It’s been a bit. But I can assure you we didn’t go on vacation. Quite the contrary.
Sam put on his white job-site helmet, rolled up the sleeves on his best gingham shirt, shimmied into his darkest pair of selvedge jeans and got to work architecting what will become our database. This will be where we store information that keeps track of the trails, steps, nodes and where the front-end requests that information from. Sam’s also working on the API to make them sing to each other in harmony. Our JSON structure will need to reflect the tree/node structure we’re currently working towards. Although we’ve been discussing the merits of other visualizations as well. Things that might lend themselves to the often chaotic nature of information gathering. We’re thinking that tree/node might be a little to straight forward.
(Side note: The vocabulary of what we’re doing is going to be important. Semantic clues are going to give the user a mental model of what’s going on. And the more clear the mental model, the less work they’ll have to do in order to start using the platform.)
I have been working on creating the Ux, front-end web development and learning d3.js, an extraordinary and powerful “data driven document” library that allows you create data visualizations like it ain’t no thang. Of course, that’s after you master it. At the current stage, I can confidently say that it is, in fact, a thang. But I’m getting there. And I’m enjoying it.
What we’re working towards is not the multi-view that we originally presented. That’s a long term goal. But for now we’ve condensed things down into a single view.
This will be easier to understand the first time someone encounters it. And it still invokes the functionality and utility that we originally set out to create.
For the final I wanted to model an existing object and then use some of the tools in Rhino to transform it. I was, obviously, drinking coffee when the idea to model a simple coffee cup dawned on me.
I was inspired to “glitch” the object by this rendering of a dresser that’s been floating around for a while.
A few iterations of the cup with different glitch motifs.
Views of the final version.
Shaded version of the final model.
Trailz (we’re taking suggestions for names…) is a project by myself and Sam Sadtler exploring better and more efficient collection and organization of the information you find and want to save on the internet.
(header image from: http://www.adventure-journal.com)
It’s about that time of the term: mid. That means critiques, check-ins, and getting shit done. So for this week’s crit we presented the first serious set of wireframes/Ux documents.
Up until this point Sam and I have been talking about the purpose of the project, what we’d like to accomplish, and developing the codebase – essentially trying to get a sense of the what the thing is and get the basic functionality working. We’ve been so focused on the technology and the idea that we haven’t stepped back to think much about what the user experience is. And that’s really what we’re trying to improve on: the user experience of finding and saving information on the internet.
So here’s our first draft…
This is the current state of your browser. Tabs. URLs. Images. Text. Video. Links to more of all of the above. Lots of information.
Some you want. Most you don’t.
These are your bookmarks. It’s a mess. A tangle of nested folders. Search is “meh”…
Most of the great stuff you’ve saved is lost in the pile of info.
Here’s your history. You’ve never looked at it. Unless you’re trying to get someone in trouble…
Trailz is a Chrome extension that helps keep track of an organize the information you find online. It’s goal is to be useful for both directed search and free exploration. When you create a trail a sidebar appears and saves your current location, which appears as a node. New nodes are appended to the end of the list. Clicking on a node takes you to that site.
When a site (or information on a site is saved) Trailz automatically generates and saves a list of relevant, searchable keywords. This makes finding trails, pages, or info easier later. And will make nesting folders unnecessary later.
Multi-valence nodes represent websites that you’ve saved multiple webpages within (e.g. www.site.com, www.site.com/about, www.site.com/imgs/coverart). Those nodes collapse and expand as needed. The filled-in node represents the page you’re currently on.
You can have multiple trails open at one time, across tabs.
If you don’t need to save an entire page, you can highlight text, select images or video and just save that information.
Whatever you save ends up on a card within your trail. A “trailhead” view is a digest view of all the site and information you’ve saved. Here you can add/remove keywords, review information, make connections, annotate, and share.
This “trailhead” view is the dashboard. It shows an overview of your trails and shows a bit of meta-data related to them.
This view is a bit more dynamic. At a glance, it shows your saved nodes in a tree structure. You can see where your research intent or approach changed.
This week I moved away from modeling from the ground up to focus on modeling the ground from up above. I’m excited to see what the capabilities and limitations of Rhino’s Heightfield tool are.
I’m interested in creating serving dishes (or molds for serving dishes) that are direct representations of the source terrain that the food being served on them comes from. For this first experiment I’m imagining serving ceviche or a thinly sliced fish on top of a rippling water surface.
I pulled a image of water from Google Image Search as the source. And after processing with Heighfield, turning the surface into a sold and cutting out the outline it looked like this.
This circular cutout is 5.1″ in diameter and about 1″ from base to the top the surface. I thought it might take a little too long on the 3D printer to be worth it, so I used the 4-axis mill. Just to get a prototype in my hands. Now that I have an object to study I can think about what additional modeling features I could only get with 3D printing.
Sam Sadtler and I continue our work on developing a smarter and more effective internet bookmarking system.
As of our first meeting, Sam and I divvied up the main components of the project. He’s taking the basic building blocks of the Chrome extension and (what we’ve been referring to as) the card structure. That’s just the combined information presented as a long-form node in the digest view of a trail. Things like URL, selected text and/or image, and user annotations. I’ve been working on implimenting a system to automatically capture relevent keywords from a bookmarked URL and create searchable tags.
I’m still wondering how I ended up with that task.
At this point we’re on track to get on track any day now. It only took me half a day to build a prototype site and get Alchemy parsing URLs and returning keywords. Now I need to link the two and figure out an extensible pipeline that passes those keywords into a searchable database of bookmarks. Plus tie-in or refactor Sam’s work to start to bring everything all together.
I’m looking forward to another interesting, productive week.
This week we’re designing and printing objects that “snap on” to other objects.
I like to have a pen handy. I’ll pull the classic pen-behind-the-ear move. Unless I’m wearing my glasses. In that case there’s just not room enough for the both of them behind my ear. So this week I designed and printed a pen holder that snaps on to the arm on my glasses for easy pen storage.
I started by modeling an approximation of the glasses arm. I measured the width and height using a caliper, but guesstimated the filleting that creates a pseudo-circular cross-section. The arm is slightly rounder towards the rear and maintains a more true rectangle shape closer to the front of the frame. I then copied this form, scaled it up, and subtracted the smaller form from the middle to give me the basic shape of my clamp.
Then I split the form to create two matching halves. Then I added a trapezoid that ran the length of one of the clamps and subtracted that from the matching rail. This would allow for a slide-on fit. Although later when I tried to reattach them around the frame, I realized that the tapered design wouldn’t allow for an easy slide. It worked, but could be improved. Also, I whipped up a simple pen holder to attach to the outside of the clamp.
A two-hour print job later…
Last week Sam and I met after a gentle nudge from Stefani. (Hi, Stefani. Thanks Stefani.)
We had a great two hour meeting over breakfast at a little Greek diner right in the neighborhood that Sam and I share. He had griddlecakes with syrup and crispy bacon. I had the breakfast special: two eggs (scrambled), home fries, toast (wheat) with jelly (grape; it’s all they have), and crispy bacon. We both had coffee. I take mine black. Sam likes his on the lighter side with a healthy addition of half and half. I got too many refills. He seemed like he had more self control.
We went back and forth for an hour or so outline ideas we were actively interested in pursuing and those that had long since disappeared into the smudged, greasy, folds of our notebooks. (Maybe his aren’t greasy. I can only speak for myself. I tend to eat popcorn when I’m thinking.) Some of the ideas we pitched to each other were the same as the initial ideas we outlined in class. Some we came up with on the spot. Some seemed emergent from our shared interests.
We were both circling one theme in particular – organization in digital “space”. My ideas were more head-on aimed at this, Sam was rounding the corner with an idea that he had been thinking about for a while. Something similar to an idea I’ve been harboring re: internet bookmarks.
At their core the ideas were asking the same questions: Can we make better use of our internet browsing? Can we make the bookmarks we save (and often never go back to) more useful? How can we better follow our own process of searching, finding, and discovery on the internet. How can we continue to explore the promises of hypertext that Vannevar Bush presented in his seminal 1945 Atlantic article “As We May Think”?
Our idea, codenamed Trailz [sic], is a browser extension that traces your browsing path and helps you seamlessly capture and organize relevant information for easy reference later.
Functionality will (hopefully) include:
- traceable search and link paths
- material extraction (photo, video, text)
- search by tag
- path sharing
- path editing
You’ve heard the rumors.
“The formula is top secret. They keep it locked in a radiation-proof bunker 60,000 ft below the surface of the middle of the Mojave desert.”
“No way. Everyone knows that they coded the recipe on a chip and implanted in the cranial cavity of an anonymous suburban house-wife. When they need it they send out high-frequency radio waves and activate her a la the Manchurian Candidate.”
“You guys are both wrong. Why would they keep the whole formula in one place? Duh. Each member of the board has memorized one part of the recipe. And once every decade they gather on a boat in international waters to initiate new board members, pass along the knowledge, and bet on endangered monkey fights.”
Who’s to say what is or isn’t true? All I’m saying is that the New York Times published a recipe back in 2014 that’s very similar to the infamous Coke. In class we took some liberties with that recipe to create our own un-cola version of “capitalism’s dirty water.” And this week was all about experimenting with flavor profiles and expanding our mind’s palate. Or our palate’s mind.
My secret ingredients? Pretzels and Twizzlers. Two matches made in heaven with Coke.
By every measure there was no way that this should have been fit for human consumption…but…it was decent! Not great. I wouldn’t serve this to anyone who wasn’t was nursing or pregnant. But it was drinkable, and even, dare I say? Enjoyable.