Google to target PR hoarders. Options

[Black_Knight]

There have been repeated small quotes and rumblings from Google of late that lead me to believe that using JavaScript to hide outbound links is no longer a safe practice.

Google engineers have stated in a few places lately that they are extending Googlebot's ability to read and understand JavaScript. Especially links in JavaScript. Initially I thought this might simply be another effort to uncover the 'hidden internet' and index more, but comments from GoogleGuy in a WMW discussion about PR Hunting Techniques indicate a far more direct application.

(Regarding Javascript links instead of classical links.)
 
Of course, folks never know when we're going to adjust our scoring. It's pretty easy to spot domains that are hoarding PageRank; that can be just another factor in scoring. If you work really hard to boost your authority-like score while trying to minimize your hub-like score, that sets your site apart from most domains. Just something to bear in mind..

The particular application/focus of that post was only further underlined by his directly following post. Use the link above to view the comment in the original context of the discussion at WMW.

[Grumpus]

It's about time. 8)

G.

[Everyman]

Those GoogleGuy comments were made on 21 May 2002 -- almost two years ago. Too much has changed since then. Google PageRank broke almost one year ago, and has yet to recover. Spam in Google has increased over the last year. Hoarding backlinks works quite well. What's new is that you can now insert targetted anchor text in those backlinks and it works about five times better.

[digitalpoint]

The date of the thread aside, one problem I would see is anyone "savvy" enough to try and conserve PageRank (definitely not something I practice , BTW), is going to be smart enough to simply put their JavaScript in an external file that is not allowed to be spidered by Google. So unless the spider breaks the sites "terms and conditions" (for lack of a better term), and disobeys the robots.txt file, the fact that Googlebot can understand JavaScript becomes a moot point if it's unable to *see* the JavaScript.

- Shawn

[Black_Knight]

The statements about Googlebot being hyped-up to better understand javascript are all recent. SES NYC statements and fresher. It's only a statement of possible intentions to doing so that took going back to past stuff.

[digitalpoint]

Agreed, I've heard the same thing as well... I think it will be good for Google to better understand a page from a fundamental level, but I don't see how they would be able to get around PageRank conservation via robot exclusion.

- Shawn

[Black_Knight]

By viewing but not serving any .js files and .css files even if protected by the robots.txt file. I think that cloaking may become more popular again, especially in the idea of cloaking content within the scripts bins.

[digitalpoint]

Agreed... And the problem is there might be legit reasons to keep Google from spidering your scripts directory. So it would be hard for Google to apply any sort of penalty for just that.

- Shawn

[Peter (IMC)]

Agreed, I've heard the same thing as well... I think it will be good for Google to better understand a page from a fundamental level, but I don't see how they would be able to get around PageRank conservation via robot exclusion.

A site has 200 backlinks and 0 outgoing links. Does that make sense? A page can have 0 outgoing links, but a complete website? That wouldn't make sense.

And why would you want to conserve PageRank? How much PageRank will you conserve anyway? Will the difference be a PR of 7 instead of 5? I doubt it.

You get high rankings for beging a hub too,... linking out is not a bad thing and doesn't hurt your ranks.

And if you want to reduce the amount of pagerank you give out from your link page, then just link to the link page from one page only. (the home page, because if you don't link to it from the home page you will have troubles getting people interested in link exchanges).

This way you conserve your site's PR and at the same time you make Google happy.

Regards,

Peter

[Ruud]

Although the quoted statement reads a bit like a threat I suspect that what Google will do is follow links no matter what - and assign those linked sites their due PR.

Unless Google becomes like a new W3C and forces everyone to change a lot of practical systems this is the only thing they can do with this kind of information.

Why?

The web is full of link directories with some kind of redirect system to enable link popularity count. I myself use phpAdsNew to serve some Commission Junction ads (experimental). They go through my ad server so I can see which campaign works where and maybe why. Some *major* sites, thing about.com, use an on-the-fly top frame script to show outbound links.

You cannot punish all these valid uses of "hidden" linking, be it JS, CGI, PHP, ASP.

The solution would be to distinguish script from URL - something that should not be that hard. Then you can follow the URL as if it had been a regular one and assign it its proper additional PR.

Stated with much voice of authority but basically just my reasoning :wink:

Ruud

[viggen]

The question that arises is why do we need
http://www.google.com/search?q=allinurl:ad...1&start=10&sa=N those kind of url's in the index?

Hello btw to every one!

cheers
viggen

[cline]

The analogy I use to explan PR to clients is that PR works just like high school social structures, where PR = popularity. The most popular kids have lots of friends (inbound links), and particularly lots of popular friends. Kids are popular within certain cliques which are formed around various interests (e.g., sports, music, theatre...) so if you narrow down to an interest (aka search term) you can get a more relevant measure of popularity.

So, what does this have to do with hoarding PR?

While it's true that the most popular kids are likely to have far more kids who claim them as friends as vice versa, your popularity is pretty questionable if you claim to have no friends.

[Black_Knight]

There have been some wider citations and comments about this discussion elsewhere around the web of late. Some of those discussions have shown a small misunderstanding of what I've said. Especially concerning my opening paragraph:

There have been repeated small quotes and rumblings from Google of late that lead me to believe that using JavaScript to hide outbound links is no longer a safe practice.

Obviously I was rather more vague that I should have been in the words "no longer a safe practice".

Hiding links with JavaScript is no longer a safe (reliable) way of hiding those links from googlebot.

Accepting that premise, there are obvious further ramifications to the fact that Googlebot could soon be following links you thought you had hidden away. There's the issue of PageRank, and the separate issue of linking to 'bad neighborhoods'. All other effects of links upon ranking will likewise have to be assessed also for links that once would have been considered safely undetectable.

Hopefully that has made this thread a lot clearer to all.

[Tree]

Thanks for that insight Black Night!
While not 'directly' related because I don't hord links, nor care that to get a site listed at google you have to be an expert in the backlinks department (reluctantly being forced to become one).
A javascript redirect on a site for members only pages, that sends non- logged in users to the index, is then logically targeted by google as the wrong kind of redirect- or indications thereof. So now we know that legitimate uses for javascript links that we don't want google to follow- will be....
Missing the good ol days when we wrote scripts for sites for our visitors, instead of for google, or we wont have any. :?

[bragadocchio]

Hi Tree,

Welcome to the forums.

I tend to try to focus upon writing pages for people. Fortunately, many of the practices that make sites rank well also make them work well for people.

I've avoided javascript links because not everyone has java loaded on their computers. Even if Google starts following those, if there are still people surfing without java, I'd likely do without.

I do like your statement about writing sites for visitors. I think we can still do that.

[projectphp]

I don't hord links, nor care that to get a site listed at google you have to be an expert in the backlinks department (reluctantly being forced to become one).

I get the sentiment, but its not entirely accurate. Backlinks are useful, no doubt, but you only need one to be listed. To rank you may need more, but to get listed is easy.

Backlinks were vital long before Google came to prominence. Backlinks are the Internet's version of roads, and without them, you will get no traffic if there was no Search Engine. I think before search engines became so popular, links were even more aggressively sought out.

So now we know that legitimate uses for javascript links that we don't want google to follow- will be.

If you want to hide stuff on the net, password protection is the only way to go. Unfortunately, there is no other way. With teh prevalence of Toolbars, finding pages listed in Search Engines (particularly Yahoo's) will be even more common, unfortunately. Now more than ever, password protecting pages is vital!!!

[Black_Knight]

However, just for preventing spidering of certain content, the easiest solution may be to simply create a members sub-directory and then exclude spiders from anything in that directory via the robots.txt

[Tree]

Please forgive the length of this post, its difficult to be comprehensive without it, and I thought it best to include a little introduction since I am new here (thus future posts should be shorter).
User-agent: Googlebot
Disallow /members/
I have this in place as well as noindex, nofollow in the meta tags.
I didn't create the site, I am taking it over and on damage control phase. The pages are already in googles index of the site ( allinurl:sitename in a google search ).
Experience: Writing programs/coding with the only computer available then ( 4 stories of vacuum tubes at HoneyWell ) when Sergey Brin and Larry Page were still in grade school. Before the Internet was invented by Vint Cerf and Bob Khan, let alone the Web (Tim Berners-Lee).
The intent of the above creations, still fresh to me, I get confused in the details at times- like focusing on the changes taking place at google were the links issue (web) seems to take greater priority lately (florida upgrade), than document contents.
Logic says that if you have outgoing links (hub vs. authority), then you should also be giving credit at google to these documents/sites. Yet this also brings us back to the core issue (google to target link hoarders), of having to write for google more than your outgoing link partners, or even your visitors, or you risk not having any ( I don't have to like the truth, to see it ).
I don't want to be a SEO specialist, unless that means valid coding, 508 Federal accessability guidelines, and other related ways of putting your visitors and their operating systems first and using that platform to put clean and clear relevant content of interest to your visitors.
I want to provide that base for my clients, and allow them to focus, then, on being experts in the contents of the site. This works well at yahoo and MSN etc. But shows poor results at google, so my clients then care less about accessibility, quality coding, quality content, when these show to be taking second place at the #1 search engine- to counting the number of quality related incoming and outgoing links (ouch!).
Leaving the door open for a search engine that places consideration for quality of a site- more important than link farming.

Consider a specialist in content like a college professor, writing a document and people looking for it, can find it- even if he isn't a specialist in getting other popular sites to link to his- because he isn't a specialist in links (google), but of what the site or page is about (content vs. site creation for google). Which if this already were the case, there would be little link hoarding, google wouldn't be providing the motive for it?
Case in point, I'd rather have created referrance links above, but dont want to harm this sites google PR with outgoing links - thus defeating the purpose of the webs intention or harming this sites PR authority if I dont.