While I've been much happier with Firefox than IE (both for browsing and for development), both browsers suck for hardcore multibrowsing (HMB), in which a user has many web pages open at once, and may open several links from any given page, in separate tabs or windows, keeping them open as separate threads of exploration or work. At my current screen resolution, both start scrolling the tab bar at 10 tabs. If you get much beyond 10 tabs, selecting a tab becomes annoying.

I wonder if any other browser (or plugin) has a better answer to this problem. A multi-row tab bar would be useful, but after more than 5 or so rows, your browsing area would start to get limited. The answer may lie in grouping tabs. If I'm on a Wikipedia or E2 binge, several tabs will probably be related to the same area of exploration, and could be expanded or collapsed somehow, based on which group had focus. Windows does this on the task bar, putting separate windows of the same application in a pull-up menu. I turn this off because it's annoying, but that may be because I have no control over what gets grouped: just because two windows are Firefox, doesn't mean they're related.

Which brings me to another feature I'd like to see. It's sort of a half-assed answer to the grouping problem: letting the user drag a tab from one window to another--wait, Firefox does allow this now, although the window is still in the original window as well, and I don't think it will transfer any "state", such as the form data I'm entering for this blog post.

Another feature that would be nice is some visualization of the relationships between tabs. I would like a tree view of what tab I opened each other tab from (even if I've already closed some of them), and I would also like to be able to tell what link on a given page the child tab came from.

So I'll let you know if I find anything with these features.

I came home from work yesterday and surfed the web for a couple hours. I had been fairly worn out, but after vegging for a while, I was relaxed. As I've said before, the web is my TV. It's something to do while I'm not doing anything. Something to distract me while I wait for the tension to work itself out.

Realizing that I was done unwinding, I stopped surfing and fired up Heroku. I think my problem with Rails so far has been that I wasn't really interested in what I was trying to build, so I didn't have much motivation to learn more Rails. So I've started on something more fun, and the difference has made a difference.

I finished my experiment with restricting my web surfing a couple weeks ago, but I still haven't posted a summary. For a few days, I felt really good, refreshed. I got a couple things done that I otherwise wouldn't have, like figuring out how to open the fluorescent light fixture in the kitchen ceiling. Eventually, feeling bored, I watched a bunch of DVDs, which probably should count as cheating. I think there's a happy medium, but I'm not sure exactly what it is. It may be to set a maximum number of hours per week when I can surf. Or it may be to try and recognize what surfing patterns are useful and what aren't. Or it may be to alternate periods of normal surfing being allowed, with periods of being "off the grid". Perhaps I'm just rationalizing why we needs the Precious.

One Web to rule them all
And, in the darkness, bind them

@justin_dz i saw the WebEx building. think i saw 2 actly. drvng thru silcnValy is like drving thru a museum of names from the web. whoh yhoo

(140 char lmt)

I took a walk today after work. I looked at the clouds some. There were pretty post-rain bluish clouds, and there were golden pre-sunset clouds shining through dark but sparse post-sunset clouds below. I like looking at clouds. I'm good at it.

I'm not so good at passing people on the street. Should I look at them or look away? Will they get mad if I look away too much? Should I say hello, even when I don't feel like it? What about when I do feel like it? If I do say hello, how far away should they be? I usually alternate between looking at them and looking away. I've heard that in the animal kingdom there is a specific meaning to looking down vs. to the side vs. above, but I'm not sure what that is.

Walking back across the bridge over the Potomac, I saw the birds again, in the tree on top of the support for the old train bridge. I thought they might be herons before, but now I'm pretty sure they're geese. They honk and fly like geese, but they're pretty dark. A girl was on the bridge taking pictures, maybe of the geese. I think I smiled and nodded, and looked down as I passed.

I e-mailed myself earlier to post this later. I'm not going to edit it much unless it really needs it. [edit: it needed it. I edited it for formatting a bit, and now I'm getting offline.]

===

first: the subject that came up when I started typing this was "agile development and memetics". I forget what I was thinking when I subjected that, but I think it had to do with an article warning against thinking you're "agile enough", and how that is an instance of the agile meme defending itself.

===

Justin twittered that a post on his work blog, "From the desktop to the web and back again," might be interesting to his personal blog readers. He was right. A couple issues which I'm interested to see how they play out in this area:

- platform independence: one of the benefits of the web is that, in theory, it doesn't matter what OS or browser a user is using. The amount of effort that goes into cross-browser compatibility shows that that's somewhat of a leaky abstraction, but for the most part your application isn't an
IE/Firefox/Safari/Opera app but a Web app. Which is good: lock-in sucks.

- Heterogeneous code: Will these hybrid web/desktop apps require, in addition to html, javascript, css, php, and sql (or insert your favorite server-side language/stack), yet another language for the desktop-side behavior. What I'd like to see is the difference being as simple as:

on(:server) do
  something
end

on(:browser) do
  something :else
end

on(:desktop) do
  your :thing
end
# not necessarily in ruby.

The no web surfing experiment has been going on for two evenings now. I'm gonna go ahead and call it.. Project Wipeout. Two things I've noticed so far: 1. more time, 2. web surfing takes energy.

Time: I've made progress on reading a couple programming books, which is something I never get around to. Incidentally, Concurrent Programming in Java at the same time interesting and puts me to sleep. I played with rails a bit. I'm not at the point of being able to do anything interesting yet, but I think I'm close.

Energy: It wouldn't be true to say I *have* more energy now, but I might once I've adjusted. The first night I couldn't get to sleep until at least 3:30. So I think this freed up some energy, despite going for a walk.

Web surfing also distracts me from the fact that it's time to go to bed.

I'd love to make this post better, but I have to go to work pretty soon.

I spend way too much time online.

But Dustin, I hear you say, in the new digital world, we are always online.

This is true, but I'm talking about something that for lack of a better term I'll call "surfing the web". My relationship to the web is what many people's relationships are to the TV. I come home from work, often tired, and plop down in front of it. I'll start reading a blog, news site, or RSS feed, or do a search, and in a few minutes, I've opened every link in a new tab, from which I follow every subsequent link.

A lot of what I read is informative and I click the links because I'm curious, so it's probably about as beneficial as channel surfing through things like Discovery, The Learning Channel, whatever channel they show that show with all the guns on. But the upshot (or downshot: I'm not sure if I'm using that word right, and I'm NOT going to google it (or better: ninjawords it)) is that I end up reading a bunch of stuff and not doing anyting. Lots of programming blogs and no programming. Lots of blogs in general and not doing anything blogworthy.

SO. For the next week, including the monday a week from today that I'm taking off, I'm going to try and limit my surfing to the half hour or so before work while I'm eating breakfast. This doesn't include all Web usage: I'm still going to do the occasional e-mail check; I'll hopefully get around to messing with Ruby on Rails using Heroku; I might blog some goings-on that I've got going on (what DO people do while they're not surfing the web, anyway?); I might even fix the fscking RSS feed. But no passive consumption.

I'll at least blog about it afterward. I'll catch you on the flipside.

I just read this HOWTO on productivity by Aaron Swartz. One suggestion I liked was to try to make dull tasks fun. Something I thought of recently was, when working with exceptionally bad source code (which I may or may not have written), is to pretend it was written by aliens. It only seems like hyperdimensional spaghetti because their brains are more advanced than ours. (Hopefully the code can be translated to become understandable by humans.) I haven't actually tried this.

I do think how fun work is, is overlooked way too often. I wonder if any studies have been done on the the quantity of fun as compared to software projects and their ability to stay on schedule, within budget, and produce quality code.

After using Google Reader for a few weeks, I realize that:

Google Reader > NewsHutch

This is evidenced by the fact that I actually continue to use GR. NewsHutch and Sage failed in that respect.

I also realize that I'm doing it wrong when it comes to RSS: Giving the full story in the RSS feed is better than giving a 250-char blurb (especially if your blurberizing software breaks when there are less-than or greater-than signs). If I were trying to monetize my blog by using the RSS feed as a teaser to drive people to the site so they can look at ads, it might be different (although I don't think that strategy works very well. On the other hand, I'm not one of the 6% of Internet users that actually clicks on ads (unless they're for web comics or t-shirts)).

In other news, I used to think I should add a draft feature to this, so I could put out posts that are longer than my attention span. I think I have a better idea now: posts with history. I can post something in its half-baked state, then edit it but the history of edits stays with the post, and it's public like a wiki article's history. What do you think?

Now I need to get enthusiastic about the PHP code it will take to actually do this. This is going to be hard.