ZydoSEO - 2:43 pm on Aug 1, 2012 (gmt 0)
When an article is published on a WP blog, I think authors frequently go back and add links to older articles/posts as they add more and more related content. Perhaps an original post about topic X mentioned topic Y in passing but didn't discuss topic Y in detail. The author might later write another post about topic Y in more detail and go back to the original post on topic X and link to it so that those readers of the old post can get more detail on topic Y from the new post if desired. What is unnatural about that? It's how a web site's content naturally evolves over time.
Posts on forums often get indexed in a matter of a few minutes. So if I post a question, it gets indexed, and then others respond with a great discussion including links to very relevant pages to support the discussion around that question, what is more valuable? Just the original question? Or the subsequent discussion including related links within that discussion to support what is being talked about?
What if a forum thread first gets indexed after 2 not so great comments... but then there are many more comments that follow later with great content and links? Are those 2 comments and any associated links contained in them in some way more valuable to users of search engines than the 100 great comments and associated links that followed simply because they were there at the random time that the engine crawled the content?
The very nature of growing a web site by adding additional pages over time absolutely demands that you add links to older, previously crawled pages. Sometimes these get added to navigational structures that make up part of the architecture of the site. But often times, not all pages on a site can be included in navigation, sub-navigation, etc. The ONLY way to make these pages crawlable is to add contectual links to them on pages previously crawled. Does it seem logical or fair that every page added to a site after the site's initial crawl have it's internal inbound links devalued?
While I agree that there "might" be certain mediums like newsletters, press releases, etc. where it may be less natural to add links later, these are likely corner cases and do not apply to most of the web. And any rule to deal with certain corner cases (i.e. a very small percentage of pages on the web) applied broadly to an entire index of the web just asking for trouble.
I could also see that there might be certain situations where they might want to implement such a rule, possibly when a particular blog or forum is known for being exploited by spammers to get followed links and the webmaster refuses to do anything about it... But in those cases it seems it would just be easier to devalue ALL outbound links from such a site rather than trying to determine based on a page's initial crawl time which to count, which to devalue.
Such a crawl time relative rule would essentually devalue all UGC and links within UCG (blog comments, forum discussions, reviews, etc.) added by users after an initial crawl of the URL. It would promote maintaining old, static, out-of-date content on the web and site designs because changes to the content, contextual links, and or internal linking structures would certainly negatively affect rankings. Essentially it would say don't publish a new site or page/post/article until it is absolutely perfect and never change it because changing it later will screw you.
I don't see any such rule as doing anything to "give users better search results" which should be the goal of any algo rule like this. And since very few pages on the web (I would guess some miniscule, tiny, tiny fraction of 1%) remain static forever after being initially crawled, such a rule would do far more harm than good to search results.
But hey... I could be wrong... I have seen engines implement stupid stuff before.