Over at macresearch.org, they've posted a nice review of BibDesk. One quote: "If you use Latex search no more, this is a killer app." Kudos to Adam and Christiaan.

I see an increasing number of Macs at research meetings I go to, and I usually can't help but peek on their dock to look for that yellow folder. I usually don't ask people if they've heard of it, but I have thought of putting a big sticker with the BibDesk icon on my PowerBook, just to see if anyone recognizes it.

Even after moving web-based mail to a separate app, I still usually have at least two browsers open at all times - Camino when something doesn't load in Safari, Firefox for tails (a microformats helper), and Safari, which I'm used to, has my bookmarks, and auto-fills forms.

Safari freezes and crashes pretty often for me, but I can't easily switch away from it because it has my history, both in actual saved history and collections of bookmarks which always seem to get dumped into an 'Import' folder when you move between browsers.

Browser history doesn't seem like a lot of data. It seems like a natural choice to keep in a central place, just like the OS X Address Book. However, I don't know of any effort that takes the user's point of view and lets us keep a separate history list, so we don't have to lose data every time we want to try out a new browser.

So - anyone want to step to the plate? Hopefully all it would take is a good API and one browser that works with it - if one WebKit-based browser (OmniWeb) and one Gecko-based browser (Camino) adopted this, it would take something big to make me use Safari again. If a browser imported other browsers' history and made it easy to save & search that history separately, I couldn't think of a reason not to try it out, and see what other features I might like.

From BLDGBLOG, some pictures and romantic musings about everyone's favorite kind of computer - big, expensive, finicky and shared: the Supercomputer.

Be sure to click through to see the picture of the MareNostrum system filling a glass box in a chapel in Barcelona. There are more great photos on photographer Simon Norfolk's site.

I wish there was more well-done photography of these systems - not all are nearly as photogenic overall as MareNostrum, but there are plenty of great details, and so many are already lost to history. My vote for favorite detail is the Cray 2 coolant waterfall:

(It's the blue glowing thing in the top center - from SpikyNorman.net)

Here are some interesting quotes from the BLDGBLOG post:

These computers, Norfolk continues, 'are omniscient and omnipresent and these are not qualities in which we find a simulacrum of ourselves - these are qualities that describe the Divine. The problem is not that these computers might one day resemble humans; it is that they already resemble gods.'

I would say Golem.

So if I were forced to take issue with the existence of these machines, it would not be because of their use in modeling new nuclear warheads - as Norfolk makes clear they do - but in something far more secondary, even faintly absurd: what I'd call the lack of a supercomputer poetics, or a more imaginative role for these machines to play in our literary and even religious lives. Oracular, Delphic, radically non-secular: they are either all or none of the above.

If we'd all agree to think clearly about climate science, there's quite a bit of Oracular prediction using supercomputers going on there. I'd love to see high-performance computing become more a part of the culture, as computer security undoubtedly has. I think such stories are unlikely to be told visually, though - it's hard to imagine how it'd be done without being either overdramatized and silly or dreadfully boring. After all, how do you make a movie out of editing text and waiting?

Now that I don't have regular access to cable TV, the sensation I felt when I noticed the announcement of NHL playoff video available on iTunes was that of Steve Jobs signing my name on a check for ten dollars and winking.

Each video is ~20 minutes of highlights, and the video quality, while suffering from the same problems I noticed previously, is pretty good.

Check out the previews here: Eastern Conference Finals and Western Conference Finals.

The old debate over donation-ware and pricing has popped up again, in response to Peter Maurer's Textpander, since expanded into TextExpander. The debate rages in comments on TUAW and a bit more on HawkWings. Both TUAW and the MacUser blog choose to call it "freeware", which is technically true, but blurs the lines in an important way.

Peter's stuff is released with no price, just a polite suggestion over on the side: "If you decide to use Witch regularly, please do consider making a donation." He doesn't say how much. In my experience, even getting a $5 donation from a happy user is really a boost. Free coffee in the morning is awfully nice - add in a comment saying what you like about my app, and it makes me want to impress you with something new.

Writing software, even small apps, is not easy. A nice polished donationware app with all the trimmings represents a serious investment of time. When I was working on BibDesk, I would hack on it for 4-5 hours at a time every couple of days. As I've discovered over the past year or so, it's not something you can really do intermittently.

I bet that many of Peter's Textpander users have real jobs. We grad students call you "real people." If a real person goes for one week at grande instead of venti, or ate ramen for a night instead of going to Outback, the change is more than enough to make the Mac software world a better place.

Why? Just think of all the bad apps you have to pay for from big companies who don't really get it, and then think whether a few bucks is worth it to support software that actually works the way you like. Your dollar might be the one that seals the decision for the next Gus Mueller, Fraser Speirs, Mike Piatek-Jimenez, Daniel Jalkut or Steve Gehrman to start developing full-time.

I'm glad to hear that Peter made some money. You might think that $30 is too much for TextExpander, but don't fault him for selling his app - he doesn't owe us anything, not after the huge gift of his free time and talent he already donated to us.

XCode 2.3 is out, so why not spend a nice May evening reading some Dev Tools release notes?

There are some excited posts about Dedicated Network Builds from people who work at Apple, but not having a cluster of build machines myself, I'm not so worked up. (Unless sourceforge gets on that for their build farm.)

Then I saw this one line buried a little past halfway through:

• Nibtool
• Added functionality for exporting and importing properties as a plist Nibtool can extract properties into a plist format that can be edited and then reimported into the nib using the new --export (-e) and --import (-i) flags.

If I read this right, it means the door is open for easy automated access to bindings and maybe a way to make editing bindings easier. I know visual programming has its benefits, but bindings are just so opaque - you see one object's binding at a time, and making mass changes to bindings is just impossible - until now, I hope.

I'm downloading it now.

I came back from a weekend camping trip to find that my post about WebMail.app got a lot of attention: TUAW, Hawk Wings, and MacUser, for starters.

Some people got what I was trying to do with it, and some didn't. The idea got passed around a bit, so a few commenters probably didn't read my original post. That's not a big deal, but I did want to clear up a few points that weren't obvious from what I did write.

I don't hate Apple Mail.

I didn't say anything about disliking or replacing Apple Mail. Mail is my primary mail client, and there's plenty about it that I like. I just use both, since I like GMail better for viewing mailing lists.

WebMail.app was not a mission statement for the web.

I like rich local clients. I don't think that apps like WebMail are the future of desktop apps.

I don't think that cloning WebMail.app for any given website/web app will necessarily be a good idea. Sometimes it will, but if you could benefit from system-specific GUI features or services like Spotlight, CoreImage, AddressBook integration, control over data backup, etc, then just another browser window won't help you. For instance - I can't think of any reason to move flickr browsing to a mini-browser like WebMail. I don't need to avoid distraction when I'm using flickr - I'm already distracted.

Also, I'm not qualified to expound on the future of the web, so let's not get ahead of ourselves there.

I wanted it to be a bad browser.

This was the part that got people who didn't read what I wrote in the original post about distraction. Yes, you could get almost the same effect by using real browsers with their toolbars off, but the whole point was to make it hard to distract yourself on the web.

There's a fine line where if you add any more features, you're re-implementing a real browser, and that's counterproductive. We can disagree on the position of that line, but I feel I hit it pretty close. For instance, if you want tabs, I think you should probably just use a real browser. And for my purposes, not being able to open links in other browsers is an important feature.

It wasn't a product announcement.

I wrote it to avoid distraction - I'm not going to be supporting it, adding features, nor will I even work very hard to get an icon. I suggest using the icon mentioned in the Hawk Wings post about it. That's probably what I'll do if I ever get around to it.

Some people suggested features that wouldn't make it more distracting, including handling other web mail sites, doing the right thing with uploads for attachments, importing contacts, a better login interface (actually, I don't get why you'd care about that), and using newer WebKit builds for better compatibility. I don't plan to do any of these, but you're welcome to. If anyone wants to use the name and take it in that direction, I'm OK with that - let me know.

I do use it, though, so if it ever breaks on a new system and there's no good replacement, I'll get it working again.

I have heard from at least one person who's building a similar app and has done some more interesting things with it, so when I find out more about that, I'll pass along a link. (But I'm not sure if it will be free or open source.)

It was super easy.

Now that I've addressed most of the criticisms that I wanted to, I want to ask that if you like this app, please reserve your praise for the Web Kit team at Apple - it took me months to realize that I wanted this app, but only minutes to build it because of all their hard work. It really was basically the demo app, and not even the one at the end of the talk. Thanks!

Webmail is really a separate application. When I'm visiting GMail, I'm checking mail, not browsing the web. So what's so bad about using a browser for this?

Minor Gripes

If I keep a browser window running with GMail, now clicking on the Safari dock icon just brings that sucker up instead of creating a new empty window.

I love the GMail key shortcuts and Safari has an annoying tendency to get the key focus wrong so I have to click in the window somewhere after moving up out of a thread to get my shortcuts back.

Safari crashes, I forget to reload GMail, I miss important messages.

Even all that isn't so bad. For a while I ran Firefox with only GMail open. I could get a GMail notifier so I won't miss messages... but:

The Real Problem

The final straw is this: Every time I check mail, I'm diving right into the world's biggest time-sink. My email isn't usually a waste of my time, but all the windows I've left floating around, my bookmarks bar, or a quick Google search are. These are the things that eat up afternoons, and webmail is a gateway to that distraction.

Browsers have lots of features that I don't need to use GMail: bookmarks, back & forward buttons, a search field, page history, a location bar, and on and on.

My Easy Way Out - the Minimalist Specialized Browser

A while back, I wrote a separate web browser just for GMail.

All it does is load GMail in a nice big window and duck out of your way. No location bar. And no bookmarks.

It says: Go ahead and follow that link your friend (or bug tracker) sent you, but to check BoingBoing, you're going to have to go over to Safari. Maybe you'll decide to go back to work instead.

It's basically the WebKit demo, except that I tried to improve the key shortcut situation a bit, and it has a progress indicator.

I've been using it for a while now, and the only features of real browsers that I miss are pretty simple to add - text find, a refresh command. I just haven't needed them that much. Sadly, one feature I'd love to add to GMail, a key shortcut to "go to inbox", eludes me, since their Javascript is pretty obfuscated. Update - "go to inbox" already exists as the sequence "gi".

Meet WebMail.app

If you want to try out this idea without the hassle of writing those ten lines yourself, get a tarball here: Webmail-1.0.tgz and let me know what you think.

The source is in there, it's BSD licensed, and I'll happily accept patches that make it more useful for email, but remember that making it more useful for general browsing is kind of not the point.

Oh, and it lacks a real icon. Sorry.

Update much later: a new version that supports printing and attaching files is available here: Webmail+printing+attaching.zip

I visit 43Folders only about once a week now - I couldn't deal with the cognitive dissonance implied by procrastinating by visiting a site about how to be more efficient and avoid procrastination.

I was really glad yesterday to see Merlin's post about faking full screen mode with a trio of useful apps. I had tried Menushade before, and didn't really see the point - the menu bar doesn't bug me - but in combination with the other two that Merlin recommends, Spirited Away and BackDrop, his scheme is a really great way to turn your computer into a tool for focus and not diversion.

At first I thought I wanted a faster auto-hide than the default 60 seconds, but after a bit of time with the delay set to zero and then a less disconcerting 10 seconds, I decided that it was confusing to lose context that quickly, so the default's been working well for me. Also, it makes diversions painfully obvious when your work-related windows blink away and you're sitting staring at ESPN. It immediately begs the question: "You've already wasted one minute - is this that important?"

I was also reluctant to hide the dock, but I tried it, and now I'm never going back. I have SpamSieve and Growl to point out important emails (checking once an hour), and for the rest of the time, my numerous diversions are better off out of sight, out of mind.

Another tactic I've started using is a dedicated browser app for Gmail. Just because I need to look at my mail doesn't mean I need to spend some time checking hockey scores - so I built a separate browser that only loads Gmail, and that temptation is gone. I'll probably post more about that sometime later, but for now, I need to get back to work.

Dear Rady School, you may be dedicated to producing business leaders who can straddle both the business world and the world of science and technology, but please ditch the little animated guy on your homepage who told me that.

Also, please read this adaptive path essay on user-centered URL design, and revisit your URLs. Most of your URLs look like this: http://management.ucsd.edu/cms/showcontent.aspx?ContentID=163

I care about 30 of those characters, including dots, one slash and protocol. There are another 31 characters that never change and don't tell me anything about the page.

For the previous URL (#163), might I suggest: http://management.ucsd.edu/facilities

And for the next number up, #164 (apparently not up yet, but it should be the web syllabus for a class), might I suggest: http://management.ucsd.edu/courses/2006/summer/mgt-111

Thanks,

mike