greetings everyone... here's something to ponder:

do you think it's important to use the "title" atribute of <a href> when linking an image?

from a usability standpoint this serves no purpose as the <img /> "alt" attribute will always take over on image hover. the user would never see the title of the <a href> for the image anyhow.

the only time an <a href> title can be seen on hover of an image is if there is no ALT attribute for the image.

assuming you do use your ALT attributes on images, i am concerned that by using the title tag anyway when linking images it could be viewed by the engines as "excessive keyword loading" and actual lower the weight of the page or worse, flag the site as being "SEO'd"

then, i wonder if it's best to use the title attribute for the linked image - or - if it's best to use the ALT attribute for image call itself.

any thoughts on this?

Can't say I've ever thought about putting the title attribute within an image...


Couldn't see why it is needed though, as you stated above, the alt does that for images. Really, it is a user thing, more than a search engine, so IMO, I would only recommend using the alt for images, and title for text. I only recommend using the title for text even when something else needs to be said about where the anchor is pointing, otherwise, not used.

I think the alt attribute will only show up in IE. In compliant browsers, you only see the alt if the image fails to load. The title, however, will pop up. If you want the little tooltip window on your images, I'd suggest putting both an alt and a title attribute, and making them identical.

Can't say I've ever thought about putting the title attribute within an image...
nope anthony ...you didn't understand the question. my concern is not doubling up on attributes for a tag. i am refering to linking images and whether or not to use the "title" attribute for <a href> if i am already using the ALT tag for the image.

I think the alt attribute will only show up in IE.
yep, you're right qwerty. after some testing, i found you comments correct. the complaint browser (in this case firefox) did not show alt attribute text on hover; however, when title-ing the link that was "wrapping" the image ...that did show.

thanks for the answer ...i am sure this will be beneficial for many peeps like me who still struggle with browser compliance and "what shows up on which browser" logic...

Title is your friend ;)

Fathom:

Has there been any evidence that you know of that shows title has a significant impact on rankings?

Title attributes don't get indexed, so they're only of usability value.

It's certainly not going to help for SEO purposes as it's not indexed by any major engine that I know of.

To be very clear - The <title> attribute for a PAGE is very important, but the "Title=" attribute for a LINK is, AFAIK, not.

I would use the ALT or the Title, but not both, for the reasons the OP outlined as concerns. And since the ALT is better supported and actually indexed, I'd personally choose ALT.

Ian

Thx Guys:

I did indeed mean the title within the links tags. I don't use it all that much right now is why I was concerned. IMO the page title is the most important factor:)

I don't see it as dangerous to use both, as I described above. It would be foolish for a search engine to view this as spamming, as it truly is a usability issue.

Here's a legitimate example: you have a page of thumbnail images which link to larger versions of those images. If you put "click image to enlarge" in both the alt and title attributes, you're covered: people for whom the image failed to load will know to click, and so will those who see the thumbnail.

Yes, you could just put that in plain text on the page, but fear of search engine wrath shouldn't be the reason to decide to go that way, IMO.

And yes, I've done this, and no, it's never caused me any problems, but I suppose it's worth noting that there's no way any manual review of such a page would conclude that I was trying to stuff in keywords. If you used the alt and title for something like "click for details on our keyword keyword keyword" you might be taking a bit more of a chance.

To be very clear - The <title> attribute for a PAGE is very important, but the "Title=" attribute for a LINK is, AFAIK, not.

Title for a page is "Title Tag" not attribute, I believe.

But yes, I was referring to a link title attribute in my post discussing their worthlessness as far as SEO goes.

Interesting... Thanks qwerty! I wasn't thinking about image only links.

Useability is a good thing. Just because something won't help you with an SEO issue doesn't mean it's not useful - and of course if it's not indexed then it's not spam (spam affects SE behaviour - this would not).

Thanks! <adds trick to web design toolbox>

Ian

everything on a page gets indexed... whether used for ordered rankings is the question.

evidence - inconclusive... but if there is usability value for the user - I truly can't see search engines saying screw usability - we only want to rank things that lack usability.

It's certainly not going to help for SEO purposes as it's not indexed by any major engine that I know of.

Is this based on assumption or insider knowledge?

Evidence of something helping is hard enough, but certainty based on lack of evidence isn't possible.

Google's cache and even text only cache both demonstrates the tool tip... if this isn't indexed - what is?

CAVEAT - W3C supports the use of title attributes and use in most elements... in my book that's 'best practices' regardless of order ranking help.

Thus - title is your friend - it will never hurt.

All you need to do to test it is to put a title attribute containing unique text onto a page, wait for the page to get indexed, then search for that exact text (in quotation marks).

I don't have a complete answer to your question regarding the cached text, since for those images on which I use both the title and the alt, they're the same, so for all I know what I'm seeing in the cached text is the alt and not the title. But I've never seen the title attribute of a text link in the cached text of a page.

I just looked at the Google text cache of a page that utilizes the "title attribute" in href tags and they DO NOT show up. The alt text shows in the text cache.

I just looked at the Google text cache of a page that utilizes the "title attribute" in href tags and they DO NOT show up. The alt text shows in the text cache.

hmmm... roll over a text link that uses title="" in text only cache - have a couple hundred websites that show the tool tip.

ahhhhh....Fathom, you are correct. I didn't understand the "on hover" part, I was looking for the text to be on the text cache page like the alt text is. I stand corrected.

All you need to do to test it is to put a title attribute containing unique text onto a page, wait for the page to get indexed, then search for that exact text (in quotation marks).

I don't have a complete answer to your question regarding the cached text, since for those images on which I use both the title and the alt, they're the same, so for all I know what I'm seeing in the cached text is the alt and not the title. But I've never seen the title attribute of a text link in the cached text of a page.

in theory sure - how long do you wait for absolution?

ordered rankings isn't about any one thing... thus total segregation to test a singular premise is flawed.

example: add a page with tons of solid text copy and good on-page optimization to test specific on page factor... but don't link to it - to ensure you have a fair unbiased test it must be orphaned - now get it to stay in google for observation - you can't.

The cached text thing merits further study, I'd say. I'm not convinced that seeing something there is enough to tell me that that something has been indexed.

For example, look at the cached text of a page that uses relative text links, and go to the source code. You're looking at a page on Google's server, but a link that's coded as <a href="/aboutus.htm"> still points to the About Us page on the site's server, not Google's.

So it's true that you can see the title attribute of a link when you mouse over it in the cached text, but I just don't know what to call that effect. I still believe it's not indexed if you can't search for it. Maybe we have to say that Google has grabbed the code, but they're not doing anything with it.

in theory sure - how long do you wait for absolution?

ordered rankings isn't about any one thing... thus total segregation to test a singular premise is flawed.

example: add a page with tons of solid text copy and good on-page optimization to test specific on page factor... but don't link to it - to ensure you have a fair unbiased test it must be orphaned - now get it to stay in google for observation - you can't.
I don't believe you need total segregation to test something like this, because you're not testing to see if something ranks well; just that it ranks. And if you use something unique, you don't have to search through 50 pages of SERPs. If anything comes up, it's what you've created -- after all, it's unique.

So take your home page, or any page that gets crawled fairly often, and put a title attribute of some nonsense phrase on one of the links. Wait until the page gets crawled again, then search on the nonsense phrase. Pretty simple.

I don't believe you need total segregation to test something like this, because you're not testing to see if something ranks well; just that it ranks. And if you use something unique, you don't have to search through 50 pages of SERPs. If anything comes up, it's what you've created -- after all, it's unique.

So take your home page, or any page that gets crawled fairly often, and put a title attribute of some nonsense phrase on one of the links. Wait until the page gets crawled again, then search on the nonsense phrase. Pretty simple. I have a home page that contains a unique phrase in a title attribute (of the href tag). It has used this phrase, as a tool tip, since the page was created and has been in the index for at least 4 months. I just searched for this phrase, in quotes, and Google returned:Your search - "forces in the Cosmos had to align to produce this" - did not match any documents. Does this prove qwerty's point? :confused:

I think that what it demonstrates (mind you, I'm guessing here) is that googlebot grabs all the code in a document when it crawls it, but google doesn't make use of all of that code. If that's true, then I guess the next question would be whether they don't use it because they can't, or because they don't see a reason to.

The code behind cached copies of pages that contain javascripts has those scripts in it, but I think we're all in agreement that google doesn't read that. The cached text of pages will show the alt attribute of non-linked images, but it's been generally believed for years that google only pays attention to the alt attribute of images that anchor links.

So maybe it's there, but the search engine is ignoring it.

it suggests that google didn't return a precise match based on a singular user agent support attribute.

inconclusive for anything else.

Is this based on assumption or insider knowledge?

It's based on evidence.

Pages that show up for other words that are on the page or in other tags that ARE indexed, don't show up for a search for the words in the link attribute tags, according to the tests I've done.

I've tested alt attributes the same way and have made different conclusions. (I.e., they are indexed only on clickable links.)

it suggests that google didn't return a precise match based on a singular user agent support attribute.

inconclusive for anything else.
Well, I've just tried the same search, using every option on FF's user-agent switcher, and got no results each time. If you know of another way to test this, please explain.

I've tested alt attributes the same way and have made different conclusions. (I.e., they are indexed only on clickable links.)

so in both instances the tests provided conclusive or inconclusive evidence for 'on-page' factors... but in both instances a 'link' was being used which also presence 'off-page' factors where the link itself tends to have far more benefit 'off-page' than 'on-page'... I would think a supporting attributes for this element would likely show results [if any] in the same manner.

Sorry fathom, you've completely lost me, buddy.

You're welcome to conclude whatever you want. I'm content in knowing that link title attributes will be of no help to my rankings.

google deprecated the value that alt="" had to 'on-page' factors for ordered rankings but the fact that there is seemingly still evidence that alt="" in conjunction with a 'link' - is still weighed most notably to added relevancy to the 'link-to' page and further text in proximity to the link is also used to weigh relevancy, supports the potential value that title="" has in ordered rankings.

It's safe to say that your testing and conclusions are accurate for testing on-page value but a link is more valuable to the page it links to than the page it is on - so is the alt="" and quite probably title="" as well.

but will be your buddy in either case! ;)

This guide is not for the soft skined. I learned a lot from this guy. He explains the difference between alt and title for images better then anybody I have seen.
http://members.optusnet.com.au/~night.owl/morons.html#alt_texts

Regardless of SEO, I think good design is something to not be overlooked.

How much it affects SERP rankins, I don't know, but I don't think it will hurt. If it isn't being used now by google, it'll probabily come in handy at some point in time with some search engine. I always use it.

If you want some text to appear when the user hovers a mouse over the image, then it's the "title" attribute that you should be using. Regardless of the behaviour of some browsers, regardless of rubbish perpetuated by some people who don't know what they're talking about, that is the purpose of the "title" attribute, not the "alt" attribute.
That about covers it, at least as far as usability is concerned. Thanks for the link.

One thing I'd argue with about what you wrote, though: you say that you always use the title attribute. I find that in most cases, there's no need for it. I only use it -- on images or text links (or anything else for that matter) when I've got a message I think it would be useful to show as a "tooltip" rather than putting it right on the page. For example, something like "link will open in a new browser window" or "page in Greek". But I guess that's just a matter of taste.

There is visible, tangible evidence of the value of alt="foo " but never has been any substantiation or evidence of the value of the title attribute. Not that it's a bad thing to use, since it's specified in the standards, but evidence of it's value for SEO has not ever been demonstrated in any fashion whatsoever.

< title="blah" ... is a tag shared by all HTML 4.x elements; however don't take my word that it's hardly useful, add one to a <table>, <tr>, <td> and that sort of tag. You'll see a tooltip when you mouseover. There's no < alt= > attribute for much else -- img and input there are. A few others but in an img tag IMHO you're wasting your time discussing it. It does the same thing as title in that context, and my guess is it's up to the browser types to decide which one they render. Not both. One tooltip is all you need.

Don K

two things -- first, I believe that if you use a title and an alt tag, the title should override the alt tag as the tooltip. I believe this is the correct behavior. IE6 complies with this, if you set an image alt tag as "this is the alt tag" and title as "this is the title" and mouse over it, you should see the title.

in the absence of a title tag, IE uses the alt tag as the tooltip -- however this is not to html spec. the only thing that's supposed to be used as the tooltip is the title. The alt tag is specifically set to show up only when the image fails to load, or when the user is viewing the site in text-only mode.

2nd thing -- I believe I heard a quote years ago that google said they index well over 100 items on a page. with that in mind, why shouldn't title be one of the things they use? even if it's .00000000000000001 % of what makes the page rank, I wouldn't be surprised at all. At the same time, to protect themselves from spam, I can see them deciding that a site will never rank for a phrase based solely on it appearing in the titles

2nd thing -- I believe I heard a quote years ago that google said they index well over 100 items on a page. with that in mind, why shouldn't title be one of the things they use? even if it's .00000000000000001 % of what makes the page rank, I wouldn't be surprised at all. At the same time, to protect themselves from spam, I can see them deciding that a site will never rank for a phrase based solely on it appearing in the titles

Interest enough I decide to have another crack at this.

I do tend to focus on the usability aspect and never did any control testing... however, I find it difficult to believe that anyone can be 100% certain that a nil return is conclusive evidence of anything.

I use the speculation of hyphen vs underscore as an example. While googleguy has pretty confirmed that hyphen is a word separator and underscore isn't - does key-phase.html actually induces better ranks for the term >keyword phrase< over key_phase.html

The difference as in the above quote has to be very small on a singular occurance.

In any case - I'll through this up shorty and we can scruntize the return.

http://forums.searchenginewatch.com/showthread.php?p=29943#post29943

I would simply place your main keywords in the hrefs and alt tags, which provide a good backup to your META tags.

Nice use of Shakespeare by the way!

Isn't the Title of a link for the vision impaired?

Also, it seems that there is some assumptions being made about SEs not using the Title text of a link. While it's true that one can place unique text in a link Title and it will not show in the SERPs, I don't think that proves the Title of a link is not used as part of a SEs ranking algo.

Well, if you're suggesting that a page about books with the word "books" in the title, the headers, the inbound links, etc. may have a slightly better rank if it also has books in the title attribute of some of those links, I suppose that's possible, but my point about using a unique term only in that one attribute was the only way I could think of of narrowing it down to just that single factor.

If there are no pages indexed that contain the string "ptmeeef" and I stick that in a title attribute only, then it seems logical to me that if it meant anything at all to the search engines, I'd get one result for "ptmeeef" when I searched on it once that page had been indexed. But if it's a factor that has no effect on its own, I don't know how setting up a page with other elements containing that string could prove that the title attribute's use had anything to do with the page's ranking.

But I'm waiting for the results of Fathom's test, and I'll gladly admit I was wrong if the test demonstrates that.

But I'm waiting for the results of Fathom's test, and I'll gladly admit I was wrong if the test demonstrates that.

I'll note that: this isn't a question of right or wrong... it's a question of whether you 'should use it or not'.

I use title="" because it make sense to use it not because it enhancing ranks.

If tomorrow 'links' didn't enhance rank results - a ton of people would immediately stop linking and looking for links... yet 'link' still makes prefect sense regardless of the added value you get from search engines or not.

No argument there. I use title attributes when I see an advantage to them from the user's perspective, and I'll continue to do so. If it turns out that they somehow help ranking, I may find myself using them more, and hoping they don't get so overused that SEs start ignoring them... at which point I'll be back to using them when I see an advantage to them from the user's perspective :)

The title attribute of a link is for the blind.

First hint of results is now appearing.

http://forums.searchenginewatch.com/showthread.php?p=29943#post29943

The title attribute of a link is for the blind.
It's certainly useful to the blind, or anyone using a screen reader, but that doesn't mean that it can't be beneficial to sighted people using standard browsers.

The title attribute of a link is for the blind.

Wonder how they see it pop up in the little bubble...
:confused:

Wonder how they see it pop up in the little bubble...
:confused:

does qwerty's post just prior to your question answer that? Will most screen readers "display" the title attribute link I wonder?

I'm making certain assumptions here, since the closest I come to having a screen reader is MS Narrator -- which apparently isn't very fond of FireFox, as it only reads a few things on the page. But I opened this page in IE with Narrator on, and when I moused over the "quote" button, I heard "tooltip - reply with quote".

The thing is, that's both the alt and title for that button, so I'm not sure which one it was reading. But I checked a text link that has a title attribute (in IE) on one of my sites, and it did read that out as a "tooltip".

As I noted earlier, on those image links to which I add a title attribute, I make it the same as the alt attribute, so I can't test to see if Narrator reads the title of images, and I don't have time right now to edit the code on any of my images to make a different title. Anybody else care to test this?

Wonder how they see it pop up in the little bubble...They can't, just as they will never see your insensitive post.

does qwerty's post just prior to your question answer that? Will most screen readers "display" the title attribute link I wonder?

The point is, that it's obviously NOT just for the blind, or they wouldn't bother to have it pop up to read....

The point is, that it's obviously NOT just for the blind, or they wouldn't bother to have it pop up to read....I cannot see where anyone stated is was "just for the blind"???

The point is, that it's obviously NOT just for the blind, or they wouldn't bother to have it pop up to read....

thank you for clearing that up! :p My "non-developerness" always gets in the way...also I have never used such a software to "see" a site....any suggestions for samples that will show code? ;)

thank you for clearing that up! :p My "non-developerness" always gets in the way...also I have never used such a software to "see" a site....any suggestions for samples that will show code? ;)In Internet Explorer > "View" > "Source". In Firefox > "View" > "Page Source" (same in Netscape, I think).

You can also use (in your browser's address bar) "view-source:http://any-domain.tld/page.htm" to see a page's source (including the source of a meta redirected page). That will not work in IE6 on Windows XP + SP2.

Hope that helps.

I cannot see where anyone stated is was "just for the blind"???

The title attribute of a link is for the blind.

You're right, you didn't say "just" but the way I read it, it was implied. I apologize if that's not what you meant.

Jill

In Internet Explorer > "View" > "Source". In Firefox > "View" > "Page Source" (same in Netscape, I think).

You can also use (in your browser's address bar) "view-source:http://any-domain.tld/page.htm" to see a page's source (including the source of a meta redirected page). That will not work in IE6 on Windows XP + SP2.

Hope that helps.

Thanks martin but I look at source every day. I think we are talking about some sort of screen and code "reader" designed to help blind people "see."

Am I lost in hyperspace here or is this conversation more confusing than it should be? :p

Thanks martin but I look at source every day. I think we are talking about some sort of screen and code "reader" designed to help blind people "see."
Your welcome Chris and sorry for the misunderstanding. When you said:
thank you for clearing that up! :p My "non-developerness" always gets in the way...also I have never used such a software to "see" a site....any suggestions for samples that will show code? ;)I let your statement "My "non-developerness"" bias my understanding of your intent (how's that for confused?).

This thread has got a bit confusing!! :confused:

Yeah, it has gotten a bit convoluted. So, to see if we can get back on track, I've got a very simple question: does anyone here have access to a screen reader, and if so, would you be willing to use it to check out one of my pages and report back on whether the reader gave you the contents of a title attribute?

Yeah, it has gotten a bit convoluted. So, to see if we can get back on track, I've got a very simple question: does anyone here have access to a screen reader, and if so, would you be willing to use it to check out one of my pages and report back on whether the reader gave you the contents of a title attribute?

yeah and let me know if there's a free one or a demo somewhere too! :D

yeah and let me know if there's a free one or a demo somewhere too! :D
I think this is what we have been looking for (web screen reader demo):

http://readplease.com/english/downloads/#rb2 (30 day demo)

Wikipedia page on screen readers (with links to software) http://en.wikipedia.org/wiki/Screen_reader

I haven't had time to download & try any of these yet.

OK, I downloaded a 30-minute version of HAL and tried it out on my page. You use the tab key to move from link to link, and when it gets onto a link containing a title attribute it says Not visited link
http... [the target url]
[the anchor text]
Press insert to click
It did not announce the title attribute.

Maybe I didn't have enough time to find whatever in the application's settings tells it to read title attributes. I don't know, since it only gave me 30 minutes to play with it before it disabled itself.