Google Patent - Artificial Anchor for a Document in the SERPs

[tedster]

How would you feel about Google adding an artificial page fragment anchor, so that they could send users to the specific part of a page that was chosen in the snippet, for instance?

With systems and methods described herein, mechanisms are provided to generate or simulate links with artificial named anchors and to allow the browser to recognize the artificial named anchor and navigate directly to the desired specific part of the target webpage even when the author of the webpage has not created a named anchor at the specific part of the webpage.

US Patent Application [appft.uspto.gov]

My first thought was "there goes my carefully designed 'above the fold' part of the page. But I can see how this would be extremely useful for the end user sent to a long page, or especially one of those monster PDF files.

< Thanks to Bill Slawski at SEO by the Sea [seobythesea.com] >

[Reno]

Most information must be read in the context of all the rest of the surrounding content. So if a person skips the first part of some explanation, I can see them missing the very "setup" that the author wanted to have understood, prior to that person arriving at the artificial anchor.

So sure, sometimes it would be fine -- "What's the modern capital of Widgetworld?"

But other times, a more indepth reading should lead to the requested query point -- "What are the historical forces that led to the founding of the Widgetworld capital?"

Am not exactly sure how Google will determine when a short answer is adequate, and when a greater understanding is required...

.....................

[ken_b]

This would probably lead to even more ads that follow the user down the page.

[MLHmptn]

I beleive this is being tested as we speak but not sure, tonight I am noticing breadcrumbs under the description snippets (on the SERPs) to which these breadcrumbs are on-page (on websites) navigation breadcrumbs. New sitelink expansion or artificial anchors or am I just confused at the moment? :>~

[gn_wendy]

What I would have liked to see is a "block" feature in WMT, like for the sitelinks. ie. being able to tell google: this is a LP, so please don't send the user to the bottom...

but I guess we can keep dreaming.

[TheMadScientist]

I'm really happy about some things...

1.) They are trying to patent it.
(I hope they are approved so no one else can do this.)

2.) I know php and my sites are already dynamic, so I can insert a large fixed footer with my ads in it based on referrer...

<?php 
if(strpos($_SERVER['HTTP_REFERER'],'google')!==FALSE) { 
 $big_footer='true'; 
} 
?> 

3.) I can ad in the footer:

If you would like to remove these ads and view the page as it is intended please click <link>here</link> to refresh, and if you would like to not see them again, please quit using Google as your search engine. We recommend you use Bing or Yahoo! instead. Thanks!

IMO: You kids at the 'plex are going a bit too far with the artificial editing of my work on my website... If I want named anchors on my page I'll put them there, thanks. If you would like to send visitors to the named anchors I've inserted feel free, that's what they're there for, but who are you to artificially adjust, edit, insert, or otherwise change my web page?

I will very gladly recommend any search engine except Google to anyone and everyone if (or when) this is implemented. I seriously, honestly hope you are approved for the patent and whether this is implemented before or after approval, I hope other webmasters recommend Bing and Yahoo! rather than Google to their visitors too...

It's fairly easy to dynamically insert some ads and text based on referrer people. Please, if you insert the ads, insert the text too so people know they're only seeing them because they used Google, and it's not normal for them to be where they are on your site...

People will think Google's got a virus or something if they see 'These ads present because you searched using Google. Use Bing or Yahoo! and they'll be removed.' enough times.

Do the people at Google actually think I'm going to let them 'jump' visitors past every ad on the page without doing something? How could I stay in business if people don't see my advertisements?

[londrum]

not sure how they are intending to actually create the anchor on the page, but if they artificially insert a <span>, or a <div>, or a <em> or something like that with the ID in it, wouldn't they be affected by the site's CSS?

if you've got <span>'s set as display:block, for example, then you could mess up the whole page.

[TheMadScientist]

In the patent application, they say it would have to be made accessible either through installation, such as the (or a) tool bar, or built into future browsers, which would would of course include chrome... and if you haven't read it yet, they want to use a custom string following the # to trigger the event, like: _g.sub.--

[signor_john]

My first thought was "there goes my carefully designed 'above the fold' part of the page. But I can see how this would be extremely useful for the end user sent to a long page, or especially one of those monster PDF files.

I'm I'm looking for "windows error message abc1234" and a discussion of that error message is buried 15 messages down in a long forum thread about reasons why Windows might crash, being taken directly to that phrase will be a huge convenience.

On the other hand, if I'm looking for "widgetville subway fares" and I'm taken to that phrase within the body text of the Widgetville Transit Authority's Subway Fares page, I'll probably be annoyed, because the whole page is about Widgetville subway fares and I'd just as soon read it from the beginning.

[dstiles]

The UK government is currently being threatened with legal action for allowing interception of web pages by ISPs for the purpose of targetted advertising (eg phorm, nebuadd).

Since google cannot modify our web sites I assume they intercept the page during download, either using their own proxy or the toolbar addon, and insert code within the page. Can anyone suggest an alternative method?

That, to my mind, comes under "interception", even if it is by the toolbar on the visitor's computer, since they will be unaware the page was tampered with from the original. If I send a web page to a visitor I do not want google (or anyone else) planting code in the middle of it. Apart from anything else it could disrupt the page presentation.

Relevant footnote: I have always included MSSmartTagsPreventParsing in my sites. This is worse since it isn't the browser just highlighting words during rendering: the page is actually altered in the proposed scenario.

[willybfriendly]

Yet another example that Google has no respect for site owners and their work.

Sounds like boundary issues [en.wikipedia.org]...

[tedster]

As I read the patent, this is not the same kind of "interception". Instead the implementation requires a specific module, and that would imply some kind of user consent at installation time, rather than mere interception of the page's delivery without permission. Here's a section from the patent application:

2... the artificial anchor comprises an instruction for an artificial anchor module in a document browser on a client device to navigate directly to the intra-document portion of the target document...

[Reno]

OK, so let's assume they do this. When I use their search and click on one of the listings, then allow the page to fully load, then do a "View Source" -- will I see this artificial anchor? Or is it going to be undetectable?

..................

[TheMadScientist]

My guess is undetectable. IMO, they'll probably have the 'installation(s)' supporting the 'unique string' they send make a change to the DOM, which does not appear in the source code of the page... think AJAX.

[dstiles]

But it is still altering MY web page, skipping perhaps large portions of relevant text, as at least one person above mentioned, and possibly presenting my document out of context.

If someone acts on the information immediately presented ("set light to the petrol") after having missed the large red text above that says, "Do not do this at home", who gets hauled into court for publishing dangerously mis-leading information (and given some of the dumb court cases in the Stella Awards, you just KNOW someone will be!).

It is MY web page and I do not want it altered. Is that not legally clear enough for google?

[CainIV]

From a conversion and user point of view this makes on pages that are segmented and laid out to offer information in steps.

I think this might work, if Google were to release this type of technology along with a tag where the user could accurately label a web page for this purpose as opposed to simply pushing it on the consumer.

[g1smd]

Maybe the answer is to clearly label every part of your page so that there is nowhere for Google to add in extra anchors.

[thor_swe]

It is MY web page and I do not want it altered. Is that not legally clear enough for google?

I can't agree. The browsers has different built-in functionality and render webpages differently. I can't really see the difference between clicking a link to reach a portion of the site and press Ctrl + f and write "set light to the petrol", the effect is the same, user reach same portion of the text. Except for the built in browsers support for webpage altering(change size, remove css/js/etc, search..) we have a LOT of addons that remove ads, change stuff, edit headers, etc.

This IS the nature of the web(describing design with words), only thing we can do is to adapt or start printing the information in books, then you know its presented to the user exactly like you want to.

[maximillianos]

If you have google code on your page (ie - adsense, analytics, search box, etc) they won't need to alter your source. They will already have a backdoor script embedded on your page to help pull off the jump.

[Solution1]

They could easily do this with their cached copy of a webpage, as they display it. They have control of the cached copy, and can insert as many named anchors in it as they want.

[londrum]

it's probably a useful idea, but i think they'll be shooting themselves in the foot when it comes to adwords/adsense earnings.
most people have their best-paying ads above the fold (because that's where google says to put them). if google suddenly starts depositing users halfway down the page their revenue is bound to take a dip.

there's also that annoying problem that anchors have when pointing to them before the page loads.
the page will frequently settle on the anchor, only to scroll on further when an image somewhere above it finally loads in.
it's confusing to people when they're dumped halfway down the page, and the relevant snippet is nowhere in site.

[httpwebwitch]

This sounds like something which is only possible via a plugin (like the Google toolbar), a proprietary browser (like Chrome), or if participating webmasters include a hotscripted link on their page (like Google Analytics). Otherwise, an inbound link doesn't have the ability to alter a document or scroll to a specified y-coordinate.

This is a neat patent. I don't particularly like it, but it is neat.

[achshar]

Hmmm when i first heard it, i felt good for the user will have reduced click to reach the info he wants but then it affects the ads.. One thing thay can do is something like that of meta noindex. They could ask the webmasters to have a meta tag or some code if they want to have user jump directly to the portion he wants. If that meta/codes isnt there. They avoid adding the artificial anchors... simple but complicated.

[fargo1999]

I hope TheMadScientist will provide a detailed tutorial here on how to exactly remove the spy-ware...

[jomaxx]

They should start closer to home, by fixing their own cache system. Not infrequently, when I am looking for a specific text string such as a name, that string is nowhere on the cached page. Even if it clearly shows up in the SERP snippet. Even if I do a View Source and search the HTML code for the text.

I don't see why it should be so hard to link to a cached version of the page that matches what shows up in the search result.

[tedster]

In those cases, have you checked the top frame (above the cached page) to see if there is a message, such as "these words only appear in links pointing to the page"? Sometimes that's the reason.

It seems to me that this proposed artificial anchor system might also help the user if it was enhanced to include such a message whenever there is no exact occurrence of the keyword on-page.

[TheMadScientist]

I hope TheMadScientist will provide a detailed tutorial here on how to exactly remove the spy-ware...

Remove the tool bar from the visitors browser and disallow chrome?

I wish there was an easy way to fight, detect, remove this garbage, but the problem is: it happens during the transaction between Google and the visitor's browser before the request is ever sent for the page, but the 'string' is sent only to the browser, not the server delivering the content.

So, the server and server-side script are useless in trying to fight it, because they won't ever receive notice there is an extra tag waiting in the browser that will take control of the delivery of the page, access the DOM and change the position the page displays at.

The best I can think of is some JS, but they have to release it and make it live so people can try to come up with a block for it.

It might be (is, IMO) legal for them to do this, but in the webmaster opinion of Google pole, I think they're way over stepping.

If I can't find an easy way to block this garbage I will probably really move my ads to the bottom of the page in a fixed footer for all visitors from Google or use a server side script and cause the page to not display the actual content for Google referrals, because in the world of careful wording, two can play...

##### ### #####

'Google Visitor...

It has been brought to our attention Google may be artificially inserting code into the web pages you access from their website through either your browser or toolbar if you have one installed. This artificial code insertion may cause the page you are trying to access to display and/or behave in a way not intended by the owner.

To ensure while visiting this page you view the page the way it was intended, in it's entirety, without anything artificial being inserted, added, edited or removed by Google, please click <link>here</link> to refresh.

Although this was not our doing, we sincerely apologize for the inconvenience.

We hope you seriously consider using Yahoo or Bing in the future so you know you are viewing the sites you visit the way the owners intended.'

##### ### #####

@ The Plex: You wouldn't be thinking about inserting your own ads on the page once the visitor gives you permission to manipulate it would you? Nah, you guys are cool. You'd never think of doing anything like that would you?

[commanderW]

Certain aspects of this discussion remind me of the time when some websites felt the need to try to control what page another website or search engine should be able to link to. The idea being that the website was constructed to lead the visitor through certain doorways and past various advertisements etc. I have not read anything about this for several years now and am under the impression such claims were dismissed in court.
Since I am now desperately trying to figure out a good way to monetize my own sites, I understand how this could seem a threat to ad revenue, if the surfer is taken past the ads directly to the text they've searched for.
However,I would like to point out that the description of the service is really only a more efficient version of clicking on a link to a long or difficult to navigate page and then hitting 'command f' (or 'ctfl f' as is mentioned in a post above). At least one word of the search term I have entered in the search box of the browser is automatically placed in the search box of the find function (at least on OS 10.3.9). This patent sounds like a simple automation of that process and ultimately, a time saver for the surfer.

RE ad techniques - I am put in mind of the TV ads here in America, which raise the volume on the TV without the viewers choice, in order to force them to hear the ad all the way in the kitchen or bathroom, where we go between segments. This infuriates me and makes me feel to be an enemy of advertising in general. At 1 AM, with TV just loud enough to hear during the show, the ads are so loud they threaten to disturb the neighbours. Extremely upsetting. I feel the same way about other kinds of extortion of behaviour when I am on a website trying to find something and realize that it has been deliberately buried in a rat maze of navigation links so I will click an ad. My immediate reaction to this is mistrust. I would never click an ad on a site like this, for fear the link would take me to a 'drive-bye download' etc.
Now, more familiar with the internet and being a website builder and owner myself, I am alittle more sophisticated. But, when all is said and done, I never ever click on ads. Because the methods have aggravated me too many times in too many ways.
Obviously, many will disagree with my prefered approach to these kinds of things, which is kinder and gentler forms of persuasion.

As for the creation of an anchor outside the page code, well Firefox can completely turn off the css or even allow you to rewrite the code for your own convenience. This is a powerfull usability tool. And what about the plug-ins that claim to block ads altogether? Since so many people have been driven to hate advertising, these plug-ins seem a far greater threat to income. Is Firefox hijacking anyones web design?

When Google (or anyone else) starts placing their own ads on my pages without my invitation (as I read an ISP in Britain dared to do), then I'll get angry. Then it'll be time for a necktie party. But otherwise, I don't see all these issues that are raised over every change or advance G makes to it's service.

[dstiles]

I wonder how they will get over the legal notices on some sites to the general intent: "You may not intercept nor alter in any way the contents of this web site".

Technically, search engines are already in breach of many issues including copyright, alteration and general content "theft" but are permitted to get away with it for implicit benefits to the web site - fair enough up to a point. Several google projects recently discussed within these pages are pushing these issues to their limits and beyond.

It is time that the obsolete robots.txt was supplemented with a legally binding document that lays out clearly, in a way readable by robots, just what can and cannot be perpetrated upon specific web sites. I have been musing on such a document for some time now, though I doubt I could ever get such a thing off the ground. One thing I have certainly decided, though: although search engines should contribute to the specification it should definitely be masterminded by webmasters and site owners. Search engines will otherwise design and word it to give themselves pretty much carte blanche.

[TheMadScientist]

As for the creation of an anchor outside the page code, well Firefox can completely turn off the css or even allow you to rewrite the code for your own convenience. This is a powerfull usability tool. And what about the plug-ins that claim to block ads altogether? Since so many people have been driven to hate advertising, these plug-ins seem a far greater threat to income. Is Firefox hijacking anyones web design?

The difference is these take an action by the end user, and it's not an automated 'our algo interpreted your query as ...blah... so your browser will now show you ...blah... on the page.'

Let's leave the ads out of this for a minute and talk about branding... The visitor skips your logo, your header, your look, your feel, and all the presentation you have to offer in the most prominent place on the page.

How can you effectively brand a website when you have no control over where the person lands? If the visitor takes action after the page loads, IOW cmd/ctrl f to search, at least you got to show them your logo, at least you had the chance to show them a bit of what else you offer besides text on 'blah'.

If Google wants to use the named anchors we do, then cool. Count them as links. Determine they are useful for visitors and we used them for a reason. Use the algo to determine where the most relevant place on the page we allow people to 'jump' to is, but don't just randomly insert them on the page.

I would like to point out that the description of the service is really only a more efficient version of clicking on a link to a long or difficult to navigate page and then hitting 'command f' (or 'ctfl f' as is mentioned in a post above).

So, since they can't determine and deliver a relevant, visitor friendly page with great content, navigation and ease of use you think this is a viable work around, which allows them to not have to do a better job of determining what's relevant and usable to show to their visitors?

I still think it's garbage...