Mon, April 18, 2005

In a few weeks it will be one year since onebee burst onto the scene with a fresh design and a few new features. I think it's been a tremendous year in terms of the content (quality and frequency) and I'm proud of the community we've built, welcoming new readers, furthering the discussion in the comments, etc. But there are a few features that never made it into the original onebee launch, and several new ideas that have come up over the past year, and now it's time for those ideas to get their due.

Beginning last Friday, I'm embarking on a sizable overhaul to most of onebee's code, making it into everything I could ever want it to be, and then (hopefully) stopping development and focusing all my onebee time on writing new content rather than hacking the code. The lion's share of the new features will only show up in the administrative interface I use to publish columns, create polls, etc., but there are quite a few goodies in store for the readers, too. (Plus, anything that makes it easier to post means better stuff being posted more often, which is a win for everyone.)

At work (and I never talk about work on my website – you can get fired for that!) we're also doing a redesign to our main site, and I had the idea that we should "blog" about the process, since we're in the communication business and a behind-the-scenes peek at our processes and decisions might provide a sort of "best practices" guide for our users. (Thanks for bearing with me; I'll try never to use management-speak like that again.) Anyway, the idea was roundly rejected (although I found out later that my boss pitched it to someone else in a later meeting, without attributing it to me). But the AccuWeather.com people are doing it, so it can't be that bad an idea.

Since there's bound to be some slowdown in output while I'm working on this, I'll try to fill the gaps with brief progress reports about how things are going and what I'm working on. It's a stark reversal from my usual secrecy about new features until the last minute, but I've decided that was a silly way to go anyway. (I mean, I gave Ari Fleischer such a hard time!)

I had last Friday off, so I set myself up for a one-man summit to get all the ideas together, develop some priorities, and put together a roadmap for this project. I pretty much sequestered myself all weekend, with a brief exception Friday afternoon for a haircut and some grocery shopping. The result was pretty impressive. All the little details – which I'd put off thinking about until I could really devote some time to them – started coming together. There are a lot of things I'm really excited about, and a lot of things that will take some hard work, and fortunately some of those things overlap or otherwise I'd just sit here, staring at the keyboard in terror and twitching nervously.

I started out by collecting all the ideas. A few were written out on paper from last year's redesign – ideas that Mom came up with, or other things I'd been thinking about. A few I had e-mailed to myself, or written on the back of a napkin and dropped in a file folder. A couple dozen I had posted to an ongoing list of ideas in a Basecamp project I'd been maintaining just for the purpose. Some were sketches of possible layouts.

I got all the ideas in front of me, then went through, one by one, writing each on an index card, and assigning a priority and a difficulty level. This will allow me to see which items need to be handled next, and each time I sit down to work I can decide if I want to tackle something difficult or if I don't have the time or brain power to do more than a few simple fixes. The index cards are sort of a low tech approach, but very valuable in that they're infinitely easy to rearrange, and you can spread them all out in front of you at once if you need to see the whole picture.

The drawback is, they don't "associate" well with one another. So many items depend on the completion of other items, or affect others because decisions made in one area will carry over into another. I had to go with a little written list on the corner of each card, detailing the main dependencies. I wish I could've stapled them together or attached them in some way, but that would really break the rearrange/spread out benefit. This way, it's a quick checklist to make sure I don't forget about some related items.

With well over 50 note cards, I quickly realized that I needed to prioritize again. (More than half of them fell into priority #2 out of 10 on the first pass.) I spread them all out in front of me, and went through again, setting a more reasonable hierarchy, then divided the cards up into a handful of sub-projects which revolve around related areas of the site. (All the design items in one category, for example, and functionality items in another.) This way, I can approach them as discrete projects and narrow my focus a little, but the priorities and related lists will keep me apprised of additional considerations that I need to keep in mind.

(Interestingly, last week as I was preparing for this undertaking, I came across the mention of David Allen's Getting Things Done on one of the sites I read. Halfway through the process on Friday, I took a look at an overview of the GTD approach, and found that I'd spontaneously recreated most of it myself. It boils down to: divide activities into smaller lists that are easier to manage, prioritize and re-prioritize frequently, focus on one thing at a time, defer decisions that can be made later. The David Allen site and the book make it feel a little cultish, but this overview is the one I read, and makes it feel more approachable. I'm not drinking any Kool-Aid, but I did appreciate the benefit of my methodical approach to organization, and I'm glad to see that some suit with a self-help book isn't offering anything that much better than what I came up with on my own.)

By this point, I'd become pretty familiar with all the major items on the index cards, so I was able to identify some overall questions that will be guiding the process: big concepts that need to be hashed out early so I know how to approach the little things. This reminds me of something that the geniuses at 37signals talked about at the "Building Basecamp" seminar I went to last summer. They refer to these as "project mantras," and not in a new-age way – just that you develop some very general guidelines so that when small, detailed questions come up, you can test them against these core beliefs to find the answer that best fits the project.

A lot of the onebee reorganization (at least on the public side of the site) has to do with how items are presented to the reader. I'm trying to streamline and simplify that, so that it's clear where to go for new things, and it's clear how to find things you're interested in. So far, my mantras list (which will grow, I'm sure), consists of:

1. You can always find it where you first saw it.

2. To the reader, each thing is just a thing.

1 means that if you're looking for a column that you first saw in April of 2005, then you'll be able to find it in the April 2005 archives, sandwiched right between the same two items it was between when you first encountered it.

2 means that – as much as I love to pontificate on the difference between a "column" and a "log entry" – to you, they're both just another "thing" on onebee. They show up, they have a couple of paragraphs of gently amusing drivel, you read them and you're done. They have minor differences in their behavior (columns can be written ahead of time and scheduled for future publication, while entries go live as soon as they're written; columns have an introductory blurb on the homepage), but those distinctions mean a lot more to me than they do to you. I should define them by their similarities, not their differences – so in the future they won't be segregated or otherwise treated in such a way as to enhance the distinction. I'll treat them more like peas in a pod.

The next thing I did was to input all the index cards into a Basecamp project. I really adore Basecamp – as I've said many times before – but I'm not yet sure whether it's the best tool for managing this particular project. I had to put some square pegs in round holes. The benefit will be that I can leave the note cards at home and access all the specifics via Basecamp anywhere I go, but I'm not sure if the flexibility will live up to the note card model, so I may have to abandon Basecamp, and just manage the project offline. I certainly like the idea of opening up a page and typing in a few comments as a progress report, and having them automatically stored for my future reference. We'll just have to see how it goes.

In the meantime, while the project is being managed through Basecamp, there's an RSS feed available of everything I do there. If you have a web browser or RSS reader capable of displaying an RSS 2.0 feed, you're welcome to subscribe to the following feed. (More on what RSS is.)

(I ask that you don't input the feed into an online reader like Bloglines, because it might be picked up by search engines and other things I don't want. If I find that the RSS feed is being used improperly, I'll have to take it down.)

Here's the link.

Most of the items are sketched out in sloppy shorthand that makes more sense to me, but if you're interested, you're more than welcome to take a peek. Of course, there will also be more periodic updates here at onebee of how the process is going.

So far, I've already tackled the first major challenge, changing the layout of the site so that the beeline (the little dotted line which divides the page vertically) will appear all the way down the page, no matter how short either column is. (Previously, it was tied to the length of the right column, which was a dumb move on my part and required me to pad the right column if it were displaying really short content.) So, that's a score! The new design won't appear on onebee until all the changes are finished, since the overhaul is being completed on a separate development copy of the site. But since, at least for now, I'm trying to do everything live with the same databases and includes, you'll note that the copyright information has moved from the bottom of the page to the bottom of the left column. Smells like progress!

© onebee · all rights reserved