hours so I take a later train. Tokyo train commuting has become a little surreal over the past week. Seats are actually available on the inner Yamanote line from Shinjuku before 9.

Most people wear face masks and don’t talk on the train but masks are still in short supply at the local drug store. I have to line up before store hours to get a box. Fortunately my partner, a doctor, bought hand sanitizer and mask sanitizer back in January before supplies disappeared. I use, and re-use 2 face masks with mask sanitizer and also have carry spray bottle of ethanol for hands. I don’t touch anything if at all possible, I lean on something instead, but the hand spray comes out the moment I go out the transit gate. There’s also my regulation hand wash and gargle routine when getting to the office, or returning home. I recommend Apple Pay Suica on Apple Watch if that’s a commute option, it removes the necessity of touching the iPhone.

If you have to commute for work like me, I suggest dressing on the warm side or have a scarf handy. Dressing for a comfortable commute is always a challenge but most cars now have a window or two open for good ventilation. On the Tokyu Ikegami line the staff are opening half the car windows 5 cm or so. It can be quite chilly even with the heat on.

Have a safe and healthy commute.

Advertisements

The Mobile PASMO Super Suica Challenge

The recently announced Mobile PASMO has some serious limitations lucidly explained in FeliCa Dude’s ‘Mobile PASMO – something we shouldn’t need‘ reddit post. It shines a light on the unfortunate petty politics of Japanese business culture, a catch-22 that ends up killing the very opportunities Japanese companies work to create. Mimicchii is a good Japanese word for it: so obsessively stuck on pointless small details that one completely misses the big opportunity. The PASMO association knows they will loose out, eventually, but hang on to their one and only advantage, commute passes, in the hope they gain a better losers bargain in the end. But how much opportunity is lost by then?

As FeliCa Dude points out, Mobile PASMO is a pointless waste of money and system resources to replicate what Mobile Suica already does:

PASMO is inferior to Suica in many respects, the idea of deploying Mobile PASMO and removing the user’s ability to choose Mobile Suica is fairly short-sighted. Such a development likely cost many hours and much money, but is effectively a boondoggle and a monument to the stubborn failure of JR and the PASMO Association to sort out a way to issue commuter passes on each other’s cards.

Taken to an extreme each transit card player would build its own mobile service but this is impossible in an era of shrinking ridership and resources.

Come together into one mobile service please…

The next generation 2 cards in 1 Suica due in 2021 aims to fix the current state of affairs. Architecturally I expect the problems will be solved, but corporate politics are another matter. JR East will have to offer enough cost saving incentives and flexible extras for the other major transit card players to host their service assets on Mobile Suica: commute plans, Shinkansen eTickets and more. It’s certainly in everybody’s best interest to do so. Time to put aside the mimicchii politics and duplication. If Japanese transit companies can’t come together to build the future, everybody loses.

In Praise of Disappearing Japan, served on pizza toast

Pizza toast and Neapolitan spaghetti made with ketchup are Kissaten staples, don’t forget the Tabasco

Alan Booth’s The Roads to Sata created a modern English language book genre, Disappearing Japan. It’s an endangered species that has been disappearing for decades, aka “the real Japan.” Whatever that is.

I read Roads to Sata long ago and liked it, but that was before I had experiences to compare, i.e. a frame of reference. Later on after many adventures working summer mountain lodges in the Southern Alps I read Booth’s companion book Looking for the Lost: Journeys Through a Vanishing Japan and liked it, much less. Travel books and writing are like travel companions, most of them drive you crazy, only a rare few make truly good life long companions. After 30 years in Japan I don’t care to travel with Alan Booth anymore or emulate his gimmicky patchwork style that mixed evocative narrative hooks with black comedy, embarrassingly wrong cultural “insights” and a decidedly narcissistic take of Japanese history that cleverly camouflages good old western style cultural snobbery as ‘outsider viewpoint’.

The Alan Booth Japan travel writing style is alive and well in Craig Mod’s I Walked 600 Miles Across Japan for Pizza Toast, it shares the same kitschy patchwork walkabout formula explaining Showa era Kissaten culture. It’s not my cup of coffee but it’s a fun informative read for people who don’t know Japan or haven’t lived here long. Brushing past a few embarrassingly bad ‘hooks’ like the Pachinco mamas leaving infants in the car parking lot reference, old timers probably see a different narrative lurking in discarded details like the abandoned but unexplained karaoke parlor coin laundry picture, a quirky Gifu sight familiar to anybody who knows that area and history.

Kissaten culture will not disappear. I like pizza toast, blend coffee, neapolitan ketchup spaghetti, green cream soda with a glow in the dark maraschino cherry, oshibori and nice clean glass ashtrays. Lots of people do and the growing number of older people in Japan with lots of time to kill need places to go. I remain hopeful that visitors to Japan will venture out of overpriced Starbucks and discover the joys of classic Japanese Kissaten.

The Second Wave of Japanese Desktop Publishing

This article was published in The Seybold Report Volume 30 Number 6, November 2000. It was a strange time when QuickDraw GX 3rd party developments like Japanese stroke fonts and plug-in font scalers were up in the air without a clear migration path to the newly announced MacOS X ATSUI Carbon framework, OpenType Japanese fonts and InDesign J had been announced but were months away from shipping, and Japanese print customers were angry with Adobe and Morisawa for forcing font upgrades on everybody. It’s a snapshot of Japanese publishing in transition as Steve Jobs was saving Apple and tossing the Japanese print market a few morsels like extended Japanese glyph sets, Adobe was caught flat footed and was furious with Apple.


Japanese DTP arrived with Apple’s NTX-J PostScript printer and Linotype’s first Japanese PostScript imagesetter in 1989. They came at the right time: The early- ’90s economy was bubbling, companies had money to burn and Japanese DTP took off. It was a young, booming market and it forgave many mistakes that would haunt the industry later.

By 1996, the go-go days were gone and they would not come back. By this time, DTP tools (Quark Xpress, Illustrator, Photoshop and, to a smaller extent, PageMaker) had captured nearly 40 percent of the production process. For a conservative industry like Japanese publishing, this was phenomenal―until compared to the West. There, in the same amount of time, practically the entire industry converted to DTP production. Japan is still about 40 percent and holding.

What happened? Three things: the economy, the failure of Western technology to address Japanese issues, and the same old cultural differences Westerners have been running into since Commodore Perry knocked down the doors in 1853. They just do things differently here, get used to it.

The economic crunch hit the publishing market hard, and it hasn’t really recovered. Consider the book and magazine market: On average, this year’s revenues are down 3.4 percent over last year.

The advertising industry has been down, but has recently shown signs of recovering: On average, revenues are up 10.4 percent so far this year.

Basically, the publishing industry overextended itself in the bubble years of the early ’90s and is still adjusting.

Some of the difficulty is the publication distribution structure. Anybody in the business will tell you there just isn’t enough shelf space and return rates are climbing. Internet publishing is not much of a factor yet. Japanese book retail laws, which prohibit stores from selling below suggested retail prices, will be relaxed soon and will put tremendous pressure on the current price structure, which supports lots of marginally profitable items

The Japanese market is also a highly segmented one. There are the major newspapers, which have very specialized workflow systems and in-house fonts. There is the book industry, which runs the gamut from standard tanko bon hardbacks to the ingenious, very popular pocket-size bunko hon paperbacks (which really have no equal in the West). There are lots of printers, using both DTP and traditional systems, that cover everything in-between. In a class of their own are manga: the hefty weekly comic books you can find on subway luggage racks, left there by salarymen too embarrassed to take them home.

Hiragino: In Japan, Mac OS X will include a family of fonts, Hiragino, with an impressive 17,000 characters.

On the whole, the industry is still in a slump and has been treading water for the last few years, with relatively little investment in new technologies such as computer-to-plate. At the same time, competition and consolidation in the printing sector have forced prices down-more than 30 percent in the past year alone-so there is a great need to reduce production costs and increase efficiency. This is the great challenge of the Japanese publishing industry. Meeting it will require new investment in hardware and software. The challenge for vendors is to provide the right products. Those who watch the market and listen to customers carefully will find opportunity.

The font technology failure
A designer once said, “Fonts are like air.” They are the environment in which DTP lives and thrives. Compared to the West, the Japanese DTP air is pretty thin.

One of the great tragedies was the refusal of Shaken KK, the largest, most popular, and perhaps most beautiful Japanese font library, to license its faces to Adobe or have anything to do with PostScript. It kept its proprietary machines and formats, losing over 40 percent of its market share (and lots of prestige and talent) in the bargain. The industry is filled with ex-Shaken people. A low point came a year and a half ago when tax raiders found several safes in the company basement filled with 10,000-yen bills; the president of Shaken, Yuko Ishii had stashed away the money in the early ’80s and was now reporting it as profit.

This sorry state of affairs had two huge repercussions. It denied a large part of the Japanese font legacy to designers working with DTP. (Imagine a world without Helvetica and Times and you get the idea.) It also left the only other competitor, Morisawa, which did license its library to Adobe, with a practical monopoly. But these were not the only problems.

Initial problems
The first implementation of Japanese PostScript Type 1 fonts had serious shortcomings. First of all, they could not be downloaded on a per-job basis, but had to reside permanently on the printer. This turned the production process upside down; service bureaus and printers were suddenly dictating to designers which fonts could and could not be used. For every PostScript device, users had to invest in font licenses― and Japanese fonts are very expensive. In a market where making a single face can easily require ¥30,000,000 (about $280,000) and consume two years, font vendors made sure they got maximum return on their investments. 

To this end, Morisawa and Adobe came up with the idea of marketing two flavors of PostScript printer fonts: low-resolution (up to 600 dpi) and unlimited. A single unlimited-resolution Japanese font costs ¥218,000 (about $2,000).

The situation became more complex when ATM J arrived on the scene in late 1993. Morisawa’s ATM fonts used hardware binding to lock the font to the hard disk. This resulted in a high maintenance operation, because hardware binding required installer floppies for each font and for every CPU. (This remains true even today.) If the user got a new hard disk, he had to send the old floppies back to the vendor and exchange them, for a fee, for new ones. Large operations had the extra burden of keeping track of the floppy that had been installed on each CPU. On top of all this, each installation took time; putting a family of 15 Kanji fonts on two or three Macs could easily eat up a day.

In addition, compared to traditional systems, these fonts had primitive typographic features: lack of pair kerning, lack of proper metrics and poor Gaiji support. (Gaiji are Kanji characters outside of the current JIS [Japanese Industrial Standard] and Unicode encoding sets and are not included in a standard font. They comprise many “unofficial” Kanji characters, mistakes and misinterpretations, and seldom-used Kanji passed down for generations, long before printing presses and governments created standards. 

These Gaiji characters are widely used in people and place names. To this day, they are a reason for publishers to hang on to their proprietary systems. Last but not least, Japanese PostScript fonts were not cross platform.

The first fix
Adobe has been trying to fix the font problem. Its first attempt was CID (character-key ID). CID divorced two-byte PostScript fonts from dependence on any encoding scheme and allowed character subsets to be downloaded to the printer on a job basis, laying the foundation for PDF font embedding in Japanese Acrobat 4.0 as well.

Unfortunately the CID font upgrade was a marketing washout. Upgrades were not cheap, and there were few benefits the end user could really see. Morisawa also decided to tweak some glyph designs and the font metrics. Designers would open old files with the new fonts and find they had to redo everything. They stayed away in droves. As a result, the installed base is still overwhelmingly OCF (original composite font), which has slowed acceptance of PDF as a major production format.

Dueling fixes
Adobe’s final solution for the Japanese font problem is OpenType. OpenType will be a very big deal and, in the long term, fundamentally change the market. OpenType finally brings true font parity between Windows and Mac. Morisawa and Fontworks are busy preparing OpenType upgrades. It marks the first time their font libraries will be available cross-platform. Both companies see Windows as an opportunity for revenue growth―something that has been missing on the Mac side for years. Users will finally be able to choose the platform that best fits their needs and budgets without worrying about font problems.

This will be especially interesting because Mac OS X and OpenType are both arriving in 2001. According to a Japanese Publishing Consortium survey in 1999, Apple has nearly 80 percent of the DTP installed base. However, a large share of the machines in the printing and publishing areas (but not in design) are pre-G3 PowerPC systems that will have to be upgraded to run Mac OS X. Apple Japan’s “We’ll let the market decide” marketing statements may become a true nightmare if the market chooses Windows instead of Mac OS X

Adobe Japan 1-4
The most important feature of Japanese OpenType will be the Adobe Japan 1-4 character collection. This is the big Gaiji fix everybody has been waiting for. AJ 1-4 addresses the Gaiji encoding problem and provides a consistent and cross-platform way for font developers to add Gaiji to OpenType fonts. Adobe says the collection is “to provide professional publishers with a glyph set that will suit their needs.” AJ 1-4 will add 6,000 glyphs. They will include additional Latin characters: macroned vowels, italic forms, fractions, third and quarter-width numerals and punctuation and Latin ligatures.

They will also offer a rich set of Kanji options:

  • Annotated forms: These are popular ‘shorthand’ glyphs for such Kanji as “corporation limited” as well as specially designed numbers:
  • Hiragana, katakana and Kanji ligatures
  • Alternate kana for horizontal and vertical writing
  • Ruby glyphs: These are special kana characters that are placed next to difficult Kanji to show readers how to pronounce them. Many software DTP packages offer this feature. However, traditional ruby glyphs are a different design so they are easy to read at the small point sizes in which ruby glyphs are always used. With OpenType, these special designs can finally be included in a Japanese font.
  • Kanji and Kanji variants
    Anyone familiar with Japanese knows the language has changed more in 200 years than English has. Particularly after World War II, many Kanji were simplified and, even though the older, more complex Kanji disappeared from newspapers, magazines and school textbooks, they still live on in books, maps and proper names. There are also many different variations of certain Kanji, such as the ‘nabe’ character for the name ‘Watanabe’ (see illustration).
Kanji glyph variations

AJ 1-4 was created by Adobe in cooperation with Morisawa, Fontworks and other industry experts. It’s not going to satisfy everybody, but it’s an excellent effort that should take good care of the majority. The only drawback is timing, coming as it does on the heels of the CID upgrade catastrophe. Adobe and Morisawa are planning to market two packages: Standard, which will have the same features as current CID fonts, and Pro, which will have AJ 1- 4 Gaiji and advanced layout.

Character layout
Advanced typography is another major OpenType feature and one that, until now, has been an elusive goal. After all, Apple has been pushing advanced typography for years, first with QuickDraw GX and now with ATSUI (Apple Text Services for Unicode Imaging) in Mac OS X. Apple is still pushing its solution; the Hiragino fonts for Mac OS X will include both OpenType layout and Apple Advanced Layout (AAT) tables.

Advanced layout is not a trivial feature to add, particularly for a complex written language such as Japanese. As noted in the Gaiji section above, there are all kinds of ligatures, vertical variants and arcane Kanji. But a well thought-out and executed set of features can vastly simplify the user experience and address many shortcomings of Japanese DTP.

One example of how this can be used is Morisawa’s new Mincho Kyoiku font. This font family has special glyphs required in Japanese school texts that help children learn how to write.

Because it is based on the standard Mincho design, Morisawa digital type manager Nobuo Tomita said, it would make sense to implement the different glyphs as variants in the standard Mincho font package instead of selling them separately as they do now.

GX legacies
At the World Wide Developer Conference and at an Apple font developers’ conference in Tokyo shortly afterward, Apple stated that Apple Advanced Typography tables were the “recommended” way to add advanced layout features. One Apple engineer described the differences between the two approaches this way. “AAT is pretty much a superset of OpenType [layout], with much better performance.

There’s no real pressure for us to do this [support OpenType layout directly in ATSUI].” The Hiragino OpenType fonts for Mac OS X have both AAT and OpenType layout tables.

ATSUI is a system-level API, while OpenType layout is an application-level API. One very nice thing about ATSUI is that application developers don’t have to know all the details of font table formats and how to process them. ATSUI just takes care of it.

OpenType layout requires the application to do all the processing, so developers have to know a lot about a language if they want the features to work right. On the Windows side, there is the OpenType Layout Services Library, which helps somewhat; but whether on Windows or Mac, application developers will not find it easy to implement OpenType layout.

Adobe, of course, has the advantage over the competition here because its own type layout library, Cooltype, is built into every Adobe application. Adobe will pull out the stops and bundle the KozukaMincho Pro OpenType Kanji font, which of course does not have AAT tables, with InDesign J, due to be shipped in the first quarter of 2001.

Unhappy developers
Apple’s AAT solution might be better technology, but some font developers are not amused. Morisawa’s Nobuo Tomita said, “Apple has not been very forward with their plans…. AAT has lots [of features] that don’t have much to do with Japanese. We plan to go with Adobe’s cross-platform solution.” If Morisawa doesn’t support AAT, then the case is pretty much closed. 

Adobe, predictably, isn’t very enthusiastic about AAT either. “I went over to Apple and told them to just kill it…. We went through this whole thing with GX already,” said Adobe’s Asian Font product manager Julie Ma. Given Apple’s lack of marketing enthusiasm for ATSUI―it is ignoring some fantastic third-party technology that makes use of it, such as Fontworks International’s stroke font technology (announced at Tokyo Seybold two years ago)―it is understandable that developers are hesitant. Quark, for one, is hedging its bets; when asked which technology it plans to go with, Masato Nishimura of Quark Japan said, “We’ll wait to hear from our customers before we make any decisions.”

Another Apple move not sitting well with font developers is the additional Gaiji characters that are part of the 17,000-character set in the Mac OS X Hiragino OpenType Pro fonts. These go well beyond the Adobe Japan 1-4 collection.

Not much is clear at this point, but the extra Gaiji appear to be another ATSUI-only feature guaranteed not to work with Adobe software (nor with any other application that does not support the Apple API). The feature by itself is not bad, but Apple has not communicated with Japanese font developers well, and that is bad. The developers need detailed information to define their own OpenType features and to make sure it all works seamlessly with Hiragino and Mac OS X, and that information has not been forthcoming. As one font developer moaned, “Just when things are looking better, this happens and everybody loses.” If Apple really wants Mac OS X to be the best publishing solution, it will have to get its Developer Relations act in high gear, and fast.

In the end, Japanese OpenType’s greatest achievement will be that it delivers most of the features the high-end customers have been asking for: Gaiji, robust cross-platform support (allowing developers to put both OpenType layout and Apple’s AAT layout in the same package), and embedding and dynamic downloading. It will finally bring the Japanese printing environment to the same level the West has had for years. As Adobe Japan’s Seiichiro Miyajima, group manager for product marketing said, “To put it simply, it will be the same as using Roman [one-byte].” The timing for Japanese OpenType is difficult, but the benefits are very real. The industry will get there―eventually.

Japanese layout and the promise of InDesign J
In order to really understand the differences between Japanese and Western layout, you have to go back to elementary school. Think back to second grade when you were just learning cursive; the teacher gave you a piece of paper with lines. This was the bottom line, that was the top line. She taught you to write along the bottom line.

In Japan, the students get a piece of paper with rows of little boxes. The teacher tells the kids to write Kanji characters in the centers of the boxes. Japanese typographic layout is exactly the same; it is all based on little boxes, known as virtual bodies. It is also called the grid system, because the middle of each box is one center point on a grid. Everything is calculated from the center; there is no baseline. The whole system makes sense when you realize it all comes from the days of block type; the virtual bodies are blocks of metal type set end to end. The vocabulary, the measurements and the aesthetics all evolved from that.

Virtual bodies
Unlike DTP layout, which is graphics-driven, Japanese text composition, called kumihan, is driven by how much text will fit in a given space. Designers know how many characters (virtual bodies) are supposed to be on a line and on a page before they start composition, and this is how they discuss layout with writers and editors. Western composition is calculated from margins, a wholly different concept.

Adobe Japan InDesign product manager Hiroshi Miyamoto, who has typesetting experience, explained the difference. “It’s very important for [Japanese] operators to know the number of characters on each page, and it’s difficult to work that way with PageMaker, Quark Xpress and Illustrator. PostScript Japanese fonts (OCF, CID, OpenType) have no virtual-body information whatsoever, and historically there is no baseline in Japanese kumihan. But since PostScript fonts have only baseline information, that’s how all DTP software developed. Japanese fonts have different baseline positioning, so that when you change the font, the line breaks change.

Traditional systems used the grid system, so as long as you kept the same size, any font line breaking stayed uniform, and even if you did change the size, the grid made it very easy to make adjustments. If you are trying to calculate grids in Xpress, the point system doesn’t match well, so you end up with too many, or too few characters on the page, and the designer compensates by incrementing in .01-point values, which is overkill…. Also it’s very difficult to handle pages that have both English and Japanese, a real problem for magazines.”

It’s not impossible to do quality kumihan with DTP. It just takes more effort than traditional systems, and everything has to be done manually, which is inefficient. For text-heavy and specialty market segments still using traditional or proprietary systems (newspapers, books, manga), the cost benefits of DTP are just not there.

Japanese engine
InDesign J, announced this past February at MacWorld Tokyo and due to be shipped in the first quarter of 2001, is the first major piece of DTP software that follows Japanese typesetting conventions. To this end, InDesign J has a special Japanese typesetting engine and Japanese layout grid. Because there is only baseline information in PostScript fonts, InDesign J calculates and creates a full-width zenkaku virtual body, and this allows the designer to calculate the number of characters in a line, on a page or in an InDesign text frame, based on point size and line spacing.

Another basic Japanese kumihan function that professionals expect to have is known as Ji-Dori-Gyo-Dori. In traditional systems the priority is: (1) Calculate additions to the previous line, if the first character of the next line is a punctuation mark, (2) Calculate the shift kana so the first character of the next line is not a punctuation mark, (3) Calculate the average spacing of each kana.

This sounds easy to do, but it’s not, for two reasons. First, as noted above, everything is calculated from the virtual body’s central grid point. Second, there are special sets of and Kanji known as “Classes.” Proprietary systems have these highly specialized line-breaking rules and Class settings built in, so they can automatically calculate and adjust the spacing very quickly.

They are also very good with mixed Japanese and English, which is half-width: two Roman characters equal the width of one Kanji. InDesign J has all these features. InDesign also has support for the kyu measurement standard used in traditional systems, where one kyu equals 0.25 millimeters. Quark Xpress supports kyu too, but Adobe promised that InDesign will have a more accurate method to calculate Japanese kyu and will eliminate errors that occur in the current versions of Xpress and PageMaker.

Quark’s reaction
Quark is not standing on the sidelines. “We foresee strong growth in the market as Japanese font technologies become richer and easier to use. More and more designers will feel comfortable using DTP and especially Quark Xpress instead of traditional typesetting methods,” Desktop Systems Product Manager Masato Nishimura explained. He could not discuss specific features of Quark Xpress 5.0J, other than saying it will not be carbon-compliant. “That version will be a release subsequent to 5.0.” But he did mention some features requested by Japanese Xpress users that will very likely appear: placing each character in a desirable position (center) even if a line contains characters of different font sizes; scaling characters by center line grid in line-width increments in vertical text; enabling users to easily keep a full-width zenkaku type setting mode for all Japanese characters and punctuation.

However, Quark will not pursue InDesign’s approach of matching all the features of traditional kumihan. Adobe has followed the Japanese Industrial Standard (JIS) X4051 typesetting and composition specification (written by an ex-Shaken employee, no less), which is the kumihan “bible.” (Follow it and you’ll never commit compositional sin.) Quark has not. “We want to give the most important features to the majority of users,” Nishimura explained.

“Many designers feel they have enough features already,” a sentiment I have heard from others in the industry. From a layout perspective, the current tools are good enough. But they are not good enough for the unconverted 60 percent who are dealing with composition rather than layout. So the big question is, will this new environment, the second wave of Japanese DTP arriving in 2001―OpenType, advanced layout, Gaiji extended character sets, cross platform Kanji, InDesign J, Quark Xpress 5.0J―finally convince the unconverted to make the change to the open standards-based digital workflow that DTP has to offer? The answer: definitely maybe.

Cultural differences
“At this point I don’t think the issue has to do with technical reasons,” Toppan Printing KK Publishing Information Center’s Shinichi Konno, told me. “It has to do with middle managers and their way of thinking. Many companies know they have to move to a digital workflow at some point; some are even feeling panicky, but those kind of people don’t want to take the responsibility for changing over.” Having worked in the Japanese publishing industry for more than ten years, I can attest to the very conservative mentality of middle managers, and I can even agree―up to a point. I have been in many situations where a company desperately wants to change but doesn’t know how. If you listen carefully and talk with the managers on a level they understand, amazing transformations can take place. But it takes time and patient effort.

There is another factor at work, too: shokunin mentality. “Shokunin” is a tricky Japanese word. The dictionary translation lists the word as artisan or craftsman, but the nuance implies a long apprenticeship, an obsession with quality and detail and pride― traits the Japanese are famous for. I call them specialists, and Japanese culture does cherish specialists. The publishing industry is full of them, and if there is one thing you don’t do, it is tell a specialist his business. But specialists tend not to see the big picture. They focus on their own little area of expertise and have trouble understanding company-wide benefits of digital workflow.

This is where Adobe is playing a smart, subtle hand with InDesign J. It is using Japanese standards such as JIS X4051 and is also using all the specialty words. The marketing material is carefully scrubbed clean of foreign words such as layout (reeauhto), and text (tekisuto), instead using native specialist words like kumihan. Even the official product name foregoes the lazy ‘J’ for Japanese that most product names go for, using the formal “Nihongo ban” instead.

Quark, too, has a solid, if slightly different, vision. “The nucleus of our strategy is media-independent publishing, the next generation publishing model that encompasses print, Web, E-books …and beyond,” Nishimura explained. He said Quark hopes to free publishers to create and manage content so they can deliver it anywhere at low cost, with a layout appropriate to each media format.

It sounds good, but here again―and this is the whole point of this article―developers simply cannot do a localization job and sling their products out in the market. It takes time, careful study of market needs, and thorough knowledge of the segments and specialties therein. For example, media asset management has not done very well as a category in Japan. But it can, as it has for CRS Systems, the Japanese localizer and distributor of Canto’s Cumulus. CRS has seen Cumulus gradually grow from a software package into a intranet integration business, as Japanese managers slowly understand the cost savings from integrated workflow. However, like the font and layout situation, it has to happen in a Japanese context to which shokunin, the specialists, and middle managers can relate. 

Conclusion
Current Japanese DTP workflow is far from perfect. It is expensive and unreliable; it has primitive typography and layout and a limited variety of Kanji fonts that lack necessary Gaiji characters. Perfect Japanese DTP workflow should have a rich set of fonts, rich typography, layout and composition, Gaiji character sets, and theability to pass files between computers seamlessly and print without any problems.

The second wave of Japanese DTP will not magically convert the unconverted. It will be a gradual change, as companies see and truly understand digital workflow as cost reduction and potential revenue growth. The highly segregated nature of the Japanese market also means there are plenty of specialty niches to fill with all manner of InDesign plug-ins, Quark Xtensions and flexible media management systems, giving the 60 percent all the more reason to convert to a standards-based digital workflow.

A friend once told me that Japanese society is like an egg. In the West, the shell is brittle. A small amount of force from the inside can break through and bring change. But the Japanese shell is tough and rubbery; a small force can only bend the walls. It takes much more force to break the Japanese egg. But once it breaks, change can come very quickly. The publishing market will take time, but it will change. And change is opportunity.


Stroke Fonts: The Future of Japanese Type?

OpenType, the Adobe Japan 1-4 Character Collection and Unicode do a good job of breaking the current limitations of Japanese fonts. Unfortunately, encoding is not the only problem. You also have to create all those extra Kanji. The proposed Adobe Japan 1- 4 character set alone is 6,000 characters. One of the glaring de- ficiencies of current outline technology is that every single Kanji must be traced and tweaked extensively. A regular Japanese font has nearly 7,000 characters; add the Adobe Japan 1-4 character set for OpenType and you have a whopping 13,000 glyphs.

Take that total and multiply by each weight (light, demi-bold, bold) that has to be designed, and you have an idea of how much work goes into font-making and why Japanese fonts are so expensive. The fonts are large—anywhere from 3 to 8 MB each—and they are not very efficient, which is one of the reasons why Japanese Multiple Master fonts will never happen.

An interesting feature of Mac OS (both 9 and X) is the Open Font Architecture (OFA) that evolved from QuickDraw GX. As the name implies, OFA is an open plug-in architecture. It works with any font technology, be it PostScript, TrueType, or a completely new technology, such as stroke fonts created specifically for Japanese, Chinese and Korean writing systems.

Recombinant parts
Anybody who has studied Chinese or Japa- nese knows that although each Kanji is unique, certain parts occur again and again, recombining to create new characters. You can get a good feel for this by looking at Chinese or Japanese calligraphy. The brush is the most natural way to write Kanji, and with a little study, you quickly comprehend the strict order of each stroke.

In a similar way, Fontworks Interna- tional broke down its Kanji fonts into parts that loosely correspond to brush stokes. These stroke parts are kept in a library that the stroke-font scaler uses to draw the character, resulting in a much smaller and more efficient font. Torsten Buck, Fontworks technical director, told me that stroke technology “allows us to do weight variations over the full range from Light through Ultra Bold without losing typographic details,” all in a 4-MB font. An equivalent PostScript Kanji font family can weigh in around 18 MB.

Where stroke technology really shines is character creation. Once a base library of parts has been cre- ated, a designer can create high quality Kanji quickly and easily. A key feature of stroke technology is that it preserves the stroke width as the part is scaled, which is impossible to do with PostScript outlines.

To create a new Kanji, you simply use a similar character, swap out the parts that need to be changed and, perhaps, fine-tune them. Stroke fonts are a great technology to break the Japanese font logjam of high cost and limited design. It could do for Japanese fonts what Fontographer did for Western typography: put real power in the hands of designers and open up the floodgates.

Apple fumbles
Unfortunately, Apple doesn’t see this as an oppor- tunity. It creates great things like ATSUI and OFA and then forgets about them, leaving third-party developers holding the baby. When Fontworks approached Adobe’s InDesign team to discuss stroke font technology, Adobe didn’t even know what ATSUI was. Clearly somebody in Apple Developer Relations wasn’t doing the job.

Fontworks’s original plan was to have full-blown Japanese stroke fonts with advanced typography and multiple weights. But in the face of Apple’s indifference, it postponed the plan and instead re- leased Gaiji Master last February. This is a slimmed-down, Gaiji-only version of 2×2, Fontworks’s heavy-duty in-house stroke font development environment. It is a fascinating piece of software, and it reminds you of the early Macintosh days when the graphics software was thrilling and cutting-edge.

E-book opportunity
One reason why Fontworks went ahead with this is that it believes E-books and E-publishing could be a huge market in Japan. And stroke fonts could be the key. As Fontworks President Ross Evans put it, “For that to happen, fonts have to be as easy to read on a screen as they are in a book, otherwise people won’t adapt to it. Fonts have to be designed for the technology.”

Japanese and Chinese Kanji are complex and don’t hold up well at small point sizes on computer screens. They break down completely with bold and italic faces. All of the currently popular Japa- nese font faces were designed for print and are not screen-opti- mized. LCD screens that display up to 210 dpi will be coming soon from Sharp. But as Evans explained, “Even that is far less than a laser printer at 600 dpi. It has taken Shaken 18 years to build up their library of 150 fonts. How are they going to optimize them for the screen? With stroke technology we can do that far, far faster than anyone can.”

He showed me demonstrations of what the technology can do. The stroke-font scaler rasterizes everything in real time, without any heavy multimedia code. The quality appears very attractive for Web publishing. Evans continued, “This is the future of type. How can you have this stuff happening in Kanji unless you have stroke fonts? Apple can do this and nobody else can … if they wanted to support the technology.”

Japanese Text Layout for the Future* (hint: there isn’t one)

I finally had time to catch Adobe Nat McCully’s ATypl Tokyo 2019 presentation. He covers the topic that I have covered in depth many times before: the (sad) state of CJK typography. As Nat points out most software developers and system engineers talk about CJK support as typography without any idea of what it means. Throwing CJK glyphs on a screen is not typography, they are not the same thing at all.

The defining feature of CJK typography and layout in general and Japanese typography in particular is that space is an essential composition element equal with text and graphics, with fine space element control way beyond a baseline. Instead of thinking about how much space should be between text, flip it around and think about how much text should be between the space. Baseline font metrics will never deliver great CJK typography because there are too many limitations. So everybody implements the missing stuff on the fly and everybody does it different. Unfortunately the irony of it all is that Adobe played a huge role in how these limitations played out in the evolution of digital fonts, desktop publishing (DTP) and the situation we have today.

QuickDraw GX was probably the only time in computer history that fonts, layout engine and the basic OS came together to solve these limitations for all language systems, all language typography as equal from the bottom up. Parts of that effort survived, such as Apple’s San Francisco variable system font based on the TrueType GX model, and the inclusion of the TrueType GX model as the base technology for OpenType Variable fonts. Nice as this is, it’s only a tiny sliver of the GX vision pie that survived, all the other baseline font metric and CJK typography limitations still exist. Outside of a handful of people like Nat at Adobe, and the Adobe CJK typography ghetto approach of keeping all the good stuff corralled in InDesign J, very little is being done to address them.

Call me a pessimist but after 20 years of watching things slide sideways, I don’t see much hope for the future evolution of great CJK typography on digital devices. Most western software development people think that having CKJ glyphs on a screen is ‘good enough’ CJK typography, end of story.

Already I see the OpenType Variable Font effort devolving into a bauble for web developer geeks, always stuck in demo-hell, never going mainstream. It is the same story for quality CJK typography on digital devices. When the current Adobe CJK leaders like McCully and Ken Lunde reach retirement age, whom have devoted their careers to fixing these problems, I think it will be the end of an era. In many ways we are already there.

Apple prides itself on having good typography but cannot be bothered with such Japanese typography basics as not mixing Gothic and Ryumin Japanese font styles seen here in the Photos app

UPDATE
Ken Lunde posted a wonderful overview of his Adobe career to date, also his ATypl Tokyo 2019 presentation.