PloneMetrics Person of the Year

It's that time of year again when I peruse the events of the past year and unilaterally designate some lucky person "Plone Metrics Person of the Year."  This award comes with only the promise that if the winner and I are in the same town at the same time, I'll buy them a beer. 

Past winners have been:

2009	Nate Aune
2008	Joel Burton
2007	Alexander Limi

The competition has been particular stiff, with outstanding contributions from a wide number of Plonistas worldwide.  It's been a eventful year for Plone and that makes 2010's throwdown especially difficult.  Noteworthy contributions during 2010 that deserve a shout-out include, but are not limited to:

• The 2010 Plone Conference people at Netsight, who did a great job with this year's event. Matt Hamilton's (general and host-specific) and Astra Fowden's summaries are worth their weight in gold. 
• The release managers for Plone 4 (Eric Steele) and 5 (Hanno Schlichting) are also high on my list.  Of course, their job wouldn't be possible without the core developers and all the Plone contributors.  Hats off to them all. 
• As always, the Foundation Board gets a great big nod of appreciation here.  Gier Baekholt (2009-10) and Calvin Hendryx-Parker (2010-11) deserve special mention for shepherding the Plone community.
• There is also some local Albuquerque talent among this year's nominees:  Michael Bernstein (Code for America fellow) and Emily Lewis (Webuquerque sparkplug).
• The Plone IRC Superstar Contest winner and runners up always deserve special praise.  The final determination has still not been announced as we go to press, but this is a reminder that the IRC community is a key part of the Plone ecosystem.   
• In a similar vein, here's a thanks to all those blogging and microblogging about Plone.  I'm particularly pleased to see diversity highlighted with a successful year of PloneChix. 
• Finally, Roberto Allende gets a big tip o' the hat as the champion of World Plone Day.  This year André Nogueira takes over the reins as we look forward to WPD 27 April.  

All that said, with this being a metrics-centric blog, I thought I'd run a few numbers and see what falls out.

As I mentioned Matt and Netsight are serious contenders.  Here's chart of World Plone Conference attendance since 2006.

Bristol turned out to be right in the middle of the range and probably would have done better had the world economy been brighter.  I regret missing the conference (yup, tight budget), but the high quality videos of the presentations has been a terrific resource that has softened the blow.  They have placed the bar high for future conferences. 

Another notable nominee for today's award is Eric Steele.  Here's the graphic that shows the impact of the Plone 4 release.  

Not only was there a huge spike in Plone.org visitors associated with the release, but a sustained increase of 15-20% that continues to this day.  (Green dashes show pre-release trend; red dotted line represents post-release trend.)

Based on that gi-normous spike, 130% above a typical Wednesday, plus the continued bump-up in Plone visitors, the Plone Metrics Person of the Year Award for 2010 goes to Eric.  By extension, this year's award goes out to all those who contributed to Plone 4 with code, testing, feedback, documentation, product upgrades, and all the myriad of details that make up a successful release. Well done, everyone!

In closing, let me paraphrase the immortal words of Bobby Flay: "So to all you awesome [Plonistas] out there, keep doing what you're doing, but ask yourself this... are you ready for a throwdown?"

Plone Mobile

I've been putting together a presentation on various web metrics for the gang at my day-job.  In particular, people are asking about Google Analytics.  As part of that, I've been poking around in a few less frequented parts of the system and some of that has carried over to inquiries about Plone.org users.  Tonight I'm going to focus on the growing mobile community of visitors. 

Above is the graph of visitors using mobile devices since GA started tracking them in November 2009.  The most notable feature is the incredible peak that marks the release of Plone 4.0.  The aftershocks lasted two weeks. 

The other significant feature is the upward trend.  There is a perceptible upward slope to the number of mobile users.  In fact, if we remove the Plone 4.0 spike, the regression analysis still gives us a slope of 0.13 (a new mobile visitor every seven days).  The regression coefficient (R^2) is only 0.65, so there's lots of noise in the data, mostly from weekday-weekend fluctuations.

To really see the trend, compare the last month with the same period a year ago.  Clearly mobile use is on the upswing.

My final note is on the distribution of mobile users.  From the city-level global map of mobile visitors to Plone.org, it's pretty obvious that mobile users are world-wide.  Of course, it comes as no surprise that mobile use tracks Plone.org general use, heavy in North America and Europe.  But there are significant hotspots in South America, the Middle East, south and east Asia, as well as Australia and New Zealand.

IdealWare CMS Report

Once again IdealWare has come out with their 2010 annual review of Plone, WordPress, Drupal, and Joomla. As last year, they are one of the few software comparisons of any type to have a completely transparent methodology. Not only is the body of the report well balanced and thought out, the detailed system-by-system accounting of the subtopics that make up their category scores makes for very worthwhile reading. Tip o' the hat again.

Also like last year, I'm taking the liberty of translating IdealWare's scoring of  None, Fair, Solid, Excellent into 0, 1, 2, 3.  The result is the following table.

	WordPress	Joomla 1.6	Drupal	Plone
Ease of Hosting and Installation	3	3	3	1
Ease of Setup: Simple Site	3	2	2	1
Ease of Setup: Complex Site	3	3	2	2
Ease of Use: Content Editors	3	2	2	3
Ease of Use: Site Administrator	3	2	2	2
Graphical Flexibility	3	3	3	3
Accessibility and Search Engine Optimization	2	3	1	3
Structural Flexiblity	2	2	3	3
User Roles and Workflow	1	3	2	3
Community/Web 2.0 Functionality	3	2	3	2
Extending and Integrating	3	3	3	3
Security	1	2	2	3
Support/Community Strength	3	3	3	3
2010	33	33	31	32
2009*	29	27	30	29

* In 2009 only 12 categories (instead of 13) were used.

Once again there is only a small spread in the scores, about 6%, well below the 10% of 2009.  All CMS's here are improving quickly.  Reviewers still hold Plone's non-LAMP hosting model against it, giving it a Fair for hosting and installation.  I beg to differ with them about setting up a simple site.  Once the software is installed, Plone's out-of-the-box feature set solves a wide range of CMS use-cases with a minimum of fuss. 

What I stated last year still stands:

"Its always nice to see good scores and the fact that Idealware chose Plone along with only 3 other serious open source systems is high praise in and of itself. Idealware should be complimented for having a transparent methodology, a relatively neutral approach, and giving good press to four worthwhile systems.

"Still, I can't stress enough that your specific, unique requirements must drive your choice for a CMS. Don't let someone else's numbers blind you to what you and your community of users need to accomplish."

I urge those shopping for a CMS to weigh the scores according to your needs and requirements.  If ease of use for content contributors, SEO excellence, and security are key features, Plone clearly comes out ahead.  Go ahead and cut-and-paste my table into a spreadsheet, change the values, play with different weights, and then be sure to take a test drive** before you make a deep and lasting commitment.  Changing CMS in mid-stream is not a simple task--take the long view and get it right the first time. 
_______________
** Try a Plone demo site of your own at Six Feet Up or request a free Plone site at Objectis. 

CMS Comparison, expanded for Plone

Ashleigh Davis put together a nice table for comparing Drupal, Joomla, and WordPress.  It seems only fair that I take that format one step further and produce an equivalent listing for Plone.

	Plone
Homepage	plone.org
About	Plone lets non-technical people create and maintain information for a public website or an intranet using only a web browser. Plone is easy to understand and use — allowing users to be productive in just half an hour — yet offers a wealth of community-developed add-ons and extensibility to keep meeting your needs for years to come.
Example Sites	Online publication: http://discovermagazine.com/ Charitable giving: http://www.oxfamamerica.org/ http://www.globalhand.org/ Government: https://www.cia.gov/ http://www.fbi.gov/
Installation	http://plone.org/documentation/manual/installing-plone http://plone.org/support/forums/setup
Ease of use	Plone requires technical expertise for advanced customization; however, it is very full featured out-of-the-box. Content contributors, editors, and readers will find Plone easy to understand and use.
Features	Blending the creativity and speed of open source with a technologically advanced Python back-end, Plone offers superior security without sacrificing power or extensibility.
Caching Plug-Ins	ZEO server (also works and plays well with CacheFu and Squid)
Best Use Cases	For web content management, portals, content integration, collaboration, social software, and enterprise content management, especially where security is a premium.

I'd appreciate comments on how to improve this summary, especially on caching plug-ins, an area where I am weak.

Plone.org, as others see it

I've been working on a Google Analytics presentation for a January symposium at my day-job.  That has led me to poke around a little more at the Plone.org stats there.  

One relatively new feature is In-Page Analytics.  It gives you as close to a heat map as you can get for free.  Percentage clicks appear as small orange bubbles above each link.  Interestingly, the top links for Plone.org are Downloads (26%), New Features (19%), Documentation (16%), and Home (8%).  That's 69% of all clicks in just four areas.  

One cute trick (at least in Firefox) is to open the site in question in another tab.  The in-page bubbles are also applied in the new tab, which let's you get a better picture of the context of the statistics.  

Another statistic tucked away in Analytics is the frequency of screen resolutions used by site visitors.  Almost two thirds of Plone.org site visitors use one of just five screen resolutions:

1280x800	15.90%
1280x1024	14.79%
1440x900	11.75%
1680x1050	11.04%
1024x768	8.86%
	62.34%

To my surprise, 1280x800 is the most frequently used screen size.  Here's a screen shot of Plone.org resized to that dimension. 

Plone

Almost all of the news and upcoming events are "below the fold" at this resolution, as is the "Plone is among the top 2%..." text block.  But the good news is that 69% of all visitor clicks are on screen at this resolution. 

Also below the fold is the quasi-mega-menu in the footer. As Jakob Nielson recently pointed out mega-menus can be done badly, but done well (see his original March 2009 piece), they can enhance user experience, reduce scrolling, and help users find what they are looking for more easily.  Just a thought...

Now let's take that same 1280x800 resolution for a stroll with some other CMS homepages. 

Drupal

Although it looks complete because the 800 px "fold" breaks in a clear area, a large bulk of material (and a jam-packed quasi-mega-menu at the bottom) needs scrolling before it can be seen.

Joomla!

 

In some ways like the Plone site, Announcements and Community News are truncated.  Curiously, there's almost 1 cm of blank background above the content header.  Perhaps that's to accommodate the snazzy "Support Joomla" diagonal banner. 

SharePoint

Not so bad at first glance, until you realize that the entire area in line with the lighter gray background band is a rotating set of images.  The only fixed content is the lengthy and hard to scan horizontal top menu and three highlighted but unclickable headings (What People are Saying, Capabilities & Benefits, SharePointSpotlight).  Download SharePoint is practically invisible in the lower right and I almost missed the little cropped thumbnails in the set of images.

The Moral of the Story
  I guess the take-away lesson from all this is to test any webpage design with a variety of screen resolutions before you lock in on something suboptimal for a large number of your site visitors.  Take advantage of In-Page Analytics and make sure that highly visited links are highly visible and don't require scrolling to locate.  The most important things in your page should be the most important things.

Plone.org and Plone.net Tag Clouds

It's certainly easy enough to generate tag clouds these days, but two places I'd not seen it were the Plone sites.  So here you have it:  Plone.org and Plone.net word clouds. 

My method was to extract as CSV the top 500 titles from Google Analytics using their Content by Title option for each site.  That was cleaned up a bit in Open Office Calc.  Then it was simply a matter of copying-and-pasting all the rows into Wordle and looking for a pleasant and readable result.  I used Wordle's default ceiling of 150 most frequent words. 

From Plone.org:

From Plone.net:

From both of these, I'd have to say that the Plone sites are doing a very good job of covering their respective domains.  Neither site has just a few terms crushing the others into the background.  Frequency distribution of terms is well graded.  There's a wide and useful vocabulary displayed in each graphic, indicating that all the key topics are indeed key topics. 

Share and enjoy!

For the Gang in Bristol

As the World Plone 2010 Conference moves into its final day (exclusive of the weekend sprints), I thought I'd type while the Plonistas in Bristol sleep.  OK, it's only midnight over there, so probably no one is really asleep.  You know what they say, "Sleep is for wusses." 

Here's a synopsis of some of the measures that I track over here at Plone Metrics, just for discussion starters. 

First off, here's the lastest Amazon Sales Rank stats.  Normally I do this quarterly, but grabbing the data a month or so early won't do any harm. 

Knox and Stahl continue to hold a good sales rank (low value) along with McKay and Redomino as well as McKay's Kindle version.  I'll post the raw data over at the Plone Marketing site later.

The LaunchPad stats are always fun.  Release 4.0.0 has already garnered 29,840 downloads of the Windows installer.  An additional 4,206 downloads of vers. 4.0.1 have taken place this month since its release on the 4th. That's 34,046 downloads for Windows. 

Hanno suggested adding PyPi numbers to the *nix installer numbers.  That turns out to be 6,388 plus 1,010 OS X downloads for 4.0.

As always, the one set of statistics sure to raise the ire of other communities is vulnerabilities reported at the National Vulnerability Database. 

	3 mo	3 yr
Plone	0	9
WordPress	2	144
Drupal	10	282
Joomla	12	484

As usual, PHP-based systems fare much worse.  Over the past 3 years systems like WordPress, Drupal and Joomla have hundreds of vulnerabilities, many of them serious.  Even in the last few months all three of those CMS's have had vulnerabilities with high levels of CVSS severity.

It's no wonder that Federal agencies in the U.S like the CIA, FBI, DHS, and DOE turn to Plone.   Internationally, nearly 300 governmental Plone sites are listed at Plone.net.  Closer to home here in New Mexico, the City of Albuquerque, Albuquerque Public Schools, and Sandia National Laboratories all use Plone. 

Another qualitative way of looking at things is to look at the CMS subway map from Real Story Group (formerly CMS Reports).  This is the June 2010 version.  The main change is the addition of SharePoint Add-ons, which doesn't impact Plone's position at the intersection of web content management, portals & content integration, and collaboration & social software. 

The main change is the addition of SharePoint Add-ons, which doesn't impact Plone's position at the intersection of web content management, portals & content integration, and collaboration & social software. By their reckoning, Drupal should more properly be over by WordPress and Fatwire where web CMS and collaboration/social software run parallel.  Plone might be better served being compared with Open Text and IBM. 

Blog Action Day and Plone

October 15 is Blog Action Day and the topic this year is global water quality.  Together with US Fund for UNICEF, they're helping to build a movement of people across the world calling on UN Secretary-General Ban Ki-moon to accelerate the UN's work to supply clean, safe drinking water to the world's poorest populations.  Several initiatives are at work:

Raise funds for water: Change.org partnered with leading organizations to enable folks to raise money to provide clean drinking water to those in need. You can directly donate through Water.org, an organization co-founded by Matt Damon. Just $25 provides clean water for a lifetime for one person.

Spread the Word: Help to spread the word about Blog Action Day 2010 across the web. Take a moment to tell your friends to sign up via email, Facebook and Twitter. The more voices that are involved in the conversation, the greater the collective impact.

What, you might ask, does any of this have to do with Plone?  Turns out my oldest active Plone portal is http://waterportal.sandia.gov.  It's a collaborative workspace for teams all over the world to share water data, hydrogeological models, and other information about a resource that can be a contentious regional issue.  Waterportal is now over six years old, still chugging along on Plone 2.5.  We have workgroups from the Middle East, Central Asia, and the United States all sharing expertise and interest in clean, safe water.

Maybe this Oct. 15 I'll do a pro bono upgrade of waterportal to Plone 3. 

Change.org|Start Petition

Windows vs *nix vs OS X

In response to last week's post, Wichert Akkerman pointed out that he thought the Windows installers always showed more downloads than Unix-based ones.  He suggested that I dig around for the actual numbers.  Never one to turn down a challenge, I extracted the numbers from Launchpad and took a look around. 

Turns out, there are some anomalies.  The data for 2.5.x are probably incomplete due to the changeover to Launchpad several years ago.  Also, Plone 3.2.1 shows almost 700,000 downloads of the Windows installer.  Plone 3.2.2 shows almost 200,00 downloads of the unified installer for *nix.  If we take the data "as is," the graph looks like this with 3.2 dwarfing everything else.

If we remove the 3.2.1 and 3.2.2 data, assuming those to be some sort of outlier, the graph looks like this. 

Either way, Wichert is right--the Windows installer is being downloaded 3-4 times for every download of the unified installer.  Only 2.5 shows the unified installer outperforming the Windows version.  BTW, OS X is always under 10%, and for most versions, under 2%. 

I leave it as an exercise for the student to interpret all this.  Does this mean lots of people download the Windows version to test the product before installing on a Unix box?  Is Enfold Systems with it's toolset aimed squarely at Plone for Windows in the sweet spot for Plone consulting?  And what is going on with downloads for 3.2.x?  Your thoughts, please. 

Plone 4 Released

It's been a red-letter week for Plone with the release of version 4.0 on Wednesday.  Typically, Wednesdays are average days at Plone.org, but this week we saw 2.5 times the normal traffic as visitors flocked to the site.  Since then, site visitor loads have been 1.5 to 2 times daily averages. 

Downloads for Plone 4.0 are significant also.  Since this week's release, almost 6,000 downloads have been logged at launchpad.net.  That's about 10% of the lifetime download total for 3.3.5.

Interestingly, the installer for Windows (.exe) is being downloaded about twice as frequently as the unified installer (.tgz) for Unix-based systems.  That will probably change with time--right now people seem to be downloading to their Windows desktops and taking 4.0 for a test drive.

The press has been good for this release with CMS Report, CMS Wire, The H Open, and MacNN all running the news in English.  Several foreign language press releases have turned up, too.  Of course, Plone shops are spreading the word as well with Jon Stahl, Enfold (including a link to the 4.0 user's guide), and Six Feet Up getting their material out the door first. 

I should reiterate Jon's note over in the Evangelism forum encouraging everyone to blog, tweet, like, digg, and otherwise spread the Plone 4 goodness.  Also, thanks to Graham Perrin for pointing out the outstanding Plone 4.0 video. And in closing, a special tip o' the hat to Eric Steele and the 4.0 developers for making this release happen. 

Amazon Sales Ranks for Plone Books

It's been a good year so far for Plone textbooks.  New titles have been rolling out steadily.  Packt currently lists 10 plus a couple related titles dealing with CMS selection and Python. 

I should mention that Packt has just opened the floor for nominations for their 2010 Open Source Awards. Get on over to http://bit.ly/deN0Cn and put a plug in for #Plone. You might just win the doorprize of a new Kindle.

Meanwhile, back on topic, Amazon's list of Plone books is several pages long and as usual I've mined the sales rank data. 

As you can see, values continue to bounce around quite a bit.  The overall trends, as one would expect, are for sale rank to get higher (fewer sales) as time passes.  That's simply because people tend to buy books when they first come out, especially IT books, which have a shelf-life determined by the version of software involved. 

The graph is getting pretty busy with all the authors I'm tracking.  The last 3 titles (Gimenez & Romero, Gross, and de Alba) are all new since my last posting, so there is no trending information available.  Stay tuned for next September's data. 

Another set of statistics that I've been tracking are the BuiltWith numbers.  Apparently, they sample a couple million websites and attempt to identify underlying technologies.  Their 10 Aug. graph looks like this:

There's an encouraging upward trend in Plone usage, if you trust their undocumented methodology.  They state that 685 of the top million websites on the Internet use Plone.  They've additionally identified 5,634 other sites in a more extensive survey. 

Never Underestimate the Web

The unlikeliest charity is probably HeroRAT, a program that uses giant, bomb-sniffing rats in Africa to de-mine areas in Mozambique. These rats can also be trained to detect TB and are being used for that purpose in Tanzania.

They received an unexpected boost this weekend when Nicholas Kristof wrote a thoughtful NY Times column on honoring dads through supporting charity instead of commercialism. As a result their HeroRATs' program received 2,969 donations and $128,239. That far outstripped their previous 341 donations and exceeds the funding goal for the entire project. I'd be willing to bet the vast majority of those contributions came from the NYT online edition, not the dead-tree version.

Closer to home and more on-point for a Plone blog, Martin Aspeli had an unfortunate and catastrophic hardware failure. Based on a simple blog posting and making use of ChipIn for accepting donations, he was able to reach his target in just 4 hours and within 48 hours had hit 143%.  (BTW, the ChipIn account is active until July 18... hint, hint.)

It's these sorts of online efforts that are redefining how charitable giving is done and allowing a new level of personal involvement and increased efficiencies.  Just as the Sumatran tsunami* and Haitian earthquake revealed the power of the Web to mobilize people, we now see that its momentum continues and that the Plone community can rise to the occasion on any scale.

Kristof's piece also points out how a single bit of good online publicity can dramatically shift one's fortune. I urge everyone to participate in evangelizing Plone.  Take advantage of the Plone marketing site at Co-Activate and join the Evangelism section of the forum. 

_______________

* It was the fact that the newly redesigned OxFam site (done with Plone, of course) was able to process an unprecedented number of donations immediately after the disaster in the Indian Ocean basin that my team realized that Plone was stable and could  scale up to handle our largest enterprise requirements.  We haven't been disappointed.

Plone Rocks

I saw an Amplicate tweet a day or two ago about "Plone sucks."  Following up on that I found that there was  "Plone rocks" side of Amplicate and quickly stumbled into a world of automated Twitter stats.  Although interesting, it has some serious flaws in understanding human language, for example, the tweet "no way Drupal rocks" was counted as a positive "Drupal rocks" score just because of the literal phrase without any understanding of context.

Apparently Amplicate has only been around since October 2009, so there's not much data to go by, but still it was interesting to poke around. 

As you can see the rocks and sucks trends are roughly even.  This is actually good news because people tend to gripe more than they praise in general (squeaky wheel and all that). I'll run the test for significant differences between the two lines some evening when I'm not sipping 18-year old single malt.

More interesting than the numbers was the strange classification system.  Plone is considered a software framework.  Sharepoint is found under software applications.  Drupal is tucked away under Internet content management systems as is Joomla.


If you're feeling good about Plone right now, tweet about it with the trigger phrase "Plone rocks" and hop on over to Amplicate to add your vote to the tally of supporters.

CMS Security

I'm still trying to catch up on blog posts after March with jury duty and emptying out the parent's home of 40 years.  For World Plone Day 2010 I thought I'd bring up one of my favorite contentious subjects:  security.  Seems like every year some pundit runs through the security arguments for CMS.  Here's my chance to get ahead of the curve.

First, the methodology.  Candidate CMS's are the top ten listed in CMS Matrix when sorted by compares (BTW, Plone comes in sixth).  The source for my security data is the National CVE and CCE Vulnerability Database, searching each candidate CMS for vulnerabilities listed in the past 3 years and past 3 months.  The highest level of vulnerability in the past 3 months were counted and totaled as a "Seriousness" score.  Here they are in alphabetical order. 

CMS	3 yr. vuln.	3 mo. vuln.	Seriousness
DotNetNuke	13	0	0
Drupal	266	16	6 medium
Joomla	426	87	36 high
Mambo	85	0	0
PHP Nuke	6	0	0
Plone	8	0	0
TYPO3	224	44	26 high
WebGUI	9	0	0
WordPress	141	6	4 high
Xoops	65	3	1 high

Second, the results.  I'm pleasantly surprised that 5 out of 10 turned up with no vulnerabilities in the past 3 months. However, of those only PHP Nuke has a better 3-year number of vulnerabilities.  TYPO3 and Joomla have a much larger number of vulnerabilities rated "high" by the National Vulnerability Database than all the rest combined. 

Typically one hears defenses like "CMS x has more vulnerabilities reported because it is more popular/has a larger install base and therefore has more eyes looking for problems."  That's the same argument Microsoft put forward for all its security flaws when compared with Linux.  You don't want to go there. 

But even putting aside the logical fallacies in such arguments, both pro and con, it's clear that systems with more security flaws require more effort to patch and lock down.  The bottom line is that more flaws mean more chances for the sys admin to slip up and more opportunities for the bad guys. 

By way of an anecdote, at my day-job our computer security people were much relieved to learn that we were using a non-PHP-based CMS.  On that note, Happy World Plone Day 2010. 

Plone and the Wonder Wheel

If you google for a term, let's say, "CMS," you'll notice a little "Show options" link at the upper left that you probably use for narrowing down date ranges or filtering news items and images.  About 3/4ths of the way down the list of option is one called "Wonder wheel."  It displays a spoked wheel labeled with, for lack of a better term, interesting links.  The usual top 10 search results appear in a column to the right.

From the CMS wheel, clicking on one of the spoke's links, for example, "Content Management Systems," opens a second wheel, to wit:

What I find interesting is the difference between the top 10 search results and the spokes, looking at which CMS's turn up within each "spin" of the wonder wheel.  At the first level ("CMS"), most results are disambiguation links.  The only specific systems listed are CMS Made Simple and concrete5, even though "microsoft cms" is one of the spokes.

At the second level, the spokes show ".net content management system," "php nuke," and "microsoft content management system."  The top 10 results now return a few general links plus Alfresco, Pligg, LightCMS, ModX, Drupal, Joomla!, and Plone.  Oh yes, there are some sponsored links at this point.  From here on out, there are always three commercial systems turning up as sponsored links. 

Beyond this, the combinations and alternative paths become numerous.  Following "enterprise content management system" takes us to a wheel with mostly general spokes, but two specific systems garner a spoke each.  Six of the top 10 results are specific systems, none Plone.

Backing up and following "open source content management system," WordPress, CMS Made Simple, and Drupal each label a spoke.  Six of the top 10 results are specific systems, one of which is Plone. 

Backing up once again and following "web content management system" gets us a wheel with only general categories for spokes but 5 out of the top 10 results are specific systems, none Plone. However, selecting the awkward "web content management system open source" spoke takes you to the first wheel in this chain where Plone has a spoke of its own. 

All this raises some questions:

• What determines whether an item appears on a particular wonder wheel?  
• What determines an item's position on the wonder wheel?  
• Why doesn't Plone have a position somewhere in the daughter wheels of "open source content management?"  
• What needs to be done to get Plone to turn up in more wheels?  

I guess it's that last question that I want to toss out to the Plone community as we count down the final hours to World Plone Day 2010. I can't say that wonder wheels are becoming the new way to search--for one thing, they're not available on Google's mobile page for my Droid.  But I can't help thinking that under the hood a better wonder wheel position translates to better SEO behavior in general.  Your thoughts, please.

Plone at my Day Job

Way back in November of 2007, I published a graph of the growth of our Plone projects at my day job.  Michael Bernstein asked me to update that graph a couple months ago.  Here it is almost May and I'm finally getting around to it.

The blue points show the total number of sites constructed over time based on the date of creation.  The yellow points represent the net number of active sites (totals sites minus ones no longer in use).  You can see that about 60% of our sites are active at any one time but that we've had very steady growth in the past 5-6 years.

Even though things are flattening out a bit, we find our current projects are much larger and have more sophisticated requirements.  Also, we've been migrating old 2.5 sites up to 3.x this past 6 months and that's kept us busy.  On top of that, we're finally getting some internal customers to think more broadly about site reuse.

For example, the site for our UNSCR 1540 training grows with each new workshop in the program.  A couple years ago each workshop would have had a separate, short-lived website.  Now one large portal serves to connect students who participate in different workshops and has a life beyond a single conference.  New students have access to previous workshop material and, like any good Web 2.0 phenomenon, the 1540 portal gains value exponentially as the user base grows. 

Google Analytics (below) shows that the 1540 portal has garnered a widespread audience.  Unfortunately, things haven't picked up in Africa, which is disappointing given that the last workshop was in Nairobi back in February. Even so, we've got a sizable piece of the world covered and that's a good thing. 

Quarterly Amazon Sales Rank Stats

It's time for me to pull myself away from the Herculean task of emptying out my parents' house and get back to our regularly scheduled program.  It's the beginning of the quarter when I usually take a moment to see how Plone titles are faring over at Amazon.com.  Here's the latest sales rank statistics for Plone texts: 

From the zigs and zags you can see that Amazon sales ranks, as always, aren't very stable.  Julie Meloni's "Plone Content Management Essentials" took a huge leap.  (Remember low sales ranks is a good thing.)  And in fact, just about all Plone texts improved their sales ranks.  Four books have sales ranks below 200,000 and only three had their rank go upwards (Rose, Cooper, Lotze and Theune).  Frankly, I'd have to say this trend, if it continues, could indicate a healthy turn of affairs just in time for Plone 4. 

Mobile Plone

A colleague sent me an article on mobile computing this afternoon and among other things, it inspired me to take a look at what's shaking with Plone and mobile devices.  Google Analytics for the past five months has been tracking mobile device stats.  You can find it under Visitors | Mobile | Mobile Devices.

Using Plone.org as an exemplar, Analytics tells us that less than 1% of visitors are using a mobile device.  Conveniently, Analytics breaks this down by operating system. Looks like iPhone is the clear winner at the moment, even with the small sample size. 

iPhone	65%
Android	10%
iPod	10%
SymbianOS	8%
BlackBerry	2%
Other	5%

One can also group by browser and here Safari comes in with a resounding 80% followed by a weak 13% for Mozilla Compatible Agents.  NetFront has 1% while Opera, IE, BlackBerry and the rest trickle in at less than 0.8% each. 

The mobile usage statistics are quite different from our average Plone.org visitor.  Pages per mobile visit average 2.5 while the site average is 4.4.  Average mobile time on site is 2 minutes vs 4.5 for the overall average.  Mobile users are not going deeply into the site and leaving quickly.  Looks like some refactoring for mobile would be helpful to this small but growing population of site users. 

There is a barely perceptible upward trend in mobile visitors over the five months for which we have data.  The daily stats are very noisy with lots of variability from day to day.  I'll continue to track this over time and see where it leads. 

Meanwhile, be thinking about your own sites and how users with mobile devices are experiencing them.  Give some thought to Jakob Nielen's remarks on mobile usability.  Mobile devices are set to become the dominant means by which a huge chunk of the world's population are getting to the Web.  Don't leave them out. 

Weekend Visitors

Now that the dust and feathers from the Great Backyard Plone Count have settled, it's time to get back to basics.  Tonight I was looking in on visitor stats for Plone.net and Plone.org.  An interesting pattern turned up that I'd missed before.  Here are two graphs from Plone.org of roughly the last six months, the first of visitors:

and the second of % new visitors:

The most obvious feature is the Holiday Dip during the week of Christmas, although new visits didn't collapse over the vacation.  However, if one looks closely (very closely, trust me on this) the "teeth" of new visitors line up precisely with the weekly minima of all visitors.  Every "% new visitor" maxima is on a Saturday or Sunday.  Every "all visitor" maxima is on a Monday through Wednesday with a deep drop-off on the weekends.

New visitors are peaking on the weekends while visitors in general are hitting the site early in the work week. 

It's 4:00 in the morning and I'm still trying to work out the implications.  Clearly Plone is an enterprise CMS, not one that caters to hobbyists.  Early each week Plone users, managers, and developers hit Plone.org with news, events, questions, and answers as they do their work and solve their problems du jour. 

But every Saturday, a group of new visitors arrive on-site.  It's only about 10% higher than the usual flow of new visitors.  Are these professional web designers and developers frustrated with their company's CMS (or lack thereof) and looking for something better?  Are they DIY'ers looking to see if they can make use of Plone for some weekend project?  Are they volunteers, now that the work week is done, helping out their favorite philanthropic endeavor? 

It may be a case of "all of the above."  I'd appreciate hearing your theories on this and certainly like to get feedback from any newcomers who first hit Plone.org on a Saturday.

Great Backyard Plone Count Results

First off, I want to thank all of you who participated in this year's GBPC.  We garnered 219 sites, which is up 19% from 2009.  We had contributions from Canada, Germany, Italy, U.K., and U.S.  Considering that Delicious has 2016 sites tagged with "plone-site" and Plone.net lists 1955, we've collected a sample of over 10% of the easily known Plone sites. 

Industries that were listed included education (53), non-profit (24), government (14), healthcare(11), small business (8), assoications (6), arts & entertainment (5), and environmental (5) plus 45 others.  Thirty were private intranets, so in that regard the GBPC was useful in getting a sample of sites that are normally out of sight. 

Of interest to me and Jukka Ojaniemi was the usability column this year.  52% of the sites reported had a self-reported rating of 3 ("Awesome"), 27% had a rating of 2 ("Alright"), and only 3% sucked.  About 19% lacked a usability score since they were confirmed from last year without further updating.  A little spot checking of sites with different scores show a fair amount of unevenness in the ratings--some clean, even elegant sites scored low while similar ones did very well.  Overall however, I'd say that Plone lends itself to uncluttered, easy to navigate sites. 

I'll be digging around in the data in the weeks ahead but for now the raw data is at the Google Docs spreadsheet.  From Twitter today Nate Aune suggested using Celery and Scrape.py to crawl and scrape for a web-wide search for Plone sites.  Expect a more complete Plone census as I work on that. 

The Great Backyard Plone Count

As promised, the Second Annual Great Backyard Plone Count has begun.  The GBPCis a shameless copy of the Great Backyard Bird Count (which will soon be active for those in North America). Just as GBBC is an effort by the Audubon Society and the Cornell Lab of Ornithology to track birds, GBPC is an effort to catalog Plone sites worldwide. Its a voluntary, self-reporting project (with all the weaknesses that go along with it) that attempts to find as many Plone sites as possible, including and especially intranet sites behind firewalls.

The Great Backyard Plone Count is an annual four-day event that engages Plone watchers of all ages in counting Plone sites to create a real-time snapshot of where the sites are across the world. This year the GBPC is February 12-15, 2009, the same dates as the GBBC (with adjustments to give our worldwide Plonistas a full 4 days in any timezone).  Anyone can participate, from beginning Plonistas to experts.  It takes as little as a few minutes on one day, or you can count for as long as you like each day of the event. It’s free, fun, and easy—and it helps Plone.  We'll be adding updated 2010 GBPC materials on this blog and on the Plone marketing site as they become available.


The tools that are now active include Google Docs or Gmail mechanisms to: 

• Use the 2010 Data Form 

• Download the Plone Checklist (download an empty table--use File | Download as--fill in all your sites at once; CSV or OpenOffice Calc preferred)

• Submit Your Plone Checklist (upload a bulk list to the GBPC e-mail address) 

• Confirm last year's sites (send the msg "Confirm All" or list inactive sites, one URL per line--you may still add new site sightings by either method above)

• Explore the 2009 Results

Thanks in advance for your participation.

GBPC on Prezi

How to ID Plone Sites

The Great Backyard Plone Count kicks off at the cusp of midnight this Friday 12 February, just like its ornithological counterpart, the Great Backyard Bird Count.  The bird watchers have it easy, only dealing with North America and the fact that few birds can be seen by the dark of night.  But Plone sites "fly" 24x7 and Plone watchers are scattered worldwide.  (Darryl Dixon of New Zealand wins the prize for being the first from that timezone to submit a "siting" last year from the easternmost timezone that participated.)

The result is that as Thursday turns to Friday in say Kiritimati, the earliest timezone, it will be midnight Wed-Thurs in Honolulu, 3:00 AM Thursday in Albuquerque, and 10:00 AM Thursday in London.  (Tip o' the hat to The Fixed Time World Clock.)  So we'll open up the online submission form at 3:00 AM MST Thursday 11 February and close it down at 3:00 AM MST Tuesday 15 February.  That way every timezone gets the full window of opportunity, even if it adds 24 hours to the worldwide event.  Plone.org shows visitors from Fiji and New Zealand (Suva and Auckland are +13 UTC), as well as New Caledonia (Noumea is +11 UTC), so we want to give everyone an opportunity to participate fully.

Last week I posted a GBPC Prezi presentation but noted that it was stripped out of Planet Plone and other feeds.  If you care, take a look at http://prezi.com/jahnf2owdk9y/.  Also, while on the subject of Prezi, I created a reusable Prezi for Plone in general, embedded at the end of this post.  Maurizio Delmonte has already translated it into Italian and it looks like Naotaka Hotta is working on a Japanese version. 

Meanwhile, Nate Aune and I have been having a discussion about how one can identify Plone sites.  Of course, the obvious ways are to look for that nice "Plone Powered" colophon and the generator tag (<meta name = "Plone - http://plone.org" >).  Another effective way to ID Plone sites is to tag them in Delicious (thanks again, Sam).  Just make sure that your Plone sites are tagged with the phrase "plone-site" and they'll turn up when we search http://delicious.com/search?p=plone-site.  Right now there are 1932 sites tagged in this manner.  If you haven't tagged your site(s), do so now and next weekend's data mining will sweep them up. 

In untagged sites that are highly customized or have a static front end, we may have to rely on other features for identification.  Lukasz developed Plone data-mining scripts (http://www.llakomy.com/articles/plone-websites-in-uk) for the UK. Builtwith is using a proprietary algorithm and some sort of sampling to track trends in web technologies (they also seem to be missing some obvious high-profile Plone sites, so I'm not too confident in their statistics).  Last year when I discussed this, there was a comment from Andreas Jung via Matt Hamilton that is germane.  They suggest that "/manage_copyright" is a useful method exposed by Zope 2 sites.  I'd love to see what other techniques people can come up with.

Especially for firewalled Plone sites deep behind proxy servers, the only way we'll ever be able to count your contribution is for you as a site developer, user, or owner to self-report.  Whatever way you identify Plone sites, take a moment next weekend to tag them in Delicious (if they're public) and submit new or confirm your previous year's sitings (active links to the tools will be Tweeted and published in this blog starting the 12th).  Get out there and count some birds, too.

Plone on Prezi

. 

The Great Backyard Plone Count 2010

I'm gearing up for this year's GBPC, The Great Backyard Plone Count in only 14 more days.  This year's event corresponds with The Great Backyard Bird Count, also taking place on the same long weekend, 12-15 February, 2010.



Last year, our first ever, we took in 184 sightings of Plone sites. Thanks to every one who participated especially to Matt Hamilton (30 sites) and D. Thomas (27 sites), who were our top contributors.  If you contributed last year, we'll set up a simple mechanism for you to simply confirm that they are still active sites without having to re-enter any data. 

One important change this year is the addition of a usability question, inspired by Jukka Ojaniemi's Plone research.  It's just a quick 1-3 scale (awful to alright to awesome), but it will help define the span of Plone quality.

Some of the tools we'll set up will include Google Docs mechanisms to: 

• Use the 2010 Data Form  (active only 12-15 Feb.)

• Download the Plone Checklist (download an empty table, fill in all your sites at once)

• Submit Your Plone Checklist (upload a bulk list to an e-mail address--active only 12-15 Feb.) 

• Confirm last year's sites (via e-mail notification--active only 12-15 Feb--you may still add new site sightings)

• Explore the 2009 Results

Considering that Plone.net lists 1950 sites, being able to track about 10% of that number, especially intranets that are otherwise invisible behind firewalls is very helpful to those of us who watch the numbers.

Plone and Disaster Relief

Watching the Hope for Haiti Now telethon, I'm reminded that one reason my shop uses Plone harkens back to the Sumatran earthquake and tsunami back in Dec. 2004.  We were babes in arms, futzing around with various home-built Python and Zope concoctions for interactive websites.  Plone 2.0.1 had come up on my radar as a promising alternative to home cooking our portals. 

Meanwhile, Oxfam America had just retooled their website using Plone with the help of Enfold Systems.  December 26 the disaster in Sumatra struck.  In a single month the new Oxfam America site processed something on the order of $14 million in charitable donations, an amazing amount.  Obviously Plone could handle the scalability problem.  (See how they did it.)  Oxfam America continues its next-generation site with the help of Jazcarta.

With that vote of confidence, we forged ahead, building numerous small to medium portals, upgrading to 2.5, and basically getting steadily smarter about the whole Python-Zope-Plone stack.  Just yesterday we retired our first portal, Sandia's Training Course on Cooperative Monitoring, due to budget reductions.  However, our second, a water monitoring and hydrogeological modeling site, continues to this day. 

Now, with the Haitian earthquake crisis, I'm reminded that catastrophes continue to plague the world and Plone continues to play a role.  Go to Oxfam America and give whatever you are able to.  If you can, give a monthly gift to help sustain their continuing efforts worldwide.