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

Monday, March 30, 2009

The CMS Websites

Tonight I was trolling around in CMS Matrix to see what interesting tidbits I could pick up. What I came across that piqued my interest was their list of links. Omitting the generalist Wikipedia, there are six featured references.

I took the time to visit each site and search for "Plone" to get an idea of the currency and thoroughness of their material. What I found was surprisingly uneven.
Resource Plone Citations Most Recent Plone Item
CMS Critic 5 03/16/09
CMS Watch™ 88 10/26/07
CMS Info 38 12/22/08
CMS Review 12 03/17/09
CMS Report 82 03/25/09
CMS Wire 120 03/04/09

CMS Watch was very out of date despite the fairly large collection of citations. CMS Wire had the largest number of items but doesn't have a date sort feature, so it would have been easy to miss something more current than their reference to the multi-site management article earlier this month.

CMS Critic only had five URLs (and one of those was their complete CMS listing), but they were very recent. Interestingly, Plone doesn't make CMS Critic's list of enterprise content management, there isn't a Plone review, and there isn't an interview with any key players. Perhaps there's a marketing opportunity here.

Saturday, March 21, 2009

Usability of CMS Home Sites redux--the new Plone.org

The much anticipated Plone.org redesign has rolled out the door this week and from what I've seen on Twitter, its a winner. From what I've seen myself, its a winner. Welcome to 2010, Plone... 2010 and beyond.

Admittedly, I've got a soft spot for Plone, using it in my day job for implementing a variety of web projects. This week its been a portal for an upcoming UNSCR 1540 workshop. Last week it was a global security website. The week before that we used it as the modern front end for a legacy distance learning tool. So how's a guy supposed to be unbiased?


Let me turn back the page a month or so to my posting on usability and see how the new Plone.org stacks up with the old. Recall that I was ranking sites based on how few of Jakob Neilsen's "Top Ten Mistakes in Web Design" are evident. Low numbers mean fewer usability mistakes and a better home site.

Mistake

Old Plone

New
Plone
Observations

Bad search


PDF files for online reading


Not changing the color of visited links


Non-scannable text 0.5
"F pattern" fully embraced
Fixed font size


Page titles with low search engine visibility 1
1
Buggy but expected to improve
Anything that looks like an advertisement


Violating design conventions

0.5
Hyperlinks within text block non-standard
Opening new browser windows


Not answering users questions 1



Targeted straight-forward references
Total
2.5 1.5


Plone. The homepage is very clean and now displays key information in the highly scannable "F-shaped" pattern that eye-tracking studies have shown is common when users are looking for something. News and upcoming events are in an unobtrusive font even though centrally placed--easy for the eyes to jump past if that's not what you're looking for. Large event graphics, which are, in a manner of speaking, advertisements from a newbie's perspective, are out of the way yet easy to find for community members (and exceptionally well integrated into the design, too). The main tree branch graphic perfectly captures the eyes' flow of vision, taking the new site visitor straight to "Explore the possibilities." On top of all that, the opening-bud theme resonates with me personally since I'm a reformed botanist.

Page titles remain tricky in the documentation and could be better for internal search engine visibility. There appears to be a bug in there--I found myself at a login screen when searching for "lock down your site."

The marketing box at the bottom of the page is gone, replaced by a concise block of text that addresses key pain points that potential Plone users might be bringing with them as they visit the site. It took me a moment to realize that the bold text represents hyperlinks--0.5 point off for not following standard hyperlink conventions here.

Plone.org now has only 1.5 of Neilsen's baddies, a 40% improvement over the previous 2.5 score and a 10% improvement based on the total scale maximum of 10.

Does anything give me real heartburn here? Nope, I'm very impressed, especially when compared with the numbers other CMS were scoring (5.0 - 6.0).

I was a little surprised when the "best security" link took me to CVE but not the pleasantly empty query page for Plone in the last 3 months or at least the NVD search page. Considering how unusable the CVE/NVD pages are, it might be better still to set up a dynamic page drawing down CVE counts for a couple dozen CMSs. (I think by making a criticism like that, I just volunteered for some Python scripting to solve the problem.)

I was a little distracted by the out of focus portions of the main graphic, but the longer I look at it, the better I like it just the way it is. There's also something just a little off about the Plone logo on the box, but its probably just the perspective being used on the angled container. Nice use of a single wet-tile graphic there--that's an often overused motif.

I saw a reference to "total world domination" with a trademark logo on the "New look" page. Not sure what that means, but I'm still leary of the concept based on my March 7 posting.

I do miss the colophon proper, if only because its a great printer's term to throw around at parties. That said, the footer is an outstanding example of how to arrange information so that its available but not in the way. Most site designers, including the old Plone.org, would highlight that material at the very top of the page and really cut down usability.

For those that haven't clicked on the "A new look for plone.org" news item and read all the way down to the "We'd like to thank the Academy" section, here's the list of those who made this possible:

  • Iain Claridge of Netsight, who did the graphic design many moons ago, and kept improving it,
  • Alexander Limi, Plone co-founder, who translated it into a simple and elegant Plone theme of real live HTML and CSS on top of xdv/Deliverance,
  • The amazing plone.org admin team, who brought plone.org into the modern age by upgrading us to Plone 3, and continued improving the site and its setup:
  • Alex Clark
  • Laurence Rowe
  • Tarek Ziadé
  • Alexander Limi
  • Steve McMahon
  • David Glick
  • Matthew Wilkes
  • Martin Aspeli
  • Graham Perrin
  • Ricardo Newbery
  • Ian Bicking,
  • Paul Everitt and the crew at OpenPlans, who created Deliverance (and its sibling: xdv), the new theming system we’re using.
  • The many people who updated, edited, rearrange and revised plone.org content in preparation for the relaunch.
In conclusion, I won't just give a tip o' the hat to the design team that rolled out the new Plone.org. I genuflect in their presence. Terrific job, everyone. Awesome!

Tuesday, March 10, 2009

Plone's Subway Station

One graphical idiom that has caught on in the last few years has been the subway map for the Internet. First popularized by iA by placing top-ranking Internet concepts on top of the Tokyo metro map, now they're working on their 4th annual production (pending any day now).

Similarly, CMS Watch has put together a vendor map of the CMS landscape. Here's a small copy (click to view the larger original).

Interestingly, Plone is tucked away in the lower lefthand corner at the intersection of the web content management, social software & collaboration, and enterprise portal lines. The main line running horizontally across the middle of the map is enterprise content managment--not surprising given CMS Watch line of business.

Back to opensource, Drupal is one stop away, but off the enterprise portal line. Joomla, Typo3, DotNetNuke, EZ Publish, and their ilk are nearby on only the web content management line. WordPress, frequently compared with CMSs like Plone, is halfway across the map on the social software & collaboration tube.

To reach other stations that share Plone's three lines, one has to travel all the way down to the main line. There we finally meet up with IBM, Microsoft, EMC, Interwoven, Vinette, and Oracle. These are all commercial, proprietary heavy hitters.

This visualization gives one the best eagle's eye view of why its so hard to compare the many vendors and their products. Systems that are frequently lumped together may be adjacent but on different lines. Distinctive core capabilities and differentiating feature sets mean that CMSs have stations throughout the requirements landscape and are adapting in different ways as that landscape shifts, expands, and evolves.

Thanks for this, CMS Watch.

---------------------------

While I've got your attention, I'd like to mention an unrelated item that also hit the web on 3 March. Its a posting on A List Apart entitled The Elements of Social Architecture. Worth the read. Nice plug for the book, too. Keep in mind the closing quote from the author, Christina Wodtke:
If we remember the social in social architecture, we can continue to make new products that delight people as well as change their lives.

Saturday, March 7, 2009

World Domination and CMS Evolution

I've been noticing that Drupal often has been sending out the message of "world domination" and that has made headlines over at CMSWire this week. I'm not sure what world domination means in terms of CMS or OSS, but it sounds like driving all the competitors out of business or into insignificant niche markets. That sounds like attempting to be all things to all people.

Let's talk ecology and evolution for a moment. There are indeed very good generalist organisms out there: think Homo sapiens and Blattella germanica. But if you actually add up biomass or count individuals or consider many other indicators of breadth of ecological success, you'll find that its probably some species of dinoflagellate, a single-celled phytoplankton that numbers in the gazillions throughout the oceans of the world. You have to be very near the absolute bottom of the food chain to be successful by that metric. In software, these are operating systems and low-level communication protocols.

The only definition of "world domination" that fits a highly evolved system is the ability to actually critically impact the global ecosystem, much as mankind is currently doing in terms of climate change and species extinctions.

Can any CMS claim in its wildest dreams to actually impact the ICT "ecosystem" in such a fundamental way? I would argue that only operating systems, transfer protocols, and maybe some database management systems have this level of world domination. World domination for a CMS must be limited to running the other CMSs off the field.

Back to ecology and evolution and being all things to all environments (or all CMSs to all people). There are basically two ways to look at environmental penetration and it has to do with r and K selection. r-selected species are good at maximizing their intrinsic rate of reproduction. These are invasive species. Think Taraxacum, the dandelion.

K-selected species are those that flourish near their carrying capacity, the limit at which needed resources are scarce. They have low intrinsic rates of reproduction but typically long life spans. These are community dominant organisms. Think Quercus, the oak tree.

r-selected organisms thrive on disturbed environments like dandelions spring up in the bare patch your dog dug up in your lawn. K-selected species are slow-growing and difficult to establish, but once they have a foothold, they hang on while hundreds of generations of dandelions come and go at their feet.

So what is Drupal, a dandelion or an oak?

I'd suggest that spreadsheets and word processors are the trees of the temperate hardwood forest that is the 2009 software environment. But like a forest, its not a monoculture, but a complex mix of oak, maple, yew, and a few dozen other aspect dominants. There is no world dominant deciduous hardwood species.

So what's going on in the forest today? -- r selection.

Google shows the following trends for Plone, Drupal, Joomla, and WordPress.



Note that when Joomla took off or even when WordPress (not really a CMS IMHO) started its rise, there was no corresponding decrease in Drupal and Plone. This is evidence that the CMS environment is not at carrying capacity. In short, CMS development and deployment isn't a zero-sum game. We're nowhere near the carrying capacity of CMSs, as is also evident by the hundreds of CMSs over at CMS Matrix.

What's the slow decline in Google searches for Plone while everyone else is increasing like dandelions in your backyard? Matt Hamilton observed that, internal to Plone.org, searches are handled by Plone itself while Joomla, et al are using Google's site search. That automatically biases the results because they're missing tons of internal searches not tracked by Google.

As Alexander Limi recently observed on the Plone msg boards, if these stats are all there is, we (Plone and Drupal) are already dead--Joomla and WordPress are far ahead. I contend that WordPress is not powerful enough to be a true CMS but can pass as a low-end one for those lacking more complex workflow and security requirements. And if I were looking for an enterprize CMS, events of the Mambo-Joomla rift back in 2005 would give me night terrors.

That leaves us back at Drupal and "world domination," trying to be all things to all people. Can one software stack be the perfect blog, wiki, community plumbing, document archive, and web publishing solution? Can one organism be the perfect cryptogam, fungus, gymnosperm, ungulate, and carnivore? There's a reason for divesity--for organisms its the complex interactions among other organisms and the environment. For software, there are equally complex interactions and a rapidly evolving environment. World domination just won't happen for any one framework or application.

Now don't ask me about convergent evolution and CMS...

Sunday, March 1, 2009

CMS Statistics at Amazon

Seems like I just turned around and now three months have passed since I did my last quarterly summary of Amazon sales rank numbers for Plone. Here's the new graph:


Here's the table:

Title Author 3/1/2009
Professional Plone Development Aspeli 249,341
A User's Guide to Plone, 2nd Ed Lotze, et al. 238,151
Building Websites with Plone Cooper 629,953
Definitive Guide to Plone McKay 446,029
Plone Content Management Essentials Meloni 786,631
Content Management with Plone: Handbook for Authors and Editors Lotze,Theune 832,108
The Definitive Guide to Plone, 2nd Ed Reale,McKay 1,027,531
Plone Live Pelletier, Shariff 1,337,175
Plone 3: A Beginners Guide Knox, et al. 89,753

Tip o' the hat to Knox, et al., where et al. is a who's who of Plone core. This much anticipated new title rolls out the door with a sales rank of about 90,000 and well ahead of the rest of the pack of Plone titles. Nearly tied for second and third are Aspeli and Lotze et al.

However, Plone pales in comparison with other CMS texts. Top-ranked at 3,816 is Angela Bryon's Using Drupal. Nearby at 4,872 is Joomla! A User's Guide: Building a Successful Joomla! Powered Website. There are SharePoint titles at 15,000 and 20,000. DotNetNuke for Dummies weighs in at 31,000. But don't feel bad, Web Content Managment for Dummies has a sales rank of only 939,917.

Another way of slicing the Amazon.com pie is to browse down through their categories: Books > Computers & Internet > Web Development > Content Management. There you'll find Cameron Cooper at #11 and Zope: Web Application Development and Content Management at #22. Amazon updates its category listings hourly, so check back at 6:00 AM and it'll be different.

To give you an idea of how much these numbers bounce around, A User's Guide to Plone: Updated for Plone 3 is now ranked 262,229, down from 238,151 when I started harvesting data this evening. Practical Plone 3 has dropped to 99,227. The data is almost to dynamic to be meaningful.

One final thing to look at tonight is what the Plone publishing trends over time have been. Starting with Definitive Guide to Plone back in June, 2004, a flurry of Plone titles hit the shelves. Then there was a long pause before Content Management with Plone: Handbook for Authors and Editors came out. Since late 2006 the books have come out on a fairly regular basis and the pace appears to be picking up.


I realize there's tons of great how-to's and tutorials over at Plone.org, but a great way to support Plone and its authors is to buy their books, even if you just donate it to the public library in your neighborhood. Go to Amazon and write a review. Give 'em a ranking (5 stars would be nice). Pre-order the 2nd addition of The Definitive Guide. Write O'Reilly and Wiley and Sons (For Dummies publishers) and request that they put out some Plone titles. Heck, you might even write your own Plone textbook, but above all, support your local authors.