Favicon for XMTS

[wsk]

vB has an automated system for favicon. ...

I'm not buying that as an answer.

What I will believe is that newer browsers are automatically looking for the favicon in the root and upon finding it will display it. W3C discourage this method: http://www.w3.org/2005/10/howto-favicon

Turns out that the mechanism I suggested in the op is only partially correct according to w3c page. On the other hand, my method is described in a Wikipedia article referenced by that very same w3c page above (http://en.wikipedia.org/wiki/Favicon. Go figure.

I suspect that a standards compliant browser correctly, will not display favicons unless directed to do so by the appropriately included code in the displayed page. Perhaps that is why I don't see the favicon.

Do a test. Put those two lines in a test page and I'll tell you if I see our favorite favicon.

[Mike1]

What I will believe is that newer browsers are automatically looking for the favicon in the root and upon finding it will display it.

Absolutely correct. Which is why we replaced the default vBulletin favicon.ico with a Saltire image and then moved it from the forum directory to the main root.

I've just visited the X Marks home page with Firefox 2.0, Internet Exploder 7.0 <shudder> and Opera 9.02. The Saltire favicon.ico is properly displayed in all three browsers.

[Mike1]

Is it safe to figure this issue has been sorted?

[wsk]

Is it safe to figure this issue has been sorted?

Were it up to me, I would say no. I would say no because the current recomendation for favicons published by w3c discourages the use of a fixed location for any document or image. Were it up to me I would make the change to comply with the w3c recomendation.

While I was at it I would also make sure that pages generated by vB complied with the DOCTYPE - this could be a vB issue that I would have no control over. I'm a bit anal that way. What's the purpose of claiming a standard if I'm not going to comply with it. W3C Validation Service

Were it up to me ...

But it's not. You are Second-In-Command. Responsibility ... big bucks ... etc.

[Mike1]

The interesting point about the entire <link rel="icon" issue is that it is a crutch to get past a shortcoming in IE. I admin a handful of boards and in every instance, IE is the only browser that could not properly display favicon. And once the page is saved in the IE Favorites folder, <link rel="icon" becomes totally unnecessary, as IE will then display it.

I ran the X Marks CMPS page through your W3C validator and quite interestingly enough, not a single lick of the vBulletin code was reported. Some older code that was brought in from modules created in an earlier version of the CMPS was all that was reported.

And the entire issue of W3C validation really fails to answer the earlier question as to whether or not your browser is displaying the favicon.

[wsk]

The interesting point about the entire <link rel="icon" issue is that it is a crutch to get past a shortcoming in IE. I admin a handful of boards and in every instance, IE is the only browser that could not properly display favicon. And once the page is saved in the IE Favorites folder, <link rel="icon" becomes totally unnecessary, as IE will then display it.

I've said it before: IE sucks; there's just no getting round that. I could go on and on about it but won't ...

I ran the X Marks CMPS page through your W3C validator and quite interestingly enough, not a single lick of the vBulletin code was reported. Some older code that was brought in from modules created in an earlier version of the CMPS was all that was reported.

Don't know what CMPS is. If the vB code is correct then good on you (or them - whomever maintains that).

And the entire issue of W3C validation really fails to answer the earlier question as to whether or not your browser is displaying the favicon.

I could be mistaken, that's easily enough done, but I thought we'd already established that my browser doesn't display the favicon. I surmised that it doesn't because pages coming from XMTS don't tell it where to find the favicon - as I believe they should. I do know that the two lines of code that I provided in the OP will cause my browser to display the favicon from my own web site so display is possible - url is, of course, different. Those two lines will allow you to put the favicon anywhere you want rather than force you to put it in the root.

Upon additional contemplation, I surmise that newer browsers display the favicon because web site authors, in way too many cases, are unaware of web standards but they know that people who view their pages like favicons so they do just enough work to put one in the root of their site and call it good. In this case browser creators are complicit in the ongoing lack of quality one finds in so many web sites. Sorry, ranting, I'll stop.

You're right, w3c validation has nothing to do with the issue to hand. Still, I am an advocate of standards and adherence thereto. I believe that by hewing to the standard the broad variety of browsers will be much more likely to display what I want displayed as I intended it to be displayed because by doing so I minimize the amount of guessing that the browser must do when interpreting my html. Of course, valid html won't overcome browser shortcomings or bugs ...

Have I answered your questions?

[Mike1]

Have I answered your questions?

And then some.


These screenshots will illustrate that Opera capably displays the favicon file, without use of the <link=rel crutch.

The following return is from THIS VALIDATOR

http://www.xmarksthescot.com/ was checked and found to have a valid favicon.ico file.
Notes:This page has a valid favicon.ico but it does not use a favicon link tag (<link rel="shortcut icon" ...> or <link rel="icon" ...>). Although this is not a problem, the favicon link tag is likely to be more compatible with upcoming browser versions.

All of which brings us back to your stated belief "that newer browsers are automatically looking for the favicon in the root and upon finding it will display it." This is all proof that you were absolutely correct in your belief.

[wsk]

These screenshots will illustrate that Opera capably displays the favicon file, without use of the <link=rel crutch.

About Opera 9.02: Agreed. That point was never in dispute. That <link rel="..." ... /> is a crutch: I guess we disagree.

... THIS VALIDATOR...

Thanks for that link. Didn't know about that. Another tool for the box.

http://www.xmarksthescot.com/ was checked and found to have a valid favicon.ico file.
Notes:This page has a valid favicon.ico but it does not use a favicon link tag (<link rel="shortcut icon" ...> or <link rel="icon" ...>). Although this is not a problem, the favicon link tag is likely to be more compatible with upcoming browser versions.

That XMTS has a valid favicon is not in dispute.

Not sure how to read the validator's note though. Newer browsers will understand the tag and act accordingly - older browsers will ignore it as they should. But does it also mean that use of the <link rel="..." ... /> is recommended?

All of which brings us back to your stated belief "that newer browsers are automatically looking for the favicon in the root and upon finding it will display it." This is all proof that you were absolutely correct in your belief.

Again, I did not think that this was in dispute.

Somehow I think that "what we have here is a failure to comunicate" (Strother Martin did say that well). When others replied to this thread saying that they could see the favicon I was done with this conversation. I recognize that I have an older browser (it ain't broke so I haven't bothered to upgrade). Since we reopened this conversation I have been advocating for adherence to the w3c standards and attendant recommendations. This I believe will benefit XMTS. The benefits will be subtle and most of us won't notice but the benefits are there nonetheless.

Let us not be adversaries.

[Bob C]

The favicon shows up just fine in IE7.