CMS Matrix Revisited

Every now and then I wander over to CMSmatrix.org and check things out.  Systems come, systems go, and for a handful, there's a steady maturation and growth.  Tonight I took the top ten CMSs as rated by "Comparisions."  Must mean something if people are usually looking at the same set of systems.

I then counted the CMS Matrix features that they lacked, as defined by a feature having a "No" or "Limited" or "Costs Extra" in a particular cell.  One minus that value divided by the total number of features is given in the Pct Features column below.  The "System Requirements" section was not tallied due to it's largely text content.

CMS	Pct Features
TYPO3 4.5.2	97%
Plone 4.0	96%
WebGUI 7.9	89%
eZ Publish 4.2	83%
Joomla 1.6.0	82%
Drupal 6.10	77%
Xoops 2.0.18	71%
Mambo 4.6.1	71%
WordPress 3.0.4	63%
PHP Nuke 6	15%

If you prefer the graphic, here's the same data set.

It's interesting to see that among top-compared CMSs in CMS Matrix, there's a surprising amount of variability.  TYPO3 and Plone top the chart above 90% and the rest trail away down to 63%.  Then PHP Nuke languishes at the bottom.

Perhaps tomorrow I'll dig out my copy of 'R' and see if these cluster in any meaningful way.  I have the information summarized by each CMS Matrix category (Security, Support, Ease of Use, Performance, Management, Interoperability, Flexibility, Built-in Applications, and Commerce), so I'll be able to look at each section by itself in the days ahead.  But one word of warning before you base a decision on these data, check the specific features you require against any candidate system.  No need to throw out a possibility, if the reason is the lack of a feature you'll never use. 

Plone Books and Plone Impact

Amazon 

A bit past due, I'm back to Plone Metrics with an update on my almost quarterly Amazon stats. 

The data set continues to grow unpredictably with 13 out of 15 texts that I track reporting sales rank numbers.  The spread between minimum (good) and maximum (not so much) is the largest ever, due to slow sales of Gross's specialized multimedia tome.  Some values are nearly constant, while others show the inevitable decline (increase in sales rank) as Plone 2 textbooks become obsolete.

For the first time ever, there are no books below (better than) 500,000.  Part of that is slow roll out of Plone 4 texts.  Aspeli's latest is included, due out in June.  Below is a chart showing publication numbers for each year since 2004. 

Impact

I had the good fortune today to attend a workshop in Santa Fe on Measuring & Mapping Science.  Many of the concepts have applicability to measures of CMS effectiveness and I'll certainly bringing these new ideas to this blog in the coming months.  While many of the presentations today concentrated on tracking the network of citations in scientific publications, others brought up novel ways of looking at usage statistics.

One important distinction was that between prestige and popularity.  Simply counting clicks, views, downloads, and so forth is a surrogate measure for something's popularity.  But the impact of something must be measured differently to get a handle on its importance or prestige. 

As a first whack at a quick-and-easy impact metric, I turned to Google Scholar to see what sort of journal citations the various CMS's are picking up.  I used the top 10 CMS's from CMS Matrix with the largest numbers of compares. 

The result has almost zero correlation, meaning that importance in terms of who's writing articles about Plone doesn't have any relationship to the number of comparisons CMS Matrix reports. Here Plone (red) has 1300 citations and almost 150,000 compares. 

Google Scholar is just a first cut at a measure of impact--citation counts are only the first step.  The context of those citations now has to be understood to learn what is a positive vs neutral or negative cite.  Because Plone is a very mature CMS, in coming weeks I'll be looking for better metrics of how it has had a significant impact over the last decade in terms of driving the CMS environment by being a leader in innovation, emergent features, and usability. 

Japanese Earthquake and Tsunami

As with the unrest in North Africa, the effects of yesterday's disaster in Japan are felt even on Plone.org.  Here's the distribution of visits to Plone.org from last year, 11 March 2010.

And here's the map of site visitors from 11 March 2011...

In the northeast of the country site visits disappeared (understandably) with only one on the west coast.  Site visits picked up significantly in the south and west. Perhaps people are using Plone to develop sites that can quickly aid in the humanitarian relief effort?  Google's People Finder is now available in Japanese and Ushahidi has an active site.  If you have an idea for making use of Plone at this time, don't be shy. 

Meanwhile, our hearts go out to the victims of this disaster. 

The Great Backyard Plone Count -- 2011

So the Great Backyard Plone Count for 2011 is officially off and running.  Things didn't exactly start out as smoothly as I'd hoped -- got a call from the nurse at my parents' assisted living facility around 7:00 this evening.  Turns out Mom had a 3-alarm gusher of a nosebleed that couldn't be controlled.  While she got an ambulance ride to Presbyterian, I drove over to the hospital.  To make a long story short, everyone was incredibly helpful and after some cauterization and a very strange inflatable gauze packing up the schnoz, we actually got Mom tucked back into her own bed by midnight.

Of course, that means I didn't get out copies of last year's "sightings" to their owners in time to verify that the sites are still active.  It also means that I'm only now turning on the data entry form.

Even with that hiccup right at the start, if you'd like to participate, especially in getting credit for hidden or obscure Plone sites, this is your chance.  Use the form below to enter information about Plone sites you are aware of, especially intranet sites that are behind firewalls. In those cases a URL is not necessary, but a site name and description would be particularly helpful. Names and personal information (e-mail, etc.) are only for internal statistical and QA use and will not be used beyond this study.  The live form can also be found at:

https://spreadsheets.google.com/viewform?key=0AjB6ixF2SsaScHktWlJpYnBXTVpWRnJzMWhoaEU4WWc&hl=en#gid=0

The form will be open for data entry until midnight Monday the 21st.  Thanks in advance. If you'd like a look at last year's data, check out the Google Docs summary sheet.



As an aside, get your binoculars out and spend 15 minutes (or more) counting the birds in your own backyard.  Then submit your observations to become part of an enormous citizen-scientist ornithological study.

The Great Backyard Plone Count

It's that time of the year -- time for both the Great Backyard Bird Count (Plone-driven, btw) and the Great Backyard Plone Count.  While the bird count is a collaborative bird-watching effort by the Audubon Society, the Cornell Lab of Ornithology, and Bird Studies Canada, the Plone count is a voluntary effort to collect data on the world-wide distribution of Plone-powered sites.  In homage to the GBBC, the Plone count is held the same weekend, which this year is Friday 18 through Monday 21 February. 

Because many Plone sites are intranets behind firewalls, they can't be located by crawling the web.  This is a chance for developers, site owners, and users to stand up and be counted.  It's a non-scientific, totally voluntary effort for self-reporting Plone-based web portals. 

I'll be distributing portions of last year's list to respective contributors for confirmation of existing sites on Thursday.  Then on Friday, the input form to the Google Docs spreadsheet will be opened up.  Anyone can then submit sightings of Plone in the wild. 

Although there's significant bias in a survey like this, the most value comes from tracking trends over time.  This is the third annual Backyard Plone Count and as the number of yearly data points increase, we'll better be able to extrapolate from the observations. 

More to come this Friday.  As they say at the sushi bar, stay tuna'd...

The Security Announcement and Mobile

An interesting spike turned up in the Plone.org visitor stats this past week.  Although the regular visitors showed just a tiny blip, mobile usage more than doubled. 

Seems that the recent security announcement got people's attention and they turned to Plone.org with their smart phones.  Based on this, it would seem prudent to design mobile style sheets to pay particular attention to time-sensitive announcements.  That's probably good advice for any website.

One other observation from the above figure:  mobile visitors to Plone.org don't show the usual weekly pattern of general visitors.  Minima are as likely to occur on a Tuesday as a Sunday for mobile.  Meanwhile, general visitors show a strong dropoff on the weekends.  This trend for mobile might just be random fluctuations due to the smaller number of mobile visitors, something that gets swamped by the huge number of non-mobile site users.

If you use Plone.org via a mobile device, I'd like to hear from you -- how and when do you turn to the website with your phone?  Please comment.

Plone in Egypt Addendum

The Internet finally came back up in Egypt today and, as part of the initial flush of activity, there were visitors to Plone.org.  Welcome back, Egypt.