So I took a 10mg dose of Ambien roughly an hour ago, and I've still been unable to sleep. What I've learned is that Ambien makes you physically tired, but it doesn't change your mental status - my mind's been racing with so many ideas and things that I just can't shut out the voices and get sleep. So even though I'm physically very very tired, I cannot fall asleep. It sucks. I'd imagine the same would be happening if I scorded modafinil - my body would be physically alert, but my mind would be shut down. Apparently kids take Ambien and stay away (to get high, I guess); I have to admit the last time this (unintentionally) happened, I apparently went downstairs and started cooking up a storm. My mom was curious what all the ruckus was, so she came downstairs. Apparently I had a very interesting conversation with her about my sister, me, and our futures (I'm normally very reserved towards my parents), which I barely remembered the next day (I at first thought I had dreamt that conversation, then she told me it was real).

What am I trying to say? There's a good chance that I'll barely remember writing any of this tomorrow. I know I'm pretty prone to sleep walking (happened many times in my youth) and sleep talking (happened many times in my youth at college). What I didn't know I was capable was of sending phone calls to my bosses at work while asleep.

Hah! Got ya scared, didn't I?

Last Sunday (I woke up at 3pm), I checked my email and noticed that Urs sent me an email that said (paraphrased, I'm on the wrong machine): "Are you ok? Your message was a bit weird and you didnt seem alright..."

I figured his email client's autofiller screwed up and responded with a message saying, "You sure it was me? I was asleep all morning!"

His response: "Yep, it was a Roy, and you're the only Roy I know!" So either another roy mis-dialed Urs, or I *really* did call him while I was asleep. Speaking of which, this website says there are only 185 Roy Kims in the US. Bollocks. My plan on taking over the world with an army of Roys equipped with Ford Taurus has failed twice this week - once since the Taurus has been discontinued, and twice because there are not enough Roys!

So what's keeping me up tonight?

Swarmf (layout layout layout!), listfoo (version 1.1), tabulas (when is it ever going to be done?), LoD (this is my venture with Matt), and "to be builts."

I don't know what you guys think about when you go to sleep, but I like to think of projects that I don't have the time or the resources to pursue, but would love to think the project through.

If you remember my post a while ago about the "perfect webmail" client, it's that sort of thing.

Last week it was managed higher education courseware. It's astouding that even today, the dominant system is BlackBoard. Universities ... bastions of intelligence and cutting edge research .... blackboard. There is an open source version called moodle, and it looks very promising. But like all open source projects, the thing lacks polish. The diffference between unpolished and polished projects is the difference between audiomatch and audioscrobbler (guess which one tanked and which one blew up).

So on better courseware management, there are 3 groups: students, teachers, and administrators. In reality, administrators don't have much to do with the l'arnin' side of things, so we don't need to focus on them. Prospective students are nice too, but they can be pushed to the side (websites != courseware management).

The problem with creating any content management system is that the system will impose itself onto the users (in some way or another). Wikis have the advantage of being *so* open that you create your own structure (whcih is a huge difference and a huge boon). MindTouch recently switched over their homepage area to Drupal, a strict CMS, and managing it is a huge nightmare. I love the wiki way, so much more.

I think the truly best software system would be to create a really flexible framework that allows feature generation to be done relatively quickly. Build simple tools and classes; build a framework for handling views and posts and gets; handle all the internationalization and database handling and sessions handling .... and then let a few smart developers build the features you want.

What I've described is (duh!) open source software and frameworks (like RoR).

My disdain towards frameworks and OSS is primarily that it's scary. It's scary to step outside of the bounds of a framework, or it's scary to have ot modify the source code of an OSS to get it to do what you want. You cannot expect to sit there after a few hours and have a full comprehensive understanding of how this particular object was written or what philosophy came into it. What do you risk by changing it?

Part of this is frustration from my job; in the beginning days, we did not fear breaking things. We took huge risks, trusted each otehr, and basically ran running down the street as fast as we could trying out features. But lately ... things must be much more stable, and that sense of chaosness is gone (which is a good thing, too).

But I haven't been able to create. Everybody enjoys teh feeling of creation - "I did this" and that type of visceral feeling is missing from my job now. I mostly do upgrades on older stuff, fixing up bugs elsewhere, and basically take the most conservative (least risk) path.

But back to my point. It'd be super sweet to build kick-ass courseware that not only students woudl enjoy, but teachers would enjoy using as well. This alludes back to Fuunk, (which is probably the most abandoned project of mine, but I spend significant amount of time thinking about it) which was going to try to bridge the gap one way by offering students excellent tools to manage their classes and professors.

What would the ideal student-oriented courseware do?

• fuunk: It lets me schedule classes using a powerful search interface that extracts in a lot of metadata from professors so I can easily get a "battlefield view" of possible classes
• It'd store information about my major and give me tracking advice on my major - it woudl also show me what would happen if I wanted to switch majors
• It would show professor ratings and comments from students
• It would show me all the books I need to buy for classes I've registered with
• it would let me create a profile of information abotu myself so i can connect with my friends
• it would let me show me what classes my friends are taking and which friends have taken my classes - it could also arrange for book transfers between friends
• A general wiki area where students can combine class notes

What would the ideal professor-oriented courseware do?

• Easily update their own personal webpage with their CV and information about their research (gotta attract students) - give them a blog! They've got the most interesting thing to say!
• Allow for creation of "class pages" which would have the default blackboard features - tests, quizzes, files, etc.
• A wiki/discussion board to encourage student participation
• Show grade breakdown of students
• Have an easy check to make sure students aren't plagiarizing from one another
• Allow professors to introduce promisings students to colleagues

This is the type of tools I can hope that one day will be built - a completely social tool that bounds beyond groups (students and teachers) and bridges gaps between them to help faciliate information. I think it's very easy to create an echo chamber with people you're comfortable with, but it gets much harder when you're trying to do it outside of your immediate group.

(APOLOGIES HERE because I know I'm rambling a lot; the fingers are just sorta typing, and I'm not really editing myself here. This is pure late-night-can't-sleep-even-after-i-took-sleeping-pills rambling)

So how can this apply to other concepts? Well, after the website makeover for the town of chapel hill website thing happened, it got me thinking about what needs a town would have in running a website. What do I, as a citizen, want from my town website?

• I'd like to be able to know all relevant news regarding the town (aggregate it from N&O, the Chapel Hill News, press releases, and local bloggers)
• I'd like to be able to access all electronic documentation (minute meetings, proposals, etc.) that is easily browsable, searchable, and readable (not just HTML, but PDF and DOC also) so I can track the status of Chapel Hill North
• I'd like to get the contact information for people easily
• I'd like a calendar of events , but not just of town council meetings. I'd like a calendar of events around Chapel Hill that *make* Chapel Hill special and unique.
• I'd like a user-editable section (wiki-ish) where we can make recommendations on local venues, restaurants, etc. Owners could submit profiles of their own places and they could have their own page up there . (I realize I'm just describing Citysearch here)
• Come election time, I'd like to easily find out who is running, where their websites are, where they stand, and what i need to do if i want to vote. The whole process, for people who haven't done it, is scary. Do I need to register? Where? By when? Maybe even throw up an easy "5 Steps to vote".
  
  Oftentimes when I'm helping people, I get too technical with my answers. I assume they know what a browser is. The same goes the other way too. Sometimes you just gotta offer the simplest and most basic of all questions.

And of course, the biggest thing I'd like: I'd like the data to be open. I would like the site to have very simple APIs for accessing the data so I can aggregate it on my own and write my own little programs to do stuff with them. I've seen some pretty cool mashups with data from police reports and crimes ... (I think WillR posted it awhile back).

I had this long dinner with Konrad about these dreams for a site like this - how freakin' wonderful would it be? And not only that, to have the whole project be open-sourced so any city could gain from this?

We both talked about putting together a small webcrew to try to build a demo of this if there was a chance we could land the Chapel Hill contract ... but time went on ... alas.

And you wonder why I can't sleep ...