google browser emulation

mimic the user experience

[plumsauce]

From reading the messages here, I sometimes wonder if Google hasn't gotten just a little too clever for themselves.

By this, I mean that sometimes I see really complicated solutions applied to easy problems.

It would be so much easier for both sides if Google on its webmaster page would just commit, without prevarication, to indexing pages according to the user experience using a browser and respect RFC's

For example, there is a lot of concern here about the effects of duplicate content.

Much discussion is had about 301/302/401/404 result codes, and meta refreshes.

People worry about the effect of having the same content at www.example.com and example.com. This is almost always done in dns with a host record for one and an alias for the other. Strictly compliant with the dns rfc, afaik. The point is, this is easily discoverable by interpreting the dns query done by the spider. If a site has done this, they have declared openly that these are one and the same site. There should be *no misunderstanding* by Google that this is an attempt at presenting duplicate content. Just merge the aliased links into the canonical links. After all, a browser will grab the dns records and reach the same pages whichever host name was used and the user would neither care nor be resentful. No redirects should be necessary, Google should just respect the information that is already declared in the dns entries.

People also worry about the effect of zero delay meta refreshes. This, again, would not be a problem if the rule was "index what the user sees". It's a very simple rule. If the page redirects, or meta-refreshes, drop the source page and index the destination page. This is *exactly* what the user is going to see. Nothing more, nothing less.

As far as dns is concerned, I use the host record, alias record approach exclusively without trying to embed special case redirects. And if Google can't figure it out, well maybe they need to take a deep breath.

for the curious:

[faqs.org...]

+++

[Brett_Tabke]

domain.com returning the same page in 200 status as www.domain.com is not related to dns issues. Niether is www.foodomain.com returning the same page in 200 status as www.bardomain.com. Both can be innocent and both can be intential manipulations of G. That's what G is up against.

[DerekH]

plumsauce wrote

>>People also worry about the effect of zero delay meta refreshes. <<

Actually, I don't. One of my domains uses a zero-second meta refresh, and google presents the source URL with the destination content, exactly as the user sees.

My only problem with it is that changes to the destination content seem to take longer to be noticed by google than on other pages.

Google also indexes the destination page separately each time it does a deep crawl, and then conveniently chooses not to display it in the SERPS after about a day, which is rather convenient too!

DerekH

[plumsauce]

Brett,

That's right it is not a dns issue. It's a google issue, or rather my issue with google, if you prefer.

I did *not* mention foodservice.com and barservice.com
returning the same page. That is a totally different usage.

The specific example was www.example.com as an alias of example.com, that is an alias to a host in the *same* zone. Perfectly valid and common usage. Moreover, it is not unexpected to the user. This has nothing to do with aliases to hosts in different zones.

In any case, this is all part of a bigger picture. That being the seeming preference for the arcane over simplicity.

DerekH
>>One of my domains uses a zero-second meta refresh, and google presents the source URL with the destination content, exactly as the user sees.

I would consider that to be broken, as the user sees, after the refresh, the *destination* url in his browser address bar as well as the destination content. That's why I think the proper behaviour is to drop the source url/source content and index the destination url/destination content.

+++

[DerekH]

I wrote
>>One of my domains uses a zero-second meta refresh, and google presents the source URL with the destination content, exactly as the user sees.

plumsauce replied
I would consider that to be broken, as the user sees, after the refresh, the *destination* url in his browser address bar as well as the destination content. That's why I think the proper behaviour is to drop the source url/source content and index the destination url/destination content.

Not actually the case here. My site uses a domain name that forwards to my content on a different website. The way it is set up, the browser address bar doesn't change - the frameset is at the first domain and the content is at the other.
The redirect is completely seamless and transparent both to the user and to google.

Although maybe this is just clouding the issue!
DerekH