juust ~ php oddities » pagerank http://www.juust.org Link theory and search engine optimization Thu, 19 Jan 2012 09:39:32 +0000 en hourly 1 http://wordpress.org/?v=3.2.1 seo tricks : old wine in new bags… http://www.juust.org/index.php/seo-tricks-old-wine-in-new-bags/2008/09/ http://www.juust.org/index.php/seo-tricks-old-wine-in-new-bags/2008/09/#comments Fri, 26 Sep 2008 05:19:05 +0000 juust http://www.juust.org/?p=160 Continue reading →]]> Get some pagerank : this trick would require tedious boring link checking, but since SeoLinx (an extension of SeoQuake) that has become a lot easier. SeoLinx shows the stats of a links target url so you don’t have to go to every page to retrieve the stats. Cool plugin. Let’s put it to some practical use.

the trick : comment on old forum threads

Once you have SeoLinx installed find an ‘old’ forum, register if you haven’t already and make sure you get a signature link. Sometimes you first have to be a member for a week or write ten posts, but once you have a sig-link you get backlinks off the forum.

Then go comment on really old forum threads.

With SeoLinx you can easily spot the juicy old threads. Old threads on for instance DigitalPoint or Webmasterworld are sometimes pagerank 3. In case of the DP post, PR2 with 8 posts at the time of writing.

Pick a forum, and browse to the last page of the threads. Hover over the thread anchor and SeoLinx shows you the pagerank of the thread page. As long as the number of posts is below (10, 16 depending on the forum settings) you can put your comments in and they will appear on the first page of that thread, that has that nice pagerank and juice.

Old wine in new bags can be a sweet thing.

the benefit

A pagerank 3 ‘targetted’ anchor is worth about $9,- a month, $100,- per year. It can take an hour to find a juicy one, but hey, $100,- value for an hours work is well worth the trouble.

I might make this a blog feature, seo tips and tricks of the month.

]]> http://www.juust.org/index.php/seo-tricks-old-wine-in-new-bags/2008/09/feed/ 2 Spidering http://www.juust.org/index.php/spidering/2008/08/ http://www.juust.org/index.php/spidering/2008/08/#comments Thu, 28 Aug 2008 00:15:44 +0000 juust http://www.juust.org/?p=127 Continue reading →]]> Someone asked about the ‘pagerank spider’, I put the code online as is, it isn’t finished and if I wanted to finish it I would make a few changes.

the main remaining issues are

  • 1 memory usage
  • 2 how to handle the www.-prefix
  • 3 indexed pages at google
  • 4 http codes

1 a big class uses a lot of memory, a mysql backed version has an extra dependency, takes longer to develop and is slower. I needed a fast spider for a quick feedback on a small site.

Check out phpDig, they have a mature open-source(?) spider with a mysql backend, and a usergroup and forum.

2 google have a section where you can choose to have all domain pages indexed represented as either juust.org or www.juust.org. It hints on that having an influence on page ranking but no actual straight forward ‘rule’. I have no idea what the actual impact is.

3 google index and cache pages when spidering other sites that link to yours. If the page the link points to was valid at the time, the page it links to is indexed and cached. Especially with files you dumped, or query-result pages, search pages, you cannot remove the cached page but it is counted to your site.

Putting search pages on ‘noindex’ is smart, especially if you use one of these funky search box gadgets in your template that can list any result, if someone queries your site for (nasty+term) and puts the query as link to your search page, once the link is followed a page from your site loaded with (nasty+term) is indexed and you cannot erase it from the cache, so then you have a problem. Put the file on robots=”noindex”, and try and confine the search to your own domain, or use a profanity filter.

4 http-codes, I checked them out for a link-validator routine two weeks ago, I might be adding that mysql backend after all, and make a more sturdy version, but not for the next few weeks.

———
Some background info
searchtools.com /robots /robot-checklist

phpDig

]]> http://www.juust.org/index.php/spidering/2008/08/feed/ 0 social bookmarking to get your site indexed http://www.juust.org/index.php/social-bookmarking-to-get-your-site-indexed/2008/08/ http://www.juust.org/index.php/social-bookmarking-to-get-your-site-indexed/2008/08/#comments Mon, 18 Aug 2008 21:53:49 +0000 admin http://www.juust.org/?p=112 Continue reading →]]> Yesterday I put one link through twitter on twemes.com and four links on del.icio.us to the links.trismegistos.net php Link Directory.

Today I googled ‘trismegistos links’ to see what the effect (if any) would be and the link I put on twemes actually shows up first in google (and top-10 frontpage, spot 6 of 40.000 results).

I also issued a 700 URL sitemap to google webmaster first, and only added the bookmarks after the sitemap was downloaded.

I was just curious which method would yield the best result, and twitter/twemes is the winner.

why bother ?

Because a test I did shows most directory sites subcategory pages (where most links are) have no assigned pagerank and if i want to run an effective directory I have to get a fix on that problem and get a fix to fix it up.

I did two tests on directory sites, where I downloaded Yahoo SiteExplorer indexed urls’ and retrieved the pagerank per url.

sites pages/site total ranked percentage
16 1000 16.000 120 0.7%
150 50 7500 200 2.5%

Roughly interpreted, per site most pages are indexed, but only about 2 per 100 pages have a pagerank value assigned.

The others have no value assigned and don’t pass any value on links on it. In all cases (except dmoz, which is ranking on most branches) it was the index page and main category pages that were ranked and the pages with links were all N/A not-available.

So testing the effect of social bookmarking on pages that would hold links is interesting.

A submission now costs 3 to 15 cts, which shows the value of links in a directory is low, and only featured links (which usually appear on the index and main category pages which do rank) are sold for $3,-/year to $40,-/permanent.

An estimate for a link on a PR3 page for a year in a directory page is $13,-/year.

If i can get 700 pages to rank PR1 and sell links for $3,-/year, 700 pages with 20 links times 3 makes $60.000,-++ a year. Compare that to $10/year for 20 links on a PR3 category page, 10 pages is $2.000/year.

And if 700 social bookmarks can make sure after a year my whole directory is ranked, indexed, brings in $60..000,- and delivers the goods (a ranking link for entrants at $3,-/year) then a month linkspamming is well worth the trouble.

Another option is reciprocals on the category page itself (from an indexed page, some link-pages are conveniently not indexed ;)

]]> http://www.juust.org/index.php/social-bookmarking-to-get-your-site-indexed/2008/08/feed/ 0 tool : pagerank per url from a sitemap http://www.juust.org/index.php/pagerank-from-sitemap/2008/08/ http://www.juust.org/index.php/pagerank-from-sitemap/2008/08/#comments Fri, 01 Aug 2008 13:41:18 +0000 juust http://www.juust.org/?p=90 Continue reading →]]> I wired a google pagerank toolbar-query snippet to a simplexml sitemap readout, and put it on a page. You can fill in a sitemap url and get the google pageranks of all ‘mapped’ urls.

It works, I stripped it down and you can download it here or on the sample page.

I mainly wanted the snippet wired to a sitemap to compare the results of my pagerank spider tool with an actual google readout. Running a sitemap through a toolbar query snippet is the fastest way.

I allready had a spider result of siteometrics (calc pr) so now I can compare it to google’s toolbar query on http://www.siteometrics.com/sitemap.xml :

google pr calc pr URL
2 http://www.siteometrics.com/
2 0.80 /index.php
0 0.32 /advertise.html
- 0.77 /recommend.php
- 0.75 /search-engine-saturation.php
0 0.75 /link-popularity.php
0 0.75 /pagerank.php
0 0.75 /bulk-pagerank.php
0 0.75 /pagerank-mult-pages.php
0 0.75 /link-pop-pagerank.php
- 0.75 /link-search-pagerank.php
0 0.75 /alexa.php
0 0.75 /bulk-alexa.php
0 0.75 /serpcheck.php
0 0.75 /keyword-research.php
0 0.67 /visitor-info.php
- 0.24 /useful-links.html
0 0.24 /contact-us.html
- 0.24 /sitemap.html
- 0.24 /privacy-policy.html

Weird result, the sitemap they issue is part old site, part new site. If you check the pageranks on the newer .php files it’s the same, though.

a quarter of the urls link into the archived site, that might cause the drop in pagerank (links to /feed and google.com on every page, see the other article on siteometrics).

for the freaks : here’s the php code (assume url is a valid sitemap-url).

  1.  
  2. $myurl=$_REQUEST['url'];
  3. $xml = simplexml_load_file($myurl);
  4. foreach($xml->url as $u) echo pagerank((string) $u->loc)."<br />";
  5. exit;
  6.  
  7. function pagerank($url) {
  8.  if (!preg_match('/^(http:\/\/)?([^\/]+)/i', $url)) { $url='http://'.$url; }
  9.  $pr=curl_getpr($url);
  10.  return $pr.';'.$url.';';
  11. }
  12.  
  13. function getch($url) { return CheckHash(HashURL($url)); }
  14.  
  15. function curl_getpr($url) {
  16.  $googlehost='toolbarqueries.google.com';
  17.  $googleua='Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.6) Gecko/20060728 Firefox/1.5';
  18.  
  19.  $ch = getch($url);
  20.  
  21.  $form="http://toolbarqueries.google.com/search?client=navclient-auto&ch=$ch&features=Rank&q=info:$url";
  22.  $cr = curl_init($form);
  23.  curl_setopt($cr, CURLOPT_FAILONERROR, true);
  24.  curl_setopt($cr, CURLOPT_HEADER, 0);
  25.  curl_setopt($cr, CURLOPT_USERAGENT, $googleua); // Spoof the user-agent
  26.  curl_setopt($cr, CURLOPT_RETURNTRANSFER, true);
  27.  $data = curl_exec($cr);
  28.  
  29.  if(!$data) {
  30.    curl_close($cr);
  31.    unset($cr);
  32.    $pr='-';
  33.    return $pr;
  34.  } else {
  35.   $pos = strpos($data, "Rank_");
  36.   if($pos === false) {
  37.    curl_close($cr);
  38.    unset($cr);
  39.    $pr='-';
  40.    return $pr;
  41.                 } else{
  42.    $pr=substr($data, $pos + 9);
  43.    $pr=trim($pr);
  44.    $pr=str_replace("\n",'',$pr);
  45.    curl_close($cr);
  46.    unset($cr);
  47.    return $pr;
  48.   }
  49.     }
  50. }
  51.  
  52. //PageRank Lookup v1.1 by HM2K (update: 31/01/07)
  53. //based on an algorithm found at: http://pagerank.gamesaga.net/
  54. //live demo: http://www.highrankforum.com/pagerank.php
  55.  
  56. //convert a string to a 32-bit integer
  57. function StrToNum($Str, $Check, $Magic) {
  58.     $Int32Unit = 4294967296;  // 2^32
  59.  
  60.     $length = strlen($Str);
  61.     for ($i = 0; $i < $length; $i++) {
  62.         $Check *= $Magic;  
  63.         if ($Check >= $Int32Unit) {
  64.             $Check = ($Check - $Int32Unit * (int) ($Check / $Int32Unit));
  65.             //if the check less than -2^31
  66.             $Check = ($Check < -2147483648) ? ($Check + $Int32Unit) : $Check;
  67.         }
  68.         $Check += ord($Str{$i});
  69.     }
  70.     return $Check;
  71. }
  72.  
  73. //genearate a hash for a url
  74. function HashURL($String) {
  75.     $Check1 = StrToNum($String, 0×1505, 0×21);
  76.     $Check2 = StrToNum($String, 0, 0x1003F);
  77.     $Check1 >>= 2;  
  78.     $Check1 = (($Check1 >> 4) & 0x3FFFFC0 ) | ($Check1 & 0x3F);
  79.     $Check1 = (($Check1 >> 4) & 0x3FFC00 ) | ($Check1 & 0x3FF);
  80.     $Check1 = (($Check1 >> 4) & 0x3C000 ) | ($Check1 & 0x3FFF);
  81.     $T1 = (((($Check1 & 0x3C0) < < 4) | ($Check1 & 0x3C)) <&lt;2 ) | ($Check2 & 0xF0F );
  82.     $T2 = (((($Check1 & 0xFFFFC000) << 4) | ($Check1 & 0x3C00)) << 0xA) | ($Check2 & 0xF0F0000 );
  83.  
  84.     return ($T1 | $T2);
  85. }
  86.  
  87. //genearate a checksum for the hash string
  88. function CheckHash($Hashnum) {
  89.     $CheckByte = 0;
  90.     $Flag = 0;
  91.  
  92.     $HashStr = sprintf('%u', $Hashnum) ;
  93.     $length = strlen($HashStr);
  94.  
  95.     for ($i = $length - 1;  $i >= 0;  $i –) {
  96.         $Re = $HashStr{$i};
  97.         if (1 === ($Flag % 2)) {              
  98.             $Re += $Re;    
  99.             $Re = (int)($Re / 10) + ($Re % 10);
  100.         }
  101.         $CheckByte += $Re;
  102.         $Flag ++;
  103.     }
  104.  
  105.     $CheckByte %= 10;
  106.     if (0 !== $CheckByte) {
  107.         $CheckByte = 10 - $CheckByte;
  108.         if (1 === ($Flag % 2) ) {
  109.             if (1 === ($CheckByte % 2)) {
  110.                 $CheckByte += 9;
  111.             }
  112.             $CheckByte >>= 1;
  113.         }
  114.     }
  115.  
  116.     return '7'.$CheckByte.$HashStr;
  117. }
]]> http://www.juust.org/index.php/pagerank-from-sitemap/2008/08/feed/ 0 seo : pagerank and serp part I http://www.juust.org/index.php/seo-pagerank-and-serp-part-one/2008/07/ http://www.juust.org/index.php/seo-pagerank-and-serp-part-one/2008/07/#comments Thu, 03 Jul 2008 15:56:41 +0000 juust http://www.juust.org/?p=13 Continue reading →]]> Today I go do something dumb, as usual ! I still had to finish my serp tool, and as I was checking out a site’s performance I really needed it, so I added a permutation routine and a mysql backend to the serp tool and tied it to a domain info class.

The serp is a three keys set with permutations,

  • single  1, 2, 3
  • double 12, 23, 13, 32, 21, 32
  • triple     123, 132, 213, 231, 312, 321

15 searches, my chosen keys are php, serp, pagerank, I get a full spread. 1500 results and  +/- 680 different hosts.

then i go do my old magic trick,

count the results, spot 1-3 = 3 points, spot 4-6 = 2 points and spot 7-10 = 1 point. The rest are 0.2 points so all domains get a count of the number of results and a sum for the points they score.

Then I know who are ‘top dog’ in my :) search engine result page.

Forums and communities have a broader spread (phpclasses, seomoz), more pages with content and more titles so they are bound to score on about every permutation and usually have a big pagerank (5 to 8), but what they gain in size they loose in strength and speed.

lets check the result set :

domain results points pagerank pages backlinks dmoz
www.siteometrics.com 16 32.8 2 37 244
seolutions.net 16 26.8 3 43 986
forums.digitalpoint.com 20 13.2 7 4354858 147000 1
www.selfseo.com 13 12 3 884 17500
www.toptenserp.com 9 8.6 2 196 763
www.seroundtable.com 9 8.2 4 15053 203000
www.prchecker.info 8 6.8 8 323 1460000
www.seochat.com 10 6.6 4 12972 558000
www.webmasterworld.com 24 6.4 7 230611 182000
marketingfeeds.nl 2 6 4 37237 6210
www.getfreesofts.com 13 5.4 4 275521 34900
www.shoemoney.com 3 5.2 6 12774 105000
video.aol.com 3 5.2 8 14073952 3300000 1
www.rankwhere.com 3 5.2 4 228 919
www.ljfind.com 3 5.2 3 1771559 1800
forum.siteground.com 9 5 7 30050 4570
www.phpclasses.org 6 4.8 6 150356 88100
www.hotscripts.com 6 4.8 7 219323 461000
hosthideout.com 7 4.6 5 161532 2150 1
www.top25web.com 4 4.4 3 143 15900
juustout.gethost.nl 8 4 0 173 184
www.webmasterforums.com 9 3.6 3 50550 45300
sitening.com 8 3.4 6 2714 29700 1
www.phplivesupport.com 3 3.4 7 1463 120000
www.webworkshop.net 4 3.4 5 119233 15100
www.google-pagerank.net 4 3.4 5 318 20200
www.database-search.com 3 3.4 3 14077 195000
en.wikipedia.org 2 3.2 9 208719522 56700000 1
www.nap.edu 3 3.2 9 2458 227000
www.php.net 1 3 9 170567 15700000
www.webopedia.com 1 3 5 35722 2170000
www.cs.und.edu 2 3 6 195 409
www.investopedia.com 1 3 4 19828 191000
www.smartpagerank.com 6 2.8 3 12051 25400
www.phpbits.info 5 2.6 2 22 2940
www.programsdb.com 13 2.6 4 236295 37700
www.serp-chem.eu 4 2.4 6 18 11400
www.pagerankcode.com 4 2.4 2 6 211
www.searchenginepanel.com 12 2.4 3 84 2590
www.devpapers.com 3 2.4 5 2196 152000
www.googlecommunity.com 2 2.2 5 251463 4150
www.webmasters.am 11 2.2 5 15587 3330
www.sitepoint.com 11 2.2 7 2134801 371000
link.ezer.com 2 2.2 4 2094 3180
forums.searchenginewatch.com 11 2.2 7 86027 142000 1
livepr.raketforskning.com 2 2.2 3 105136 122000
searchengineland.com 10 2 7 41187 442000 1
www.talkdigger.com 1 2 6 743479 145000

…which supports the idea that you don’t need massive backlink counts and even pagerank-0 (my other site (gloat)) is not any hindrance. Pagerank distribution in small sites is easier to manage so a few backlinks will do.

Next edition : seo : pagerank and serp : part II I pick a few small sites of the list and spider them, assert their link structure, retrieve the backlinks and spider the linking pages to see what links point directly to which urls, and make a “push-and-juice” analysis of the actual strength of these pages. Then I go relate that to the place on the search engne result page and estimate what is needed for a top-10 spot.

enough nonsense for today, here is a commercial message from Aunty Google :
[ad#test]

]]> http://www.juust.org/index.php/seo-pagerank-and-serp-part-one/2008/07/feed/ 0