Read on Twitter
Hi guys over at @screamingfrog!
Been readin& #39; your article about your link score calculation and how it relates to PageRank and I have a couple of concerns. Let me just list& #39;em for you:
Continues
https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Down pointing backhand index" aria-label="Emoji: Down pointing backhand index"> https://twitter.com/mjcachon/status/1255740224895844353">https://twitter.com/mjcachon/...
Been readin& #39; your article about your link score calculation and how it relates to PageRank and I have a couple of concerns. Let me just list& #39;em for you:
Continues
1. You do not seem to take into consideration links& #39; position in the DOM.
Not all links in a given url "weight" the same. PageRank implies quantity and quality (otherwise there& #39;d be no need to iterate) dimensions. Some kind of correction should be applied here.
Continueshttps://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Down pointing backhand index" aria-label="Emoji: Down pointing backhand index">
Not all links in a given url "weight" the same. PageRank implies quantity and quality (otherwise there& #39;d be no need to iterate) dimensions. Some kind of correction should be applied here.
Continues
2. You don& #39;t seem to take into account external links for "your PR" calculation.
Then, unless I have no backlinks to my site, the calculation is flawed. The more backlinks, the more flawed.
You need both internal & external links to calculate any url& #39;s PR.
Continueshttps://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Down pointing backhand index" aria-label="Emoji: Down pointing backhand index">
Then, unless I have no backlinks to my site, the calculation is flawed. The more backlinks, the more flawed.
You need both internal & external links to calculate any url& #39;s PR.
Continues
3. You say: damping coefficient controls the amount of PR that is passed.
Damping factor actually affects how much PR is passed, but does not control it.
DF is a randomization margin allowed to the algo to solve dangling nodes & disconnected graph problems.
Continueshttps://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Down pointing backhand index" aria-label="Emoji: Down pointing backhand index">
Damping factor actually affects how much PR is passed, but does not control it.
DF is a randomization margin allowed to the algo to solve dangling nodes & disconnected graph problems.
Continues
4. Your calculation performs 10 iterations in total.
Why 10? AFAIK there is not a fixed amount of iterations until the result becomes stationary.
Is that a performance related limit?
Final thoughtshttps://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Down pointing backhand index" aria-label="Emoji: Down pointing backhand index">
Why 10? AFAIK there is not a fixed amount of iterations until the result becomes stationary.
Is that a performance related limit?
Final thoughts
Don& #39;t get me wrong, I do love your software. It& #39;s great.
I also understand that it is almost impossible to have a somewhat accurate approximation to real PR values. Way too complicated. And you effort deserves all my credit, honestly.
However...https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Down pointing backhand index" aria-label="Emoji: Down pointing backhand index">
I also understand that it is almost impossible to have a somewhat accurate approximation to real PR values. Way too complicated. And you effort deserves all my credit, honestly.
However...
I just feel like users deserve to have a more detailed explanation of this metric.
If they rely on it as a form of PageRank equivalent, they will be misled.
Yes, you call it "internal" PageRank for a reason. But it doesn& #39;t work like that in the real world.
Just my thoughts.
If they rely on it as a form of PageRank equivalent, they will be misled.
Yes, you call it "internal" PageRank for a reason. But it doesn& #39;t work like that in the real world.
Just my thoughts.
Oh, and by the way, I might as well be mistaken. Please correct me as necessary, of course.
Thank you!
Thank you!
