User-Agent Detection

General question about two step analysis, matchers and other general API questions
sm_support2
Posts: 294
Joined: Mon Jun 20, 2011 5:04 pm

User-Agent Detection

Postby sm_support2 » Fri Jun 24, 2011 2:03 pm

In a few examples, I have noticed the same device sending different user agents which appear quite similar but are mapping to different devices using the WURFL API:

[1]
NokiaX2-00/5.0 (p) Profile/MIDP-2.1 Configuration/CLDC-1.1 Mozilla/5.0 AppleWebKit/420 (KHTML, like Gecko) Safari/420
--> Maps to "nokia_x2_00_ver1"
NokiaX2-01/5.0 (07.10) Profile/MIDP-2.1 Configuration/CLDC-1.1 Mozilla/5.0 AppleWebKit/420+ (KHTML, like Gecko) Safari/420+
--> Maps to "generic_web_browser"

[2]
SonyEricssonU8a/R2EA; Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0 Profile/MIDP-2.1 Configuration/CLDC-1.1) AppleWebKit/525 (KHTML, like Gecko) Version/3.0 Safari/525
--> Maps to "nokia_generic_series60"

SonyEricssonU8i/R2AA; Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0 Profile/MIDP-2.1 Configuration/CLDC-1.1) AppleWebKit/525 (KHTML, like Gecko) Version/3.0 Safari/525
---> Maps to "sonyericsson_u8i_ver1"

Can anyone help me with understanding this behavior? thanks.

Regards,

sm_support2
Posts: 294
Joined: Mon Jun 20, 2011 5:04 pm

Re: User-Agent Detection

Postby sm_support2 » Fri Jun 24, 2011 2:04 pm

> In a few examples, I have noticed the same device sending different user
> agents which appear quite similar but are mapping to different devices using
> the WURFL API:


yes. It is not a bug. It's a feature. The value of the WURFL API is its attempt
to avoid false positives, i.e. recognize a device instead of another. Unless
WURFL is reasonably confident that a matched device is close enough (same make
and model, for example), then the API prefers to give up and return a generic
match. This way, adopters can easily detect devices which need to be added in
WURFL and avoid a false sense of security that everything is being matched
correctly.
This is the reason why two-step UA analysis was introduced 3 years ago.

>
> [1]
> NokiaX2-00/5.0 (p) Profile/MIDP-2.1 Configuration/CLDC-1.1 Mozilla/5.0
> AppleWebKit/420 (KHTML, like Gecko) Safari/420
> --> Maps to "nokia_x2_00_ver1"
> NokiaX2-01/5.0 (07.10) Profile/MIDP-2.1 Configuration/CLDC-1.1 Mozilla/5.0
> AppleWebKit/420+ (KHTML, like Gecko) Safari/420+
> --> Maps to "generic_web_browser"

This is correct. The Nokia matcher will use RIS with first slash index as the
threshold. For WURFL these are different devices and both should be added to
WURFL. If the devices are really the evolution one of the other, than letting 01
fall back into 00 is a good idea.


>
> [2]
> SonyEricssonU8a/R2EA; Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0
> Profile/MIDP-2.1 Configuration/CLDC-1.1) AppleWebKit/525 (KHTML, like Gecko)
> Version/3.0 Safari/525
> --> Maps to "nokia_generic_series60"
>
> SonyEricssonU8i/R2AA; Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0
> Profile/MIDP-2.1 Configuration/CLDC-1.1) AppleWebKit/525 (KHTML, like Gecko)
> Version/3.0 Safari/525
> ---> Maps to "sonyericsson_u8i_ver1"
>

same here


Who is online

Users browsing this forum: No registered users and 10 guests