"Count what is countable. Measure what is measureable. What is not measureable, make measureable." -- Galileo

Wednesday, October 31, 2007

Plone and Drupal at m.odul.us

I came across a long posting comparing Drupal and Plone at m.odul.us dated 2 Sept. 2007. The author does a good job of examining key CMS components and features. He's clearly spent some time working with both systems, so he can speak with some authority. One sign that Matt is not blowing smoke is that he is knowledgeable about ArchGenXML, something many superficial reviewers miss. For me its one of the key tools that I rely on every week. Another good sign about this review is that community is rated highly. I find this aspect of open-source an often overlooked part of the picture.

The m.odul.us review has garnered a handful of comments, but it probably deserves more. For those who are looking at any CMS, not just Plone and Drupal, a read-through will give you a good idea of what a sensible evaluation must touch upon. Just to give you an idea of the breadth of Matt's piece, here are his headings:
  • Easy Publication to the Web
  • Flexible Information Architecture
  • Extensibility
  • PHP vs Python
  • LAMP vs Zope
  • Scale, Speed, and Deploying
  • Hosting Requirements
  • Security
  • Community
  • Contributed Add-ons
  • Documentation and Community Support
In the end Matt's conclusions are sound and I share his opinion that you must tie your decision to your requirements.

Drupal is a community content management system, and it largely farms out other tasks (e.g., asset management, CRM, etc.) to other web apps, choosing instead to integrate with them. This is a reasonable stance; Drupal is a CMS, and it integrates into a larger world of Apache served web applications.

Plone, however, looks at a most web application problems as specific instances of content management. Thus, with Plone, is it not absurd to develop extensions to handle your asset management system or your email newsletters (complete with server). Plone is built on a powerful application server and it makes sense to leverage it and your data once you're used to programming for Plone.


Tuesday, October 30, 2007

Plone and mojoPortal at Packt

Packt Publishing announced their "Best Other Open Source CMS" award today and the judges gave it to mojoPortal, a system with which I have no experience. Plone came in second with Silva third. Alas, there is no cash prize for runners up in this category.

I would like to know more about their methodology because they state:
"mojoPortal’s ease of use, set of relevant tools and plugins and also the fact that it is cross platform, made it stand out above the rest."
Certainly Plone is platform independent, so that by itself shouldn't have been the deciding factor. That leaves ease of use and tools/plugins. Hard to tell exactly what they're looking for based on that cryptic statement.

I will agree that usability has become a big issue for me based on my research last spring in which a good chunk of the non-IT staff in my center were found to be "connected but hassled." See the Pew Internet & American Life Project Internet typology questionnaire for more info.

Monday, October 29, 2007

Plone Metrics Goes on Walkabout

Things will get a little patchy over the next two weeks--I'm off to Amman, Jordan to host a workshop for Middle Eastern NGOs, teaching them the ins and outs of Plone for non-profits. We'll see if Plone features for building community translate into a robust network of regional partners working on civil society initiatives. Please see http://wacsi.unm.edu

We'll have representatives from Egypt, Saudia Arabia, Syria, and, of course, Jordan. And as always, the infrastructure constraints will be a big part of things. For the moment I imagine UNM's Plone server will be able to handle the influx of new users and new instances. Eventually, if we're truly successful, we'll need to find a dedicated server for pro bono Plone services. Also, we'll have to upgrade to 3.0 quickly.

Stand tuna'd...

Meanwhile, watch http://www.packtpub.com/award to see who comes out on top for Tuesday's "Best Other Open Source CMS" category.

Friday, October 26, 2007

UC Davis CMS Initiative and Plone

I see that UC Davis has been evalutating CMS with an aim towards developing a campus-wide framework. Please see http://pubcomm-29.ucdavis.edu/cmssurvey/ for the results of their survey of CMS at institutes of higher learning.

Its an interesting survey and I was surprised to see that do-it-yourself CMS predominated with 18 responses. Plone was next with 13, Drupal at 10, and Joomla tied with Zope at 8 each. Lots of material at the UC Davis project pages, so feel free to poke around and get a feel for their methodologies.

Wednesday, October 24, 2007

Professional Plone Development

Martin Aspeli's new book just arrived from Amazon today. Even though I taught at the College of Santa Fe tonight until 9:30, I ripped into it and read the first 22 pages.

Nicely written. Clean style. I enjoyed Limi's Foreword. I learned a thing or two about Zope 2 and Zope 3 plus their relationship to Plone. All that in 22 pages. Gonna be a worthwhile read.

I believe it was Joel Burton who observed that the number of books published about a software system is a good indicator of the health and widespread adoption of that system. With Martin's latest entry into the CMS publishing lists, Plone continues to hold its own. Its clear that the number of books published and the number of books sold is an interesting integration of the number and level of activity of both the development community and the user community. Its difficult to reconcile this with other more common means of mensuration.

What does it mean that Plone has far more non-English translations of books out there? It speaks to Limi and Aspeli's observations that Plone the CMS is an artifact of Plone the community.

Now, off to measure that community....

Monday, October 22, 2007

Thoughts from Today's New York Times

Today's New York Times has a piece about methodologies (registration required) for counting site hits as it pertains to the online advertising business. At issue is the discrepancy between server log numbers and sampling panels. In the end,
“The irony is we’ve always called for more measurement,” said Stephen Kim, director for global trade marketing at Microsoft Digital Advertising Solutions. “Now we’re getting it, but many people are somewhat frozen in how to deal with having more measurement.”
_______________

Also on the NYT website today was the article on online libraries. Turns out many libraries are turning down offers by Microsoft and Google to digitize their collections and instead going with the Open Content Alliance, a nonprofit effort aimed at making their materials broadly available.
_______________

Finally, as an aside to the difficulties of measuring social and economic phenomena, NYT discussed the recent report from the Head of Reconstruction Teams in Iraq. One conclusion was that "the program still has not developed concrete methods to measure the effects of the teams on progress in the country."
_______________

What's all this to do with Plone metrics? First, that even with exact server data, we don't know what people are actually doing and sampled panels don't necessarily do a better job. What does it mean if someone at Los Alamos National Laboratory downloads a Plone installer? What does it mean when I have 15 students at my College of Santa Fe class download the installer? Apples and oranges again.

Secondly, the digital libraries observations point out that guaranteed open source is a significant component to IT decisions these days. The Documentum/Alfresco case study will be a fascinating Master's Thesis someday for someone studying the early history of the Internet and the relationship between COS and open source.

Finally, the report from Iraq speaks to the difficulty of measuring effectiveness. This takes me back to my early posts about Bullock's dissertation on integrated measures of effectiveness. I'll continue to pursue that as I slowly fill in the matrix with data and normalize the data.

Sunday, October 21, 2007

Vote for Plone

Vote for Plone in the Best Other Open Source CMS category of the Packt Publishing CMS Awards. Although its a non-scientific poll, Packt's results do get some press every year. Its probably more important to vote in the nomination rounds earlier in the year to make sure that Plone makes it into all appropriate categories. Voting runs until Friday October 26, so hop on it.

Thursday, October 18, 2007

Validation, Reassurance, Recognition

From Sharon Melnick's perceptive piece in today's HuffPo, here's an excerpt that has relevance to the Plone software community, not just individuals. I have taken editorial license with it in order to emphasize its appropriateness to the Plone community of users and developers. My comments in brackets [...].

Many [software systems] are seeking approval on a daily basis, though they may not be consciously aware of it. From in-depth conversations ... as a business coach and trainer, many report typically engaging in the following behaviors to seek validation, reassurance, or recognition:

• working obsessively in order to get a "pat on the back" from a ... client
• saying 'yes' to everyones' requests but not finishing what you need to do
• exhausting yourself being perfect to make sure others think well of you
• asking others' opinions even though you know in your gut what to do
• worrying about 'politics' and what others think about you
• stealing credit from others

Equally frequent are behaviors in which people avoid or procrastinate in order to prevent other people from being able to criticize or reject them:

• have good ideas but don't assert them ...
• procrastinate so ... work can't be commented on
• stay mired in comfort zone of details instead of thinking strategically
• avoid direct feedback

All of these behaviors put energy and attention into managing other people's perceptions of [one's system]. This is how people act when they have doubts about their value.

I'm pleased to report that I see little of this in the Plone community--contributors may work compulsively, but its not for that bigger paycheck, certainly not in the open-source world. Intrinsic value of the job keeps people at it because its the right thing to do. While its true, in the long haul recognition of good work as part of an open-source project does translate into opportunities, whether a new job, better consultancies, or other perks. But this doesn't mean that Plone is running solely on ego power and that's a good thing.

Tuesday, October 16, 2007

Plone and Human Rights

I thought I'd give a plug tonight to Tom Moroz from the Open Society Institute (OSI--not to be confused with the Open Source Initiative). He presented at Plone 2007 and was summarized in Maurits van Rees blog. To reiterate the key elements in common of both open source and open society are:
  • Knowledge is provisional and fallible. You have to check it. You have to be able to check it. An open source release is not going to be perfect in the first release. That release brings more people to the software and they check how good the software/knowledge is.
  • Responsiveness: fix reported bugs quickly. Get a deep understanding of the problem and then fix it, which is the same in open society.
  • Transparency: the code is open or the information is open. We publish numbers of how much is being spent where to rebuild Iraq.
  • Pluralistic and multi-cultural. We respect minorities. (Plone cares about internationalisation, ed.)
  • High degree of responsibility. The individual should interpret her value. No-one is forcing anyone to do anything. You feel responsible anyway. You are involved and want to be a good member of the community.
  • Freedom and human rights are at the foundation.
  • Social mobility and a matter of openness. It does not matter where you come from.

Monday, October 15, 2007

Plone at Penn State

Here's another bit of well written analysis of CMS from WebLion at Penn State. They put Plone out in front of Drupal, but more importantly, they qualify their key features and why it meets their requirements. For those who just want the bottom line, WebLion concludes:

Plone is ideally suited for these use cases:

  • Large-volume public-facing internet sites
  • Intranet sites and sites containing sensitive data
  • Sites hosting specialized web applications
  • Research sites containing protected research data
  • Hosting multiple sites without sharing databases
This gets back to the message I've been harping on--understand your users, your requirements, your constraints, and own needed features. With hundreds of CMS out there, one of them is right for you. Look at similar organizations that may share your use case. Contact them. Most will be happy to share their decision-making process with you. If their environment and use case is close to yours, by all means follow their lead.

Sunday, October 14, 2007

Plone and DISCOVER Magazine

CMSWire has a recent piece about Gartner's evaluation of commercial CMS. They also have the first installment of an article describing how DISCOVER Magazine came to retire its old web environment and set up a Plone-based web publishing system. The article is profusely illustrated with actual content examples from DISCOVER. Their key points (and I quote):
  • DISCOVER chose Plone for its ease-of-use, robust feature set, strong open source community, reduced vendor lock-in, outstanding customizability, standards compliance, reputation, economics, and high performance.
  • Before Plone, it took DISCOVER up to three weeks to publish a magazine issue online. With Plone, that has been cut to as few as three days.
  • Plone’s WYSIWYG editor greatly facilitates the management of the site for non-technical users and enforces design consistency.
  • Plone automatically generates related articles for each of the thousands of articles on the site. No manual intervention is required.
  • Video and multi-media content is easily managed
  • Thousands of articles were successfully and automatically migrated from DISCOVER’s old CMS into Plone with no loss of inbound links.
It reminds me of the day when we were just starting out testing Plone and I learned of Oxfam's Plone site. It went online in November 2004, the great Sumatran Earthquake struck in late December, and within a month Oxfam had processed over $14 million in donations. Its that kind of real-world story that makes you realize that this dog will hunt.

For those who need more than the Oxfam story and the DISCOVER example, take a stroll through http://plone.net/sites. There you'll find ACM SIGGraph and a host of other powerful organizations and companies who are using Plone.

Saturday, October 13, 2007

InfoWorld Rating of Open Source CMS

Thanks to Carlos de la Guardia for spotting the very recent InfoWorld piece pitting Alfresco, Joomla, DotNetNuke, Drupal, and Plone. They were very upfront about their methodology, which although explicit lacks detail below the top level of score aggregation.

InfoWorld reviewers scored each contender in eleven areas:
  1. Availability
  2. Ease of use
  3. Interoperability
  4. Management
  5. Performance
  6. Security
  7. Reliability
  8. Scalability
  9. Setup
  10. Support
  11. Value
But alas, they only report scores in ease of use, features, scalability, security, management, and value. Of course, since these are all open source, availability ought to be 100% for all. Performance, reliability, setup, and support need to be addressed. As I've stated before, mere feature counts misses the fact that you should be matching features to requirements. In this case, I'd guess that Alfresco, built by fugitives from Documentum, is in fact a document management system, not strictly a CMS meant for online community development.

Friday, October 12, 2007

Conferences as Metrics

I spent a good bit of last night watching streaming video from Naples. Looks like it was an awesome venue. From the Conference website I note that 360 people attended. Then it occurred to me that number of conferences (local, regional, and global), number of people attending, quality of conference website, and so forth is an interesting integration of the strength of the core development team and dedication of the high visibility installations.

The Plone 2007 Conference was streamed, already has a number of Technorati posts, has some video bits already posted (congrats to the new Foundation Board), a strong Second Life virtual conference presence, 63 presentations (all of whose abstracts are online), and filled four rooms for three days (not counting pre-conferance training and post-conferance sprints). There is active Twittering from the conference and links to online transcripts (many have been uploaded as blogs by the presenters). Permanent online video archives will appear soon, if last year's experience from Seattle holds true. (Speaking of which, if you watch only one Plone Conference video, it should be last year's talk by Eben Moglen. Inspiring!)

Out of curiosity, I went looking for the last DrupalCon, which was last month in Barcelona (good choice, guys). I came away less than overwealmed: birds of a feather sessions canceled, no microphones at some sessions, negative vibes about posting presentations to SlideShare, but their video archives are up and running. I believe they had 84 formal presentations planned and those covered a wide range of ground for their community.

Next, being a glutton for punishment, I went looking for the last SharePoint Conference. Eventually I found it at http://technet.microsoft.com/en-au/bb498198.aspx , but it was a non-trivial exercise. Of course, comparing commercial with open source is a huge apples-and-oranges problems. How can http://sharepoint.microsoft.com/sharepoint/default.aspx be compared with Drupal.org or Plone.org?

This all speaks to the fact that conference attendence, frequency, number of sessions, quality (hard to quantify), and so on are proxies for a software system's acceptance and momentum. As of yet I don't know what it means that Drupal had AV problems but 21 more presentations, so stand by as I look at some ways to normalize these measures. Its even a more difficult problem when factoring in commercial off-the-shelf software.

Thursday, October 11, 2007

PHP and Python Beauty Contest

This morning I thought I'd like to make some observations about Allan Benamer's comment, to wit: "I think that PHP is adopted over Python because PHP IS better for whatever reason. I think there is a wisdom to the masses."

While there is evidence that crowds converge on useful optima, many's the time I've watched groups settle on a sub-optimal solution for basically historical reasons. Re: Software examples, my favorite is the prevalence of MS Word, which has high usage largely due to its tight coupling with Windows. Ask a professional word processor who is fluent in a number of systems and you'll hear them rave about WordPerfect. Heck, back in the 80s Sandia National Laboratories homed in on Mass-11 (Mass who?) as their idea of the best word processor when WordPerfect 4.0 was available.

But perhaps the best all-time example of mass convergence on a seriously sub-optimal solution for accidental reasons is the keyboard I'm typing on--QWERTY. Dvorak keyboards are known to be faster and less error prone, but QWERTY takes the day, not because in any sense it is better, but because it was an earlier standard probably built to deal with jammed letters. Had any jammed keys on your computer lately?

Back to Benamer, PHP is not necessarily better, but it is manifestly more popular. My first thoughts on the matter (at 4:18 AM, so I may recant this by the clear light of day) is that PHP is tightly linked to Perl-like programming mindsets, while Python gets people to move into different paradigms, which may be uncomfortable.

Another thought is that like Word vs WordPerfect, Drupal vs Plone comparisons are missing the fact that Word has a different underlying framework--hide the codes--while WordPerfect is a fully transparent show the codes system. I think something like that is going on with CMS these days.

One place where I see PHP making hay is in the free hosting realm. You have to really look to find a free Zope/Plone host, while there are many servers out there that will let you have a basic LAMP site for nothing. That's slowly changing and as costs for servers, RAM, and drive space all come down (meaning capacities go up), Plone's heavier footprint becomes less of a matter.

That said, I should mention that a couple of my CSF students took off and set up Zope/Plone on their home machines, configured their routers to deal with Comcast correctly, and are now hosting Plone-driven personal sites off their home systems. Enfold's Windows installer makes the installation trivial.

Tuesday, October 9, 2007

Plone in the Real World

I continue to work with Plone during the day and think about metrics at night. Alas, tonight I've got to get an early posting in--video-conference with Amman tomorrow morning at 5:00 AM. Yep, with 9 timezones difference, any f2f communications require getting our shiny white fundaments out of bed by 5 in the morning or staying up until at least midnight. Especially during Ramadan, work schedules in the Middle East are shortened--no breakfast, no lunch, no tea breaks. Basically, you're done with your 8 hrs by 3:00 and plenty tired for low blood sugar.

Today I took Hovie Hawk's advice (hovie.com) and jinned up an alternative WACSI page with graphics in a grid instead of pages of text. I also found the problem with 2.1's RSS script. Easy fix. Now the Google Reader is happy as a clam with our site's feeds. The slides for the two days of NGO training in Jordan are coming together. Just need a few more screen captures of user/group administration, customization, and Plone use of CSS to finish off the set. 50 slides = 6 hours with plenty of hands-on exercises.

The Quest for the Holy Grail

I continue to work with Plone during the day and think about metrics at night. Alas, tonight I've got to get an early posting in--video-conference with Amman tomorrow morning at 5:00 AM. Yep, with 9 timezones difference, any f2f communications require getting our shiny white fundaments out of bed by 5 in the morning or staying up until at least midnight. Especially during Ramadan, work schedules in the Middle East are shortened--no breakfast, no lunch, no tea breaks. Basically, you're done with your 8 hrs by 3:00 and plenty tired for low blood sugar.

Today I took Hovie Hawk's advice (hovie.com) and jinned up an alternative WACSI page with graphics in a grid instead of pages of text. I also found the problem with 2.1's RSS script. Easy fix. Now the Google Reader is happy as a clam with our site's feeds. The slides for the two days of NGO training in Jordan are coming together. Just need a few more screen captures of user/group administration, customization, and Plone use of CSS to finish off the set. 50 slides = 6 hours with plenty of hands-on exercises.

Monday, October 8, 2007

Plone Advocacy

By now you may have noticed that this blog has a teeny, tiny bit of Plone advocacy in it. However, just as in my presentation to InterLab last week, the bottom line is to make sure you match requirements with systems. That includes feature comparisons to see that tools can deliver needed objectives, assessments to see that staff and consultants can use software systems, and a cold, hard objective review of your ICT environment.

There is no magic online feature matrix, no silver-bullet white paper, no error-free consultancy, no unambiguous, simple answer. A few blogs have attempted to approach the problem (for example, see Benamer and Burton's discussion), but there appears to be no general consensus.

For now I'll heft a toast to the lucky ones in Naples at the World Plone Conference and sit tight until the videos come online in a few weeks. And I think my strategy of measuring progress towards a successful Plone end-state (a widely adopted, stable yet evolving system) is more worthwhile than many other approaches, especially in that it encourages transparency and points towards similar efforts for other software systems.

Sunday, October 7, 2007

Download History

Here's the latest compilation of download stats from SourceForge since the dawn of time. The all-time high in Oct. 2006 was just after the 2.1.4 rollout in Sept. 2006.

Other peaks coincide with releases: Mar. 2005 is four months after 2.0.5 and Apr. 2004 was probably due to 2.0.

The most recent peaks align with the version 2.5.2 activity early last year. As with vulnerability data, 3.0 hasn't been out long enough to make a significant appearance. Look for a spike after Naples and towards the end of the year as 3.0 usage picks up.

Saturday, October 6, 2007

Some Trends

I've started filling in the matrix of effectiveness discussed two weeks ago. One of the most telling items is the MITRE CVE list of vulnerabilities. They list 3 vulnerabilities that are relevant to Plone version 2.5, all of them of medium severity. (3.0.1 is too recent to have any meaningful data--they list no vulnerabilities.) By comparison Drupal has 73(!), SharePoint has 12 (5 of them of high severity), and Mambo has 122(!).

Additionally, I plotted the dates of the significant releases to come up with this graphic. It shows a remarkably steady trend of evenly spaced releases. The ticks on the x-axis are approx. 6 months apart. Tip o' the hat to all the developers who make this happen.



Of course, the big upward jumps are major releases and I've been so presumptuous as to put in a trendline. If 3.5 comes out next July, it will be well ahead of the regression model.

Friday, October 5, 2007

Content Management at the Dept of Energy

I recently attended the DOE Internet Workshop at Los Alamos National Laboratory (InterLab). To my surprise, there still isn't a consensus on an enterprise-wide CMS. Islands of Drupal, WordPress, and others are out there. Some labs use home-built CMS, others Oracle, and still others SharePoint, which probably has the widest installation base. Problem is that SharePoint is tricky to use as an extranet and really doesn't work and play well with non-IE browsers. And don't even get me talking about custom workflows...

My "Plone at Sandia" presentation was well received and may have made a few take another look. Probably also got a number of folks to take a second look at Python. After all, what does Google (a Python shop) know that we don't?

Nov. 1 I'm off to Amman for a WACSI workshop (wacsi.unm.edu) where I'll be training representatives from Middle East NGOs on Plone. We're expecting Jordanian, Syrian, Saudi, and Egyptian participants. Should be an excellent event.