Source

Classic web advertising for banners (in some cases text links) are still paid by impressions. Like the banners on the left column of Facebook pages, everytime user changes a page, an ad is shown, and the advertiser is charged per 1000 impressions. Everybody knows that there is nothing interesting with this.

Facebook, some weeks ago has changed their photo gallery to an ajax photo gallery.
Everytime you see a photo and start browsing to the other photos, without refreshing the page, an Ajax call was done and new photo was shown on the very same page. There was no refresh, therefore the rest of the page -as well as the banner on the left- were not changed.

Facebook recently switched back to their old way. One photo is 1 page view again. I wondered why? It was really obvious. They’ve lost major page views with just this change. On the old system, people were rapidly viewing photos spending (in most cases) no more than 15 seconds per photo. For an album of 20 photos, it was an easily generated 20 page views for them. New system, slowed down the page views but now the users were spending more time on the same page with the same banner. Still with the old metric, 1 page view for 20 photos.

Web metrics, still measured with page views in most cases. Starting with the use of RSS, this metric started to crack down. 100k subscribers to an RSS feed, and only a few thousand user on the website, seemed like a website has no traffic.

We still measure advertisements with page views. This kills the use of ajax for most actions for the sake of doing a page view. There must be a good comprimise in between.

Finetune.com, when they’ve first launced, the website was all ajax. Every page’s content was loaded with ajax and there was only 1 page view for the user. The rest was stream of data.

Songza.com, you are always in the same page but you do a lot of actions.

definr as well brings all it’s content with ajax on the very same page.
You may easily find more examples like these.

Also, you can’t really refresh the advertisment block on a page with every X amount of ajax calls done. Advertisment networks like Google’s Adsense strictly forbindens the use of this kind of stuff to generate page views. They think it’s fake page views. In fact it’s not.

All these said, web 2.0 needs a new metric to measure. I don’t know how they are going to call it but XHRs should be accounted for.
Here are my candidates for the new metric.

XPU: XHRs per User. Measures, how many ajax calls done by user per session.
XPP: XHRs per page. Measures, how many ajax calls done in a page.

These still will not invalidate the page view metric. These are just multipliers or supplementary.

So how would the advertisers pay with these metrics? How should this translate to the old world?

Lets keep the numbers small just for this example’s purpose:

Site A, a classic website with 100 users and 1000 pages has 200 page views per user per month.
Site B, a web 2.0 website with 100 users and 10 pages has 10 page views per user per month, 200 XPU and 20 XPP.

With the old metric Site A is more successfull.
With the new composite metric, Site B is more successfull.

Site A has (100 users * 200 pages) = 20K pageviews per month
Site B (100 users * 10 pages * 200 XPU) = 200K pageview per month.

Advertising on a page with higher XPP should be much more expensive then a static page because, the refresh rate of the page is low and possibly user will have more time to notice the banner on the same page.

Honestly, I would be scared to put an advertisement on Facebook. Since there are too many users and there is no open metric to measure how much time user spends on a page, your advertisement dollars would be wasted in a minute, without getting enough attention from the users.

The new ajax photo gallery of facebook was more advertiser friendly. User views 10, 20 photos on the very same page, possibly having more time to notice the advertisement when finished viewing all the photos.

Facebook switched back to their old photo gallery now. Possibly all the advertisment contracts were made with CPM and they were loosing money.

This is what I think:

• Community:
  • Trust the members’ input.
  • Allow the members to be known and get credit by measuring their contributions.
  • Allow other members in the community to measure and respond to contributions.
  • Enable members’ knowledge base to evolve as processes and concepts change.
  • Enables user to embrace the content ownership
• Content
  • Make it easy to contribute to your knowledge base and make it accessible to others.
  • Reviews, Ratings, Comments, Focus: Freedom of expression
  • Free content
  • Using the power of the crowd
  • Originality
  • Community decides what they want to see on the site
• Context
  • Displaying the information based on reality of the user or page
  • Giving the value to people near you (in taste or physical location)
  • Provides personalization of the content
• - Commerce
  • Adapting non-revenue-generating website into that generates revenue
  • Via affiliates programs
  • Without disturbing the community
  • Should be well hidden between the lines
  • $$$
• Personalization
  • Allowing the user to tailor the site in accordance with their tastes and preferences.
  • Caters the sense of ownership / ease of use
• Vertical Search
  • Introducing the context into search (showing results that are more relevant to user)
  • Introducing other sites (possibly affiliate sites/competitors) into search.

Some blackhats, hoping to get the page rank benefits of a .gov site, hacked the forum of the site and placed a gateway like page. I guess search results come from google as well.

Check it out yourself.
nevadacityca.gov

PS: the keyword I was looking was the name of one of my sites, not cealis lol.

I’m not sure if this is on purpose, but it’s most likely that google is also moderating the biggest moderated directory on the net. Hehe that’s a good one.

See it yourself (I’m not sure for how long this links will be alive)
Dmoz link : updated around June 4th
Google’s Link: doesn’t exist as of June 5th

Also I’ve found if you search for url that exist on dmoz’s certain category, google returns a result but the link is simply for oh for.
See the first link.

Last notes: Yahoo’s directory doesn’t have the category too. I kind of understand it since the category is under Business branch and it’s a paid service for yahoo to appear under that branch.

Are Scrapers Exploiting the sitemap.xml file?

Many people seem to posting saying after adding sitemaps they are suffering a problem with content. Could sitemaps.xml be beeing abused. Is the new content title and meta tag scraped before the sitemap is submitted to google by sitemap generators?

A month ago, when ask.com posted about the sitemap.xml in the robots.txt, I posted about how this can new feature can be abused and how you should protect sitemaps. I even posted this on the webmaster’s world as well… No one cared then, now they are suffering.

What can I say more, thanks for the sunday night drama.

Simply you add a line into your robots.txt saying

Sitemap: <sitemap_location></sitemap_location>

This part is really cool but for site harvesters it’s an unbelivable tool. So you can handover the key to your site and web harvesters can crawl your site really easily because probably you’ve put all your site’s pages into your sitemap.

Sounds like a good plan in an ideal world. With all the cloakers and content scrapers you must be really smart not to be ripped apart.

My suggestion is to know who you’re serving the sitemap. Currently Google, Yahoo and Ask is supporting this sitemaps.xml and no other site has anything to do with it.
Here is a simple check you can add in the begining of your sitemap thing:

< ?php
    function botIsAllowed($ip){
        //get the reverse dns of the ip.
        $host = strtolower(gethostbyaddr($ip));
        $botDomains = array('.inktomisearch.com',
                                     '.googlebot.com',
                                     '.ask.com',
                             );

        //search for the reverse dns matches the white list
        foreach($botDomains as $bot){
            if (strpos(strrev($host),strrev($bot))===0){
                $qip= gethostbyname($host);
                return ($qip==$ip);
            }
        }
        return false;
}

if (!botIsAllowed($_SERVER['REMOTE_ADDR'])){
    echo "Banned!";
    exit;
}
?>

I’m sure everyone can get the idea of reverse dns and forward dns checking.
If I missed any decent site that uses the sitemaps let me know.

Note: If you’re still using static sitemaps (!) you can just include the xml after the code.

Here is the related US code:

§ 2252C. Misleading words or digital images on the Internet
(a) IN GENERAL.—Whoever knowingly embeds words or digital images into the source code of a website with the intent to deceive a person into viewing material constituting obscenity shall be fined under this title and imprisoned for not more than 10 years.

(b) MINORS.—Whoever knowingly embeds words or digital images into the source code of a website with the intent to deceive
a minor into viewing material harmful to minors on the Internet shall be fined under this title and imprisoned for not more than 20 years.

(c) CONSTRUCTION.—For the purposes of this section, a word or digital image that clearly indicates the sexual content of the
site, such as ‘sex’ or ‘porn’, is not misleading.

(d) DEFINITIONS.—As used in this section—HR 4472 EAS.

(1) the terms ‘material that is harmful to minors’ and ‘sex’ have the meaning given such terms in section 2252B; and ‘‘(2)the term ’source code’ means the combination of text and other characters comprising the content, both viewable and nonviewable, of a web page, including any website publishing language, programming language, protocol or functional content, as well as any successor languages or protocols.’’.

(b) TABLE OF SECTIONS.—The table of sections for chapter 110 of title 18, United States Code, is amended by inserting after
the item relating to section 2252B the following: 2252C. Misleading words or digital images on the Internet.’’.

hehehe. Next time you post the goatse man or the bath tub girl, remember this…

It’s a stupid keyword match, not a context match. This means you can’t really post anything about those keywords even if you’re writing against or for.

Here is the list I’ve found that is banned. Maybe there are more.

Update: I’ve found this one which is really weird: iraqi dinar

iraqi dinar
anabolic steroid
buy diethylpropion
buy ephedra
buy generic viagra
buy levitra
buy pain relief
buy phentermine
buy prescription
buy re
buy viagra
cable descrambler
casino online
casino roulette
child pornography
designer replica
fake degrees
fake diploma
find pharmacies
get medications
get viagara
internet casino
internet gambling
online blackjack
online casino
online craps
online kasino
online kasinos
online poker
online roulette
online slots
play online casino
play online poker
play roulette online
play texas holdem
pokeronline
replica designer handbags
replica handbags
replica watch
sportbetting
sportsbook