On 23/12/11 11:39, Dustbin wrote:
Edmund wrote:
Dustbin wrote:
It should not be necessary to spoof firefox.
If people wrote their code to work to TCP/IP standard protocols SM would
work.
Do you mean putting FireFox in the UA string? If so, that has (AFAIU)
nothing to do with the TCP/IP standard protocols, but the way the
website detects browsers and adapts the website to the user
depending on the UA.
Since (AFAIK) most sites cater to IE, Chrome and Firefox, they care
only about those browsers. Even though SeaMonkey also uses the same
engine as FireFox, our users don't get the same treatment as
FireFox users. If the websites detect the engine instead of the
browser string, it would make it simple and we can just use
SeaMonkey in the UA.
If I've erred.. any corrections appreciated.
It is not that you have erred it is that they should not be abusing the
system's facilities to engineer code for a limited number of browsers.
The code should work equally on all browsers. That is the point of
standards. Writing code with a preference for one or two browsers is wrong.
It is notable that Netscape/Mozilla/SeaMonkey has been around since the
dawn of HTTP and, therefore, is one of the earliest browsers; yet coders
are favouring far more recent browsers such as FireFox and Chrome.
[...]
You are right, of course: sites, and the people who write them, should
be sniffing for specific features rather than for specific versions of
specific browsers. We at SeaMonkey tried for years to educate them to do
just that — there is still an open meta bug about these very attempts,
at https://bugzilla.mozilla.org/show_bug.cgi?id=334967 — but it's a
losing battle. When Fennec and Camino added a Firefox version to their
UA strings, SeaMonkey followed suit, and between one day and the next
90% of all broken sites started working again with our (new) default UA
setting.
The advantage over overriding the UA string by means of the UA Switcher
(or by setting general.useragent.override) is that as we update from one
browser version to the next, the default UA string reflects the new
version: for instance, before this "Advertise compatibility" existed, I
had set a User Agent Switcher setting for the sites which required it;
it is still there in my UA Switcher choices, but of course it hasn't
been updated, and this is what it says:
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.6pre) Gecko/20091125
SeaMonkey/2.0.1pre, a sibling of Firefox/3.5
as compared with the default setting of my current SeaMonkey build, viz.
Mozilla/5.0 (X11; Linux x86_64; rv:12.0a1) Gecko/20111226 Firefox/12.0a1
SeaMonkey/2.9a1
As you can see, the latter faithfully reflects the characteristics of my
present build (and the "Firefox/12.0a1" part ought to be unnecessary, it
can be reconstructed from the presence of "Gecko/" and of "rv:12.0a1",
but it says to moronic sites which don't know any Gecko browser other
than Firefox that anything that Firefox 12.0a1 can browse, so can I),
while the former is hopelessly out of date, reflecting a build which is
not supported anymore, and a hardware platform which isn't anymore what
I'm using.
BTW, AFAIK there are no real "standards" about the UA string, any
standards about it are de-facto: compare with Konqueror:
Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/533.3 (KHTML, like Gecko)
konqueror/4.7.2 Safari/533.3
which BTW manages to "spoof" both Gecko and Safari, and maybe (I'm less
sure) Google Chrome; or with Lynx:
Lynx/2.8.7rel.2 libwww-FM/2.14 SSL-MM/1.4.1 OpenSSL/1.0.0e
which didn't even add the "Mozilla/?.? (" string at the start, which IE
added long ago in order to spoof Netscape, which was still, at that
time, the market leader.
Best regards,
Tony.
--
Join the march to save individuality!
_______________________________________________
support-seamonkey mailing list
[email protected]
https://lists.mozilla.org/listinfo/support-seamonkey
