I recently had the pleasure of visiting the Edward Gorey House. I was particularly pleased to see some of his rough work. Before he got going with the pen and ink, Gorey would do a basic forms layout on paper. What you see in his initial scribbling is the genesis of form. This becomes more impressive as some unfinished works showed the finished and the rough there all in one.

I think that anybody who touches ink to paper thinks about layout and composition. You have a little spark of an idea and you start fleshing it out. And then you realize you have a space to reckon with and you need to do something with that space.¹ But you don’t necessarily want to lard it up; well, maybe you do if you’re Martin Handford creating a Where’s Waldo? So you think about what belongs and what doesn’t. What creates harmony, what furthers discord.

At work in the Gorey sketches are the slow ruminations of a planner. Not all people need to work this way. Frank Gehry’s doodles don’t indicate much of a process to me at all; I’m sure it’s there, but I have a hard time differentiating it from chaos. Process is understandable, chaos isn’t.

Not all draft work requires significant planning. Christopher Bowns spells out his preference panel interface design (via df.) in a series of steps. The process, consciously articulated here, is one of iteration. What Christopher Bowns describes is a series of discovering deficiencies and attempting solutions. It’s a little meandering, but I think it’s generally indicative of how people work. Computers support this approach much better than the relatively unforgiving ink.²

If you’re very familiar with the problem domain you probably don’t roll out ideas that have known deficiencies–at least not without good reason. Lucky for us Christopher Bowns took screenshots along the way. Articulating these deficiencies through example is much different (and more accessible) than something like the Yahoo! Design Pattern Library. The Yahoo! library is great, but it starts from the point of success and appropriate usage. Usability is inextricably linked with the unusable. In other words, you often learn more about what works (and why) from what doesn’t.

Notes

You can be much more daring with Photoshop than you can with oils. Screw up? Ctrl+z. But because you can be much more daring, you’re not really as daring. #

Gorey, always particular (he hand lettered his manuscripts), was choosey about the forms of his books. Web-pages and dialog boxes don’t come in 8 1/2″ x 11″. But UIs have their own set of conventions and constraints. #

Adobe recently announced that Google and Yahoo! (Where’s Microsoft?) will now be able to crawl Flash.  From what I can tell this announcement was met with some enthusiasm and some mehs.

Jon Gruber offers the following view:

It’s completely closed and opaque. Adobe is only providing the magic recipe to Google and Yahoo; all other search engines remain locked out.

Adobe to their credit say:

We are initially working with Google and Yahoo! to significantly improve search of this rich content on the Web, and we intend to broaden the availability of this capability to benefit all content publishers, developers and end users.

In general I’m a fan of the “develop it, get it out the door, fix it, and get it out the door again” philosophy. But I hope that by “broaden the availability of this capability to benefit all” it means the rest of us dopes get it in short order too.

According to the natural language processing blog:

Computational Linguistics is open access.
Just officially announced. Minor details: as of Mar 2009 (first issue next year), there will be no print version (electronic only) and will be open access.

Cheers to the folks at the ACL and The MIT Press (my old employer).

I’ve been running across a lot of squirrels on the internet lately. 

This is the Conversation Rate squirrel.  As if you couldn’t tell from the bling.  I really don’t know what to make of a squirrel that can be hired for $50K a month.

On the more serious side, the squirrel fish is not a squirrel (er, fish?) to be messed with.

This squirrel (er, fish?) can crack some nuts.  In the past I’ve found that Safari’s JavaScript implementation leaves IE (decent with innerHTML, slow as molasses in January with DOM manipulation) and Mozilla’s SpiderMonkey in the dust.  To see the WebKit team beat their own benchmarks soundly is really quite a feat.

According to this Readability Test, the website you are now reading is more challenging that Reader’s Digest (I do love their jokes).  However I am still bested by The Cat Who Dropped a Bombshell.

Play hard, accept defeat when it comes-that’s my moral.

Or at least so says Mike Davidson:

This sounds overly simplistic, but I really do think Apple just split the mobile world into two choices: settle for a free phone or buy an iPhone. There just aren’t many reasons to do anything else.

While this does sound overly simplistic, most of the best-selling phone section at Amazon can be had for less than a penny (including some pretty nice Nokias).

Judah Phillips points to a trend of ISP employing packet sniffers on user traffic:

Site owners collect data about what you do on ONE site (or a portfolio of their sites). ISP’s collect data about what you do on EVERY site you visit. As I understand it, some of these companies create an anonymous profile of your surfing activity by assigning a unique key to your browser.

If this technology is being widely employed by ISPs, then there’s a whole additional layer of people observing our behavior.

In addition to highlighting some interesting possibilities regarding this data capture, he proposes some interesting options for consumers:

• Move to an obvious “opt-in” model with full disclosure. Tracking via “deep packet inspection” should be an all opt-in model. If you want anonymous data from your browser collected so that you can be behaviorally targeted, then you should opt-in to be. Right now, it’s seems to be all opt-out. You probably don’t know if it’s being done to you. It’s buried in fine print you’ve probably never read. Is that your fault you didn’t read the fine print? Yeah, but the point is it shouldn’t be buried in the fine print…
• Provide me with access to the data collected. If I opt-in, I should be able to see the data collected from my browser. It’s very simple. I demand to see what you are collecting about my browser. If you are building a profile, then I demand to see the data collected in the profile. If it’s all anonymous, then explain how it is in detail, and then follow rule #1.
• Enable me to edit or prevent the data from being collected. If I opt-in, I want to be able to edit or prevent certain types of data from being collected. If you’re tracking my browser, alert me before the data is transmitted, so I can decide if I want to share it. If a profile is built, I want to be able to edit it!
• Let me opt-out at any time EASILY. If I’ve opted in, and I’m unhappy with the service, allow me to opt-out simply. Having to set an opt-out cookie on my browser is absolutely and completely absurd. I want to be able to fully opt-out at the ISP level, just once forever, not at the browser level every time cookies are deleted. Make it easy and permanent, not easily deletable.
• Disclose who you sell my data too. Like online list rentals, the next step in all this ISP profiling is selling the data to third-parties. Let me know what you’re doing with my data-before you do it- so I can opt out or prevent it from being sold to parties to which I don’t want it being sold.

I think all of these are solid ideas, but they point to a larger picture. ISPs are important because they can monitor all non-encrypted traffic across your total browsing experience. But do the processes of data collection and the rights of data holders change just because of their power? Shouldn’t the same standards hold across the board? I think the five points he listed above would be a good starting point for a lot of discussion concerning online data capture.

Note: If you’re resistant to the idea of packet sniffing, you can: run an SSH tunnel to a trusted proxy, download Tor, or pay for a service like Anonymizer.

For better or worse Django doesn’t have any real blogging software. The mailing list is peppered by green practitioners asking “what’s the best django blogging software.” To which one of the old dogs on the list will lift it’s head, look left and right, reply is that it’s “trivial in Django to write your own” and sink back down to the floor.

The lack of a first class blog unit has become a bit of a problem though. There’s an empty spot in the trophy case. The question will invariably pop into people’s heads, “So do I use Django for my website or Wordpress?” While a perfectly fine question, it belies the fact that your comparing frameworks and an application and wondering which one is better. Are there ‘Wordpress developers’? And if so, do you really want to be one?

So it came with some excitement to learn that Django now has a fairly advanced blogging app, byteflow. Among the selling points:

“It has very clean codebase and developers, which are struggling to keep it so.”

Heh. That’s a dig at someone, but I’m not sure exactly who. Anyway, struggle on comrades.

My question is that with a contingent of great hackers why hasn’t anybody released a Django based blogging platform up until now. One reason is that by making it trivial to write a blog, nobody wanted to do more than the trivial work required to get it going. And great hackers don’t want to release trivial code. Would Beethoven want to release that jingle he wrote for AutoZone?

Perhaps a deeper reason is that blog/cms/site-swallowing software is an unruly beast with complex data flows resulting in spaghetti code. Exactly the kind of molasses you’d rather not swim in.

The reason that I care about this issue, is that there’s really much more you can do with Python and Django with ease than you can with Wordpress and PHP. The Wordpress page definitely looks a lot sexier than the Byteflow page (as do their exemplars). But beware.

Oh, if Python installs (setup.py, cough) went easier and mod_python didn’t require apache fiddling (more difficult in a shared hosting setting), we’d probably have universal health care coverage by now.

Everyday you have thousands of people installing Wordpress, now maybe you can staunch the flow and get them on the Python/Django path.

MUTO a wall-painted animation by BLU from blu on Vimeo.

Google recently announced that they were retiring support for OAI-PMH in Sitemaps:

We’ve found that the information we gain from our support of OAI-PMH is disproportional to the amount of resources required to support it. Fewer than 200 sites are using OAI-PMH for Google Sitemaps at the moment.

Amen. As someone who once implemented an OAI-PMH service I can say that the web does it what it needs to do quite well without overblown standards. Committee-based ham-fisted over-architectures have their place, the dustbin of history.