308166 - Firefox 1.5 beta 1 reports wrong user agent string in http_user_agent

Status: RESOLVED DUPLICATE of bug 307808

(Firefox :: General, defect)

Description

[Dixson]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b4) Gecko/20050908 Firefox/1.4
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b4) Gecko/20050908 Firefox/1.4

The version information reported in the user agent string for Firefox 1.5 beta 1
is outdated.  For instance it reports:

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b4) Gecko/20050908 Firefox/1.4

when instead it should be something like

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b4) Gecko/20050908
Firefox/1.5b1

It is very important that the version information be correctly conveyed in the
user agent string, even for a beta release.  Thank you.

Reproducible: Always

Steps to Reproduce:
1. self explantory
2.
3.

Actual Results:  
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b4) Gecko/20050908 Firefox/1.4

Expected Results:  
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b4) Gecko/20050908
Firefox/1.5b1

IIRC over the past year or so I've frequently seen the user agent string report
an outdated version number in a beta or prerelease build. Fortunately I think it
was always corrected once the software was released. However it is important for
this version information to be updated even in beta builds.

Is there some sort of check in / QA procedure even for beta builds that can be
added/modified to check off as a step prior to an alpha/beta release that the
user agent version information has been correctly updated?  That would be great!
 Thanks all.

Comment 1

[Asaf Romano (gone)]

This is by design - we're using 1.4.x as the internal version number for the
beta releases (e.g. 1.4.1 for beta 2).

Comment 2

[Dixson]

(In reply to comment #1)
> This is by design - we're using 1.4.x as the internal version number for the
> beta releases (e.g. 1.4.1 for beta 2).

I can understand that reasoning. However this results in misleading information
about the browser version a visitor has.

Once an official beta is packaged and branded such as "1.5 beta 1" its only
appropriate that the user agent string reflect that this is indeed Firefox
1.5b1, as it is technically no longer 1.4.  Even though the internal code is
using 1.4.1, to the outside it should still reflect its real version.

I'm respectfuly requesting that this please be taken into consideration as
Firefox rolls out future alphas, betas and updated builds.

Thank you.

Comment 3

[Jo Hermans]

(In reply to comment #2)
> (In reply to comment #1)
> > This is by design - we're using 1.4.x as the internal version number for the
> > beta releases (e.g. 1.4.1 for beta 2).
> 
> I can understand that reasoning. However this results in misleading information
> about the browser version a visitor has.

That's the whole point. You are *not* running Firefox 1.5 (not released yet),
and it's not Firefox 1.0.* either.

Comment 4

[Dixson]

So if I WAS running 1.5 than it should say 1.5 right?

Then logically it only follows that if I am running 1.5 beta 1, than it should
say 1.5 beta 1.

Clearly this build is indeed an official 1.5 beta 1 and branded this way all
over the web, in the docs, and even in the Help About screen.

So, since I am running 1.5 beta 1, it is only appropriate that the user agent
report this as 1.5b1, not 1.4.

This is the standard practive that all other web vendors follow as well.  For
example IE 7.0 beta reports itself in the user agent string as "7.0b1", not as
"6.1" or "6.9".

Likewise when Opera puts out a beta, say Opera 8.0 beta 1, they use "8.0b1", not
"7.9" etc.

There should be procedures in place in the future to ensure that before an
official beta build is released that the usre agent string accurately reflect
the same information as the beta is publicly branded.  In this case it should
indeed say 1.5b1 and not 1.4. 

Comment 5

[Dixson]

Quick ipdate - I meant to say IE 7.0 beta is using "7.0b" not "7.0b1" but this
doesn't change my point.

Comment 6

[Kevin Brosnan]

*** This bug has been marked as a duplicate of 307808 ***