Re: Proposal of @ua

On Nov 24, 2007, at 4:21 AM, David Woolley wrote:

> Authots who want to write for non-IE browsers are in the minority.

That was true once, but I see little evidence of it today. I use  
Safari for virtually everything, and it is mostly rare to come across  
some site that doesn't work in it. If it does have a problem in  
Safari then FireFox works almost always. It is very rare for me to  
come across an IE-only site these days in general use. Things written  
for corporations are still quite often IE6-only (or do work, but look  
like crap in other browsers), but they are the exception to the rule.  
Even very IE-centric designers will usually at least preview their  
pages in FireFox.

In fact, in most of the blogs and such where people are writing about  
Web design and CSS, IE is considered a necessary evil to be dealt  
with, and often accounts for a lot of extra wasted time when trying  
to turn nice, working designs into something that IE can handle. For  
instance, take a look at this page, and read some of the comments:

http://www.sitepoint.com/blogs/2007/11/22/in-all-fairness-…-internet- 
explorer-still-stinks/


> They may be the most likely ones to use @ua, although I think some  
> of the majority might consider providing a very bland style sheet  
> for anything other than the latest two versions of IE.
>
>>> How do determine which browsers are the "major browsers" anyway?
>>> Can you trust the server's agent log stats given that most  
>>> browsers are
>>> compelled to lie about what they are?
>> I trust them well enough. I use what information I have available  
>> to me.  I look at my Google Analytics, which I believe does its  
>> best to
>
> I don't know how prevalent user lying is in this respect, but the  
> reason that non-IE browsers have configurable user agent strings is  
> to allow users to get round white listing.  At least for Lynx, it  
> used to be a common question as to what was the best value to put  
> in the user agent string to make sites believe you were IE.  There  
> was a feeling that, whilst it would have been dishonest, or even  
> illegal, for the out of the box version to pretend to be IE, it was  
> perfectly reasonable for end users to set it to exactly match IE.   
> You can probably assume that Lynx is seriously underestimated in  
> browser statistics, even if the true number is also small.

The world has changed since then. The vast majority of the people  
using sites such as an online banking site are non-technical, unlike  
in the early days. A lot of them don't even know what an "Internet  
browser" is (they just "launch the internet" from their Start menu),  
have never heard of Opera, don't know what an operating system is (or  
what version of Windows they are using, or how Apple computers work  
differently), and would never dream of changing their options or  
preferences, much less disguise their UA string.

But even if you are correct, it is like the Nielson's ratings: in the  
absence of more accurate information, you use what you have  
available. We also have information from our Call Center that implies  
that these are the prevalent browsers.