homepage Welcome to WebmasterWorld Guest from 54.197.94.241
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member
Visit PubCon.com
Home / Forums Index / Marketing and Biz Dev / Link Development
Forum Library, Charter, Moderators: martinibuster

Link Development Forum

This 78 message thread spans 3 pages: < < 78 ( 1 [2] 3 > >     
Anti-Comment-Spam Tag Exploits
Set up Recip-checkers to look for rel=nofollow?
nuevojefe




msg:415448
 7:27 am on Jan 19, 2005 (gmt 0)

So who's thowing a little extra code in their bot to make sure a recip link is flagged for review if it contains rel="nofollow"?

Isn't it great to have yet another reason to crunch through the source of link partners? :-S

 

RonPK




msg:415478
 1:53 pm on Jan 21, 2005 (gmt 0)

I got this bookmarklet / favlet working in IE6, Firefox and Opera:

javascript:var a=document.getElementsByTagName('a');function nfc(){for(var i=0;i<a.length;i++){if(a[i].rel.toLowerCase()=='nofollow') {a[i].style.backgroundColor='#f00';a[i].style.border='2px solid #000';a[i].style.color='#fff';}}}nfc();

The results on Wikipedia pages that fed my site's PR are shocking.

Hanu




msg:415479
 1:59 pm on Jan 21, 2005 (gmt 0)

Nice! What does the!important do? How about
text-decoration:blink;

?

wruppert




msg:415480
 2:02 pm on Jan 21, 2005 (gmt 0)

Andre -

The value of "rel" can be a space separated list. Better to use this to match in FF:


a[rel~="nofollow"] {
...your favorite highlighting...
}

Macro




msg:415481
 2:15 pm on Jan 21, 2005 (gmt 0)

RonPK, could you explain? I don't follow js (call me Googlebot :))

creative craig




msg:415482
 2:28 pm on Jan 21, 2005 (gmt 0)

There is a post over at serountable where rustybrick had spoken to the head of search at Ask Jeeves regarding why they had not gone public with their support for the new tag:

It's a good idea and we wholeheartedly support the fight against spam.

On the other hand, when it comes to our particular engine, we didn't need to make a snap decision here. The nofollow idea is more urgent for Google (and those with similar approaches) than for Ask because they use global popularity (PageRank) while we use the local popularity approach pioneered by Teoma. I'm sure we'll add support for the new tag at some point in the near future if it makes sense. Blogs are a great source of authoritative information, regardless of their global pop ranking, which is what we pride ourselves on finding for our users.

Seems like they have it under control?

RonPK




msg:415483
 2:33 pm on Jan 21, 2005 (gmt 0)

How to create a favlet / bookmarklet:

Put the code in a href attribute
<a href="javascript:[..]">hi hi</a>
Make sure all the code is on one line.
Open that page in your fav browser.

case Opera: right click on the link > Add link to bookmarks > tick the box 'Show on personal bar'

case IE6: drag & drop the link into the Links toolbar

case Firefox: drag & drop the link into the Bookmarks toolbar

Clark




msg:415484
 2:42 pm on Jan 21, 2005 (gmt 0)

I see nothing on FF. Do I have to reboot or something (can't right now, too many windows open)

badtzmaru




msg:415485
 2:51 pm on Jan 21, 2005 (gmt 0)

You can only spoof an IP in situations where you do not need any data returned. Since you are trying to receive a web page in response, IP spoofing is not feasible. However, I am wondering if it would be feasible to examine the page out of the google cache. That seems like it would be the best way to see exactly what a given page is serving up to google.

notsleepy




msg:415486
 2:53 pm on Jan 21, 2005 (gmt 0)

Custom link checker is the only way to go. Why waste time looking at a link requester's website with special highlighting when your link checker has already scanned it for you and told you that a "nofollow" attribute exists, or the page is disallowed in robots.txt, or the page is cloaked for the general public, or a link to your site exists but its commented out, or there is no link to your site at all!

The point is, its much easier for me to told from my app that this link request has a status of "Reciprocal Exists with nofollow Attribute".... DELETE

My link checker has saved me tons of time.

notsleepy




msg:415487
 2:57 pm on Jan 21, 2005 (gmt 0)

I am wondering if it would be feasible to examine the page out of the google cache. That seems like it would be the best way to see exactly what a given page is serving up to google.

Most sites that do cloaking place the meta "noarchive" tag so Google will not cache the cloaked page.

Conard




msg:415488
 4:15 pm on Jan 21, 2005 (gmt 0)

Checking links with a link checking script or program is or was a good idea.
What if all of the onsite links to the link page containing YOUR link are behind rel="nofollow" links?
Your link would check as good but the page will not be indexed.

graywolf




msg:415489
 4:45 pm on Jan 21, 2005 (gmt 0)

Seems that Wikipedia have fully adopted the nofollow tag and have placed it on all outbound links!

Now there's a change that will send little ripples throughout the entire web.

PatrickDeese




msg:415490
 4:50 pm on Jan 21, 2005 (gmt 0)

RonPK - I just wanted to say thanks - that little favlet is great.

Who's next? DMOZ?

notsleepy




msg:415491
 4:52 pm on Jan 21, 2005 (gmt 0)

What if all of the onsite links to the link page containing YOUR link are behind rel="nofollow" links?
Your link would check as good but the page will not be indexed.

You just made my life much more difficult Conrad. :) You are right. Now I have to figure out how to check all onsite links to the links page.

Clark




msg:415492
 5:21 pm on Jan 21, 2005 (gmt 0)

All this will do is separate the men from the boys. It will be a little more work to be on the higher level, but the pros will probably find life easier.

DaveAtIFG




msg:415493
 6:50 pm on Jan 21, 2005 (gmt 0)

Exploits
The no follow attribute is already showing up on the far reaches of the internet. Unfortunately, most of those sites have nothing to do with blogs, guestbooks, diaries, wikis, or forums.

Initially, Google could use the tag to identify heavily SEO'd sites I suppose. :)

iblaine




msg:415494
 6:58 pm on Jan 21, 2005 (gmt 0)

Seems that Wikipedia have fully adopted the nofollow tag and have placed it on all outbound links!

A little piece of me died when I heard the news, hah.

On a more serious note, I'm worried about link partners that abuse the nofollow tag. It would be very easy to cloak the nofollow tag and make it nearly impossible to detect.

Macro




msg:415495
 7:04 pm on Jan 21, 2005 (gmt 0)

>> A little piece of me died when I heard the news

My favourite "other" source for one way links doesn't seem to be following wikipedia :)

Clark




msg:415496
 7:11 pm on Jan 21, 2005 (gmt 0)

Too bad I never heard of wikipedia being a good source until now. Hey, where do I get plugged in haha?

bostonscott




msg:415497
 7:28 pm on Jan 21, 2005 (gmt 0)

I'm not suggesting that this will happen, but entertain this thought experiment (from Google's perspective)...

Problem: Too many webmasters are manipulating our PageRank technology by engaging in link swaps for the sole purpose of increasing their rankings. They are getting increasingly savvy at doing this.

Solution: Let's create a Prisoner's Dilemma of sorts. Let's allow webmasters to post some hidden data on their site (that competing webmasters cannot see), which allows them to stop our BOT from giving positive points to those sites that they link to. This way, when one website engages in a link swap with another, one will not really know whether the link partner is giving out a link, or just taking the inbound link. We'll even play with the back links look-up to prevent sites from verifying the honesty of their link partners. This will increase the number of total link requests from various webmasters initially because sneaky webmasters will attempt to obtain thousands of one way in-bound links. Then, everyone will get frustrated because so many sites won't reciprocate, and bogus link swaps will die down... and links will become a better indicator of site popularity.

Maybe this is unrealistic, but when I heard about this Google/MSN/Yahoo change, I wondered if this was their first step towards ending link abuse. In the long-term this would be good, because it would level the playing field and force websites to compete for inbound links by offering valuable and unique content. Bottom line: The quality of user experience goes up, search engine usage stays strong and grows, search engines make more money, advertisers get more business… market forces leads to more efficiency.

martinibuster




msg:415498
 7:48 pm on Jan 21, 2005 (gmt 0)

>>>I wondered if this was their first step towards ending link abuse.

If links were a commodity before, this nofollow thing just created a bigger market. Makes clean inbounds more valuable.

SuzyUK




msg:415499
 8:18 pm on Jan 21, 2005 (gmt 0)

I see nothing on FF. Do I have to reboot or something (can't right now, too many windows open)

Yes you do if using the userContent.css as a persistent solution.. see this thread [webmasterworld.com] for more details of two ways to do this in Firefox.

Makes clean inbounds more valuable

is this is the "time" they are playing for until xlinks are adopted/supported more widely?

Suzy

nuevojefe




msg:415500
 8:30 pm on Jan 21, 2005 (gmt 0)

Most sites that do cloaking place the meta "noarchive" tag so Google will not cache the cloaked page.

Exactly.

So, let's just say this:
Your script must check google for cache:URLofRecipLinkPage then if found, check for your link on that page. If it is found but with rel=... then flag it. If it is not found at all then it should be rescheduled for a later check. If the page never shows up in cache (30 day limit or whatever) then they're either cloaking or the page your link is on isn't getting any internal link love. Ditch 'em.

This takes longer and sucks because now you're involving google and automated queries for something that you could previously do without involving google.

graywolf




msg:415501
 8:47 pm on Jan 21, 2005 (gmt 0)

If this thing went all the way to the nth degree the only people left with clean links would be SEO's with IBLN's.

[edited by: graywolf at 9:27 pm (utc) on Jan. 21, 2005]

ogletree




msg:415502
 9:18 pm on Jan 21, 2005 (gmt 0)

I wonder if DMOZ is thinking about using this. That would be interesting. Of course you have to remember that the amount of web publishers that build sites with an se in mind is a tiny tiny fraction of web publishers.

SuzyUK




msg:415503
 9:27 pm on Jan 21, 2005 (gmt 0)

graywolf.. yes could get weird :)

then the SE's could theoretically just reverse their algos ~ if they thought the nth degree outweighed the previous problem, of course then we're back to where it started...

caveman




msg:415504
 10:01 pm on Jan 21, 2005 (gmt 0)

Is there *anyone* out there still wondering whether or not this was a bad idea or not?

I keep wanting to put my tin hat on, 'cause from the moment this was announced, the only way it make sense is if G has some insideous S.P.E.C.T.R.E.-like plan, with a goal of:

a) more easily ID'ing spammers,
b) reducing the value of links so they could later say that they are reducing their emphasis on PR (and blame it on the widespread use of nofollow),
c) ruling the world,
d) all of the above.

Not sure of how they plan to achieve option c) yet.

nofollow...nofollow...nofollow...nofollow...nofollow...nofollow...nofollow...nofollow...nofollow...nofollow...nofollow...nofollow...nofollow... ;-)

Kirby




msg:415505
 11:47 pm on Jan 21, 2005 (gmt 0)

>Not sure of how they plan to achieve option c) yet.
They are starting small... invaded Tahoe first.

saoi_jp




msg:415506
 11:09 am on Jan 22, 2005 (gmt 0)

If "nofollow" attribute is indeed for links we don't necessarily trust...then aren't we saying that the nofollow links are not trustworthy? i.e. out of our control? So wouldn't a page full of "nofollow"s look like a link free-for-all?

The biggest thing the announcement did for me was to let me know that the other two SEs also have PageRank values assigned and weighed. I guess I always knew this, but I never spent time trying to calculate it. (If this new attribute was *just* about Google, then why would the other two jump on the bandwagon as well?)

Clark




msg:415507
 4:46 pm on Jan 22, 2005 (gmt 0)

Maybe this whole thing was an attempt to get Brett to allow links in webmasterworld ;)

rytis




msg:415508
 5:05 pm on Jan 22, 2005 (gmt 0)

Y! & MSN joining G does not necessarily imply they are using mechanism similar to PageRank in their algo.

Few other link benefits, that may be affected by new tag (and may or may not be used in algo's) are: simple link count to page and anchor text affecting rank for anchor keywords.

I understand it as a tool into hands of site-owners, allowing to suggest to SEs that the link should not receive a vote for SE ranking. How SEs implement this, well I guess they are not going to tell us.

This 78 message thread spans 3 pages: < < 78 ( 1 [2] 3 > >
Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Marketing and Biz Dev / Link Development
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved