| 7:54 am on Jan 19, 2005 (gmt 0)|
| 7:59 am on Jan 19, 2005 (gmt 0)|
...who then cloak to serve you a different page without the nofollow on it ;)
| 8:03 am on Jan 19, 2005 (gmt 0)|
Better one. :-)
| 8:08 am on Jan 19, 2005 (gmt 0)|
...so my bot has to masquerade as g-bot...
Anywho, what are your thoughts on how the content of these links will now weigh in the on-page?
If for instance a scraper site uses this tag in its links to the results it has scraped in order to not pass PR to those sites and instead pass the PR on to other internal pages (or wherever he/she wants), will that reduce any positive effect on rankings that are gained by having these links pointing outward? Conversely, will this not just prevent pr leakage (and ranking benefit for the link receiver) but also prevent a site from being penalized for linking to a bad neighborhood?
| 8:13 am on Jan 19, 2005 (gmt 0)|
>>>but also prevent a site from being penalized for linking to a bad neighborhood?
Wow, you're really thinking tonight!
What that means is that you can do a massive link campaign, drop in the nofollow tags, and receive credit for a ton of inbounds that may be perceived as one ways.
I wonder if the search engines have anything in place to handle that.
| 8:18 am on Jan 19, 2005 (gmt 0)|
It would get even better if you could link to authorities and start getting some hub weight but not pass any PR or further strengthen those positions of theirs that you so covet.
I don't think this attribute will be that naively implemented though.
| 8:21 am on Jan 19, 2005 (gmt 0)|
|Wow, you're really thinking tonight! |
yea, unfortunately 12:00am is about when the juices start flowing. That explains why I only get about 4-5 hours sleep every night. I was thinking just this morning (nice 45 minute client call at 7:45am) how my New Year's resolution should have been to attempt to actually go to sleep the day before I wake up each morning instead of that same day.</off-topic>
| 8:12 pm on Jan 19, 2005 (gmt 0)|
I think that this would make a nice tool by itself. Especially for people who are in competitive industries and have thousands of link partners, you just can't hire someone to check the code.
A handy utility would be a compact application like Xenu Link Checker, but one that checks for that specific tag in the code then creates two reports.
The first report would be macro, and it would search all of your backlinks and report the websites containing that code.
The second report can be a micro report that lists all links for a particular website that carry the nofollow tag.
| 8:15 pm on Jan 19, 2005 (gmt 0)|
|...so my bot has to masquerade as g-bot... |
And masquerade it's IP address...
| 8:30 pm on Jan 19, 2005 (gmt 0)|
I they are smart (they are of course), they will figure out a way to recognise reciprical links where one partner has the 'nofollow' attribute.
Site A: <a href="site b.com">
Site B: <a href="site a.com" rel="nofollow">
Although B will never pass PR to A, maybe the link from A to B will be devalued. Perhaps it should be weighted as somewhere between a one-way link and a reciprical one?
I think (hope) that Google have thought this through. I am personally not convinced that the "nofollow" attribute means that the link is going to be ignored entirely. It may not pass PR, but it's existance might downgrade the PR transfer from the same site linked out to.
This is going to be interesting.
| 8:32 pm on Jan 19, 2005 (gmt 0)|
|And masquerade it's IP address... |
Thats not possible right? Or is it?
| 8:39 pm on Jan 19, 2005 (gmt 0)|
spoofing ips is possible. and usually illegal...
| 8:50 pm on Jan 19, 2005 (gmt 0)|
|spoofing ips is possible. |
Feel free to post your solution for masquerading as the Gbot from a valid G addy across the various relays that make up the Internet.
| 8:52 pm on Jan 19, 2005 (gmt 0)|
I don't think there exists a person who would post such a thing on any public forum...
| 8:53 pm on Jan 19, 2005 (gmt 0)|
LOL - nice try, continue with your purely theoretical conjecture though, that's what this thread is all about.
| 8:58 pm on Jan 19, 2005 (gmt 0)|
No, it's what your tangent is about.
| 10:43 pm on Jan 19, 2005 (gmt 0)|
I was thinking the same thing his morning. How does an outbound link with the nofollow tag affect your hub and authority scores.
I'm also amused that on Google's blog none of the links go straight out. MSN's blog links out striaght as does Yahoo's blog
| 1:20 pm on Jan 20, 2005 (gmt 0)|
Any off the shelf tools at this level?
Or is everything custom?
| 2:37 pm on Jan 20, 2005 (gmt 0)|
|I'm also amused that on Google's blog none of the links go straight out. MSN's blog links out striaght as does Yahoo's blog |
They must have heard about PR leakage, next they'll be submitting to the search engines!
| 2:46 pm on Jan 20, 2005 (gmt 0)|
Any off the shelf tools at this level?
Or is everything custom?
Well there are a few different link exchange management scripts but for the most part, to get the security measures needed, you pretty much have to custom build it.
Most of the scripts out today are all encrypted so you can't just hack them and add the regex to pick up the new attribute, and most of the scripts out don't really do a very good job of looking for the other things webmasters try to use to cheat.
| 6:18 pm on Jan 20, 2005 (gmt 0)|
Would it be possible for someone to write an extension for Firefox that would identify these new links and give them a special highlighting or designation of some kind?
Then you'd just have to check the page and look for the flag, OR check google's cache and see if it does the same thing.
| 6:49 pm on Jan 20, 2005 (gmt 0)|
Remember that "nofollow" has been around for years on the page level (meta).
Cloaking for it has been used and will continue to be used forever.
Next pagerank update is going to be really interesting as excited as y'all are ;)
| 7:13 am on Jan 21, 2005 (gmt 0)|
Oh they very much do pay attention: [google.com...]
scroll down to the middle - it will accept the meta for "robots" or "googlebot"
| 10:11 am on Jan 21, 2005 (gmt 0)|
Specifically on nofollow (i know they acknowledge noindex) I just saw a mention (that suprised me) that said they didn't.
I'll try to find it.
| 12:20 pm on Jan 21, 2005 (gmt 0)|
Seems that Wikipedia have fully adopted the nofollow tag and have placed it on all outbound links!
IMO defeats the point of it being an authoritative site if it does not pass that authority on.
| 12:40 pm on Jan 21, 2005 (gmt 0)|
> identify these new links and give them a special
> highlighting or designation of some kind?
Use Opera and write a custom CSS. With the press of the control-G key you could toggle in your sheet and easily highlight the no follow attribute.
| 1:04 pm on Jan 21, 2005 (gmt 0)|
|identify these new links and give them a special highlighting or designation of some kind? |
Bookmarklet outlines <a>'s with rel="nofollow":
(I posted this in supporters too, but it probably makes sense to cross post to the public side)
| 1:41 pm on Jan 21, 2005 (gmt 0)|
If someone is going to go to the trouble of IP-driven cloaking to hide a rel=nofollow attribute, there's nothing to say that they couldn't feed the SEs a page with a NOINDEX on it, or a page without your link appearing at all.
Personally, I don't believe that Google et al will treat the "nofollow" attribute as meaning "this link doesn't exist". I can certainly imagine them using it in identifying reciprocal links, if that is important to their algorithm. And if some sites seem to have an unusual number of nofollow links, I can imagine some kind of PR calculation adjustment.
IMO, the only way for a webmaster to ensure that outbound links aren't considered in some manner or other is to hide them completely by cloaking and feeding the SEs pages without the links. Whether this is worth the risk is questionable and depends on what the site is trying to accomplish.
| 1:44 pm on Jan 21, 2005 (gmt 0)|
If you're using Firefox, add the following lines to your usercontent.css:
| This 78 message thread spans 3 pages: 78 (  2 3 ) > > |