Hi,
On Jul 10, 2005, at 1:18 AM, Mohit Sindhwani wrote:
> Are they complimentary or competitive?
To date, QR and Semacode haven't been competing head to head because QR
is only available in Japan and Semacode is mainly available outside
Japan. Simply because we run on Symbian smartphones and now MIDP Java
ME phones, whereas QR runs on Japanese phones. My todo list for the
company does say "port to DoJa" somewhere in there though.
On Jul 11, 2005, at 12:39 AM, Mohit Sindhwani wrote:
> Another simple question - for the "typical" QR or Semacode, would a VGA
> camera be good enough? That's the easily available standard in
> Singapore, though 1MP is now common on newer phones.
About VGA resolution. Well, Java ME camera access (MMAPI) doesn't
specify a minimum resolution and a lot of MIDP phones only offer really
low resolutions like 160x120 (or maybe a bit more). So, in order to
work on Java phones we obviously had to make changes to our image
recognition algorithms to read semacode tags well from such small
images. But remember that the native resolution of the camera might be
VGA or higher but the MIDP apps only get a tiny fraction of that, it's
downsampled. The device manufacturers probably have their reasons for
doing that.
Umm... what else?
On Jul 11, 2005, at 9:53 PM, Curt Sampson wrote:
> The problem with Semacode is that it has a single defined markkup:
> the contents are a URL. You can't change this to have further markups
> without breaking backwards compatability. (Though potentially you could
> introduce new URL types, such as a URL pointing to or even containing
> address book information.)
Yeah, we specify that a semacode tag can only contain a URL. But now,
you see that there's a good reason for that. When you get loose in your
spec you get the situation like in Japan where the 3 operators all have
fragmented specs for how to encode other kinds of information. Semacode
is all about interoperability and future proofing. Besides, why encode
the vCard directly when you can put in a URL to a vCard that's online?
Saying that QR codes can encode more kinds of information is
ridiculous, since a URL can address ANY kind of information. There's
lots of kinds of URLs, too, like tel:NNNNNNN which will make the phone
dial a number, or sms:XXXX which can encode a number and a message to
send via SMS. Instead of trying to assume that we are smarter than the
IETF and the W3C and so on, we instead take advantage of their
expertise in designing the URL system which is already very very
flexible.
Getting back to the levels thing,
On Tue, 12 Jul 2005, Gerhard Fasol wrote:
> YES! QR code has NOT been developed for mobile phones!
Gerhard pointed out that QR Code was not developed for mobile phones.
True! It was developed for auto parts stamping. But the same name "QR"
is used for the mobile system which is a layer above and adds URL and
vCard encoding etc. Semacode has the same relationship with Data Matrix
which is a hugely popular 2d optical bar code outside Japan, used for
many of the same purposes and more. It's also an ISO standard with all
the necessary IP in the public domain.
Speed. Our new "arthur" image recognizer which has been seeded to our
developers and friendly people is a heck of a lot faster than the one
used in version 1.1. In native code (like on Symbian) we get sub-second
recognition. The "arthur" code is used in the Java version which is out
already. Since it's running on the Java interpreter it's not as fast as
the native version, but on the other hand it runs on many, many more
phones.
> My experience of QR in Japan is that the keitai cameras are far more
> sophisticated and uisng QR is so easy - my experience here in Europe
> with
> semaCode is that one has to take the picture 2 or 3 times before it
> works
> and is slow in reading the content.
My business developer Maciek was in San Francisco at JavaOne doing
demos with a Sony-Ericsson S710a (Java ME version) and he was excited
because it didn't misread once. That makes me happy :-)
--simon
--
Simon Woodside - Founder
Semacode Corporation
http://semacode.com
Received on Wed Jul 13 23:12:40 2005