Gallery2 and SEO - Tips and suggestions

Updated Feb-05-2007

I don't know if all of this has been covered (can't get the new search to work for me) but I have some general suggestions for the SEO (search engine optimizng) of Gallery2.

In no particular order:

1)Title bar

The title of each album should be in the title bar along with the photo title (when viewing photos). This can be accomplished by modifying the <title> tag in the appropriate tpl page.

2)Mod_Rewrite

Use dashes (-) instead of underscores (_), this way older search engines (and some new ones) will treat the url as separate words.

3)index.php

Apply the mod to use index.php instead of main.php. This is very important! Redirects kill rankings (especially 302's). Also, without this your site will never get assinged the appropriate Page Rank from Google. Update: This mod doesn't work anymore so if you are able to, which you most likely are, modify httpd.conf to add main.php before index.php and all should be well with the world once again.

4)Correct error pages

The default error page generates an HTTP 200 response code (OK), the correct codes should be used or the spiders will just keep indexing the error page. Update: This was fixed in the Gallery code, I believe, correct me if I'm wrong.

5)Duplicate content

Having multiple sizes of an image is great, but to a spider, it looks like copies of the same page. It's best to modify the tpl's to load the full size images in a new window, without the template includes. Duplicate content penalties are common and VERY hard to fix.

6)Album title when viewing photo's

Having the album title on the page with the photo title will GREATLY increase keyword density and weighting.

7)<h1> and <h2> tags

Adding the <h1> tag and formatting to the CSS and using it for the album and picture titles when viewing albums and photos will add to keyword weight. May help, may not but it's worth a shot.

8)Slideshows and other "features"

See duplicate content above.

9)File and album naming

Prepare a naming convention and stick to it throughout your gallery. Photo and album titles should be clear and concise. "Won't someone please think of the spiders?"
Example: Album name: Album-One Album title: Album One Photo name: Photo-001 etc.

10)Preparation is better than perspiration

If you're going to do any of this, do it BEFORE you make your site available to the outside world. Once the search engines latch on to something, it's difficult to make them let go. Especially if it's wrong. This is important so that you don't end up with a bunch of non-existent pages in the search engines when you change the names 20 times.

There is probably more that I'm missing here, but that covers the most important that I can think of and have done to my site to aid the little spiders in moving around my gallery.

If you have any to add or disagree with any of the above, please post and explain your position.

Added:

I've tried to follow this thread for the last year or so, but for some reason it only emails me updates when it feels like it heh, so here goes some Q&A

Question: Index.php always showing instead of mydomain.com/
Answer: There was a change to the Gallery code that ruled out the hack mentioned in the first itteration of this post, now your best bet is to modify httpd.conf to make main.php appear before index.php. This will not affect the way any other pages load, just the directories that contain a main.php.

Question: Do you have a list of changes, or a patch of some sort to implement?
Answer: No, you have to do what is best for your site. Most of the changes are made to individual .tpl files and are quite trivial. But I will add some examples to the end of this post to help out. (And no, Coppermine is not better).

Question: Do bots pick up a session ID when browsing my site?
Answer: Yes and no. There is a built in function to prevent bots from being issued sessions IDs, which fails as soon as certain links are followed, register and login come to mind. Prevent bots from parsing these types of links. Browse around your site as if you were a guest and make note of any links that cause the address to get information appended to the end and then track down what you can do to prevent it (disable registering or at least hard link it with a nofollow).

Question: Is mysite.com/mygallery/picture1.jpg.html?higlightid=3295 the same as mysite.com/mygallery/picture1.jpg.html?
Answer: No. Google will see them as two separate addresses pointing to the same content and label one as duplicate, perhaps both.

Question: If Google picked up some pages that I don't want there anymore (mysite.com/mygallery/picture1.jpg.html?higlightid=3295) can I use the removal to to get Google to drop the listing?
Answer: Yes, but it may take the rest of the site with it. But, the tool only works for 180 days (exactly) then the page will re-appear. Your best bet is to 404 that page until Google forgets about it.

Some quick examples:

And for crying out loud ---- MAKE A BACKUP!!----

==HighlightId==

Open modules/core/templates/blocks/BreadCrumb.tpl

Remove arg3="highlightId=`$theme.parents[parent.index_next].id`"

and

Remove "arg3="highlightId=`$theme.item.id`"

==Picture name in the resize link==

Open modules/core/templates/blocks/PhotoSizes.tpl

Find this block

      <a href="{g->url arg1="view=core.ShowItem" arg2="itemId=`$theme.item.id`"
	arg3="imageViewsIndex=`$theme.sourceImageViewIndex`"}">
	 {$smarty.capture.fullSize}
       </a>

Add {$theme.item.title|markup} thusly:

      <a href="{g->url arg1="view=core.ShowItem" arg2="itemId=`$theme.item.id`"
	arg3="imageViewsIndex=`$theme.sourceImageViewIndex`"}">{$theme.item.title|markup}
	 {$smarty.capture.fullSize}
       </a>

Add some text to make it even spiffier, and make it open in a new window without a template like this:

        <a href="{g->url arg1="view=core.DownloadItem" arg2="itemId=`$theme.item.id`"
        }" target="_blank">View {$theme.item.title|markup} full size
        {$smarty.capture.fullSize}
       </a>

==Make links out of thumbnail titles==

(This is for Matrix, poke around other themes to see how they handle titles)

Open album.tpl

Find

		{if !empty($child.title)}
		<p class="giTitle">
		  {if $child.canContainChildren}
		  {g->text text="Album: %s" arg1=$child.title|markup}
		  {else}
		  {$child.title|markup}
		  {/if}
		</p>
		{/if}

Change to:

		{if !empty($child.title)}
		<p class="giTitle">
		  {if $child.canContainChildren}
               <a href="{g->url arg1="view=core.ShowItem" arg2="itemId=`$child.id`"}">
		  {g->text text="Album: %s" arg1=$child.title|markup}
		  {else}
               <a href="{g->url arg1="view=core.ShowItem" arg2="itemId=`$child.id`"}">
		  {$child.title|markup}</a>
		  {/if}
		</p>
		{/if}

Just your standard <a href=""></a> and the 'ol {g->url}, pretty easy to make it work with other themes.

---Hint, take out Album: and you'll feel better about yourself.

==Peer List==

One thing that annoys me is seeing Snowypicture#13.jpg reduced to sno...13.jpg, so change it!

Open modules/core/templates/blocks/PeerList.tpl

Find entitytruncate:14 (there are 2)

Change the 14 to however many characters you think your file names will grow to.

========

Again, if you see any mistakes, please point them out. I'll continue to keep an eye on the thread and update as necessary. Good luck!

Redirects are death in Google. It sees the 302 as spam, even if it's just a naming thing. Their algo just doesn't know how to distinguish, yet anyway. It would be extremely beneficial to make index.php the default, especially for less experienced webmasters. PageRank is also url dependant, which means if all your inbound links point to http:\\yourdomain.com and it redirects to http:\\yourdomain.com\main.php, the PR won't fully transfer, if at all.

And thanks for the kudos, always happy to help, especially when you guys have given so much!

bmsstore
we love you too. i'm sure you'll be proud of your rants when you find them in a few years with google

@topic:
most of the above SEO recommendations could be hacked into G2. a diff / detailed instructions / a patch would speed up getting changes into G2. so if someone has implemented some of those recommendations, why not share?
else: feel free to file a feature request...

I really like G2 - it looks really good intergrated into my wordpress install - but search engine friendly its not. All of the points in the first post are valid and easily changed, but as long as the URL contains the g2_GALLERYSID part its never going to be very search engine friendly. I've spent the best part of today trying to find a way to remove it, but my lack of php ability really let me down.

If anyone has a way to remove the g2_GALLERYSID from the URL - please post it, I don't mind an ugly hack or even a suggestion of where to start.

I know this post sounds a bit down, but really its not I think G2 is great - its just that I'm facing having to move back to a sub standard solution if I can't get this sorted.

valiant,
Wasn't google one of those that gets kinda upset when you treat their bot differently form other users?

stephen_3,
Some people seem to have cookies disabled. The standard session id method in PHP is that the session id variable is empty when cookies are enabled. You getting them means you should allow cookies to get ris of them. (valiants reply in different words)
______________________
I made a theme for G2, try it

Thanks RwD, I have cookies enabled so that I don't get the session id, the problem is that I don't believe that spiders allow you to set cookies and therefore they get the the session id. I have no problem with my users or myself getting sessions, but i don't want google spidering the same pages multiple times with different session ids.

Rwd---- appropos your comments about keeping robots out... use dynamic <robots.txt> files.

* The static form of the <robots.txt> file is often the wrong flavour, it works the wrong way around. It automatically and indiscriminately lets in every visiting spyder - without fail - every time - all the time. If you've done your homework you will have coded your <robots.txt> file to include all the places that are inappropriate for indexation. So not only does the visiting spyder know that you're letting absolutely everybody and everything into your site but you've also automatically informed them as to the whereabouts of stuff you'd prefer not to be broadcast. It's all a bit dissatisfying. Wandering through your access logs, as you do, can often leave you with a bad taste in the mouth when you check up the wherewithall of the latest visiting spyder;~/ This usually means then chasing through a myriad of sub-directories rewriting all the .htaccess files - as appropriate. There's NOTHING like KNOWING that you're destined for sessions of repetitive stable door closing to increase your self-esteem!

* After some overlong experiences of spyder hunting/bashing/blocking I turned things around. I coded my site to work the logic the other way. My sites automatically reject new spyders without fail - every time - all the time. They do this without revealing all the sub-directories that should not be spydered even by white-listed spyders. White-listed spyders are provided with appropriate list of disallowed sub-directories. Thus these robots.txt files can be considered as being dynamic.

* Briefly - use PHP to pretend to run a 'normal' robots.txt but whose content is dynamically rendered.

* In detail - amend your existing .htaccess and robots.txt files thus:

[.htaccess]
<Files robots.txt>
     ForceType application/x-httpd-php
</Files>

[robots.txt]
<?php
	$disallow=array("/noindex", "/html", "/assets", "/private", "/cgi-bin", "/pdf-noindex");
	$disallow_none = TRUE;
	$agent = 0;
	header('Content-type: text/plain');
	$robots = array("Yahoo! Slurp", "ConveraCrawler", "Googlebot", "Googlebot-Image", "http://www.almaden.ibm.com/cs/crawler", 
	"ia_archiver", "IRLbot", "JemmaTheTourist", "Knowledge.com", "MJ12bot", "MojeekBot", "Mozdex", "msnbot", "NutchOrg", 
	"pipeLiner", "psbot", "searchengineworld", "snap.com", "Speedy Spider", "SurveyBot", "Teoma");
	foreach($robots as $rob) {
		$pos = substr_count($_SERVER['HTTP_USER_AGENT'], $rob);
		if ($pos) {
				if (!$agent) {
				echo "User-agent: *\n";
					if (in_array($rob, $robots)) {
						foreach($disallow as $dis) {
							echo "Disallow: $dis\n";
						}
					$agent++;
					}
				}
		}
	}
	if ($agent == 0) {
		if ($disallow_none) {
			echo "User-agent: *\n";
			echo "Disallow: /\n";
		}
	}
?>

The above <robots.txt> 'file' whitelists a sample bunch of fairly well known spyders. It provides them with the expected list of disallowed sub-directories. All other spyders get a very short file that disallows everything. The onus is on you to check the likely provenance of newcomers for possible white-listing. BTW they ALWAYS come back for another sniff around no matter what...

----best wishes, Robert

Your file says "to all robots, disallow everything in the root and beyond (except for robots.txt of course)". Properly programmed robots will abide by this and not index anything on your site. Good robots will even de-index everything previously indexed... I have accidently discovered this to my cost with Google after innocently 'protecting' a single blogging page with noindex/nofollow on the actual page. Google promptly delisted my entire site and page rating. Oh and FWIW stay away from migratory birds and chickens!!!

----best wishes, Robert

Maybe this will clarify things...

1) whitelisted spyder example
googlebot visits and requests the robots.txt file
my site recognises its (whitelisted) signature and
dynamically responds with encouraging content
to index everything but the following sub-directories...
User-agent: *
Disallow: /noindex
Disallow: /html
Disallow: /assets
Disallow: /private
Disallow: /cgi-bin
Disallow: /pdf-noindex

2) unlisted/blacklisted spyder example
evilharvestor visits and requests the robots.txt file
my site does not recognise this spyder and
dynamically responds with discouraging content
to index nothing at all...
User-agent: *
Disallow: /

That's it... automatically;~)

----best wishes, Robert

Valiant,
Google just spidered my gallery and as you said it wasn't served session id's - which is nice. Why some site's have session ids in their indexed URL's? - Im wondering how g2 identifies spiders - if its by ip then its feasible that google have added an extra range, if its by user agent could it be that google has spidered without it?
Now Im waiting for Yahoo and msn to spider to see how their handled.

stephen_3:
from modules/core/classes/GallerySession.class

    /**
     * Return the id of the search engine currently crawling the site by
     * analyzing the current request.
     *
     * @return string the crawler id, or null if it's a regular user
     */
    function identifySearchEngine() {
	if (!isset($_SERVER['HTTP_USER_AGENT'])) {
	    return null;
	}
	$userAgent = $_SERVER['HTTP_USER_AGENT'];
	if (strstr($userAgent, 'Google')) {
	    return 'google';
	} else if (strstr($userAgent, 'Yahoo')) {
	    return 'yahoo';
	} else if (strstr($userAgent, 'Ask Jeeves')) {
	    return 'askjeeves';
	} else if (strstr($userAgent, 'msnbot')) {
	    return 'microsoft';
	}

	return null;
    }

so no, we don't identify by IP, we parse the userAgent string for google, yahoo, etc.

Yes there is: http://gallery.menalto.com/node/32349?highlight=index%2Cmain

For your URL Rewrite issue, please start a new topic and include the information requested (you'll see that when you start a new topic). You shouldn't even need to look at a page for the URL Rewrite module to work. If it's an album you see an error with and you are manually entering the URL, you need to put a trailing slash on the URL (www.example.com/v/G2album/ not www.example.com/v/G2album)

I'm not a rewrite rule expert, but if you start a new topic with your problem, the URL Rewrite module developer will probably be able to help you. Be sure to be clear in your subject of what your problem is "Gallery doesn't work" doesn't help
____________________________________________
Like Gallery? Like the support? Donate now!!! See G2 live here

Spiders cannot crawl content they cannot access. So if they don't have the username and password to login and crawl the site unrestricted or as that user, then they can't see the content. Example, if you make it so users have to login to see any content a search spider will never know of any of the content you have in your gallery.
____________________________________________
Like Gallery? Like the support? Donate now!!! See G2 live here

Thanks for answering so quickly.

Continental: How did you solve the slideshow problem in your robots.txt file?

Done and done
http://72.14.203.104/search?q=cache:-D2X2R6si2kJ:gallery.menalto.com/node/36854+bmsstore+site:menalto.com&hl=en&gl=us&ct=clnk&cd=1

most of my album pages are indexed:
http://www.google.com/search?hs=vGj&hl=en&lr=&client=firefox-a&rls=org.mozilla%3Aen-US%3Aofficial&q=site%3Awww.themolavis.com&btnG=Search

[img]http://www.nukedgallery.net/signature.jpg[/img]

forumposters-----
What or which file? Whatever, there's a better way to
stop any robot. Use the URL rewrite module and switch
the position of the 'slideshow' element of the path
from the end to the front...
slideshow/%path%/slideshowapplet.html
...then the robots' exclusion always works:
Disallow: /slideshow
----best wishes, Robert

Oh, another thought, I guess from what has been said here that Google won't exactly love permalinks from the point of duplication???

---
Robert Hollingworth
www.roberthollingworth.co.uk

You are correct. Google will index both directories and find duplicate pages, then one or both sets are lkely to go supplemental. I turned permalinks off for my site and use custom 301 redirects when content moves. The whole time my site was being built google was indexing it, even though I had a noindex tag in robots.txt. When google updated it's index there were hundreds of broken urls as I had changed the url structure on my site several times. It took a lot of time to clear it up as custom 301 redirects had to be written to match every broken URl pattern and rewrite then to the current URL pattern.

Andrew
New York Photography

Yeah, but it doesn't actually remove it. It just hides it for 180 days, then it pops back up.

You can hose the entire site with that so your best bet is just wait for googlebot to figure it out.

ichthyous,

What do you mean by your change number "2"

==================================
Website Design and SEO | Poker Picture Gallery | Free Desktop Wallpapers

I have index.php set as the default for my site, but that's not really what I meant. Gallery app itself uses index.php when it generates the urls. I had to change the actual code to remove that. It's not something that is inherently bad, but since all my incoming links pointed to my domain I didn't want all the internal links pointing to index.php. I had posted about duplicate content issues with Google and slideshows...the new exlide flash slideshow module should take care of that...if you can get it to work.

My Photos of Spain Gallery
My Photos of Italy Gallery
My Photos of Greece Gallery

Yes, GoogleBot is that dumb.

Yes, these days Google is very touchy about dupe content so why risk it? Flash slideshows are much prettier and no dupe content issues at all. Once you start seeing a good number of pages going supplemental it's very hard and time consuming to track it and turn it around.

Netscan, I'm not sure I really follow your last post. I don't have any main.php/index.php redirect as I have swapped the files. I did a lot of editing in one of the class files to remove the index.php entirely from my links. All of my internal links point to just the domain name.