[TYPO3-core] Status of RFC: #9046: Nested domain linking with domain record on root page

Ernesto Baschny [cron IT] ernst at cron-it.de
Thu Oct 8 09:02:41 CEST 2009


Tobias Liebig schrieb:

>>> Say we have a tree like this:
>>>
>>> yahoo.com
>>>   - finance (no special subdomain, accessible only through yahoo.com)
>>>   - weather (weather.yahoo.com, no force flag set)
>>>     * europe
>>>     * asia
>>>   - sports (sports.yahoo.com, force flag set)
>>>     * soccer
>>>     * tennis
>>>
>>> Now say you are on the european weather page using
>>> yahoo.com/europe.XYZ.0.html, and want to link to the finance page: You
>>> wouldn't have to prefix any domain, because you are already on
>>> yahoo.com.

>> But you can prefix it with the domain without doing any harm. So that
>> the europe-page is always the same in the cache, no matter how you
>> access it (yahoo.com/europe.XYZ.0.html,
>> weather.yahoo.com/europe.XYZ.0.html). Links pointing to some page that
>> doesn't share our own sys_domain get the prefix, the ones that share our
>> sys_domain (in this case the weather-start-page, europe and asia) get
>> relative links.

> If you call "weather.yahoo.com/europe.XYZ.0.html" you don't want to left
> the site "weather.yahoo.com".
> Even if the target page (e.g. asia) is (technically) reachable by
> "yahoo.com/asia.XYZ.0.html" *and* "weather.yahoo.com/europe.XYZ.0.html",
> you do not want to create a link to an external page (yahoo.com) if the
> target is reachable within the current domain "weather.yahoo.com"

This link (europe > asia) is *always* relative because of the "no force"
flag, which means it won't ever change domains here.

> Search Engine may value an external link differently then an internal link.
> 
> On the other hand your right, it *may* do no sense to make one page
> reachable through two different domains (duplicate content?).
> 
> As both points are right (as i think), we decided to let the user choose
> and implemented the "forced flag".

In my view this could be solved as creating a redirect to the right
domain as soon as you hit a page in a domain that is not forced that has
a "brother" that is forced. Once you redirect to that forced domain, all
links in this subdomain are relative, and again links pointing to other
places of the site are "external".

Anyway most of us will use realurl and have a base href set to the "main
domain" anyway, so the landing page domain is only hit once.

>> True, but I really don't see many people using not "forcing" a domain,
>> as this will give penalities in Google (SEO), because you are providing
>> the same content through different URLs. So this cannot be the
>> recommended behaviour anyway.
> 
> i think there are two use cases of domain records (after this issue s
> solved)
> - if you set the forced flag, you creating "sub sites" which works as a
> standalone page (like pages below "weather" are always reached by
> weather.yahoo.com)

In the example from Ingmar "weather" is the non-forced one. You mean
"sports" (forced), right?

> - if you don't set the flag, you just create an "entry point" in a
> larger page.
> 
> many pages may want to use the first case. maybe we should set the
> forced flag by default? but this may change the behaviour of the domain
> records..

SEO wise it might be better to do a redirect to the "real" domain on the
"non-forced" case, which means you reach the landing page through the
weather.yahoo.com domain, but then get redirected to the
"yahoo.com/weather" address right away.

>>> You have to keep in mind that one page can be accessed through different
>>> domains.
>>
>> I already did that, and I think that we only need to differenciate it
>> like I explained above so that we always get the same page in the cache
>> no matter how it was accessed.

> and thanks, Ernesto, for troubling on this issue here. This will help to
> get another point of view on this issue and help to do the things right :-)
> 
> If somebody else want to comment: do not hesitate.

I wanted to intervene in the discussion, because I feared that we will
end up caching more than we want and if we cache the same content for
every domain you have. We end up having the same "problems" we have with
the "baseUrl" and the conditions you need to use to set it depending on
the domain you are.

Or maybe both problems can be combined in order to have a SEO optimized
experience without any configuration need. I think TYPO3 should provide
this out-of-the-box and not come with some setup that is not SEO attractive:

- avoid duplicate content
- redirect with 301 if you reach through a domain that is only there to
reach an entry point instead of showing the page
- keep links inside the "forced domain" relative
- always prepend the "other domain" if you leave the forced-domain-space

And then there is the backwards compatibility "cow". But we can avoid
that if we simply add a new configuration option to use the "new style"
and have it enabled on new installs by default. Old sites work as before.

Cheers,
Ernesto


More information about the TYPO3-team-core mailing list