IE 11 User-Agent String?

[AceInfinity]

And Internet Explorer, as though it wasn't already difficult enough to work with for web development, has thrown a few other wrinkles at web developers it seems. I was fooling around with some PHP, and noticed that my old user agent checker doesn't work for IE 11. Why? Because it's now designed to disguise itself almost fully, as Mozilla Firefox. No more "MSIE".

Mozilla/5.0 (Windows NT 6.3; Trident/7.0; rv:11.0) like Gecko

User-agent string changes (Windows)
Internet Explorer 11?s Many User-Agent Strings - IEInternals - Site Home - MSDN Blogs
IE11 to appear as Firefox to avoid legacy IE CSS - Neowin

They have also seemingly thrown away the IE commentary logic too, as it wouldn't work for me. Although this raises a question, are they adopting Firefox's set of standards now? Or are we just supposed to assume that any browser browsing the site may be IE, and use only standard practices that include compatibility for Firefox and MSIE all in one?

This is a little bit of an upset I could imagine for web developers out there. Browser compatibility was difficult enough, and as every browser comes out with new versions, the standards seem to become more convoluted and less standard (by definition anyways).

 

[cluberti]

If you're driving webpages based simply on browser version, and not on features checking, you're already technically doing it wrong for most cases anyway.

Optimizing content for different browsers: the RIGHT way - Web Education Community Group

 

[AceInfinity]

I understand that, but browser detection for information retrieval, is a scenario where I really need to check the browser being used. I had a script that would display a page of information about the user browsing it, and I needed to determine the browser in one section. How do you match the user agent to determine the browser if they disguise it to be Firefox? :noidea:

Actually, re-reading my post, I did actually mean for this to be separate from my topic statement lol:

They have also seemingly thrown away the IE commentary logic too, as it wouldn't work for me. Although this raises a question, are they adopting Firefox's set of standards now? Or are we just supposed to assume that any browser browsing the site may be IE, and use only standard practices that include compatibility for Firefox and MSIE all in one?

This is a little bit of an upset I could imagine for web developers out there. Browser compatibility was difficult enough, and as every browser comes out with new versions, the standards seem to become more convoluted and less standard (by definition anyways).

Just as a tidbit of information stating my grief with each browser's way of doing things differently, which consequently always seems to make a web developers life more difficult.

I do not however, have code in the logical format of:

if (IE)
{
    //display this
} else ...

lol :lol:

IMHO, this just screws up IE. I know many other information retrieval scripts use and rely on the user agent for collecting data. I wonder how things like Google Analytics will respond to this, as I have no idea how they determine browser information...

 

[cluberti]

As someone who's had to support IE (from the MS perspective) both from a functionality and evangelism standpoint, I'm glad the UA changed. There are so many instances of web pages rendering like crap in IE not because the browser was incapable of doing it, but because web devs (in general) just wrote crap pages for it.

Also, IE11 does have one small indication that it's IE11, it's just not "MSIE" anymore. If you know what you're looking for, you can still easily denote IE11 from the UA String. I don't recommend you do this, of course, but it can still be done.

 

[AceInfinity]

As someone who's had to support IE (from the MS perspective) both from a functionality and evangelism standpoint, I'm glad the UA changed. There are so many instances of web pages rendering like crap in IE not because the browser was incapable of doing it, but because web devs (in general) just wrote crap pages for it.

Also, IE11 does have one small indication that it's IE11, it's just not "MSIE" anymore. If you know what you're looking for, you can still easily denote IE11 from the UA String. I don't recommend you do this, of course, but it can still be done.

Maybe I'm confused...

"There are so many instances of web pages rendering like crap in IE not because the browser was incapable of doing it, but because web devs (in general) just wrote crap pages for it." -- Are you saying that IE is conforming to other browser's standards here? I don't see how a user agent change would affect the rendering methods of IE. And as you say, it wouldn't be standard practice for someone to develop a page based on the User-Agent for browser detection that decides how to render a page. :S

I don't recommend you do this, of course, but it can still be done.

I only mentioned it was being used for an information page. What other methods of browser detection would you recommend? :huh:

 

[cluberti]

No, I'm saying developers serving pages based on UA strings (rather than what the browser is actually capable of) is the reason for the UA change (dropping MSIE) in the first place.

 

[AceInfinity]

No, I'm saying developers serving pages based on UA strings (rather than what the browser is actually capable of) is the reason for the UA change (dropping MSIE) in the first place.

Ahh... Well I've never done this, and I've never seen or heard of a web developer doing this. IMO though IE should've just conformed to the web standards that other browsers use, but that's just my idea... This seems like a really weird "fix" for a browser. The only thing I've seen is the <! if [IE ... lines, which they seemingly got rid of as well, probably for the same reason you claim.

Those were useful however, from my experience, because the only browser I've come across that could be so different for browser optimization within 3 or less version changes would be IE lol. I've always been pulling my hair out just trying to get IE to display uniformly between it's latest versions.

 

[cluberti]

It does - that's the whole idea of "web standards". There's no such thing as a browser's standards, there's just standards (and in different stages of acceptance or "baked"). IE11, for instance, supports a whole swath of "web standards", but not some of the more instantly new (and not yet *actual standards*) "standards" that a lot of web devs talk about. The use of the word "standards" when it comes to the web is actually an abused, misused word, because most of the "new, shiny" things developers write their code for (and browsers like Chrome and Firefox support quite quickly) aren't actually standardized yet, and are subject to change (and Chrome, Firefox, et. al. change to handle the new track if a "standard" changes during it's development lifecycle, or provide a vendor-flagged version of that feature instead, but that means code written to the "standard" previously breaks, or you have to write to each browser like it's 1999 again). Knowing this, and knowing that IE has the benefit and curse of being a 10-year (or close to it) supported product, Microsoft can't just change course with a version of IE mid-stream - once released, it's way of rendering things MUST stay the same. As such Microsoft's browser generally only supports "standards" that are actually that, or are at least in the final stages of becoming one.

I'm not saying one way is better than another, but I'm saying the use of the word "standards" in "web standards" is already a joke, and people judge IE harshly because of it, when in fact it's the browser that is consistently *the most standards-compliant*, because it supports (fully) actual standards (in fact, a good portion of the tests for each finalized or finalizing standard come from *Microsoft*).

 

[AceInfinity]

If that was the case we would all live in an ideal world. I don't think IE comments were ever part of the web standard as it was only used by IE.

because most of the "new, shiny" things developers write their code for (and browsers like Chrome and Firefox support) aren't actually standardized yet.

Can you provide me with an example? I know Chrome was one of the first to support HTML5, but undoubtedly, that was to-be a standard at some point anyways.

I'm not saying one way is better than another, but I'm saying the use of the word "standards" in "web standards" is already a joke, and people judge IE harshly because of it

IMO, not many of the other browsers get such a bashing, solely because they conform to the standards that other browsers could consider a standard. IE is the "odd man out" usually for this very same reason. There aren't too many 'standards' that other browsers implement, that IE does not, while IE also seems to have proclaimed 'standards' that are not standards for any other browser.

A Week With Internet Explorer: Not the Browser You?ve Always Despised

Now there are a lot of articles like this, and not just because "IE" has a reputable title, but because it's been the persistent pain for lots of web developers over many other browsers that most seemingly have no issues with.

 

[cluberti]

The point isn't something not becoming a standard (although there have been things - just look at ways to (try) to embed video without controls before HTML5 for instance), it's that things change over time between something being proposed to be a standard, and it actually becoming one. If you write a browser that you update every 6 weeks, this isn't as big a problem as one that will live for 10 years. If I wrote a browser that I'd update every 6 - 8 weeks, I'd simply implement the standard as proposed, and hope for the best (and trumpet how great my browser was at "standards" compliance, even if that's really a farce). If something changes, I change my browser in a future rev, and hey, it's still "compliant!" (and you can start to see my jaded nature here - the definition of "standard" means something everywhere *but the web*, where it seems to mean something slightly different). To be fair, Microsoft tries to hit the big ones with every major IE browser release, but if something is still in flux it makes it harder to embed that into a product if you think it's going to change. Hopefully that makes sense.

As to what IE has as a "standard" that other browsers do not, short of ActiveX (and there was a reason for that being implemented at one point - not knowing why would show your age ) you will be hard-pressed to find anything IE implements as non-standard. That's why I mentioned that most standards tests that are used to show implementations of standards by the actual standards bodies (I'm not talking about ACID tests, I'm talking about actual standards-compliance testing) are provided by Microsoft, not Google or Mozilla or Opera. You don't find that slightly interesting at least?