37 replies to this topic

[Bootfit]

Mmmmmmmmm

Imagine 150 words amongst say 5k of code and then place the same 150 words among 5Mb of code.
Is that statement still true ???

Completely agree with the sentiment, couldn't be bothered arguing the toss though.

[piskie]

I don't really have an opinion on code

That's where I come from. However it's such an easy ask to keep code bloat down that I play it that way. Not a religion, but something I allow for just in case ..............

[Jill]

Mmmmmmmmm

Imagine 150 words amongst say 5k of code and then place the same 150 words among 5Mb of code.
Is that statement still true ???

My opinion is that it would still be true.

[piskie]

While we're there, is there any influence on Spidering Frequency by massive and complicated Code Bloat.

[chrishirst]

Nope. Crawlers request a URL, server responds with a data stream of a hundred/hundred and fifty Kb a second or more.

Crawl frequency is more about the links that point to a particular URI.

[piskie]

I was thinking on the lines of Googlebot/s retrieving pages by the millions and billions. Therefore speed and resources become a factor in determining the shopping list cyclic time. Stripping out and storing the wanted bit, must take more resources and processing time if there is huge bloat and Javascript stuff.

That's why I wondered if Google might just be a bit influenced by retrieval times of the Larger File sized pages with little text to fetch.

[Purple3]

Ive got a couple of sites which have suffered significantly over the last 2 months - dropping 20 places or more .

I think I know why and ive put it down to the "over use of galleries" - The galleries are large say 60 images plus in each niche sector - Alt tags have been used . The Galleries are always in the top 10 viewed pages so didnt seem any harm though on reflection Google might now be viewing them as over optimised . Does this logic sound right & if so what's the best remedy . I dont want to drop the Galleries as visitors like them

[qwerty]

Images should always have alt attributes (I can't believe I've actually beaten Chris to the punch on this, but it's not a tag). I guess the question is whether you abused those attributes by stuffing them with keywords, or if there's something else you did on the page, or if what you're seeing in your rankings has nothing to do with on-page optimization at all.

Can you give us a few examples -- just describe a few of the images and then tell us what you've got for their alt attributes?

By the way, when you say that these sites have suffered by dropping in the rankings, has this had any effect on traffic?

[Purple3]

traffic has been affected - by about 10% - cant give any further info though other that each image had a short description ie product , product feature , location . The Image Attribute had a shortened version of this description -

When the Galleries were set up it was clear they were working in SEO terms

[mge]

Could there be a penalty for simply optimizing?

I'm the webmaster of a site for a small company, and a week or two ago I carefully went through all the titles, meta descriptions and keywords. I decreased the length of the titles and descriptions, and reduced the number of keywords so that all of them really captured the essence of the page. The following week our page rank dropped from 3 to 2.

It might not be directly related, but I do wonder. Also, our website is a simple HTML website, and although I keep it updated constantly, I wonder if it would be better to switch to a CMS like Wordpress.

[chrishirst]

I was thinking on the lines of Googlebot/s retrieving pages by the millions and billions. Therefore speed and resources become a factor in determining the shopping list cyclic time. Stripping out and storing the wanted bit, must take more resources and processing time if there is huge bloat and Javascript stuff.

Way back last century that was the case. Only the first 100/110 Kb of a document was retrieved from the server.

That's why I wondered if Google might just be a bit influenced by retrieval times of the Larger File sized pages with little text to fetch.

But how would they "know" there was "little text" until they had retrieved it?

The server reports the size of the datastream in the response headers so a user agent knows when to disconnect, but it doesn't report what percentage is text and what percentage is markup or script code.

Edited by chrishirst, 09 August 2012 - 09:45 AM.

[chrishirst]

Could there be a penalty for simply optimizing?

I'm the webmaster of a site for a small company, and a week or two ago I carefully went through all the titles, meta descriptions and keywords. I decreased the length of the titles and descriptions, and reduced the number of keywords so that all of them really captured the essence of the page. The following week our page rank dropped from 3 to 2.

It might not be directly related, but I do wonder. Also, our website is a simple HTML website, and although I keep it updated constantly, I wonder if it would be better to switch to a CMS like Wordpress.

Assuming you mean PageRank rather than "where your URLs rank in the results"

PageRank (SGB PR) is absolutely nothing to do with any of that.

It might not be directly related, but I do wonder. Also, our website is a simple HTML website, and although I keep it updated constantly, I wonder if it would be better to switch to a CMS like Wordpress.

What creates your documents and URLs is nothing to do with it.

All a CMS does is make it (sometimes) easier to create and manage your site and its content.

Edited by chrishirst, 09 August 2012 - 10:19 AM.

[mge]

Assuming you mean PageRank rather than "where your URLs rank in the results"

PageRank (SGB PR) is absolutely nothing to do with any of that.


What creates your documents and URLs is nothing to do with it.

All a CMS does is make it (sometimes) easier to create and manage your site and its content.

Thank you for your comments. I think I was worrying a hair too much about the silly green bar, as well as our Alexa ranking, which both seem to be heading in the wrong direction. But where it really counts, our website seems to be doing well. We're ranking high for keywords that reflect our content. I find that we rank especially well with photos posted of our projects (we're a metal fabrication and millwright company).

[piskie]

Clearly the first retrieval is as you say Chris, however if the Code : Text ratio is abysmal, then could it just have some influence on the frequency of future visits.

[chrishirst]

So according to your theory, the URLs that make up a "photo gallery" of a website will probably get crawled at a lower frequency than the content URLs.