This site is rarely updated. benbrophy.com is more up-to-date. - Ben
Sections and ad hoc groups
There is a lot of work going on Sakai to bring sectioning and collaborative groups into Sakai. These items have been on the Sakai tools team requirements list for Sakai since the very first meeting in Indianapolis, February 2004. we're now defining functionality, but I think a lot of confusion arises in the design from collapsing sectioning and ad hog groups into one user interface.
First some definitions:
Section: Sections are primarily an administrative device. Sections are assigned instructors (Section leaders). Section data is often supplied by the registrar, and students often register for a section when they register for the class. Sections tend to be tied to a meeting time and place. Sections are used to distribute work in the gradebook, announcements and email are often sent to just one section, and assignments may have varied due dates based on section meeting times.
Ad hoc groups: Ad hoc groups are primarily a pedagogical device. Ad hoc groups don't have a formal leader, in fact students may form groups on their own. Ad hoc groups are usually used to foster collaborative work. Ad hoc groups of students might have a shared workspace, or turn in an assignment together.
Section/Ad hoc group Bundle: It's hard to get a good label for this concept. One class may have several bundles of sections or Ad hoc groups. For example, a class might have a bundle of 4 sections, a bundle of 20 project teams (ad hoc groups), and a bundle of 8 lab sections. There may be rules around the bundle - for example it sections might have a maximum size of 30 and students may switch while the labs project teams have a maximum size of 5 and the membership is assigned by the TA.
Sections and Ad Hoc Teams answer very different user goals. The rules that define them are quite different. In order to simplify the interface and make life easier for our users it may make sense to treat them as separate tools, and prioritize them separately.
Note that MIT & Stanford took created sectioning tools. We didn't support Ad hoc groups, just sections. In addition we only had one section bundle. The class can set up sections only - no labs or other types of sections. (some one from Stanford may correct me on this)
Sectioning is baseline required functionality for a course management system. MIT, Berkeley and Stanford have all said they cannot release without. It's my impression that people are genuinely surprised to learn that Sakai doesn't have sectioning yet. So it's a high priority, but it's not exciting. No one will say "Sakai is innovative because it has sectioning." But if it's not there they will say "We can't use Sakai because it doesn't have sectioning."
Ad hoc groups are exciting - when we call Sakai a Collaborative Learning Environment, this is the sort of thing we have in mind. If we deliver this, it will enable a lot of exciting possibilities in new tools. At MIT we've heard calls for this sort of functionality for years. Unfortunately it's been pushed of as we work on basic administrative tools, like sectioning.
Tags: sakai sectioning
Road Trip
Big chunk of the ed tech and multimedia staff at MIT got on a minibus and took a road trip down to Yale today. This was a reciprocal visit, the Yalies had come to see us a few months ago. We talked about Sakai and uPortal, of course, since both schools are involved in that work. Yale is already piloting Sakai and plans to pilot another 100 classes this Fall - much larger pilot than MIT is planning.
One benefit of the trip were seeing the way their teams work, reinforcing some professional contacts. I'm very interested in the work going on at Yale's library and I got sneak peak at some proposals in that area. It was also fun to have a change of pace, and spend a few hours driving around in the nerd bus with our team.
Being open in open source work
Here's an interesting blog entry by Graham Attwell titled "Moodle and Sakai - Open Source alternatives?"
He notes the difference in the perceived openness between the Moodle and Sakai communities. I've seen some of the hostility toward Sakai here and there around the web, compared to the affection felt for Moodle.
Sakai planned in advance to have the initial "seed" application built by a few "core" universities who had each built their own course management systems in the past. The application would then be handed over to a foundation and become a more trditional open source project. We are nearing the end of those initial two years.
It's interesting that Sakai has been so good at generating positive buzz, and also at generating criticism. I think the secret to both was having a closed core group that reveals it's progress periodically.
Going forward it seems that it's going to be really improtant for Sakai to successfully transform itself from a closed core group to a truly open open source project. I'm reasonably optimistic - the change is starting already. Sakaipedia is a good indicator of Sakai's growing openess.
Tags: sakai opensource
The MIT License
When I release Linkwalla via sourceforge, I'm going to use the the MIT License. I used to just have it in the public domain, but this license offers the most of the benefits of the public domain, but gives me a little protection as well.
By way, here's why I'm calling it "linkwalla." "-walla" is a common suffix in Hindi, usually used to denote a persons occupation. For example "dhobi" mean's laundry, so a "dhobiwalla" is a launderer. "Chai" is tea so the "Chaiwalla" runs teashop, the "taxiwalla" drives a taxi, etc. So it seemed right for a an application whose job is to manage links.
It's also a film term meaning "a sound effect imitating the murmur of a crowd in the background." Which seems like an OK metaphor for the constant stream of URLs I see passing through my inbox.
Tags: MIT opensource linkwalla Hindi
Return of the PHP Hobbyist
I recently announced a beta version of my linklist-building PHP app Linkwalla, and implied that I'd be stopping development. The part about stopping development was pure poppycock, like promising that I will never itch a mosquito bite. In fact I already went back and improved the RSS feed.
I decided to go on sourceforge and see who else is working in this area. There were quite a few. The main ones being:
PList, which looks nifty but requires a MySQL database. My vision is database-free - totally XML based. The installation instructions should be: Just unpack it and it works.
PHP Linklist, which is XML based but seems to be more geared to static link lists, than the date based technorati-style link list with RSS feed that I envision.
So I think I'm going to start my own sourceforge project. Who knows maybe if I ask for help someone will pitch in. Basically I have fun doing it, and see no reason to deny my nerdy pleasures.
Tags: linkwalla simplexml php sourceforge opensource
Flash Fatigue
It didn't take long. I got really tired of Flash this week. My big 3 complaints:
Scripts all over. I had actionscript on this button and that button, scripts on several frames in their own "as" layer, scripts on movieclips, scripts on frames within movie clip timelines. It's crazy to keep those things all over the place.
Windows everywhere. I am accustomed to coding in BBEdit, where I've got a window full of text an little tray of the related text files I'm working on. Pretty simple. In Flash there are so many palettes, plus the stage and the timeline and my super sized script window, that I need an external monitor just to fit them all.
Compiling. What a drag to have to compile a swf file every time I want to see how it's going. Especially as the swf file gets bigger.
Admittedly most of my trouble comes from being new to something. I would find ways to keep my actionscript external and edited in BBedit, and mostly in one place, and written in a way that makes compiling faster. But that's not easy, and I'm also still troubled by using a development platform owned by a corporation. It just doesn't feel right.
I think I'm better off redoubling my efforts to improve my DHTML skills, and couple those with PHP to make dynamic web apps.
Tags: Flash actionscript
I am a statistic
I took this survey for bloggers being one by a rade student at the media lab. If you have ablog you might enjoy taking it, too.
Craig @ work
Craig Counterman has joined the still nascent MIT blogosphere. This is great because Craig is my most trusted source for technical wisdom, I can't count the number of epiphanies I've had talking to Craig.
He's also extremely funny. One of his inaugural entries on Beaver Engineering has forever changed the way I see beavers (and software engineers).
Tags: MIT Counterman Blog Beaver
The endangered blink tag
Wow! You always here the sort of cliched jokes about using the <blink> tag. But it's been years and years (and years) since I've actually seen one in the wild. But check out Ann's Home Stay! Not only is there a blink tag, it's being used for the site's navigation. What a find.
(Web design aside, I am seriously fantasizing about a trip to Ann's Home Stay. I loved Kerala when I was there, and the place looks swell.)
Slideshow pro discusses new Stellar tool
We're planning a very limited pilot of a tool that uses images in Stellar this Fall. Based on early feedback from the faculty, it sounds like there is a need for tool to assemble and present slideshows. This is interesting - it's related to the demise of the slide projector (Kodak has stopped production of their projectors and their slide film). I'll be blogging about slideshows a lot in the next few months. Jean and I met with David Jones, a professor of the History of Science. David taught STS.003, the Rise of Modern Science, in the Spring. He builds image-rich lectures, using Apple Keynote to arrange and annotate the images he has found on the web. Like everyone we've heard from he's a heavy user of Google Images. He also finds images through various government agencies, the Smithsonian, the National Library of Medicine and through contacts with other faculty.
The biggest drawback to his current method is vagueness around intellectual property. He mentioned concerns and confusion around copyright issues several times. He has no way of knowing if it is OK for him to distribute the images has has found on the web without permission. Most of the images are pre-1920s or are taken from government agencies, so are theoretically in the public domain, but it's never absolutely clear. As a result of this confusion, he has a beautiful collection lecture notes that were not accepted into OCW because of IP concerns.
The other major issue he faces is image quality. Google images are often of inferior quality, and it is difficult to sort out the bad ones. Images from other sources are often too high in quality, and he needs to down-sample on his own using Apple Preview - a process he doesn't quite understand.
He has 1,710 images related to STS.003 in a folder on his desktop. He titled them himself and uses Apple Spotlight to find them. He does this because he often reuses the images, and doesn't want to have to find them online again.
He also uses video clips, and would love to be able to search for historic moving images.
In his slides he often juxtaposes multiple related images - though he says this is a feature he could live without. He does this because it's easy in keynote and it cuts down on the total number of slides. Also in some cases several images on one page can be used to tell a story linking them together.
In additional to the descriptive text students see on each slide, he keeps notes for himself to use in the during his lecture. Having descriptive text appear with the slides when students review the lecture notes is important to him.
During the midterm exam he showed 5 images he'd used during the lectures and asked student to identify 3 out 5 of the images. He said this made it clear who was paying attention during lectures.
In the Fall he's teaching the History of Medicine, and plans to make the presentations for this class image-rich as well. He said be willing to review our progress and do some testing of the tool we create later this summer.
Based on this conversation, it seems clear Stellar Slideshow should:
- Make copyright information clear (this may be the greatest success factors for our effort)
- Export IP info to OCW
- Resize images appropriately for the screen.
- Allow instructors to add text notes to the image
- Provide a single search field (like Google)
- Make it possible to search through all of the images a professor has used in the past
Cookie Monster
Here's an interesting article on cookies (the browser kind, no the yummy kind). See, Marketers love them, but consumers are getting into the habit of blocking them are or deleting them all regularly.
I am one of those consumers. Much to the irritation of anyone who borrows my laptop, I Have my machine set to confirm all cookies as the come in, and I delete most of them. It's little slow doing it that way, but I learn a lot about how cookies set cookies. I am the kind of person the marketing could work on - the sites could convince me i should take their cookies.
I don't think the marketing will work. Because unlike me, most people don't want to sort their cookies separating good from bad, so they just delete them all on a regular basis.
One thing that worries me is this:
The tool uses features in Macromedia Inc.'s popular Flash software, which is used for designing and viewing animated online ads, to secretly make backup copies of a user's cookies before they are deleted.
I've also noticed Flash being used to get around pop-up ad blockers. I'm afraid that if this abuse continues, people may start blocking Flash content. Since I'm sort of romancing the idea of Flash development right now, I'd hate to see that happen.
Linkwalla: Simple link list
I'm releasing my latest output of my PHP/XML hobby. I'm calling this thing "Linkwalla" - it's a way to keep a list of links on your website.
Download Linkwalla (zip - 30k)
From the Read me:
README - LINKWALLA - Version 0.1 - June 17, 2005
Hi! Thanks for checking out this script. I am releasing it to the
Public Domain, meaning you can do what ever you like with it. I'd
be honored if you use it and mention my name, but that is purely
optional.
REQUIRES: PHP 5
FEATURES:
- Build a list of links. Sort of like del.icio.us for your own
website, but without the social tagging.
- Add edit and delete from the list
- Runs off a XML file - no DB needed
- Comes with a javascript bookmark for adding links on as you
see them
- Creates an RSS feed of the links as you update the list.
INSTALLATION: Just put everything on the server as is, and it
will work. The default password is "pw." You can change it at
the top of the "dw_functions.php" file.
DIRECTIONS: Hopefully pretty self explanatory. Click "log in"
at the top of the page to get started (default password = "pw").
Then you'll see links to to add remove and edit links. The
bookmarklet will make you're life much easier.
This is the Beta release, but don't expect any new releases soon. I'm going to take a break from PHP for a while and maybe get back into Flash ActionScript. But you never know.
Tags: PHP XML simplexml link script
Steve Jobs Commencement speech
Steve Jobs (who never graduated from college himself) gave this fantastic Commencement speech at Stanford. I'm all inspired. I've always been interested in mortality as a motivator (there's the secret message behind that tattoo on my arm), and he put's it well. Kudos to his speechwriter.
HarvestRoad Hive
I saw a demo of the HarvestRoad (ASX:HRD) Hive repository today. Hive is a digital repository aimed at learning, created in Australia with large installations scattered around the world. I've seen 3-4 digital repositories demoed in the last few years, and this is definately the most impressive.
Tags: HarvestRoad Hive repository OKI
Wrapping up: Product, platform, style guide
I've come to realize that much of the of the concern over the Sakai style guide comes from the misconception that the style guide is the Sakai's total solution to user interface design. In fact it's just a guide to help designers in disparate locations create tools that are consistent with one another in their layout and interaction.
The style guide is just a tool, and using it certainly doesn't guarantee a good user experience. The efforts in Sakai by and large have been architecture driven, and deadline driven - UI has been low on the list of priorities. As a result many of the tools in Sakai feel like "proof of concept" tools. They weren't designed and tested before coding began. Applying the style guide to them retroactively is putting lipstick on a pig. Even now there is a rush to create new sectioning and resource tools for v2.1 that will mean developing those tools without user input or user testing.
The rush is understandable - the framework has to be a priority because Sakai won't work without it. And the deadlines are very very tight. But I hope as we move into 2006 we can start taking a look at the functionality in Sakai and take the time to do it right.
This relates to the earlier posts on whether Sakai is a product or a platform. Most of the work is going on at the platform level. This explains the 'proof-of-concept' feel of the tools in place now. Unfortunately users don't interact with a platform, they interact with the tools, so the tools are what people talk about.
I hope that as the platform stabilizes, the larger organization of schools working on Sakai will be able to collaborate on creating tools for Sakai, which can work together well thanks to sharing a style guide. We'll hopefully be developing lots of tools at MIT - I hope that the tools we get from other schools will not only be able to 'plugin' technically, but also work seamlessly in terms of user experience.
I think I'll take a break from writing about Sakai for a few days...
Tags: Sakai UI styleguide
Style Guide
Somebody added this note to the User Interface page in Sakaipedia (note the page is in a Wiki so the quote may change):
Right now the sakai style guide is about defining interface conventions. This is a good thing for the programmers and for the consistency about tools, but it says nothing of visual style or the overall user experience we want our students and faculty to experience.
The style guide's primary aim is to provide the people who use Sakai with a consistent user experience. So the rules around interaction are very much in the tradition of design patterns. For example the style guide entry on forms describes where and how to arrange the buttons:
- Under the form, aligned left.
- Submit button left most, cancel button right most.
- The submit button highlighted with a thick dark border.
- The cancel button takes you back to the referring page, without saving data.
That's not very exciting stuff, but it's handy for designers (and programmers) who want their tools to be consistent with the rest of Sakai. I know from experience people get driven mad by buttons being in a different places and order as they move through an application.
Even though that's not exciting, I think it's fair to say that it does say something about visual style and overall user experience. I found it very useful in designing the gradebook, and heard good things from other UI designers applying it to Sakai tools. I wish the style guide had been part of Sakai 1.5, and that it were more thoroughly applied to Sakai 2.0.
I'm eager to hear any specific feedback on the style guide. I'm watching this page closely, hoping to get some constructive criticism.
Tags: Sakai UI styleguide
Sakai Test and Quiz
Stanford's Daisy Flemming demoed Sakai's Test & Quiz tool (aka Samigo) this morning. Samigo has a complex user interface - but the pay off for that is that's it's got some complex functionality, especially it's templates and question pools.
Templates are a set of default settings for new assessments. One could set up templates for pop quizzes, for self-tests, for open surveys, etc. Question pools are, well, collections of questions. Those questions can also included user feed back based on the answer given.
It seems like the complex and configurable nature of Samigo's templates and question pools could be great for instructional designers, or particularly motivated instructors. I'm hoping we'll see some interesting reusable materials come out of Samigo.
Tags: Sakai Samigo assessment
Great taste! Less filling!
I've had some great responses from my post on whether Sakai is platform or product.
Florian Gnägi, lead developer for OLAT, wrote in on the product side. He pointed out that "Users should be able to export a course and send it to a friend to import on his own installation and it should just work." But that becomes difficult if there is no standardization around tools, the friend in his example would have to have the same toolset enabled to make the import work. He also pointed out what waste of effort it is to have a community duplicating efforts by developing 6 discussion tools with overlapping features.
I also saw this response from Michael Feldstein, writing in on the Platform side. He's got the dream of Sakai serving as a common base for all sorts of Ed Tech tools, that instructors could use to build on. "Their goal should be (like Moodle) to lower the skill requirements for tool building to a point where technically-inclined teachers can participate."
I still want both. I want good tools and a platform to build on. In a way I see the tools as part of the platform. Take the gradebook: It's a tool, but it's also a standard service. As a budding tool developer I benefit from having a standard gradebook to stick my grades in. Or at least in knowing that whatever gradebook is there will use a standard grading API.
The Moodle example is interesting. I recently set it up here to experiment with, and it didn't come with 5 discussion boards, just one. I'm sure there are other Moodle discussion boards out there, but someone picked one to include with the download.
And hey, what's with all this "Sakai brahmins," "top-down driven and CIO-driven" and "technocracy of developers" stuff? Does Sakai have an image problem or what? It seems like every open source project has people who decide what get's included in a release and what isn't (Martin Dougiamas isn't a brahmin?). I know I am of signifacntly lower caste.
Tags: sakai opensource moodle olat
Search widget
This site has an interesting search widget: http://www.dealazon.com/
Click in the search box and the advanced search features appear. It looks like a nice way to stop the advanced search features from cluttering the UI, but having them available without needing to jump to a new page.
Here's some commentary from the designer. This might be handy for the Stellar Image browser we're developing.
I found this via my Technorati tags. This is just why I added them.
Usability Test Archives
Jakob Nielsen's Alertbox today recommends that all usability groups keep an archive of old reports. He's got a good point, but it doesn't strike me as an urgent priority. It would be hard to pull new information from those old usability reports after the site has been revised. If usability suggested some form labels were unclear, and we changed the labels, that old information isn't too fresh any more. Usability reports are really interesting in their specificity to a specific application in a specific release.
I do wish our usability here at MIT kept those reports, mostly so we would have a record of the reasoning behind some of our changes. But it would be so labor intensive to pull that reasoning out, I'm not sure we'd do it.
This one note, though, does strike me as urgent:
Individual findings can be generalized to usability guidelines once you observe them repeatedly. One of the most powerful ways to increase your usability group's productivity is to develop customized guidelines for your specific type of user interface.
It's these guidelines, kept in something like, say, a style guide that really help streamline the development process. Also, at the SEPP conference last week, several people new to the projects asked for the justification behind style guide decisions. That rationale wasn't recorded, unfortunately. it was based on the combined heuristic reviews of UI designers and usability professionals from 4 universities. Wouldn't it be great though, to have an archive of user test reports to point skeptical developers at.
Tags: usability Nielsen documentation
Platform or Product?
Is Sakai an open platform for developers or an integrated product for instructors and students? For me that is the big unanswered question from last week's Sakai conference.
It seems like the optimal answer would be "both." Downloading Sakai should mean you get a place to upload files, a calendar, discussion board, gradebook, membership list, etc., and that those tools work well together (e.g. an assignment's due dates from the gradebook also show up in the calendar, or that an new file posted can be publicized with an announcement).
Sakai should also have the APIs in place so that skilled individuals or small teams can build innovative tools on top of Sakai - and be able to take advantage of the infrastructure provided by the core tools. A new lab notebook tool could take advantage of the class membership list, for example.
The conference ended with a Q&A session with the Sakai board members. I asked how decisions about what's included as the 'core tools' will be made. The response I got from one board member was "I'd like to see Sakai include six discussion boards. The user can try them all and decide which they like best." No other board member disagreed.
That's been bugging me ever since. Even assuming that each school's IT department will pick a default set of tools (a possibility mentioned by the board member), does it make sense to have Sakai include every tool built? Can there be no standards for inclusion? If I'm developing a new attendance-taking tool that I want to be able to post grades, what do I do if there are 6 gradebooks?
The keynote speaker at the conference was Brian Behlendorf talking about how they run the Apache project. Apache was frequently touted during the conference as a governance model that could be used by Sakai. But Apache is a very different product than Sakai - the user interface for Apache is a plain text configuration file. It's users are the same people as it's developers. I think the people looking to Apache as a model for Sakai see Sakai as a platform, not as a product.
The same logic goes behind the idea of shipping Sakai with 6 discussion boards. Don't like any of them? Build your own! The assumption is that the user is a software developer.
I think a shared software development platform for educational technology would be a great thing - I'm not knocking that. But it will be a shame if Sakai becomes another first rate open source software project with a second rate user interface.
Tags: Sakai opensource
UI Design Patterns
Oliver Heyer pointed me to a white paper put out by Yahoo on using design patterns for UI design. Yahoo has a variety of teams working in a large somewhat decentralized organization (e.g Yahoo! Movies, Yahoo! Sports, Yahoo! Travel, etc.). The Sakai is similar in that it has a diverse widely distributed group of designers working on a variety of interconnected tools that plug into the same framework.
The paper has influenced my thinking on the Sakai style guide (Sakai UI design patterns? Sakai Interaction Guide? What's the best title for this document?) I hope we can clarify the style guide by expressing it as a set of design patterns.
Tags: designpatterns Sakai Yahoo UI
Great Java Code talk
I'm attending a talk by Ray davis from UC Berkeley on how the gradebook was written, going through all of the tricks needed to do good coding using JSF and Sakai APIs and such. It's really interesting, and the reason is that Ray is is a really smart intelligent presenter. I know very little about Java, but he's talking in away that I can learn some concepts, and some one who actually is a Java developer would learn a lot. This talk should really be seen by any programmer doing tool development for Sakai. Unfortunately, I think the wrong people are here, because this talk is head ti head with some other big programming sessions, like the meeting of the cross language DG.
Speaking of Discussion Groups, the UI DG was an exciting conversation today. There seem to be lot of people in the SEPP who want Sakai to use a User Centered Design Process, in which tools are designed and tested with the people who will use our application before they are coded (sounds obvious, but it's not the norm).
Sakai Baltimore
I'm in Baltimore at the Sakai conference. There are 415 people attending this time, about 100 more than anyone had expected, so it's quite packed. When we presented the gradebook yesterday it was standing room only, with about 15 people standing in the hallway craning their necks trying to catch a glimpse. It's not a very comfortable to run a meeting, but it doesn't make the meeting feel more exciting, as if the whole world were talking about Sakai.
I got a look at Melete. It would be nice to move away from branding individual tools. Melete ought to be called "lesson builder" just as Samigo is now called "test & quiz." Melete a good way to build sequential lessons, with built-in date release dates to enforce pacing. It seems like it would be especially useful for distance learning. Melete's not integrated with other Sakai tools yet, which will is what will really make it spectacular (and quite Moodle-like).
The real advantage for these conferences is the chance to meet so many people working in my field. The little 5 minute chats I have with people from schools from around the world will inform my work in the coming months. Plus I meet people like Jeff Kahn, an OKI developer living in Atlanta who will be helping us build in an image repository browser for Stellar. Out conversation opened my eyes to lot of new possibilities.
Brian Behlendorf is up now talking about how the Apache foundation works, so I'm signing off for now.
Intel chips
OK, so Apple announced they are going to use Intel chips, something people have talked about for years. I don't really know what that means. I mean, I know but I'm not sure how it effects me. I am worried they are doing it so they can build even more DRM into their systems (as suggested at Wired). Kottke reassured me it's going to be OK.
I just hope I don't have see a tacky "intel inside" logo on the case of my next PowerBook.
Sakai Face Book
I'm attending the Sakai Summer Conference in Baltimore next week. The organizers put together a little facebook application to let every one post their pictures and introduce themselves. It is kind of interesting to look at now, especially the things people say they are interested in. I'm sure I'll continue to use it during the conference.
This little app looks like it was thrown together in PHP just for this event. There is no authentication, and once you add a photo you can't edit it (which is a shame, i'd like to change my interest list). But that is what is so cool about it - PHP is quick and flexible and they can put something together, and use it for a little while with a limited community. it's obviously popular, and adds a lot of value.
What a shame it is that you can't do something like this in Sakai. If you could, you'd be able to take advantage of Sakai's authentication, so that part would come free. I am eager for the day (which I'm sure will come eventually) when i can write a useful little app in PHP and plug it right in to Sakai.
Whisper reincarnated: Wu Wei
I just have to close the loop on these Whisper CMS posts. Adam Newbold quickly released a new even simpler CMS - wu wei (or download the zip from my site). Wu wei's code is easy to read an easy to use. I miss the Textile support that is built into Whisper (even though I prefer markdown). Maybe I'll learn how to add it in some day.
Tags: opensource cms whisper wu-wei