Comments on: Circumventing the Roman Alphabet URL http://www.davejenkins.com/2010/07/04/circumventing-the-roman-alphabet-url/ Ecommerce Strategy in Asia Thu, 12 Jan 2012 01:45:22 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Japan and twitter: why such a hit? » Dave Jenkins http://www.davejenkins.com/2010/07/04/circumventing-the-roman-alphabet-url/comment-page-1/#comment-38705 Japan and twitter: why such a hit? » Dave Jenkins Tue, 17 Aug 2010 14:47:44 +0000 http://www.davejenkins.com/?p=220#comment-38705 [...] Input tools – Twitter seems tailor made for mobile devices: short, text-centric, quick input.  For a networked society that already preferred the mobile phone-based Internet access, twitter drops right into place, often much handier than email.  I’ve discussed this before. [...] [...] Input tools – Twitter seems tailor made for mobile devices: short, text-centric, quick input.  For a networked society that already preferred the mobile phone-based Internet access, twitter drops right into place, often much handier than email.  I’ve discussed this before. [...]

]]> By: Dave http://www.davejenkins.com/2010/07/04/circumventing-the-roman-alphabet-url/comment-page-1/#comment-37848 Dave Wed, 23 Jun 2010 03:35:15 +0000 http://www.davejenkins.com/?p=220#comment-37848 Brilliant. I knew there would be someone out there smarter than me to clear this up. Arigatou, Inoue-san. Brilliant. I knew there would be someone out there smarter than me to clear this up. Arigatou, Inoue-san.

]]> By: Eido Inoue http://www.davejenkins.com/2010/07/04/circumventing-the-roman-alphabet-url/comment-page-1/#comment-37846 Eido Inoue Wed, 23 Jun 2010 03:14:03 +0000 http://www.davejenkins.com/?p=220#comment-37846 Japanese language URLs exist and are supported in Japan under the .jp and .com domains (try going to ). There are five reasons why ads show search term queries (in order of importance): #1. Inputing English and especially symbols such as slash ("/"), colon (":") is difficult on a ten-key (non-QWERTY) phone, which is still the most preferred input hardware for Asian mobile phones. For example, in a worse case scenario on a handset I won't mention, it takes 31 key presses to input the url . To input "グーグル" takes 14 worse case, but probably less due to auto-completion. (If you ever wondered why people search for "google" on Google, this is why: ironically, it's easier to get to "google.com" by _searching_ for it than inputing the URL directly). #2. Firing up the search widget on a typical Japanese mobile phone is a one (short or long) button press, at most two. Firing up the manual URL input function is usually a few more keystrokes and buried in a sub-menu (many people don't even know how to input a URL on their phone even though they use the browser!). #3. It's easier to stay at the top of search than you think; in addition to SEOs, you can buy (heaven forbid!) ads linked to keywords, which a professional company that is putting an ad out is going to do. Also, mobile phones often have a walled-garden search and ads, and getting exclusive rights to a keyword in special sections, etc., is easier than general search. #4. Punycode (used for non-ASCII domain names) support and/or non-English URL support is sketchy and sometimes non-existent in older browsers and especially older and more primitive mobile phones, and sometimes only partially supported (i.e. it will accept the URL in non-English, but will immediately convert it into the unreadable/too-long ) #5. Putting QR codes on train adverts (though they exist) is frowned upon because the appearance of taking photos with your camera on a crowded train is bad manners (some phones in Japan will make a shutter sound* even when in QR-code reader mode and your phone is muted, so it's harder to get a QR code covertly than you think) * The iPhone and Android-based phones sold in Japan are specially modified to make a shutter sound even when muted. Japanese language URLs exist and are supported in Japan under the .jp and .com domains (try going to ). There are five reasons why ads show search term queries (in order of importance):

#1. Inputing English and especially symbols such as slash (“/”), colon (“:”) is difficult on a ten-key (non-QWERTY) phone, which is still the most preferred input hardware for Asian mobile phones. For example, in a worse case scenario on a handset I won’t mention, it takes 31 key presses to input the url . To input “グーグル” takes 14 worse case, but probably less due to auto-completion. (If you ever wondered why people search for “google” on Google, this is why: ironically, it’s easier to get to “google.com” by _searching_ for it than inputing the URL directly).

#2. Firing up the search widget on a typical Japanese mobile phone is a one (short or long) button press, at most two. Firing up the manual URL input function is usually a few more keystrokes and buried in a sub-menu (many people don’t even know how to input a URL on their phone even though they use the browser!).

#3. It’s easier to stay at the top of search than you think; in addition to SEOs, you can buy (heaven forbid!) ads linked to keywords, which a professional company that is putting an ad out is going to do. Also, mobile phones often have a walled-garden search and ads, and getting exclusive rights to a keyword in special sections, etc., is easier than general search.

#4. Punycode (used for non-ASCII domain names) support and/or non-English URL support is sketchy and sometimes non-existent in older browsers and especially older and more primitive mobile phones, and sometimes only partially supported (i.e. it will accept the URL in non-English, but will immediately convert it into the unreadable/too-long )

#5. Putting QR codes on train adverts (though they exist) is frowned upon because the appearance of taking photos with your camera on a crowded train is bad manners (some phones in Japan will make a shutter sound* even when in QR-code reader mode and your phone is muted, so it’s harder to get a QR code covertly than you think)

* The iPhone and Android-based phones sold in Japan are specially modified to make a shutter sound even when muted.

]]> By: Dave http://www.davejenkins.com/2010/07/04/circumventing-the-roman-alphabet-url/comment-page-1/#comment-37666 Dave Sun, 13 Jun 2010 15:48:19 +0000 http://www.davejenkins.com/?p=220#comment-37666 Jon, I think URLs will be non-roman within a couple of years-- the marketing pressure to line it up with the real name of the company is too strong. Kirk-- I also agree that the DNS, and the bass-ackwards notation of sub.domain.tld.country needs a serious re-think. Unfortunately, I also think that this re-think will come with Internet2 and IP6, whenever those get here (just in time to play some DukeNukem Forever, maybe?). Jon,

I think URLs will be non-roman within a couple of years– the marketing pressure to line it up with the real name of the company is too strong.

Kirk– I also agree that the DNS, and the bass-ackwards notation of sub.domain.tld.country needs a serious re-think. Unfortunately, I also think that this re-think will come with Internet2 and IP6, whenever those get here (just in time to play some DukeNukem Forever, maybe?).

]]> By: Jon Jensen http://www.davejenkins.com/2010/07/04/circumventing-the-roman-alphabet-url/comment-page-1/#comment-37607 Jon Jensen Fri, 11 Jun 2010 01:07:18 +0000 http://www.davejenkins.com/?p=220#comment-37607 That's pretty interesting, and is a sensible solution to the problem of Roman-only URLs. Do you think the marketers are likely to switch to URLs instead of search terms even if the TLDs are still Roman, like .co.jp? Or are they likely to wait till full non-Roman TLDs are available? That’s pretty interesting, and is a sensible solution to the problem of Roman-only URLs.

Do you think the marketers are likely to switch to URLs instead of search terms even if the TLDs are still Roman, like .co.jp? Or are they likely to wait till full non-Roman TLDs are available?

]]> By: Kirk http://www.davejenkins.com/2010/07/04/circumventing-the-roman-alphabet-url/comment-page-1/#comment-37604 Kirk Thu, 10 Jun 2010 23:20:52 +0000 http://www.davejenkins.com/?p=220#comment-37604 Its a natural progression, as the web internationalizes fully, the dominance of roman characters will fall. That is a nice trick though, its almost like a hash tag in print advertising to generate a unique ID to correlate search ads on, rather than a specific characteristic of the product. It seems like this would be most effective on established brands, and it would likely cause a conflict with new products, if the ad worked well, the hash tag becomes the brand, should you change the hash tag, or someone else pick it up, you are back at square one. Hopefully the migration from the roman character monopoly will have the knock-on effect of getting a thorough redesign of DNS done. I long for a future where infinite character sets are allowed and pakistan can decide to block youtube without ruining it for the rest of us. And crypto would be nice, but thats probably a stretch. Its a natural progression, as the web internationalizes fully, the dominance of roman characters will fall. That is a nice trick though, its almost like a hash tag in print advertising to generate a unique ID to correlate search ads on, rather than a specific characteristic of the product. It seems like this would be most effective on established brands, and it would likely cause a conflict with new products, if the ad worked well, the hash tag becomes the brand, should you change the hash tag, or someone else pick it up, you are back at square one.

Hopefully the migration from the roman character monopoly will have the knock-on effect of getting a thorough redesign of DNS done. I long for a future where infinite character sets are allowed and pakistan can decide to block youtube without ruining it for the rest of us. And crypto would be nice, but thats probably a stretch.

]]>