General Questions
Topic: myphpnuke not search engine friendly

Hey guys.

Well I have been looking into the myphpnuke and search engines, and how well they will work together. What I have discovered is that they dont go together.

All of the major search engines from my research do not acknowledge the "?" when calling articles, items, etc. This means that your site is not getting indexed, and all of the content is not getting into the engines. which is not a good thing when you have a content rich site.

Since my background is in website submissions and not PHP I have no idea what the solution is. However I know that all the sharp PHP people here probably will

Since it would appear that the php when it is passing parameters is what loses the spider, can we not do it differently. For example

http://www.site.com/myscript.php?myvalue=1

can it be changed to look like below which is spider friendly, and is apparently quite doable.

http://www.site.com/myscript.php/myvalue/1

If somebody can tell me how I can modify myphpnuke so that it calls the pages without the "?" get function I would appreciate it, either PM me or post here, or email me

Eman

Look forward to some feedback.

Its not a myphpnuke problem.
That is how php behaves. Also just about all the major programming languages behave in the same manner.

They all use ? and & in the url bar...

These are not static pages that a spider can look at and index. The pages change all the time.

In order to really work, the spiders has to have access to the database, which they dont.

There was an article I read a few years ago on this matter, and the conclusion of that article then was that spiders were obsolete in that manner because it physically cannot keep up with dynamic content.

-----------------
trini0...
mPN2 Test Bed

Thanks for the reply.

AS far as I can see you are correct. Back in the day the engines could not keep up with the dynamic content because it did not see the "?" get variable as a valid referrer, and so the engines have been programmed to ignore all urls that have it in it.

However it is my understanding that by creating variables for the $path_info you will be able to create spider friendly data. So long as we can call the dynamic data and retrieve it without passing a parameter in the HREF(which engines ignore) we should be good.

I am about 99% positive we can make myhpnuke spider friendly, but I sadly lack the talents to do it.

Look forward to more feedback
Eman

I think that is impossible. Without a parameter, then php (or any code engine) has no grounds to execute code according to the url.

I may be wrong, but if you or someone else could show me that it can be done, Ill look at it.

Quote:

On 2002-05-13 09:58, Emancipator wrote: So long as we can call the dynamic data and retrieve it without passing a parameter in the HREF(which engines ignore) we should be good.

-----------------
trini0...
mPN2 Test Bed

I am playing with it as we speak Trini0, I really would like to figure out how to get it to work, since I know its doable, from talking with friends who use other PHP scripts.

I guess the closest example I have is what I am working on right now on my site. Again I am not a php programmer, so god knows this is not right

currently when I click on an article link on my front page the link is:

http://edenenergy.com/html/search.php?query=&topic=31

It would be nice if it could look like this:

http://edenenergy.com/html/search.php/query/topic/31

I am not sure how this can be done, and anybody who visits that link will see you get a few errors in the page, none of the graphics load, etc. Quite obviously because the code was not designed to work that way.

I am going to keep mucking with it since I now have been emailed some code on how to change PHP_INFO which apparently makes it work.... I just need to find where to edit it..... can you tell I dont know PHP very well Look forward to feedback.

I know that there are a few ecommerce programs that I have trained on that are database driven but allow you to "publish" a html version of your site. The reason this can be done is two-fold:
1. After the site is designed or updated publishing to html reduces the cpu overhead.
2. Search engines.

Maybe there could be a easy way to publish a weekly html newsletter page likeGeekDudes Newsletter, and allow search engines to crawl and index that page. After several weeks you would have several pages that have been indexed and all would point back to your main page.

I haven't played around with the MPN newsletter feature but this might work.

Any other ideas.

Newt-dog

-----------------
Hey! visit PhantomCow.com sometime and Chat!

Newt thanks for the feedback. I wish i could myself figure out a good way to do it. I would like if I did not have to do a newsletter to build the sites popularity in the engines, especially considering its not overly relevant to the topic of the sites I design(newsletters are not common).

Newt do you happen to know if a spider can actually access the data in a PHP site, I know its dynamic, but I am trying to find a way that the spiders will actually follow it.


According to the site www.zend.com it is doable, and I have been doing a fair bit of reading there. I guess the big issue is, how easy is it, and is anybody willing to work on this project with me.

Thanks for the great feedback, looking forward to more.

Eman

Newt

I checked out that geekdudes site, I like the HTML idea, but I still wish we could link to our dynamic content in a fashion that the engines will actually index it. Rather then using the "?" get function. These parameter calls are an instant signal for the spider to stop indexing.

I like the concept Newt, just wish that the dynamic content that gets put in the site could be used to the advantage of the site. If we could develop a way to make it work myphpnuke would be used widespread in the corporate market, which is where I dwell.

Eman

There is one prog that comes to mind, that I tried about a year ago when I was young in PHP. Ez-Publish. Their url were like that, that confused me.
Where at zend.com are you seeing this information.

Let me look at it and see whats involved...

-----------------
trini0...
mPN2 Test Bed

Actually, this is NOT a PHP issue at all. It is a web server issue. Apache on Linux, for instance, stops parsing the URL as soon as it finds finds the first non-directory component.

Example:

http://test.mydomain.com/phptestdir/phptest.php/abc/123

Apache/linux stops when it gets to the phptest.php and executes it. The remainder of the line is available (I forget which variable: $PHP_INFO ???).

Unfortunately, this is NOT universal. M$oft IIS does NOT do this so it makes writing portable applications very difficult if you also want them to be search-engine friendly.

I believe that there is an option under IIS that DOES enable this kind of behavior, but that is an IIS system-level property... not one available to just anyone to enable.

Then there are the associated issues of whether you run PHP In-Process or as a CGI..... The parameters are NOT manifested in a consistent manner.....

I gave up on it for myOOPS.... at least for the short run.