4 Jun 2012 14:57
SVN: [115476] trunk/wikistats/reportcard
<ezachte <at> svn.wikimedia.org>
2012-06-04 12:57:59 GMT
2012-06-04 12:57:59 GMT
https://www.mediawiki.org/wiki/Special:Code/MediaWiki/115476 Revision: 115476 Author: ezachte Date: 2012-06-04 12:57:59 +0000 (Mon, 04 Jun 2012) Log Message: ----------- last revision of scripts for old style report card? Modified Paths: -------------- trunk/wikistats/reportcard/ComScoreTop1000.pl trunk/wikistats/reportcard/RT_yyyy_mm.html trunk/wikistats/reportcard/ReportCardExtractWikiCountsOutput.pl trunk/wikistats/reportcard/ReportCardGenerateHtml.pl Modified: trunk/wikistats/reportcard/ComScoreTop1000.pl =================================================================== --- trunk/wikistats/reportcard/ComScoreTop1000.pl 2012-06-04 10:41:35 UTC (rev 115475) +++ trunk/wikistats/reportcard/ComScoreTop1000.pl 2012-06-04 12:57:59 UTC (rev 115476) <at> <at> -15,10 +15,10 <at> <at> my $do_normalize = $true ; my $no_normalize = $false ; - $file_csv_ref = "Reference sites UV_(Oct 09 - Dec 10).csv" ; - $file_csv_top = "Top 1000 poperties, UV trend_(Oct 09 - Dec 10).csv" ; - $file_html_ref = "ComScoreReferenceSites-2010-12.htm" ; - $file_html_top = "ComScoreTop1000Properties-2010-12.htm" ; + $file_csv_ref = "Reference sites UV_(Dec 10 - Feb 12).csv" ; + $file_csv_top = "Top 1000 poperties, UV trend_(Dec 10 - Feb 12).csv" ; + $file_html_ref = "ComScoreReferenceSites-2012-02.htm" ; + $file_html_top = "ComScoreTop1000Properties-2012-02.htm" ; # note normalization needs overhaul: * 30/actual days in month does not work for UV's <at> <at> -59,6 +59,7 <at> <at> while ($line = <CSV>) { chomp $line ; + $line =~ s/"(\d+),(\d+),(\d+)"/$1$2$3/g ; $line =~ s/"(\d+),(\d+)"/$1$2/g ; $line =~ s/"([^"]*)"/$1/g ; <at> <at> -199,8 +200,8 <at> <at> "$out_license" . "</small>\n" ; - $delta_rank_year = "&Delta Year = Rank ${months [$#months-12]} ⇒ ${months [$#months]}" ; - $delta_rank_month = "&Delta Month = Rank ${months [$#months-1]} ⇒ ${months [$#months]}" ; + $delta_rank_year = "Δ Year = Rank ${months [$#months-12]} ⇒ ${months [$#months]}" ; + $delta_rank_month = "Δ Month = Rank ${months [$#months-1]} ⇒ ${months [$#months]}" ; $delta_perc_year = "% Year = Growth ${months [$#months-12]} ⇒ ${months [$#months]}" ; $delta_perc_month = "% Month = Growth ${months [$#months-1]} ⇒ ${months [$#months]}" ; <at> <at> -249,7 +250,7 <at> <at> $html .= "<table border=1>\n" ; $html .= "<tr><td colspan=99 align=left class=l><h2><font color=#C00000>Complete list</font></h2></td></tr>" ; $html .= "<tr><th class=l colspan=2>Property</th><th class=c colspan=3>Rank</th><th class=c colspan=2>Growth</th><th class=c>Rel. pop.</th><th class=c colspan=99>Monthly data: Unique Visitors (count x 1000) <small><small>+ Rank + Rang change + Growth percentage</small></small></th></tr>" ; - $html .= "<tr><th class=l>Site (*)</th><th class=c>Type</th><th class=c>Rank</th><th class=c>&Delta Year</th><th class=c>&Delta Month</th><th class=c>% Year</th><th class=c>% Month</th><th>WM=100</small></th>" ; + $html .= "<tr><th class=l>Site (*)</th><th class=c>Type</th><th class=c>Rank</th><th class=c>Δ Year</th><th class=c>Δ Month</th><th class=c>% Year</th><th class=c>% Month</th><th>WM=100</small></th>" ; for ($i = $#months; $i >= 0 ; $i--) { $html .= "<th class=c>${months[$i]}</th>" ; } $html .= "</tr>" ; <at> <at> -423,7 +424,7 <at> <at> $html .= "</tr><tr><td colspan=99> </td></tr>" ; $html .= "<tr><td colspan=99 align=left class=l><a id='fast' name='fast'></a><h2><font color=#C00000>Fastest risers</font> <small>$threshold_filter+ rise in ranks in a year</small></h2></td></tr>" ; $html .= "<tr><th class=l colspan=2>Property</th><th class=c colspan=3>Rank</th><th class=c colspan=2>Growth</th><th class=c>Rel. pop.</th><th class=c colspan=99>Monthly data: Unique Visitors (count x 1000) <small><small>+ Rank + Rang change + Growth percentage</small></small></th></tr>" ; - $html .= "<tr><th class=l>Site</th><th class=c>Type</th><th class=c>Rank</th><th class=c>&Delta Year</th><th class=c>&Delta Month</th><th class=c>% Year</th><th class=c>% Month</th><th>WM=100</small></th>" ; + $html .= "<tr><th class=l>Site</th><th class=c>Type</th><th class=c>Rank</th><th class=c>Δ Year</th><th class=c>Δ Month</th><th class=c>% Year</th><th class=c>% Month</th><th>WM=100</small></th>" ; for ($i = $#months; $i >= 0 ; $i--) { $html .= "<th class=c>${months[$i]}</th>" ; } $html .= $html_fast_risers ; Modified: trunk/wikistats/reportcard/RT_yyyy_mm.html =================================================================== --- trunk/wikistats/reportcard/RT_yyyy_mm.html 2012-06-04 10:41:35 UTC (rev 115475) +++ trunk/wikistats/reportcard/RT_yyyy_mm.html 2012-06-04 12:57:59 UTC (rev 115476) <at> <at> -402,7 +402,7 <at> <at> <!-- Reach: Page Requests: Tabs --> <td align=left valign=middle> <!== OUT SC ==> - <img src='charts/{{yyyy}}-{{mm}}/Page-Views-Total-Indexed.png'> + <img src='charts/{{yyyy}}-{{mm}}/Page-Views-Total-Indexed-Wikis.png'> <!== OUT D ==> <!== TABS {20|log scale|linear scale|indexed per wiki|indexed per project|Mobile|M&S} ==> <!== TAB {21|<img src='charts/{{yyyy}}-{{mm}}/Page-Views-Total-Absolute-Log.png'><br>} ==> Modified: trunk/wikistats/reportcard/ReportCardExtractWikiCountsOutput.pl =================================================================== --- trunk/wikistats/reportcard/ReportCardExtractWikiCountsOutput.pl 2012-06-04 10:41:35 UTC (rev 115475) +++ trunk/wikistats/reportcard/ReportCardExtractWikiCountsOutput.pl 2012-06-04 12:57:59 UTC (rev 115476) <at> <at> -5,12 +5,11 <at> <at> $trace_on_exit = $true ; ez_lib_version (2) ; - $month_last = "12" ; - $year_last = 2010 ; + $month_last = "4" ; + $year_last = 2012 ; + $month_start = $month_last ; + $year_start = $year_last - 3 ; - $month_start = "1" ; - $year_start = 2008 ; - $m_start = &months_since_2000_01 ($year_start, $month_start) ; $m_last = &months_since_2000_01 ($year_last, $month_last) ; $m_last_12 = $m_last - 12 ; <at> <at> -49,19 +48,19 <at> <at> sub ParseArguments { -# my <at> options ; -# getopt ("io", \%options) ; + my <at> options ; + getopt ("io", \%options) ; -# die ("Specify input folder for projectcounts files as: -i path") if (! defined ($options {"i"})) ; -# die ("Specify output folder as: -o path'") if (! defined ($options {"o"})) ; + die ("Specify input folder for projectcounts files as: -i path") if (! defined ($options {"i"})) ; + die ("Specify output folder as: -o path'") if (! defined ($options {"o"})) ; -# $path_in = $options {"i"} ; -# $path_out = $options {"o"} ; + $path_in = $options {"i"} ; + $path_out = $options {"o"} ; -# die "Input folder '$path_in' does not exist" if (! -d $path_in) ; -# die "Output folder '$path_out' does not exist" if (! -d $path_out) ; + die "Input folder '$path_in' does not exist" if (! -d $path_in) ; + die "Output folder '$path_out' does not exist" if (! -d $path_out) ; - $path_in = "w:/# out bayes" ; + $path_in = "w:/# out bayes/csv_report_card" ; $path_out = "w:/ <at> report card/data" ; print "Input folder: $path_in\n" ; <at> <at> -108,7 +107,7 <at> <at> while ($line = <CSV_IN>) { ($language,$date,$counts) = split (',', $line, 3) ; - + next if $language =~ /^zz/ ; next if $language eq 'commons' and $project ne 'wx' ; next if $language eq 'sr' and $project eq 'wn' ; # ignore insane bot spam on <at> <at> -144,6 +143,7 <at> <at> $totals {"$f,$m"} += $fields [$f] ; $totals_project {"$f,$m"} {$project} += $fields [$f] ; + $totals_project {"$f,$m"} {$all_projects} += $fields [$f] ; # print "TOTALS $f $m = . " . $totals {"$f,$m"} . "\n" ; <at> <at> -199,6 +199,7 <at> <at> { chomp $line ; ($language,$date,$reguser_bot,$group,$counts) = split (',', $line, 5) ; + next if $language =~ /^zz/ ; next if $language eq 'commons' and $project ne 'wx' ; # commons also in wikipedia csv files (bug, hard to cleanup, just skip) # next if $language eq 'commons' ; # ignore editor count on commons alltogether, most are already counted for other project <at> <at> -272,7 +273,6 <at> <at> ($language,$date,$counts) = split (',', $line, 3) ; if ($language ne "commons") { next ; } - if ($date eq "00/0000") { <at> fields = split (',', $counts) ; <at> <at> -330,6 +330,7 <at> <at> print "Write file '$file_csv_out'\n" ; open CSV_OUT, '>', $file_csv_out ; $output = "" ; + foreach $f (1,2,3,4,6,11) # new editors, editors_gt_5, editors_gt_100, articles, new articles, edits { <at> <at> -338,7 +339,12 <at> <at> $line = ",Total," ; for ($m = $m_start ; $m <= $m_last ; $m++) - { $line .= $totals {"$f,$m"} . "," ; } + { + if ($f <= 3) # new editors, editors_gt_5, editors_gt_100, + { $line .= $totals_project {"$f,$m"} {$all_projects} . "," ; } + else + { $line .= $totals {"$f,$m"} . "," ; } + } # growth in one year if ($totals {"$f,$m_last_12"} != 0) <at> <at> -390,6 +396,7 <at> <at> if ($f <= 3) # 0 = Contributors, 1 = New Wikimedians, 2 = Active Editors (5+ edits), 3 = Very Active Editors (100+ edits), { $output .= ",Note: All projects does not include Commons\n" ; } $output .= "$csv_recent_months,%inc year, %inc month\n" ; + foreach $project (sort {$totals_project {"$f,$m_last"} {$b} <=> $totals_project {"$f,$m_last"} {$a}} <at> projects) { # next if $project eq 'commons' and ($f ==2 or $f == 3) ; # (very) active editors no longer counted for commons <at> <at> -426,6 +433,8 <at> <at> # print "$index $f: $key -> ${values_f_12 {$key}}\n" ; ($project,$language) = split (",", $key) ; +if ($f == 2) +{ print "project $project $language\n" ; } $language_name = $out_languages {$language} ; if (($project ne "wp") && ($project ne "wx")) { $line = "$index,$language_name " . &GetProjectName ($project) . "," ; } <at> <at> -465,6 +474,7 <at> <at> $output .= "\n,${out_report_descriptions [$f]} - Indexed - Per Project\n" ; $output .= "$csv_recent_months,%inc year, %inc month\n" ; + foreach $project (sort {$totals_project {"$f,$m_last"} {$b} <=> $totals_project {"$f,$m_last"} {$a}} <at> projects) { # next if $project eq 'commons' and ($f ==2 or $f == 3) ; # (very) active editors no longer counted for commons Modified: trunk/wikistats/reportcard/ReportCardGenerateHtml.pl =================================================================== --- trunk/wikistats/reportcard/ReportCardGenerateHtml.pl 2012-06-04 10:41:35 UTC (rev 115475) +++ trunk/wikistats/reportcard/ReportCardGenerateHtml.pl 2012-06-04 12:57:59 UTC (rev 115476) <at> <at> -9,8 +9,8 <at> <at> use Getopt::Std ; # !! adapt these for every run !! - $p_year = 2011 ; - $p_month = 12 ; + $p_year = 2012 ; + $p_month = 2 ; $debug = $false ; <at> <at> -91,40 +91,40 <at> <at> # !! This is rather crummy if not outright stupid Q&D way to collect variable data, data need to be externalized !! -## if ($2011_12) +## if ($2012_02) ## { # projects <at> rank = qw ( 5th x 0 0 th) ; # All - Web Properties - Unique Visitors - <at> reach = qw ( 31.7 x 1.8 -1.3 %) ; # All - Reach Percentage by Region - <at> visitors = qw ( 457,063,000 m 15.6 -3.7 %) ; # All - Unique Visitors by Region - <at> page_requests = qw (16,365,306,811 b 17.3 -6.1 %) ; # All - copy/calc manually monthly total and monthly and yearly growth from 1st column (Sigma) of http://stats.wikimedia.org/EN/TablesPageViewsMonthly.htm (Wikipedia only is good enough) - <at> commons_files = qw ( 11,974,179 m 50.1 1.8 %) ; # Commons - Binaries per month - Absolute - <at> article_count = qw ( 20,845,401 m 18.6 1.9 %) ; # Wp - Starting Sep-2010 Wikipedia articles only / Article count (official) - Absolute - <at> new_articles = qw ( 12,759 k 50.0 67.2 %) ; # Wp - New articles per day - Absolute - <at> edits = qw ( 11,636,349 m 1.0 4.7 %) ; # Wp - Edits per month - Absolute - <at> new_editors = qw ( 15,181 k -9.1 -6.4 %) ; # All - New editors - Absolute - <at> active_editors = qw ( 83,293 k 1.6 0.1 %) ; # All - Active editors - Absolute - <at> very_active_editors = qw ( 11,041 k 2.7 2.8 %) ; # All - Very active editors - Absolute + <at> reach = qw ( 32.8 x 4.5 -0.4 %) ; # All - Reach Percentage by Region + <at> visitors = qw ( 475,699,000 m 25.4 -1.3 %) ; # All - Unique Visitors by Region + <at> page_requests = qw (18,127,912,107 b 16.2 0.4 %) ; # All - copy/calc manually monthly total and monthly and yearly growth from 1st column (Sigma) of http://stats.wikimedia.org/EN/TablesPageViewsMonthly.htm (Wikipedia only is good enough) + <at> commons_files = qw ( 12,367,426 m 32.1 1.5 %) ; # Commons - Binaries per month - Absolute + <at> article_count = qw ( 21,158,235 m 18.5 1.3 %) ; # Wp - Starting Sep-2010 Wikipedia articles only / Article count (official) - Absolute + <at> new_articles = qw ( 8,449 k -6.1 -36.7 %) ; # Wp - New articles per day - Absolute + <at> edits = qw ( 12,422,452 m -2.5 6.8 %) ; # Wp - Edits per month - Absolute + <at> new_editors = qw ( 15,333 k -14.7 -8.8 %) ; # All - New editors - Absolute + <at> active_editors = qw ( 85,163 k -1.7 -3.6 %) ; # All - Active editors - Absolute + <at> very_active_editors = qw ( 10,913 k 0.0 -6.8 %) ; # All - Very active editors - Absolute # http://stats.wikimedia.org/EN/TablesPageViewsMonthly.htm push <at> rank, "1|Site Rank<br>#1: 5th position will be stable for long time: differences with those ranked 4th and 6th are considerable." ; - push <at> reach, "2|Reach<br>#2: Change in reach in 3 years: (avg Oct-Dec 08 -> avg Oct-Dec 11) pp= percentage points: <br>" . - "# Europe +9pp, (38%->47%), North-Am +10pp (35%->45%), Asia +1pp (16%->18%),<br>" . - "# Latin-America +6pp (35%->41%), Middle East-Africa +2pp (30%->32%),<br>" . - "# India +16pp (20%->35%), World +4.8pp (27.8%->32.7%)" ; - push <at> visitors, "3,4|Unique Visitors<br>#3: 8% higher than average for last 12 months (422M).<br>" . -# " 4.9% higher than previous month (which was also an all time high)<br>" . - "#4: Change in UV's in 3 years: (avg Oct-Dec 08 -> avg Oct-Dec 11): <br>" . - "# Europe +68%, North-Am +46%, Latin-Am: +103%, Asia: +58%, World +69%<br>" . - "# India +164%, Middle East-Africa: +180%" ; - push <at> page_requests, "5,6,7,8,9|Page Requests<br>" . - "5: <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyCombined.htm'>Fastest and slowest growth in page requests in top 25 Wikipedias in 2011</a>:<br>" . - " Fast: Chinese +323%, Persian +114%, Ukranian +69%, Korean +65%, Arabic +43%,<br>" . - " Indonesian +40%, Hebrew +40%, Russian +38%, Thai +35%, Commons +33%.<br>" . - " Slow: Finnish +2%, Hungarian +5%.<br>" . - "6: <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyAllProjects.htm'>Projects ranked by growth in page requests in 2011</a>:<br>" . - " Wikiquote: +127%, Wiktionary: +57%, Commons: +33%, Wikibooks: +26%, Wikisource: +23%,<br>" . - " Wikipedia: +16%, Wikinews: +12%, Wikiversity: +5%<br>" . - "7: Numbers have been corrected for 88 hours of missing traffic data.<br>" . - "8: Wiktionary traffic back to normal after massive botnet spam in Oct/Nov.<br>" . + push <at> reach, "2|Reach<br>#2: Change in reach in 3 years: (avg Dec-Feb 08/09 -> avg Dec-Feb 11/12) pp= percentage points: <br>" . + "# Europe +8.3pp, (38%->47%), North-Am +10.6pp (36%->46%), Asia +1.4pp (16%->18%),<br>" . + "# Latin-America +5.7pp (31%->37%), Middle East-Africa +3.1pp (29%->32%),<br>" . + "# India +15.1pp (20%->35%), World +4.8pp (27.8%->32.6%)" ; + push <at> visitors, "3,4|Unique Visitors<br>3: 9% higher than average for last 12 months (436M), even on the shortest month of the year.<br>" . +# " 1.1% higher than previous record month Oct 2011<br>" . + "#4: Change in UV's in 3 years: (avg Dec-Feb 08/09 -> avg Dec-Feb 11/12): <br>" . + "# Europe +60%, North-Am +48%, Latin-Am: +98%, Asia: +53%, World +64%<br>" . + "# India +165%, Middle East-Africa: +149%" ; + push <at> page_requests, "5,6,7,8|Page Requests<br>" . +# "5: 58% of page requests to all Wiktionaries combined came from one ip address,<br>" . +# " which spammed Special:Random page on most languages.<br>" . + "#6: Fastest and slowest growth in page requests in top 25 Wikipedias in 2011:<br>" . + "# Fast: Chinese +323%, Persian +114%, Ukranian +69%, Korean +65%, Arabic +43%,<br>" . + "# Indonesian +40%, Hebrew +40%, Russian +38%, Thai +35%, Commons +33%.<br>" . + "# Slow: Finnish +2%, Hungarian +5%, see <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyCombined.htm'>details</a><br>" . + "#7: Projects ranked by growth in page requests in 2011:<br>" . + "# Wikiquote: +127%, Wiktionary: +57%, Commons: +33%, Wikibooks: +26%, Wikisource: +23%,<br>" . + "# Wikipedia: +16%, Wikinews: +12%, Wikiversity: +5%, see <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyAllProjects.htm'>details</a><br>" . # " Reassesment may follow later.<br>" ; # "6: Note that total requests <b>from mobile devices</b> is even much higher, approx half go to regular site" ; # "6: <a href='http://stats.wikimedia.org/wikimedia/squids/SquidReportClients.htm'>Total file requests from mobile devices for all Wikimedia projects</a> in March: 7.2%!<br>" . <at> <at> -135,19 +135,31 <at> <at> # "#8: Page views for Chinese Wikipedia went from 52M in January to 163M in April.<br>" . # " #Unique visitors from China stayed low as usual, even dropped 6% from January to April.<br>" . # " #(Chinese community said year ago they had issues with Google ranking, maybe these were fixed?)<br>" . - "#9: Page requests have been normalized to 30 days (Jan*30/31, Feb*30/28, Mar*30/31, etc)<br>" ; + "#8: Page requests have been normalized to 30 days (Jan*30/31, Feb*30/28, Mar*30/31, etc)<br>" ; # push <at> commons_files, "7|Commons Files<br>7: Recent growth in " . # "<br> Probably related to Wiki Loves Monuments contest (WLM)</a><br>" ; - push <at> article_count, "10|Article Count<br>10: Some Wikipedias contain massive amounts of bot created stubs,<br>" . - " (e.g. Dutch Wikipedia 42% of total, Volap\xFCk 92% of total)<br>" . - " which makes article count a questionable metric of community effort.<br>" . - " see also <a href='http://stats.wikimedia.org/EN/BotActivityMatrixCreates.htm'>new report</a> on bot induced article creation activity per wiki." ; + push <at> article_count, "9|" ; +# push <at> article_count, "9|Article Count<br>#9: Some Wikipedias contain massive amounts of bot created stubs,<br>" . +# "# (e.g. Dutch Wikipedia 42% of total, Volap\xFCk 92% of total)<br>" . +# "# which makes article count a questionable metric of community effort.<br>" . +# "# see also new report on <a href='http://stats.wikimedia.org/EN/BotActivityMatrixCreates.htm'>bot induced article creation activity per wiki.</a>" ; ## "9: Seven Wiktionaries in top 25 Wikimedia projects" ; -push <at> new_articles, "10|New Articles Per Day<br>" . - "10: On Dutch Wikipedia <a href='http://nl.wikipedia.org/wiki/Speciaal:Bijdragen/Joopwikibot'>Joopwikibot</a> created roughly 164k of trivial taxonomy stubs in last 3 months." ; +push <at> new_articles, "9|Article Count, New Articles<br>9: Data for January (February will be added very soon)" ; +push <at> edits, "10|" ; +push <at> new_editors, "10|" ; +push <at> active_editors, "10|" ; +push <at> very_active_editors, "10|Edits, New Editors, (Very)Active Editors<br>10: Provisionary data, final data will arrive soon (< 1 % change)" ; + +#push <at> new_articles, "10,11|New Articles Per Day<br>" . +# "10: Sharp drop in overall number of new articles on all Wikipedias combined mostly due to bot inactivity<br>" . +# " on Dutch Wikipedia, which in Dec accounted for 33% of all new articles.<br>" . +# "11: Still this month the number of wikis with exceptionally large numbers of new articles rose sharply:<br>" . +# " New articles per day for Dec->Jan: Slovene 33->312, Bashkir 4->308, Malagasy 3->232,<br>" . +# " Latin 45->215, Armenian 31->123, Fiji Hindi 22->83<br>" . +# " For comparison, two large Wikipedias with normal article creation rate: Russian 348, Spanish 285" ; ## push <at> edits, "6|Edits<br>#6: Over the last 3 years there is fairly consistent growth in manual, registered edits.<br>" . ## "# Net growth in constructive edits is less clear, as this metric includes most reverting edits." ; ## " Strong one-monthly dip in July due to World Cup Socker?." ; <at> <at> -169,8 +181,169 <at> <at> ## } +### if ($2012_01) +### { # projects +# <at> rank = qw ( 5th x 0 0 th) ; # All - Web Properties - Unique Visitors +# <at> reach = qw ( 33.2 x 2.1 1.5 %) ; # All - Reach Percentage by Region +# <at> visitors = qw ( 482,157,000 m 16.5 5.5 %) ; # All - Unique Visitors by Region +# <at> page_requests = qw (18,054,255,294 b 16.5 10.3 %) ; # All - copy/calc manually monthly total and monthly and yearly growth from 1st column (Sigma) of http://stats.wikimedia.org/EN/TablesPageViewsMonthly.htm (Wikipedia only is good enough) +# <at> commons_files = qw ( 12,200,544 m 48.7 1.9 %) ; # Commons - Binaries per month - Absolute +# <at> article_count = qw ( 21,158,235 m 18.5 1.3 %) ; # Wp - Starting Sep-2010 Wikipedia articles only / Article count (official) - Absolute +# <at> new_articles = qw ( 8,449 k -6.1 -36.7 %) ; # Wp - New articles per day - Absolute +# <at> edits = qw ( 12,422,452 m -2.5 6.8 %) ; # Wp - Edits per month - Absolute +# <at> new_editors = qw ( 16,170 k -16.0 2.1 %) ; # All - New editors - Absolute +# <at> active_editors = qw ( 88,548 k -3.2 6.7 %) ; # All - Active editors - Absolute +# <at> very_active_editors = qw ( 11,730 k -0.5 6.3 %) ; # All - Very active editors - Absolute +# # http://stats.wikimedia.org/EN/TablesPageViewsMonthly.htm +# push <at> rank, "1|Site Rank<br>#1: 5th position will be stable for long time: differences with those ranked 4th and 6th are considerable." ; +# push <at> reach, "2|Reach<br>#2: Change in reach in 3 years: (avg Nov-Jan 08/09 -> avg Nov-Jan 11/12) pp= percentage points: <br>" . +# "# Europe +9pp, (38%->47%), North-Am +10pp (35%->45%), Asia +1pp (17%->18%),<br>" . +# "# Latin-America +6pp (32%->39%), Middle East-Africa +2pp (31%->32%),<br>" . +# "# India +15pp (20%->36%), World +4.7pp (27.9%->32.6%)" ; +# push <at> visitors, "3,4|Unique Visitors<br>3: 13% higher than average for last 12 months (428M).<br>" . +# " 1.1% higher than previous record month Oct 2011<br>" . +# "#4: Change in UV's in 3 years: (avg Nov-Jan 11/12 -> avg Nov-Jan 11/12): <br>" . +# "# Europe +66%, North-Am +47%, Latin-Am: +105%, Asia: +54%, World +68%<br>" . +# "# India +157%, Middle East-Africa: +173%" ; +# push <at> page_requests, "5,6,7,8|Page Requests<br>" . +# "5: 58% of page requests to all Wiktionaries combined came from one ip address,<br>" . +# " which spammed Special:Random page on most languages.<br>" . +# "#6: Fastest and slowest growth in page requests in top 25 Wikipedias in 2011:<br>" . +# "# Fast: Chinese +323%, Persian +114%, Ukranian +69%, Korean +65%, Arabic +43%,<br>" . +# "# Indonesian +40%, Hebrew +40%, Russian +38%, Thai +35%, Commons +33%.<br>" . +# "# Slow: Finnish +2%, Hungarian +5%, see <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyCombined.htm'>details</a><br>" . +# "#7: Projects ranked by growth in page requests in 2011:<br>" . +# "# Wikiquote: +127%, Wiktionary: +57%, Commons: +33%, Wikibooks: +26%, Wikisource: +23%,<br>" . +# "# Wikipedia: +16%, Wikinews: +12%, Wikiversity: +5%, see <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyAllProjects.htm'>details</a><br>" . +## " Reassesment may follow later.<br>" ; +## "6: Note that total requests <b>from mobile devices</b> is even much higher, approx half go to regular site" ; +## "6: <a href='http://stats.wikimedia.org/wikimedia/squids/SquidReportClients.htm'>Total file requests from mobile devices for all Wikimedia projects</a> in March: 7.2%!<br>" . +## " See also <a href='charts/2011-04/RatioMobileTraffic.png'>Plot for both metrics.</a><br>" . +## "7: Note: Requests from mobile devices to the non-mobile site, which are redirected to the mobile site<br>" . +## " are probably counted twice. Follow up requests from the mobile device via internal links<br>" . +## " are not affected. Number of double counts not yet known.<br>" . +## "#8: Page views for Chinese Wikipedia went from 52M in January to 163M in April.<br>" . +## " #Unique visitors from China stayed low as usual, even dropped 6% from January to April.<br>" . +## " #(Chinese community said year ago they had issues with Google ranking, maybe these were fixed?)<br>" . +# "#8: Page requests have been normalized to 30 days (Jan*30/31, Feb*30/28, Mar*30/31, etc)<br>" ; +## push <at> commons_files, "7|Commons Files<br>7: Recent growth in " . +## "<br> Probably related to Wiki Loves Monuments contest (WLM)</a><br>" ; +# push <at> article_count, "9|Article Count<br>#9: Some Wikipedias contain massive amounts of bot created stubs,<br>" . +# "# (e.g. Dutch Wikipedia 42% of total, Volap\xFCk 92% of total)<br>" . +# "# which makes article count a questionable metric of community effort.<br>" . +# "# see also new report on <a href='http://stats.wikimedia.org/EN/BotActivityMatrixCreates.htm'>bot induced article creation activity per wiki.</a>" ; + +### "9: Seven Wiktionaries in top 25 Wikimedia projects" ; +#push <at> new_articles, "10,11|New Articles Per Day<br>" . +# "10: Sharp drop in overall number of new articles on all Wikipedias combined mostly due to bot inactivity<br>" . +# " on Dutch Wikipedia, which in Dec accounted for 33% of all new articles.<br>" . +# "11: Still this month the number of wikis with exceptionally large numbers of new articles rose sharply:<br>" . +# " New articles per day for Dec->Jan: Slovene 33->312, Bashkir 4->308, Malagasy 3->232,<br>" . +# " Latin 45->215, Armenian 31->123, Fiji Hindi 22->83<br>" . +# " For comparison, two large Wikipedias with normal article creation rate: Russian 348, Spanish 285" ; +### push <at> edits, "6|Edits<br>#6: Over the last 3 years there is fairly consistent growth in manual, registered edits.<br>" . +### "# Net growth in constructive edits is less clear, as this metric includes most reverting edits." ; +### " Strong one-monthly dip in July due to World Cup Socker?." ; +### "#13: Average monthly manual edits by registered users for all Wikipedia's combined, in millions<br>" . +### " #2006 ⇒ 2010: 7.7 ⇒ 9.9 ⇒ 11.5 ⇒ 12.4 ⇒ 12.7" ; +## push <at> new_editors, "8|New Editors Per Month<br>" . +## "8: Continued decline in new editors at large wikipedias, but sharp rise at Commons (WLM)" ; +### " yet with 8120 best result in last 10 months (average last 12 months: 7805)" ; +### " Arguably slowing influx of editors can partly be attributed to (multi-factorial) <a href='http://en.wikipedia.org/wiki/Market_saturation'>saturation process(es)</a><br>" . +### " But 19% drop for Wikipedias in half year (comparing 3-monthly averages) is not consistent with that.<br>" . +### "#7:WMF recently commissioned in depth study of editor activity trends, which is ongoing." ; +### "#8: Since a few months editors on Commons are no longer included in overall editor total,<br>" . +### "# on the assumption that most of these also edit on one or more other projects.<br>" ; +### " #Detection of double counts between any projects and languages is planned for late 2010." ; +## push <at> active_editors, "7|(Very) Active Editors<br>" . +## "7: Recent spike in (very) active editors on Commons probably caused by Wiki Loves Monuments project (temporary), but also by improved upload facilities (permanent)." ; +## push <at> very_active_editors, "7|" ; +## " Rise in many wikipedias outweighed <a href='charts/2011-09/Editors-Active-Indexed-Wikis.png'>slow decline</a> (-10% in 3 years) in English, German, Japanese" ; + +### } + +### if ($2011_12) +### { # projects +# <at> rank = qw ( 5th x 0 0 th) ; # All - Web Properties - Unique Visitors +# <at> reach = qw ( 31.7 x 1.8 -1.3 %) ; # All - Reach Percentage by Region +# <at> visitors = qw ( 457,063,000 m 15.6 -3.7 %) ; # All - Unique Visitors by Region +# <at> page_requests = qw (16,365,306,811 b 17.3 -6.1 %) ; # All - copy/calc manually monthly total and monthly and yearly growth from 1st column (Sigma) of http://stats.wikimedia.org/EN/TablesPageViewsMonthly.htm (Wikipedia only is good enough) +# <at> commons_files = qw ( 11,974,179 m 50.1 1.8 %) ; # Commons - Binaries per month - Absolute +# <at> article_count = qw ( 20,845,401 m 18.6 1.9 %) ; # Wp - Starting Sep-2010 Wikipedia articles only / Article count (official) - Absolute +# <at> new_articles = qw ( 12,759 k 50.0 67.2 %) ; # Wp - New articles per day - Absolute +# <at> edits = qw ( 11,636,349 m 1.0 4.7 %) ; # Wp - Edits per month - Absolute +# <at> new_editors = qw ( 15,181 k -9.1 -6.4 %) ; # All - New editors - Absolute +# <at> active_editors = qw ( 83,293 k 1.6 0.1 %) ; # All - Active editors - Absolute +# <at> very_active_editors = qw ( 11,041 k 2.7 2.8 %) ; # All - Very active editors - Absolute +# # http://stats.wikimedia.org/EN/TablesPageViewsMonthly.htm +# push <at> rank, "1|Site Rank<br>#1: 5th position will be stable for long time: differences with those ranked 4th and 6th are considerable." ; +# push <at> reach, "2|Reach<br>#2: Change in reach in 3 years: (avg Oct-Dec 08 -> avg Oct-Dec 11) pp= percentage points: <br>" . +# "# Europe +9pp, (38%->47%), North-Am +10pp (35%->45%), Asia +1pp (16%->18%),<br>" . +# "# Latin-America +6pp (35%->41%), Middle East-Africa +2pp (30%->32%),<br>" . +# "# India +16pp (20%->35%), World +4.8pp (27.8%->32.7%)" ; +# push <at> visitors, "3,4|Unique Visitors<br>#3: 8% higher than average for last 12 months (422M).<br>" . +## " 4.9% higher than previous month (which was also an all time high)<br>" . +# "#4: Change in UV's in 3 years: (avg Oct-Dec 08 -> avg Oct-Dec 11): <br>" . +# "# Europe +68%, North-Am +46%, Latin-Am: +103%, Asia: +58%, World +69%<br>" . +# "# India +164%, Middle East-Africa: +180%" ; +# push <at> page_requests, "5,6,7,8,9|Page Requests<br>" . +# "5: <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyCombined.htm'>Fastest and slowest growth in page requests in top 25 Wikipedias in 2011</a>:<br>" . +# " Fast: Chinese +323%, Persian +114%, Ukranian +69%, Korean +65%, Arabic +43%,<br>" . +# " Indonesian +40%, Hebrew +40%, Russian +38%, Thai +35%, Commons +33%.<br>" . +# " Slow: Finnish +2%, Hungarian +5%.<br>" . +# "6: <a href='http://stats.wikimedia.org/EN/TablesPageViewsMonthlyAllProjects.htm'>Projects ranked by growth in page requests in 2011</a>:<br>" . +# " Wikiquote: +127%, Wiktionary: +57%, Commons: +33%, Wikibooks: +26%, Wikisource: +23%,<br>" . +# " Wikipedia: +16%, Wikinews: +12%, Wikiversity: +5%<br>" . +# "7: Numbers have been corrected for 88 hours of missing traffic data.<br>" . +# "8: Wiktionary traffic back to normal after massive botnet spam in Oct/Nov.<br>" . +## " Reassesment may follow later.<br>" ; +## "6: Note that total requests <b>from mobile devices</b> is even much higher, approx half go to regular site" ; +## "6: <a href='http://stats.wikimedia.org/wikimedia/squids/SquidReportClients.htm'>Total file requests from mobile devices for all Wikimedia projects</a> in March: 7.2%!<br>" . +## " See also <a href='charts/2011-04/RatioMobileTraffic.png'>Plot for both metrics.</a><br>" . +## "7: Note: Requests from mobile devices to the non-mobile site, which are redirected to the mobile site<br>" . +## " are probably counted twice. Follow up requests from the mobile device via internal links<br>" . +## " are not affected. Number of double counts not yet known.<br>" . +## "#8: Page views for Chinese Wikipedia went from 52M in January to 163M in April.<br>" . +## " #Unique visitors from China stayed low as usual, even dropped 6% from January to April.<br>" . +## " #(Chinese community said year ago they had issues with Google ranking, maybe these were fixed?)<br>" . +# "#9: Page requests have been normalized to 30 days (Jan*30/31, Feb*30/28, Mar*30/31, etc)<br>" ; +## push <at> commons_files, "7|Commons Files<br>7: Recent growth in " . +## "<br> Probably related to Wiki Loves Monuments contest (WLM)</a><br>" ; + +# push <at> article_count, "10|Article Count<br>10: Some Wikipedias contain massive amounts of bot created stubs,<br>" . +# " (e.g. Dutch Wikipedia 42% of total, Volap\xFCk 92% of total)<br>" . +# " which makes article count a questionable metric of community effort.<br>" . +# " see also <a href='http://stats.wikimedia.org/EN/BotActivityMatrixCreates.htm'>new report</a> on bot induced article creation activity per wiki." ; + + +### "9: Seven Wiktionaries in top 25 Wikimedia projects" ; +#push <at> new_articles, "10|New Articles Per Day<br>" . +# "10: On Dutch Wikipedia <a href='http://nl.wikipedia.org/wiki/Speciaal:Bijdragen/Joopwikibot'>Joopwikibot</a> created roughly 164k of trivial taxonomy stubs in last 3 months." ; +### push <at> edits, "6|Edits<br>#6: Over the last 3 years there is fairly consistent growth in manual, registered edits.<br>" . +### "# Net growth in constructive edits is less clear, as this metric includes most reverting edits." ; +### " Strong one-monthly dip in July due to World Cup Socker?." ; +### "#13: Average monthly manual edits by registered users for all Wikipedia's combined, in millions<br>" . +### " #2006 ⇒ 2010: 7.7 ⇒ 9.9 ⇒ 11.5 ⇒ 12.4 ⇒ 12.7" ; +## push <at> new_editors, "8|New Editors Per Month<br>" . +## "8: Continued decline in new editors at large wikipedias, but sharp rise at Commons (WLM)" ; +### " yet with 8120 best result in last 10 months (average last 12 months: 7805)" ; +### " Arguably slowing influx of editors can partly be attributed to (multi-factorial) <a href='http://en.wikipedia.org/wiki/Market_saturation'>saturation process(es)</a><br>" . +### " But 19% drop for Wikipedias in half year (comparing 3-monthly averages) is not consistent with that.<br>" . +### "#7:WMF recently commissioned in depth study of editor activity trends, which is ongoing." ; +### "#8: Since a few months editors on Commons are no longer included in overall editor total,<br>" . +### "# on the assumption that most of these also edit on one or more other projects.<br>" ; +### " #Detection of double counts between any projects and languages is planned for late 2010." ; +## push <at> active_editors, "7|(Very) Active Editors<br>" . +## "7: Recent spike in (very) active editors on Commons probably caused by Wiki Loves Monuments project (temporary), but also by improved upload facilities (permanent)." ; +## push <at> very_active_editors, "7|" ; +## " Rise in many wikipedias outweighed <a href='charts/2011-09/Editors-Active-Indexed-Wikis.png'>slow decline</a> (-10% in 3 years) in English, German, Japanese" ; + +### } + + + ### if ($2011_11) ### { # <at> rank = qw ( 5th x 0 0 th) ; # Web Properties - Unique Visitors <at> <at> -1277,7 +1450,8 <at> <at> { $line =~ s/\{\{yyyy\}\}_\{\{mm\+1\}\}/{{yyyy+1}}_{{mm+1}}/ ; } # Q&D temp fix } -# $no_upd = "<font color=#C00000><small>No data yet for English Wikipedia.</small><\/font>" ; + $no_upd_feb = "<font color=#C00000><small>Data for January, February will be added soon.</small><\/font>" ; + $no_upd_feb2 = "<br><font color=#C00000><small>Provisionary data, final adjustment will follow soon. (< 1% change)</small><\/font>" ; if ($true) # test ? { <at> <at> -1286,12 +1460,12 <at> <at> $line =~ s/H2 (PAGE REQUESTS)/A[$1] H2 {${page_requests[0]} billion|Page Requests for All Wikimedia Projects}/ ; $line =~ s/H2 (WEB PROPERTIES)/A[$1] H2 {${rank[0]} in rank|Web Properties - Unique Visitors}/ ; $line =~ s/H2 (COMMONS FILES)/A[$1] H2 {${commons_files[0]} million|Binary Files}/ ; - $line =~ s/H2 (ARTICLE COUNT)/A[$1] H2 {${article_count[0]} million|Wikipedia Articles $no_upd}/ ; - $line =~ s/H2 (ARTICLES PER DAY)/A[$1] H2 {${new_articles[0]}|New Wikipedia Articles Per Day $no_upd}/ ; - $line =~ s/H2 (EDITS PER MONTH)/A[$1] H2 {${edits[0]} million|Wikipedia Edits Per Month $no_upd}/ ; - $line =~ s/H2 (NEW EDITORS PER MONTH)/A[$1] H2 {${new_editors[0]}|New Editors Per Month for All Wikimedia Projects $no_upd}/ ; - $line =~ s/H2 (ACTIVE EDITORS)/A[$1] H2 {${active_editors[0]}|Active Wikimedia Editors for All Wikimedia Projects <small>(5+ edits per month)<\/small> $no_upd}/ ; - $line =~ s/H2 (VERY ACTIVE EDITORS)/A[$1] H2 {${very_active_editors[0]}|Very Active Editors for All Wikimedia Projects <small>(100+ edits per month)<\/small> $no_upd}/ ; + $line =~ s/H2 (ARTICLE COUNT)/A[$1] H2 {${article_count[0]} million|Wikipedia Articles $no_upd_feb}/ ; + $line =~ s/H2 (ARTICLES PER DAY)/A[$1] H2 {${new_articles[0]}|New Wikipedia Articles Per Day $no_upd_feb}/ ; + $line =~ s/H2 (EDITS PER MONTH)/A[$1] H2 {${edits[0]} million|Wikipedia Edits Per Month $no_upd_feb2}/ ; + $line =~ s/H2 (NEW EDITORS PER MONTH)/A[$1] H2 {${new_editors[0]}|New Editors Per Month for All Wikimedia Projects $no_upd_feb2}/ ; + $line =~ s/H2 (ACTIVE EDITORS)/A[$1] H2 {${active_editors[0]}|Active Wikimedia Editors for All Wikimedia Projects <small>(5+ edits per month)<\/small> $no_upd_feb2}/ ; + $line =~ s/H2 (VERY ACTIVE EDITORS)/A[$1] H2 {${very_active_editors[0]}|Very Active Editors for All Wikimedia Projects <small>(100+ edits per month)<\/small> $no_upd_feb2}/ ; $line =~ s/TRENDS UNIQUE VISITORS/TRENDS {$trend_one_year|${visitors[2]}%}{$trend_one_month|${visitors[3]}%}/ ; $line =~ s/TRENDS PAGE REQUESTS/TRENDS {$trend_one_year|${page_requests[2]}%}{$trend_one_month|${page_requests[3]}%}/ ; <at> <at> -1559,7 +1733,8 <at> <at> { my $path_out = shift ; - $notice_synopsis = "<font color=#008000><b>Please keep in mind December has one day more than November, which influences some of the metrics.</b></font><br>" . + $notice_synopsis = "<font color=#008000><b>Please remember subsequent months can have a different number of days, which influences<br>" . + "monthly change percentages. Leap year February still has 7% less days than January</b></font><br>" . "<font color=#008000>Exception: page request counts are always normalized to 30 day months.</font>" ; open SYNOPSIS, '>', "$path_out/RC_${p_year}_${p_month_d2}_synopsis.txt" ;
RSS Feed