wmlprogramming : Messages : 31807-31836 of 31836

If I own a photography website and I have a mobile version there are two things that Bolt and the likes are doing to disregard my right as the owner of the website:

1. They have made it impossible for me to deliver my mobile-optimized content

and

2. They transcode and compress my data against my will, so the photos that I'm showcasing will look like crap on the device

Thanks,

Steve Kamerman

On Wed, Dec 23, 2009 at 5:26 PM, Christopher Haslage <chaslage@...> wrote:

�

Luca,

Wait, I am confused with the last paragraph, Luca.� What do you mean disregarding the right of content owners?

Chris

Luca Passani wrote:

Steve, you are of course right. Bolt is some kind of operamini. Here is 
the root of all unidentified Bolts.

http://db.wurflpro.com/device/show/25147

I have added the UA strings you indicated in the other email. I also 
gave it a default screen size of 240x320, which is probably a good 
choice for users of that browser.

...and yes, this browser, just like OperaMini, has a lot in common with 
evil transcoders (mainly, shifting the focus on the end-user and 
disregarding the right of content owners).

Thanks

Luca



Steve Kamerman wrote:

Luca, once you have time we should probably open up a discussion about
the BOLT browser and how to handle it since it's not just a Blackberry
browser. I've been trying to figure out how to get the original UA
from its request header, but I don't have a device to test it with and
the emulator (http://boltbrowser.com/demo/ 
<http://boltbrowser.com/demo/>) is not sending an original
ua.

Here's some good info on the BOLT browser: 
http://wapreview.com/blog/?p=2598 <http://wapreview.com/blog/?p=2598>

Also, according to their website, the BOLT browser does NOT change
anything on the requested site, it only COMPRESSES the data (still a
transcoder by definition in my opinion).

Thanks,

Steve Kamerman

On Wed, Dec 23, 2009 at 1:50 PM, Steve Kamerman 
<stevekamerman@... <mailto:stevekamerman%40gmail.com>> wrote:

I did notice there are two BOLT browser UAs in the WURFL. The reason
they are not being detected in Chris' case is that his UA has "BOLT"
in the UA, not "Bolt" as it exists in the WURFL. Here's what I'm
using to remedy the problem for now:

Patch File: bolt.xml
--------------------------------------
<?xml version="1.0" encoding="utf-8"?>
<wurfl_patch>
<devices>
<device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
BOLT/1.601) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt0"/>
<device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
BOLT/1.600) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt1"/>
<device user_agent="Mozilla/5.0 (X11; 78; CentOS; US-en)
AppleWebKit/527+ (KHTML, like Gecko) Bolt/0.920 Version/3.0
Safari/523.15" fall_back="bolt_browser" id="terawurfl_bolt2"/>
</devices>
</wurfl_patch>
-------------------------------------

Chris, while we are working on getting these into the WURFL, you can
copy the text above and paste it into a file called "bolt.xml" in the
data directory, then edit TeraWurflConfig.php and add ";bolt.xml" to
the end of the $PATCH_FILE directive, then reload the WURFL. Here's
how it should look in the config:

/**
* Filename of patch file. If you want to use more than one,
seperate them with semicolons. They are loaded in order.
* ex: $PATCH_FILE = 'web_browsers_patch.xml;custom_patch_ver2.3.xml';
* @var String
*/
public static $PATCH_FILE =
'custom_web_patch.xml;web_browsers_patch.xml;bolt.xml';

You must use the latest WURFL with this patch, otherwise you risk
breaking the fallback tree for these devices.

Thanks,

Steve Kamerman

On Wed, Dec 23, 2009 at 1:24 PM, Luca Passani

<luca.passani@... <mailto:luca.passani%40gmail.com>> wrote:

Steve Kamerman wrote:

Luca, are you tracking this conversation? How do you intend to handle
BOLT UA Matching? Are you intending to add BOLT to the WURFL in the
near future? Does anyone know what browser BOLT is a descendant of
(i.e. WebKit or another WURFL ID)?

Yes, I am tracking. I'll catch up with this (and other stuff) during
Xmas and decide how to best handle this. It is totally possible that a
blackberry_webkit family (or something along this line) is introduced.

Thanks

Luca



------------------------------------

As of July 14 2005, it's much easier to be banned from WMLProgramming!
Please fail to read

http://tech.groups.yahoo.com/group/wmlprogramming/ 
<http://tech.groups.yahoo.com/group/wmlprogramming/> before you 
post.Yahoo! Groups Links

        

------------------------------------

As of July 14 2005, it's much easier to be banned from WMLProgramming!
Please fail to read http://tech.groups.yahoo.com/group/wmlprogramming/ before you post.Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/wmlprogramming/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/wmlprogramming/join
    (Yahoo! ID required)

<*> To change settings via email:
    wmlprogramming-digest@yahoogroups.com 
    wmlprogramming-fullfeatured@yahoogroups.com

<*> To unsubscribe from this group, send an email to:
    wmlprogramming-unsubscribe@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/

Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

#31835

From: "malcolmbox" <malcolm.box@...>
Date: Wed Dec 23, 2009 3:09 pm
Subject: Re: Desktop browser detection

malcolmbox

Offline

--- In wmlprogramming@yahoogroups.com, Marc Guay <marc.guay@...> wrote:
>
> Hi everyone,
>
> I have a user agent, coming from IETester's IE6: "Mozilla/4.0
> (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR
> 3.0.4506.2152; .NET CLR 3.5.30729)", which would seem to match the
> web_browsers_patch.xml definition of IE6: "Mozilla/4.0 (compatible;
> MSIE 6.0;", bt it's still being detected as a wireless device.  Any
> ideas why that would be happening?

Not a direct answer to your question, but something I've found successful with
detecting desktop web browsers is to pre-filter before WURFL.

Since all desktop browsers can be assumed to come from Windows, Mac, Linux, BSD,
Solaris, then look for those OSes in the UA, and if found then just hardwire to
"this is a desktop, go crazy with that thar HTML", otherwise pass to WURFL to
see what it can do.

This idea isn't mine - it's shamelessly borrowed from 
http://smartmobtoolkit.wordpress.com/2009/01/26/not-device-detection-javascript-\
perl-and-php-code/

Hope this might be some help.

Malcolm

#31834

From: "ayyappa_s_thndrkiss" <ayyappa_s_thndrkiss@...>
Date: Wed Dec 23, 2009 2:29 pm
Subject: Chinese mobiles and mobile networks

ayyappa_s_th...

Offline

Dear all,

This is my first message to this group. First of all thanks a lot to moderator
for adding me up in this group.

Here is my queries.

1. Is china mobiles networks blocking the useragent ?

2. The useragent send to the server are in different encoding for Chinese
mobiles.

Please let me know which one is true. If the 2nd one is true how to extract the
useragent ? will it be in English or Chinese ?

Reply | Forward | Messages in this Topic (1)

#31833

From: Christopher Haslage <chaslage@...>
Date: Wed Dec 23, 2009 2:26 pm
Subject: Re: Best Detection Method

chaslage...

Offline

Luca,

Wait, I am confused with the last paragraph, Luca.� What do you mean disregarding the right of content owners?

Chris

Luca Passani wrote:

Steve, you are of course right. Bolt is some kind of operamini. Here is the root of all unidentified Bolts.
http://db.wurflpro.com/device/show/25147
I have added the UA strings you indicated in the other email. I also gave it a default screen size of 240x320, which is probably a good choice for users of that browser.
...and yes, this browser, just like OperaMini, has a lot in common with evil transcoders (mainly, shifting the focus on the end-user and disregarding the right of content owners).
Thanks
Luca
Steve Kamerman wrote:

Luca, once you have time we should probably open up a discussion about
the BOLT browser and how to handle it since it's not just a Blackberry
browser. I've been trying to figure out how to get the original UA
from its request header, but I don't have a device to test it with and
the emulator (http://boltbrowser.com/demo/ <http://boltbrowser.com/demo/>) is not sending an original
ua.
Here's some good info on the BOLT browser: http://wapreview.com/blog/?p=2598 <http://wapreview.com/blog/?p=2598>
Also, according to their website, the BOLT browser does NOT change
anything on the requested site, it only COMPRESSES the data (still a
transcoder by definition in my opinion).
Thanks,
Steve Kamerman
On Wed, Dec 23, 2009 at 1:50 PM, Steve Kamerman <stevekamerman@... <mailto:stevekamerman%40gmail.com>> wrote:

I did notice there are two BOLT browser UAs in the WURFL. The reason
they are not being detected in Chris' case is that his UA has "BOLT"
in the UA, not "Bolt" as it exists in the WURFL. Here's what I'm
using to remedy the problem for now:
Patch File: bolt.xml
--------------------------------------
<?xml version="1.0" encoding="utf-8"?>
<wurfl_patch>
<devices>
<device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
BOLT/1.601) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt0"/>
<device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
BOLT/1.600) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt1"/>
<device user_agent="Mozilla/5.0 (X11; 78; CentOS; US-en)
AppleWebKit/527+ (KHTML, like Gecko) Bolt/0.920 Version/3.0
Safari/523.15" fall_back="bolt_browser" id="terawurfl_bolt2"/>
</devices>
</wurfl_patch>
-------------------------------------
Chris, while we are working on getting these into the WURFL, you can
copy the text above and paste it into a file called "bolt.xml" in the
data directory, then edit TeraWurflConfig.php and add ";bolt.xml" to
the end of the $PATCH_FILE directive, then reload the WURFL. Here's
how it should look in the config:
/**
* Filename of patch file. If you want to use more than one,
seperate them with semicolons. They are loaded in order.
* ex: $PATCH_FILE = 'web_browsers_patch.xml;custom_patch_ver2.3.xml';
* @var String
*/
public static $PATCH_FILE =
'custom_web_patch.xml;web_browsers_patch.xml;bolt.xml';
You must use the latest WURFL with this patch, otherwise you risk
breaking the fallback tree for these devices.
Thanks,
Steve Kamerman
On Wed, Dec 23, 2009 at 1:24 PM, Luca Passani

<luca.passani@... <mailto:luca.passani%40gmail.com>> wrote:

Steve Kamerman wrote:

Luca, are you tracking this conversation? How do you intend to handle
BOLT UA Matching? Are you intending to add BOLT to the WURFL in the
near future? Does anyone know what browser BOLT is a descendant of
(i.e. WebKit or another WURFL ID)?

Yes, I am tracking. I'll catch up with this (and other stuff) during
Xmas and decide how to best handle this. It is totally possible that a
blackberry_webkit family (or something along this line) is introduced.
Thanks
Luca
------------------------------------
As of July 14 2005, it's much easier to be banned from WMLProgramming!
Please fail to read

http://tech.groups.yahoo.com/group/wmlprogramming/ <http://tech.groups.yahoo.com/group/wmlprogramming/> before you post.Yahoo! Groups Links


------------------------------------
As of July 14 2005, it's much easier to be banned from WMLProgramming!
Please fail to read http://tech.groups.yahoo.com/group/wmlprogramming/ before you post.Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/wmlprogramming/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/wmlprogramming/join
(Yahoo! ID required)
<*> To change settings via email:
wmlprogramming-digest@yahoogroups.com wmlprogramming-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to:
wmlprogramming-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/

--
Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

#31832

From: Luca Passani <luca.passani@...>
Date: Wed Dec 23, 2009 2:18 pm
Subject: Re: Best Detection Method

luca_passani

Offline

Steve, you are of course right. Bolt is some kind of operamini. Here is
the root of all unidentified Bolts.

http://db.wurflpro.com/device/show/25147

I have added the UA strings you indicated in the other email. I also
gave it a default screen size of 240x320, which is probably a good
choice for users of that browser.

...and yes, this browser, just like OperaMini, has a lot in common with
evil transcoders (mainly, shifting the focus on the end-user and
disregarding the right of content owners).

Thanks

Luca



Steve Kamerman wrote:
>
> Luca, once you have time we should probably open up a discussion about
> the BOLT browser and how to handle it since it's not just a Blackberry
> browser. I've been trying to figure out how to get the original UA
> from its request header, but I don't have a device to test it with and
> the emulator (http://boltbrowser.com/demo/
> <http://boltbrowser.com/demo/>) is not sending an original
> ua.
>
> Here's some good info on the BOLT browser:
> http://wapreview.com/blog/?p=2598 <http://wapreview.com/blog/?p=2598>
>
> Also, according to their website, the BOLT browser does NOT change
> anything on the requested site, it only COMPRESSES the data (still a
> transcoder by definition in my opinion).
>
> Thanks,
>
> Steve Kamerman
>
> On Wed, Dec 23, 2009 at 1:50 PM, Steve Kamerman
> <stevekamerman@... <mailto:stevekamerman%40gmail.com>> wrote:
> > I did notice there are two BOLT browser UAs in the WURFL. The reason
> > they are not being detected in Chris' case is that his UA has "BOLT"
> > in the UA, not "Bolt" as it exists in the WURFL. Here's what I'm
> > using to remedy the problem for now:
> >
> > Patch File: bolt.xml
> > --------------------------------------
> > <?xml version="1.0" encoding="utf-8"?>
> > <wurfl_patch>
> > <devices>
> > <device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
> > BOLT/1.601) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
> > Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt0"/>
> > <device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
> > BOLT/1.600) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
> > Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt1"/>
> > <device user_agent="Mozilla/5.0 (X11; 78; CentOS; US-en)
> > AppleWebKit/527+ (KHTML, like Gecko) Bolt/0.920 Version/3.0
> > Safari/523.15" fall_back="bolt_browser" id="terawurfl_bolt2"/>
> > </devices>
> > </wurfl_patch>
> > -------------------------------------
> >
> > Chris, while we are working on getting these into the WURFL, you can
> > copy the text above and paste it into a file called "bolt.xml" in the
> > data directory, then edit TeraWurflConfig.php and add ";bolt.xml" to
> > the end of the $PATCH_FILE directive, then reload the WURFL. Here's
> > how it should look in the config:
> >
> > /**
> > * Filename of patch file. If you want to use more than one,
> > seperate them with semicolons. They are loaded in order.
> > * ex: $PATCH_FILE = 'web_browsers_patch.xml;custom_patch_ver2.3.xml';
> > * @var String
> > */
> > public static $PATCH_FILE =
> > 'custom_web_patch.xml;web_browsers_patch.xml;bolt.xml';
> >
> > You must use the latest WURFL with this patch, otherwise you risk
> > breaking the fallback tree for these devices.
> >
> > Thanks,
> >
> > Steve Kamerman
> >
> > On Wed, Dec 23, 2009 at 1:24 PM, Luca Passani
> <luca.passani@... <mailto:luca.passani%40gmail.com>> wrote:
> >> Steve Kamerman wrote:
> >>>
> >>> Luca, are you tracking this conversation? How do you intend to handle
> >>> BOLT UA Matching? Are you intending to add BOLT to the WURFL in the
> >>> near future? Does anyone know what browser BOLT is a descendant of
> >>> (i.e. WebKit or another WURFL ID)?
> >>
> >> Yes, I am tracking. I'll catch up with this (and other stuff) during
> >> Xmas and decide how to best handle this. It is totally possible that a
> >> blackberry_webkit family (or something along this line) is introduced.
> >>
> >> Thanks
> >>
> >> Luca
> >>
> >>
> >>
> >> ------------------------------------
> >>
> >> As of July 14 2005, it's much easier to be banned from WMLProgramming!
> >> Please fail to read
> http://tech.groups.yahoo.com/group/wmlprogramming/
> <http://tech.groups.yahoo.com/group/wmlprogramming/> before you
> post.Yahoo! Groups Links
> >>
> >>
> >>
> >>
> >
>
>

#31831

From: Christopher Haslage <chaslage@...>
Date: Wed Dec 23, 2009 2:15 pm
Subject: Re: Best Detection Method

chaslage...

Offline

Steve,

Your wish is my command. :)

Regular Layout:

Request Headers
array ( �
'Host' => 'www.tera-wurfl.com', �
'Accept-Encoding' => 'deflate, gzip', �
'User-Agent' => 'Mozilla/5.0 ( Windows; U; Windows NT 5.1 ; en-US; BOLT/1.601) �AppleWebKit/530+ ( KHTML, like Gecko) Version/4.0 �Safari/ 530.17', �
'Accept' => 'text/xml,application/xml, application/xhtml+xml,text/html;q=0.9 ,text/plain;q=0.8 ,image/png,*/*;q=0.5 ', �

'X-Forwarded-For' => ' 206.53.147.7 ', �
'Via' => '1.0 �us1. mysmallplanet.com:3128 ( squid/2.7. STABLE7) ', �
'Cache-Control' => 'max-age= 259200 ', �
'Connection' => 'keep-alive',
)

Mobile layout:

Request Headers
array ( �
'Host' => 'www.tera-wurfl.com', �
'Accept-Encoding' => 'deflate, gzip', �
'Accept' => 'text/javascript,text/ecmascript,application/x-javascript,text/html,application/vnd.wap.xhtml+xml,application/xhtml+xml,text/css,text/vnd.wap.wml,application/vnd.wap.wmlc,application/vnd.wap.wmlscriptc,application/java-archive,application/java,application/x-java-archive,text/vnd.sun.j2 me.app-descriptor,application/vnd.oma.drm.message,application/vnd.oma.drm.content,application/vnd.wap.mms-message,application/vnd.wap.sic,text/x-co-desc,application/vnd.oma.dd+xml,*/*', � 'http' => '//www.blackberry.net/go/mobile/profiles/uaprof/8330 /4.5.0. rdf', �
'Pragma' => 'no-cache', �
'User-Agent' => 'BlackBerry8330 / 4.5.0.138 �Profile/MIDP-2.0 �Configuration/CLDC-1.1 �VendorID/105 ', �
'X-Forwarded-For' => ' 206.53.147.7 ', �
'Via' => '1.0 �us1. mysmallplanet.com:3128 ( squid/2.7. STABLE7) ', �
'Cache-Control' => 'no-cache, max-age= 259200 ', �
'Connection' => 'keep-alive',
)

Chris

Steve Kamerman wrote:

Chris, the BOLT browser is not currently in the WURFL and its
"desktop" user agent is loosely matched to an existing desktop user
agent. �Can you please go to�http://www.tera-wurfl.com/m/�on your
device and post the Request Headers for me? �Also, please change the
browser to mobile mode and paste those headers.
Here's some info from BOLT tech support: "You can view the original
user agent string for your mobile using BOLT by switching to mobile
layout. To enable mobile layout, select "Preferences" from the "Menu",
then select the "Mobile Layout" check box."
I think I will handle this like Opera Mini, with it's own user agent matcher.
Luca, are you tracking this conversation? How do you intend to handle
BOLT UA Matching? Are you intending to add BOLT to the WURFL in the
near future? Does anyone know what browser BOLT is a descendant of
(i.e. WebKit or another WURFL ID)?
Thanks,
Steve Kamerman
On Wed, Dec 23, 2009 at 2:35 AM, Christopher Haslage
<chaslage@...> wrote:

Yes, it's the newest from a few days ago. And indeed Tera WURFL.
HTTP Agent: �Mozilla/5.0 ( Windows; U; Windows NT 5.1 ; en- US; BOLT/1.601) �AppleWebKit/530+ ( KHTML, like Gecko) �Version/4.0 �Safari/ 530.17
Here is what it reports back:
Is it From .MOBI: �No (com)
Is it Wireless: �No �(Redirect to normal WWW)
Phone Type: �- �()
Preferred Markup: �html_wi_oma_xhtmlmp_1 _0
Dual Orientation: �No
XHTML Support Level: �1
Telephone Code: �tel:
WAP Support: �No
Resolution: �90 �x 40
Thanks,
Chris
Steve Kamerman wrote:
Chris, if the Bolt is showing up as a non-wireless device, then there is something wrong. �Have you updated your WURFL lately? �A new release came out recently. �The reason the screen size is wrong is because the device wasn't properly detected. �Do you have the User-Agent for this device? �Can you please go to your web admin page and go to the WURFL Test Script and paste the user agent in, then send us the output? (I'm assuming you're using Tera-WURFL)
Thanks,
Steve Kamerman
On Tue, Dec 22, 2009 at 11:14 PM, Christopher Haslage <chaslage@...> wrote:

All,
So I am slowly working on detection methods, I even made it so anyone coming from a .mobi site will always get a mobile page.� Though now I get to WURFL and depending on detection methods for a regular website.
So, of course there is easy indicator, namely:� is_wireless_device
However, I noticed in Bolt for the Blackberry the detection shows as not wireless but the screen size is 90x40.� Which raises my next question...
Should I also look at screen size as a good indicator of content as well?� I know that Bolt doesn't show the right information, because the Blackberry is 320x240 and it's definitely on a wireless device.� Anything below 640x480 maybe?
I guess I am asking what would be the best mix of checks?
Thanks!
Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

--
Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/


------------------------------------
As of July 14 2005, it's much easier to be banned from WMLProgramming!
Please fail to read http://tech.groups.yahoo.com/group/wmlprogramming/ before you post.Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/wmlprogramming/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/wmlprogramming/join
(Yahoo! ID required)
<*> To change settings via email:
wmlprogramming-digest@yahoogroups.com wmlprogramming-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to:
wmlprogramming-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/

--
Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

#31830

From: Steve Kamerman <stevekamerman@...>
Date: Wed Dec 23, 2009 10:54 am
Subject: Re: Best Detection Method

steve_bond_k...

Offline

Luca, once you have time we should probably open up a discussion about
the BOLT browser and how to handle it since it's not just a Blackberry
browser.  I've been trying to figure out how to get the original UA
from its request header, but I don't have a device to test it with and
the emulator (http://boltbrowser.com/demo/) is not sending an original
ua.

Here's some good info on the BOLT browser: http://wapreview.com/blog/?p=2598

Also, according to their website, the BOLT browser does NOT change
anything on the requested site, it only COMPRESSES the data (still a
transcoder by definition in my opinion).

Thanks,

Steve Kamerman

On Wed, Dec 23, 2009 at 1:50 PM, Steve Kamerman <stevekamerman@...> wrote:
> I did notice there are two BOLT browser UAs in the WURFL. �The reason
> they are not being detected in Chris' case is that his UA has "BOLT"
> in the UA, not "Bolt" as it exists in the WURFL. �Here's what I'm
> using to remedy the problem for now:
>
> Patch File: bolt.xml
> --------------------------------------
> <?xml version="1.0" encoding="utf-8"?>
> <wurfl_patch>
> � � � �<devices>
> � � � � � � � �<device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1;
en-US;
> BOLT/1.601) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
> Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt0"/>
> � � � � � � � �<device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1;
en-US;
> BOLT/1.600) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
> Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt1"/>
> � � � � � � � �<device user_agent="Mozilla/5.0 (X11; 78; CentOS; US-en)
> AppleWebKit/527+ (KHTML, like Gecko) Bolt/0.920 Version/3.0
> Safari/523.15" fall_back="bolt_browser" id="terawurfl_bolt2"/>
> � � � �</devices>
> </wurfl_patch>
> -------------------------------------
>
> Chris, while we are working on getting these into the WURFL, you can
> copy the text above and paste it into a file called "bolt.xml" in the
> data directory, then edit TeraWurflConfig.php and add ";bolt.xml" to
> the end of the $PATCH_FILE directive, then reload the WURFL. �Here's
> how it should look in the config:
>
> � � � �/**
> � � � � * Filename of patch file. �If you want to use more than one,
> seperate them with semicolons. �They are loaded in order.
> � � � � * ex: $PATCH_FILE = 'web_browsers_patch.xml;custom_patch_ver2.3.xml';
> � � � � * @var String
> � � � � */
> � � � �public static $PATCH_FILE =
> 'custom_web_patch.xml;web_browsers_patch.xml;bolt.xml';
>
> You must use the latest WURFL with this patch, otherwise you risk
> breaking the fallback tree for these devices.
>
> Thanks,
>
> Steve Kamerman
>
> On Wed, Dec 23, 2009 at 1:24 PM, Luca Passani <luca.passani@...> wrote:
>> Steve Kamerman wrote:
>>>
>>> Luca, are you tracking this conversation? �How do you intend to handle
>>> BOLT UA Matching? �Are you intending to add BOLT to the WURFL in the
>>> near future? �Does anyone know what browser BOLT is a descendant of
>>> (i.e. WebKit or another WURFL ID)?
>>
>> Yes, I am tracking. I'll catch up with this (and other stuff) during
>> Xmas and decide how to best handle this. It is totally possible that a
>> blackberry_webkit family (or something along this line) is introduced.
>>
>> Thanks
>>
>> Luca
>>
>>
>>
>> ------------------------------------
>>
>> As of July 14 2005, it's much easier to be banned from WMLProgramming!
>> Please fail to read http://tech.groups.yahoo.com/group/wmlprogramming/ before
you post.Yahoo! Groups Links
>>
>>
>>
>>
>

#31829

From: Steve Kamerman <stevekamerman@...>
Date: Wed Dec 23, 2009 10:50 am
Subject: Re: Best Detection Method

steve_bond_k...

Offline

I did notice there are two BOLT browser UAs in the WURFL.  The reason
they are not being detected in Chris' case is that his UA has "BOLT"
in the UA, not "Bolt" as it exists in the WURFL.  Here's what I'm
using to remedy the problem for now:

Patch File: bolt.xml
--------------------------------------
<?xml version="1.0" encoding="utf-8"?>
<wurfl_patch>
	 <devices>
		 <device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
BOLT/1.601) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt0"/>
		 <device user_agent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
BOLT/1.600) AppleWebKit/530+ (KHTML, like Gecko) Version/4.0
Safari/530.17" fall_back="bolt_browser" id="terawurfl_bolt1"/>
		 <device user_agent="Mozilla/5.0 (X11; 78; CentOS; US-en)
AppleWebKit/527+ (KHTML, like Gecko) Bolt/0.920 Version/3.0
Safari/523.15" fall_back="bolt_browser" id="terawurfl_bolt2"/>
	 </devices>
</wurfl_patch>
-------------------------------------

Chris, while we are working on getting these into the WURFL, you can
copy the text above and paste it into a file called "bolt.xml" in the
data directory, then edit TeraWurflConfig.php and add ";bolt.xml" to
the end of the $PATCH_FILE directive, then reload the WURFL.  Here's
how it should look in the config:

	 /**
	  * Filename of patch file.  If you want to use more than one,
seperate them with semicolons.  They are loaded in order.
	  * ex: $PATCH_FILE = 'web_browsers_patch.xml;custom_patch_ver2.3.xml';
	  * @var String
	  */
	 public static $PATCH_FILE =
'custom_web_patch.xml;web_browsers_patch.xml;bolt.xml';

You must use the latest WURFL with this patch, otherwise you risk
breaking the fallback tree for these devices.

Thanks,

Steve Kamerman

On Wed, Dec 23, 2009 at 1:24 PM, Luca Passani <luca.passani@...> wrote:
> Steve Kamerman wrote:
>>
>> Luca, are you tracking this conversation? �How do you intend to handle
>> BOLT UA Matching? �Are you intending to add BOLT to the WURFL in the
>> near future? �Does anyone know what browser BOLT is a descendant of
>> (i.e. WebKit or another WURFL ID)?
>
> Yes, I am tracking. I'll catch up with this (and other stuff) during
> Xmas and decide how to best handle this. It is totally possible that a
> blackberry_webkit family (or something along this line) is introduced.
>
> Thanks
>
> Luca
>
>
>
> ------------------------------------
>
> As of July 14 2005, it's much easier to be banned from WMLProgramming!
> Please fail to read http://tech.groups.yahoo.com/group/wmlprogramming/ before
you post.Yahoo! Groups Links
>
>
>
>

#31828

From: Luca Passani <luca.passani@...>
Date: Wed Dec 23, 2009 10:24 am
Subject: Re: Best Detection Method

luca_passani

Offline

Steve Kamerman wrote:
>
> Luca, are you tracking this conversation?  How do you intend to handle
> BOLT UA Matching?  Are you intending to add BOLT to the WURFL in the
> near future?  Does anyone know what browser BOLT is a descendant of
> (i.e. WebKit or another WURFL ID)?

Yes, I am tracking. I'll catch up with this (and other stuff) during
Xmas and decide how to best handle this. It is totally possible that a
blackberry_webkit family (or something along this line) is introduced.

Thanks

Luca

#31827

From: Luca Passani <luca.passani@...>
Date: Wed Dec 23, 2009 10:18 am
Subject: Re: Wurfl Detection Errors

luca_passani

Offline

Do you have UAProf for these devices?

Luca

MDAguete wrote:
>
> Hello,
>
> I've found three errors in wurflpro with the following user agents
> collected from my web server logs
>
> * LG-GT500 -> In wurfl.xml the user agent is like LG/GT500
> "LG-GT500/v10a Browser/Teleca-Q7.1 MMS/LG-MMS-V1.0/1.2
> MediaPlayer/LGPlayer/1.0 Java/ASVM/1.1 Profile/MIDP-2.1
> Configuration/CLDC-1.1"
>
> * Samsung I8910 -> Not matches in wurflpro an device is
> samsung_i8910_ver1
>
> Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0
> Samsung/I8910-ORANGE/BVAMNIF1; Profile/MIDP-2.1 Configuration/CLDC-1.1
> ) AppleWebKit/525 (KHTML, like Gecko) Version/3.0 Safari/525
>
> * LG-KF700 -> Not matches in wurflpro an device is lg_kf700_ver1
>
> LG-KF700 Browser/Obigo-Q05A/3.12 MMS/LG-MMS-V1.0/1.2 Java/ASVM/1.1
> Profile/MIDP-2.1 Configuration/CLDC-1.1
>
> Regards.
>
>

Reply | Forward | Messages in this Topic (2)

#31826

From: "MDAguete" <manuel@...>
Date: Wed Dec 23, 2009 10:12 am
Subject: Wurfl Detection Errors

MDAguete

Offline

Hello,

I've found three errors in wurflpro with the following user agents collected
from my web server logs

* LG-GT500 -> In wurfl.xml the user agent is like LG/GT500
"LG-GT500/v10a Browser/Teleca-Q7.1 MMS/LG-MMS-V1.0/1.2 MediaPlayer/LGPlayer/1.0
Java/ASVM/1.1 Profile/MIDP-2.1 Configuration/CLDC-1.1"

* Samsung I8910  -> Not matches in wurflpro an device is   samsung_i8910_ver1

Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0 Samsung/I8910-ORANGE/BVAMNIF1;
Profile/MIDP-2.1 Configuration/CLDC-1.1 ) AppleWebKit/525 (KHTML, like Gecko)
Version/3.0 Safari/525


* LG-KF700 -> Not matches in wurflpro an device is lg_kf700_ver1

LG-KF700 Browser/Obigo-Q05A/3.12 MMS/LG-MMS-V1.0/1.2 Java/ASVM/1.1
Profile/MIDP-2.1 Configuration/CLDC-1.1


Regards.

Reply | Forward | Messages in this Topic (2)

#31825

From: Steve Kamerman <stevekamerman@...>
Date: Wed Dec 23, 2009 9:09 am
Subject: Re: Best Detection Method

steve_bond_k...

Offline

Chris, the BOLT browser is not currently in the WURFL and its
"desktop" user agent is loosely matched to an existing desktop user
agent. �Can you please go to�http://www.tera-wurfl.com/m/�on your
device and post the Request Headers for me? �Also, please change the
browser to mobile mode and paste those headers.

Here's some info from BOLT tech support: "You can view the original
user agent string for your mobile using BOLT by switching to mobile
layout. To enable mobile layout, select "Preferences" from the "Menu",
then select the "Mobile Layout" check box."

I think I will handle this like Opera Mini, with it's own user agent matcher.

Luca, are you tracking this conversation?  How do you intend to handle
BOLT UA Matching?  Are you intending to add BOLT to the WURFL in the
near future?  Does anyone know what browser BOLT is a descendant of
(i.e. WebKit or another WURFL ID)?

Thanks,

Steve Kamerman

On Wed, Dec 23, 2009 at 2:35 AM, Christopher Haslage
<chaslage@...> wrote:
>
>
>
> Yes, it's the newest from a few days ago. And indeed Tera WURFL.
>
>
> HTTP Agent: �Mozilla/5.0 ( Windows; U; Windows NT 5.1 ; en- US; BOLT/1.601)
�AppleWebKit/530+ ( KHTML, like Gecko) �Version/4.0 �Safari/ 530.17
>
> Here is what it reports back:
>
> Is it From .MOBI: �No (com)
> Is it Wireless: �No �(Redirect to normal WWW)
> Phone Type: �- �()
> Preferred Markup: �html_wi_oma_xhtmlmp_1 _0
> Dual Orientation: �No
> XHTML Support Level: �1
> Telephone Code: �tel:
> WAP Support: �No
> Resolution: �90 �x 40
>
> Thanks,
>
> Chris
>
> Steve Kamerman wrote:
>
>
> Chris, if the Bolt is showing up as a non-wireless device, then there is
something wrong. �Have you updated your WURFL lately? �A new release came out
recently. �The reason the screen size is wrong is because the device wasn't
properly detected. �Do you have the User-Agent for this device? �Can you please
go to your web admin page and go to the WURFL Test Script and paste the user
agent in, then send us the output? (I'm assuming you're using Tera-WURFL)
> Thanks,
> Steve Kamerman
>
> On Tue, Dec 22, 2009 at 11:14 PM, Christopher Haslage
<chaslage@...> wrote:
>>
>>
>>
>> All,
>>
>> So I am slowly working on detection methods, I even made it so anyone coming
from a .mobi site will always get a mobile page.� Though now I get to WURFL and
depending on detection methods for a regular website.
>>
>> So, of course there is easy indicator, namely:� is_wireless_device
>>
>> However, I noticed in Bolt for the Blackberry the detection shows as not
wireless but the screen size is 90x40.� Which raises my next question...
>>
>> Should I also look at screen size as a good indicator of content as well?� I
know that Bolt doesn't show the right information, because the Blackberry is
320x240 and it's definitely on a wireless device.� Anything below 640x480 maybe?
>>
>> I guess I am asking what would be the best mix of checks?
>>
>> Thanks!
>>
>>
>> Chris Haslage
>> Programmer
>> Delta Media Group
>> tel: 1.866.233.9833 x205
>> web: http://www.deltagroup.com/
>
> --
> Chris Haslage
> Programmer
> Delta Media Group
> tel: 1.866.233.9833 x205
> web: http://www.deltagroup.com/
>
>

#31824

From: Paul McMahon <paul@...>
Date: Wed Dec 23, 2009 6:35 am
Subject: Re: Re: Japan phones

paulmc3

Offline

Thanks Luca, I'm really excited to see that this data will be
integrated into the base WURFL.  Once this data is in the base WURFL,
we'll just add the new handsets to the WURFL directly.

Unfortunately, audio & video isn't an area that I have so much
experience in, and at first glance at the documentation, I find it
pretty confusing.  However, based on a cursory look I've enhanced our
patch with some sane values for the base fallbacks.

2009/12/21 Luca Passani <luca.passani@...>:
>
> Thank you, Paul.
>
> WURFL has not been as popular in Japan as it has been in GSM and US, so
> device information that people working in Japan can collect is likely to
> be more accurate that what one could find in WURFL (at least up until
> after I will integrate your patch).
> In short, I will go for your classification and we can figure out the
> browser right after that.
>
> Question for you: are there some general capabilities about audio/video
> playback/streaming formats for Japanese phones? 3GPP? Real? RTSP?....
>
> Arigato
>
> Luca
>
> Paul McMahon wrote:
>>
>> Hi Luca,
>>
>> 2009/12/20 Luca Passani <luca.passani@...
>> <mailto:luca.passani%40gmail.com>>:
>> > 1) which of those top-20 devices is missing?
>>
>> As far as I'm aware, none of them are missing, as we are generating
>> this patch from the carrier provided data, though I have not verified
>> this by hand. As mentioned in my previous post, the device Erik
>> thought was missing is included in the patch.
>>
>> > 2) do you guys agree with the families of Japanese phones identified in
>> > the patch? kddi_wap20_generic and docomo_2_0_browser_ver1 for example...
>> >
>> > should I adopt those in WURFL or stick to the current breakdown?
>>
>> docomo_2_0_browser_ver1 should definitely be included in the patch.
>> New docomo phones, starting from summer of this year, include the
>> "i-mode browser 2.0", which has some notable capabilities over the
>> previous i-mode browser, including cookie, css, and limited javascript
>> support. As such, it is a different beast then previous docomo
>> phones, and should be treated as such.
>>
>> kddi_wap20_generic is a trickier case, as the base wurfl has it
>> falling back on the open wave browser, which is not incorrect,
>> however, it also doesn't capture that it is a kddi device, which is
>> more important to developers of Japanese mobile sites. I guess the
>> wurfl tries to only focus on device specific, not carrier specific
>> information, so from that position perhaps the base wurfl is more
>> correct, however, in Japan there is no separation between device and
>> carrier, as devices are designed for specific carriers, and as such
>> function similarly. However, the real reason why I chose to do it
>> this way in the patch was that the carrier information from which this
>> patch was generated did not include the open wave browser version, so
>> I could not determine which version of the open wave to fallback on.
>> I suppose I could have checked what was the value in the base wurfl,
>> if present, and used that instead, and had all the other devices fall
>> back on some version of the open wave browser. If anyone is up for
>> making that modification, the source for our generator for the patch
>> is available at http://github.com/pwim/jawurflex.
>> <http://github.com/pwim/jawurflex.>
>>
>>
>
>
>
> ------------------------------------
>
> As of July 14 2005, it's much easier to be banned from WMLProgramming!
> Please fail to read http://tech.groups.yahoo.com/group/wmlprogramming/ before
you post.Yahoo! Groups Links
>
>
>
>

#31823

From: Christopher Haslage <chaslage@...>
Date: Tue Dec 22, 2009 11:35 pm
Subject: Re: Best Detection Method

chaslage...

Offline

Yes, it's the newest from a few days ago. And indeed Tera WURFL.

HTTP Agent: �Mozilla/5.0 ( Windows; U; Windows NT 5.1 ; en- US; BOLT/1.601) �AppleWebKit/530+ ( KHTML, like Gecko) �Version/4.0 �Safari/ 530.17

Here is what it reports back:

Is it From .MOBI: �No (com)
Is it Wireless: �No �(Redirect to normal WWW)
Phone Type: �- �()
Preferred Markup: �html_wi_oma_xhtmlmp_1 _0
Dual Orientation: �No
XHTML Support Level: �1
Telephone Code: �tel:
WAP Support: �No
Resolution: �90 �x 40

Thanks,

Chris

Steve Kamerman wrote:

�

Chris, if the Bolt is showing up as a non-wireless device, then there is something wrong. �Have you updated your WURFL lately? �A new release came out recently. �The reason the screen size is wrong is because the device wasn't properly detected. �Do you have the User-Agent for this device? �Can you please go to your web admin page and go to the WURFL Test Script and paste the user agent in, then send us the output? (I'm assuming you're using Tera-WURFL)

Thanks,

Steve Kamerman

On Tue, Dec 22, 2009 at 11:14 PM, Christopher Haslage <chaslage@deltagroup.com> wrote:

�

All,

So I am slowly working on detection methods, I even made it so anyone coming from a .mobi site will always get a mobile page.� Though now I get to WURFL and depending on detection methods for a regular website.

So, of course there is easy indicator, namely:� is_wireless_device

However, I noticed in Bolt for the Blackberry the detection shows as not wireless but the screen size is 90x40.� Which raises my next question...

Should I also look at screen size as a good indicator of content as well?� I know that Bolt doesn't show the right information, because the Blackberry is 320x240 and it's definitely on a wireless device.� Anything below 640x480 maybe?

I guess I am asking what would be the best mix of checks?

Thanks!

�
Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

--
Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

#31822

From: Steve Kamerman <stevekamerman@...>
Date: Tue Dec 22, 2009 9:56 pm
Subject: Re: Best Detection Method

steve_bond_k...

Offline

Chris, if the Bolt is showing up as a non-wireless device, then there is something wrong. �Have you updated your WURFL lately? �A new release came out recently. �The reason the screen size is wrong is because the device wasn't properly detected. �Do you have the User-Agent for this device? �Can you please go to your web admin page and go to the WURFL Test Script and paste the user agent in, then send us the output? (I'm assuming you're using Tera-WURFL)

Thanks,

Steve Kamerman

On Tue, Dec 22, 2009 at 11:14 PM, Christopher Haslage <chaslage@...> wrote:

�

All,

So I am slowly working on detection methods, I even made it so anyone coming from a .mobi site will always get a mobile page.� Though now I get to WURFL and depending on detection methods for a regular website.

So, of course there is easy indicator, namely:� is_wireless_device

However, I noticed in Bolt for the Blackberry the detection shows as not wireless but the screen size is 90x40.� Which raises my next question...

Should I also look at screen size as a good indicator of content as well?� I know that Bolt doesn't show the right information, because the Blackberry is 320x240 and it's definitely on a wireless device.� Anything below 640x480 maybe?

I guess I am asking what would be the best mix of checks?

Thanks!

�
Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

#31821

From: Christopher Haslage <chaslage@...>
Date: Tue Dec 22, 2009 8:14 pm
Subject: Best Detection Method

chaslage...

Offline

All,

So I am slowly working on detection methods, I even made it so anyone coming from a .mobi site will always get a mobile page. Though now I get to WURFL and depending on detection methods for a regular website.

So, of course there is easy indicator, namely: is_wireless_device

However, I noticed in Bolt for the Blackberry the detection shows as not wireless but the screen size is 90x40. Which raises my next question...

Should I also look at screen size as a good indicator of content as well? I know that Bolt doesn't show the right information, because the Blackberry is 320x240 and it's definitely on a wireless device. Anything below 640x480 maybe?

I guess I am asking what would be the best mix of checks?

Thanks!

Chris Haslage
Programmer
Delta Media Group
tel: 1.866.233.9833 x205
web: http://www.deltagroup.com/

#31820

From: Marc Guay <marc.guay@...>
Date: Tue Dec 22, 2009 6:45 pm
Subject: Re: Desktop browser detection

mguay11

Offline

Thanks for the help guys.  I wonder if Microsoft is currently trying
to researching a way to make walking down the street and breathing
more complicated?

Fistfuls of hair,
Marc

#31819

From: Steve Kamerman <stevekamerman@...>
Date: Tue Dec 22, 2009 6:13 pm
Subject: Re: Desktop browser detection

steve_bond_k...

Offline

Marc, this IE detection problem never ceases to annoy me! �I've compiled a list of the first 55 characters of IE user agent strings. �Please goto your web administration console and click on Generate Patch File, then paste these strings in there (if you already have the Firefox ones then delete them first), then press Generate Patch File and reload your WURFL file.

Mozilla/5.0 (Windows; U; Firefox Mozilla/6.0 (Windows; U; Firefox Mozilla/5.0 (X11; U; Linux i686; Firefox Mozilla/6.0 (X11; U; Linux i686; Firefox Mozilla/5.0 (X11; U; Linux x86_64; Firefox Mozilla/6.0 (X11; U; Linux x86_64; Firefox Mozilla/5.0 (Macintosh; U; Firefox Mozilla/6.0 (Macintosh; U; Firefox Mozilla/5.0 (X11; U; FreeBSD Firefox Mozilla/6.0 (X11; U; FreeBSD Firefox Mozilla/4.0 (compatible; MSIE 4.01; Digital AlphaServer Mozilla/4.0 (compatible; MSIE 4.0; Windows NT; ....../1 Mozilla/4.0 (compatible; MSIE 6.0; America Online Brows Mozilla/4.0 (compatible; MSIE 6.0; AOL 8.0; Windows NT Mozilla/4.0 (compatible; MSIE 6.0; AOL 9.0; Windows NT Mozilla/4.0 (compatible; MSIE 6.0; CS 2000 6.0; Netscap Mozilla/4.0 (compatible; MSIE 6.0; CS 2000 6.0; NS Conn Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0) Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0) Blac Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0; .NET Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0; Goog Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1) Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; HbTo Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; MyIE Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1) Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; (R1 Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; SV1; Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; Win6 Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; WOW6 Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 6.0; ru) Mozilla/4.0 (compatible; MSIE 6.0; Windows NT) ::ELNSB5 Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1) Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; FunW Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; Mozi Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; sbcy Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; {CEF Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.2; .NET Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.0; SLCC Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trid Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; WOW6 Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0; SLCC Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0; Trid Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0; Win6 Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0; WOW6 Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; Trid

I really hate to do it like this, but without adding even more MSIE logic to the UserAgentMatcher it is difficult to match some of them.

By the way, this is the culprit that was messing up the detection of your user agent:

WURFL ID: �htc_st6356_ver1_submsie

User Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; Windows Phone 6.5)

It was being matched to "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; " which is an acceptable match. �Why oh why do they have to make stupid user agents like that?

Thanks,

Steve Kamerman

On Tue, Dec 22, 2009 at 6:55 PM, Marc Guay <marc.guay@...> wrote:

�

Hi everyone,

I have a user agent, coming from IETester's IE6: "Mozilla/4.0
(compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR
3.0.4506.2152; .NET CLR 3.5.30729)", which would seem to match the
web_browsers_patch.xml definition of IE6: "Mozilla/4.0 (compatible;
MSIE 6.0;", bt it's still being detected as a wireless device. Any
ideas why that would be happening?

Using Tera-WURFL 2.0.
Marc

#31818

From: "manmountainmorgan" <manmountainmorgan@...>
Date: Tue Dec 22, 2009 4:16 pm
Subject: Re: Desktop browser detection

manmountainm...

Offline

Hi,

I've also had numerous problems with web browser detection especially with
IE6.0, which unfortunately refuses to die quietly.

The problem seems to be (after hours of stepping through java code) where the
MSIE matcher grabs a set of filtered devices. Being IE 6.0 it's string used
frequently in many winmob devices and so the matcher ends up matching what it
thinks is the best match, but is in fact not. Browsers are a lot harder to tie
down as every .NET version or toolbar seem to want to change the UA string.

I got round it by altering Handler.java in the wurfl package to have an updated
mobileBrowsers string:

private static final String[] mobileBrowsers = new String[]{"Tablet", "Mobile",
"Wireless", "Palm", "Blazer", "NetFront", "Symbian", "Bolt", "Iris", "Maemo",
"CLDC", "BREW", "MIDP", "PocketPC", "Windows CE"};

This cuts down the number of obviously incorrect matches considerably.
inaddition, you should also add some entries to the patch file with OS version
in the string such as:

<device user_agent="Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1)"
fall_back="msie_6" id="msie_6_xp"/>

This gives a better base string to match against, without the .NET stuff. So far
this is working for me.

Hope this helps you out.

morgan

--- In wmlprogramming@yahoogroups.com, Marc Guay <marc.guay@...> wrote:
>
> Hi everyone,
>
> I have a user agent, coming from IETester's IE6: "Mozilla/4.0
> (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR
> 3.0.4506.2152; .NET CLR 3.5.30729)", which would seem to match the
> web_browsers_patch.xml definition of IE6: "Mozilla/4.0 (compatible;
> MSIE 6.0;", bt it's still being detected as a wireless device.  Any
> ideas why that would be happening?
>
> Using Tera-WURFL 2.0.
> Marc
>

#31817

From: Marc Guay <marc.guay@...>
Date: Tue Dec 22, 2009 3:55 pm
Subject: Desktop browser detection

mguay11

Offline

Hi everyone,

I have a user agent, coming from IETester's IE6: "Mozilla/4.0
(compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR
3.0.4506.2152; .NET CLR 3.5.30729)", which would seem to match the
web_browsers_patch.xml definition of IE6: "Mozilla/4.0 (compatible;
MSIE 6.0;", bt it's still being detected as a wireless device.  Any
ideas why that would be happening?

Using Tera-WURFL 2.0.
Marc

#31816

From: "igorlemasson" <le_masson@...>
Date: Mon Dec 21, 2009 8:56 am
Subject: Re: HTC Imagio UA & UAProf

igorlemasson

Offline

Hi Nicola

These information were added in the last wurfl file (December 15, 2009).
Look for the device id "htc_xv6875_ver1".

BR,
Igor

--- In wmlprogramming@yahoogroups.com, "nicolacramarossa" <nicolacramarossa@...>
wrote:
>
> Hi everyone,
> i need to add a device to the DB, does anyone have the Ua and UAProf of the
HTC Imagio XV6975?
>
> thank you
> Nicola
>

Reply | Forward | Messages in this Topic (2)

#31815

From: "erikh2000" <erikh2000@...>
Date: Tue Dec 22, 2009 1:14 am
Subject: Re: Japan phones

erikh2000

Offline

Paul, thanks for the correction.  I will tell you that the existence of the
Mobalean patch weighed heavily in my company's choice of using WURFL.  So cheers
to Mobalean!

Posting my researched list of popular Japanese phones in case anyone else is
interested in the phones I checked.

From intomobile.com article bestselling mobile phones in Japan for July, 2009 (
http://www.intomobile.com/2009/08/18/iphone-3gs-32gb-is-the-no-1-best-selling-ha\
ndset-in-japan-for-the-month-of-july.html ):

   1. Apple iPhone 3GS 32GB (Softbank (OTCPK: SFTBF)) - Main
   2. Sharp SH-06A (Docomo) - Mobalean
   3. Panasonic (NYSE: PC) 830P (Softbank) - Mobalean
   4. Sharp SH-05A (Docomo) - Mobalean
   5. NEC N-08A (Docomo) - Mobalean
   6. Sony-Ericsson Premier3 (au) - Mobalean
   7. Sharp SH-02A (Docomo) - Mobalean
   8. Casio W63CA (au) - Mobalean
   9. Apple iPhone 3GS 16 GB (Softbank) - Main
  10. Sharp SH001 (au) - Mobalean

From wirelesswatch.jp article bestselling mobile phones in Japan for 2008: (
http://wirelesswatch.jp/2009/05/14/top-10-japanese-mobile-phones/ )

   1. P906i (Docomo) - Mobalean
   2. Wooo k tie W53H (au) - Mobalean
   3. P905i (Docomo) - Mobalean
   4. W61CA (au) - Mobalean
   5. AQUOS k tie W61SH (au) - Mobalean
   6. N906i&#956; (Docomo) - Mobalean
   7. SH906i (Docomo) - Mobalean
   8. SH905i (Docomo) - Mobalean
   9. 812SH (Softbank) - Mobalean
  10. Very easily phone IV (Docomo)- Mobalean

#31814

From: "erikh2000" <erikh2000@...>
Date: Tue Dec 22, 2009 1:25 am
Subject: Re: Japan phones

erikh2000

Offline

That is wonderful news, Luca!  My company is developing a high-traffic website
for a Japanese technology company.  We will be testing with the major carriers
in Japan, over a variety of devices, and any profile corrections we come across
can be contributed back in to the database

-Erik

--- In wmlprogramming@yahoogroups.com, Luca Passani <luca.passani@...> wrote:
> I have decided to include this patch in WURFL.

#31813

From: Luca Passani <luca.passani@...>
Date: Mon Dec 21, 2009 9:04 pm
Subject: Re: Re: Japan phones

luca_passani

Offline

Just so you know, the Mobalean guys use their own Ruby API which seems to be much less strict than the regular Java and PHP API you can download from the WURFL site (the one developed by WURFL-Pro). In particular, the Ruby API lets the patch override UA string and fall-back, which is something I never allowed in the official APIs because of a few good reasons.

In practice, this means that you probably won't be able to apply the patch as-is. Anyway, give us a few days, and you'll be able to get the same info out of regular WURFL

Cheers

Luca

On Mon, Dec 21, 2009 at 3:22 PM, Cristiano Diniz da Silva <cristiano@...> wrote:

�

Hi guys

My initial problem is that I don't have many experience with Japanese phones and carries and now I'm getting a lot of traffic from them, so supporting the phones for me is the least minimal.

Luca decision of supporting a patch for the Japanese phones in WURFL could only make me happier. This will truly make anyone that deals with Japanese phones easier.

As I have mentioned before, I have very little experience with Japanese phones, and I will tag along you guys on whatever decision is made, but here is one thing that might be helpful. One of the things that brought very good response from the Japan users was that on the application I'm using an emoji map for some modules on the homepage and specifically for the footer, the emojis are associated with the keyboard function (for instance emoji to show 1 is, by phone, associated with keypad 1, and that enables the link). Maybe supporting a basic emojis list for the generic devices could be a good thing too.

Anyway, great news about the patch for the Japanese phones.

Note: I was about to use this one: http://blog. mobalean. com/2009/ 06/10/mobalean- releases- japanese- wurfl-patch- ruby-libraries

Before knowing that WURFL will also have a patch.

Regards

�
Cristiano Diniz da Silva
Web Development & Solutions
http://www.mcloide.com
http://mcloide.wordpress.com
p. + 1 954 840 4010
c. + 1 561 809 1928
Coral Springs, Fl
PHP5 Zend Certified Engineer
Before printing, just keep in mind your commitment with the environment!

From: Paul McMahon <paul@...>
To: wmlprogramming@yahoogroups.com
Sent: Sun, December 20, 2009 9:57:57 PM
Subject: Re: [wmlprogramming] Re: Japan phones

�

Hi Luca,

2009/12/20 Luca Passani <luca.passani@ gmail.com>:
> 1) which of those top-20 devices is missing?

As far as I'm aware, none of them are missing, as we are generating
this patch from the carrier provided data, though I have not verified
this by hand. As mentioned in my previous post, the device Erik
thought was missing is included in the patch.

> 2) do you guys agree with the families of Japanese phones identified in
> the patch? kddi_wap20_generic and docomo_2_0_browser_ ver1 for example...
>
> should I adopt those in WURFL or stick to the current breakdown?

docomo_2_0_browser_ ver1 should definitely be included in the patch.
New docomo phones, starting from summer of this year, include the
"i-mode browser 2.0", which has some notable capabilities over the
previous i-mode browser, including cookie, css, and limited javascript
support. As such, it is a different beast then previous docomo
phones, and should be treated as such.

kddi_wap20_generic is a trickier case, as the base wurfl has it
falling back on the open wave browser, which is not incorrect,
however, it also doesn't capture that it is a kddi device, which is
more important to developers of Japanese mobile sites. I guess the
wurfl tries to only focus on device specific, not carrier specific
information, so from that position perhaps the base wurfl is more
correct, however, in Japan there is no separation between device and
carrier, as devices are designed for specific carriers, and as such
function similarly. However, the real reason why I chose to do it
this way in the patch was that the carrier information from which this
patch was generated did not include the open wave browser version, so
I could not determine which version of the open wave to fallback on.
I suppose I could have checked what was the value in the base wurfl,
if present, and used that instead, and had all the other devices fall
back on some version of the open wave browser. If anyone is up for
making that modification, the source for our generator for the patch
is available at http://github. com/pwim/ jawurflex.

#31812

From: "copluk@..." <copluk@...>
Date: Mon Dec 21, 2009 7:46 pm
Subject: Re: UTF-8 Bom

mobivid

Offline

----- Original message -----
> Hi everyone! I’m pretty new to mobile development and I’m hoping I had smilar problem with pocket ie. Css file had to be ansi or utf8 in order to get style applied

someone on
> this list can shed some light on an issue I’ve been experiencing on some devices
> and can’t for the life of me figure out what exactly the problem is.
>
> The issue in question is specifically popping up on the LG vx8600 & vx8700, but
> may be more. I have a wap page that upon viewing on the device the first
> character on the page is the i with two dots, all contents after that is
> present, except any styling that should be applied isn’t. At first I thought
> this was a character encoding issue as the i with two dots is the first symbol
> in the utf-8 bom mark. However after further investigation, I’m now not so sure
> this is the case & was wondering if anyone has run into similar issues before?
>
> Further Background Info:
>
> - The page in question is an Aspx page written in Visual Studio 2008
>
> Tests I have Run:
>
> - Created 2 stripped down test pages, one w/ the BOM, & one without…
> the device rendered both w/ styling perfectly.
>
> o Which leads me to feel this is more of a Visual Studio &/or IIS issue
>
> - Ensured the server is encoding the files as UTF-8, receiving requests
> as UTF-8, and sending responses as UTF-8
>
> - Ensured the page displays properly on other devices
>
> So has anyone on this list experienced something similar or have any advice on a
> new test to try and pinpoint where the error is occurring? Any tips/suggestions
> would be appreciated.
>
> Thanks,
> Nick

#31811

From: Nick Hilem <nhilem@...>
Date: Mon Dec 21, 2009 6:20 pm
Subject: UTF-8 Bom

nickhilem

Offline

Hi everyone! I’m pretty new to mobile development and I’m hoping someone on this list can shed some light on an issue I’ve been experiencing on some devices and can’t for the life of me figure out what exactly the problem is.

The issue in question is specifically popping up on the LG vx8600 & vx8700, but may be more. I have a wap page that upon viewing on the device the first character on the page is the i with two dots, all contents after that is present, except any styling that should be applied isn’t. At first I thought this was a character encoding issue as the i with two dots is the first symbol in the utf-8 bom mark. However after further investigation, I’m now not so sure this is the case & was wondering if anyone has run into similar issues before?

Further Background Info:

- The page in question is an Aspx page written in Visual Studio 2008

Tests I have Run:

- Created 2 stripped down test pages, one w/ the BOM, & one without… the device rendered both w/ styling perfectly.

o Which leads me to feel this is more of a Visual Studio &/or IIS issue

- Ensured the server is encoding the files as UTF-8, receiving requests as UTF-8, and sending responses as UTF-8

- Ensured the page displays properly on other devices

So has anyone on this list experienced something similar or have any advice on a new test to try and pinpoint where the error is occurring? Any tips/suggestions would be appreciated.

Thanks,
Nick