I was at the IdeaCity conference this past week, Michelle was there as an invited speaker and gave the first public presentation on this research. It's a great talk: http://www.ideacityonline.com/...
From the wikipedia article: "Gas dynamic cold spray (GDCS) is a coating deposition method developed in the mid-1980s in the Soviet Union."
Nice. I've never heard of this technology. For a while I was working for a small startup doing some mechanical design and was doing research into exotic fabrication technologies. I came across some truly great technologies -- hydroforming, which uses hydrolic pressure to form metal into deep draw forms and Explosive Forming which, yes, involves using explosive to shape metals were two of my favourites.
I know the team member who first suggested this research. As a kid, he was obsessed with spinning tops, bicycle wheels and everything else he could find that spins really fast. Looks like that passion of his spun out of control as he grew older!
I saw him speak several years ago. He came across as one of the most humble, thoughtful, engaged, humorous people I've ever met.
Great cover for a Bond villain.
I'd like to continue growing it out and developing it. I definitely need to transition it to, at the minimum, a template based system, if not a full CMS. Hand-coding the html/css is a good way to understand the subtles of the box model, responsive design, etc, but it doesn't scale.
L
I just finished a hand-coded revision to my personal website. It's small and modest and after going at it for about a week, it was clearly time to switch over to a CMS.
This is the fourth one I've come across that comes highly recommended (see also, Stacey, Bootstrap, and, obviously, Wordpress). What are the differentiators? How do you decide?
There's a nice New Yorker podcast from a couple of years ago that discusses what went into picking the name: http://www.newyorker.com/online/2011/10/03/111003on_audio_colapinto .
It was done by Lexicon Branding, who actually write code to break up words into phonems and then remix those sounds into new words. The program spits out lists of candidates that are then vetted by the linguists at Lexicon. I found it a really interesting discussion.
Since these techniques rely on bio luminescence, can the light generated from the neural activity travel to and stimulate the retinal cells?
Can the animal... see itself thinking?
A totally different approach, if you like programming is processing. If you're not a serious coder, it's got a light-weight IDE and is specifically designed to be a gateway drug for artists/designers to the world of computational design. If you're already comfortable with Java, you can import the processing core into eclipse and really go hammer and tongs at it. It's got a ton of libraries, several of which are specifically devoted to creating STLs, meshes, NURBs and exporting that geometry out to STL, dxf, dwg, etc. Take a look at toxilibs for some particularly cool functionality with meshes and physics libraries. There is also a good online forum.
Not familiar with computational design? Check out some of the examples on the processing gallery.
If you are planning on doing a lot of complex curved work, you might want to look at Rhino -- it's a highly capable surface modeler. I often describe it as "like autoCAD but inherently 3d and without the suck." The basics can be learned very quickly and you can do relatively sophisticated work with it shortly there after. It's actually kind of a pleasure to work with. Version 5 is just out and I haven't seen it, but I can't imagine it makes a huge dent in some of the UI and paradigm limitations of that package, though. The big, and ultimately frustrating limitation of Rhino is that it's a surface modeller. It's great for sketching ideas out and exploring potential and complex geometries, but if you want to 3d print something, you need to make sure you have a water-tight model, which can be a pain. Making changes to models after the fact can also be a pain.
If you are planning to send drawings out to contractors for fabrication work, or ever want to work in an architecture office, autoCAD is (god help you) still a viable route to do that sort of work. It's industry standard, the only problem is that it's paradigm for drawing is one that was bad even by the standard of R&D prototypes done in the 60s -- it's basically the equivalent of an electronic drawing board. It brings some level of intelligence to the problem of drawing, but it fundamentally doesn't understand that two orthogonal views of the same object should be linked.
Solidwork is a great option if you're willing to put in a lot of time to understand it's paradigm: everything is about defining geometrical constraints and relationships. It's all number and sketch driven. The joy of it is that if you set things up to be reconfigurable, you can change tweak fundamental elements of your model. The horror of it is when you need to make a change that you didn't anticipate, the complex network of relationships that you've created can totally break. This is often not actually as bad as it sounds -- the relationship networks can be repaired and there are best practices and design patterns for setting up assemblies and parts, but there's definitely a bit of a feeling of trepidation on opening a complex solidworks model and thinking about making a change to it after weeks or months of work. After more than a year and a half of working with this package, I went out and bought one of those mega Bible books on it, and it's been very rewarding to page through. As I said, a great option *if* you want to invest the time in learning a deep deep tool. The other thing to understand is that Solidworks is a solid-modeler: it inherently understands that the world is composed of closed, water-tight objects. It's associative and parametric, which are technical terms related to the fact that you can build associations via sketch geometry and drive dimensions and other elements via numbers (extending all the way to creating a link with a spreadsheet, if you want). It's also fairly easy to get drawings out of it, if you ever do want to communicate with a fabricator or someone else: it'll do the projections for you. It's definitely not as aggressively NURBSy as Rhino (NURBs are the name for the splines that construct curvy surfaces) but you can do curved and organic shapes with it. Just be prepared to go down another rabbit hole -- there's an entire solidworks bible devoted just to surfacing techniques.
I know some people who prep laser cutting files in either Illustrator or InkScape. Personally, the though of doing serious cad work in Illustrator makes my teeth hurt. And I'm a proficient Illustrator user.
I've been lucky to generally have access to commercial grade cad packages. My very limited experience with free and open-source CAD suggests it's skill a bit of a ghetto. This may have changed in the past year or so -- if people can recommend a genuinely good, O.S. package, by all means jump into it. But if it hasn't, I would suggest that learning any CAD package, even the relatively user-friendly Rhino, is a pretty big
Slashdot Mirror
I was at the IdeaCity conference this past week, Michelle was there as an invited speaker and gave the first public presentation on this research. It's a great talk: http://www.ideacityonline.com/...
From the wikipedia article: "Gas dynamic cold spray (GDCS) is a coating deposition method developed in the mid-1980s in the Soviet Union." Nice. I've never heard of this technology. For a while I was working for a small startup doing some mechanical design and was doing research into exotic fabrication technologies. I came across some truly great technologies -- hydroforming, which uses hydrolic pressure to form metal into deep draw forms and Explosive Forming which, yes, involves using explosive to shape metals were two of my favourites.
I know the team member who first suggested this research. As a kid, he was obsessed with spinning tops, bicycle wheels and everything else he could find that spins really fast. Looks like that passion of his spun out of control as he grew older!
I could you could say it... spun out of control.
I saw him speak several years ago. He came across as one of the most humble, thoughtful, engaged, humorous people I've ever met. Great cover for a Bond villain.
I'd like to continue growing it out and developing it. I definitely need to transition it to, at the minimum, a template based system, if not a full CMS. Hand-coding the html/css is a good way to understand the subtles of the box model, responsive design, etc, but it doesn't scale. L
Seems to be PHP.
I just finished a hand-coded revision to my personal website. It's small and modest and after going at it for about a week, it was clearly time to switch over to a CMS.
This is the fourth one I've come across that comes highly recommended (see also, Stacey, Bootstrap, and, obviously, Wordpress). What are the differentiators? How do you decide?
Are the orbital dynamics harder if you orbit Europa?
Using new techniques the time for news event to slashdot front page has been decreased to four days.
There's a nice New Yorker podcast from a couple of years ago that discusses what went into picking the name: http://www.newyorker.com/online/2011/10/03/111003on_audio_colapinto . It was done by Lexicon Branding, who actually write code to break up words into phonems and then remix those sounds into new words. The program spits out lists of candidates that are then vetted by the linguists at Lexicon. I found it a really interesting discussion.
Probably more likely: "I wish I could figure out how to eat/fuck that thing over there.."
Since these techniques rely on bio luminescence, can the light generated from the neural activity travel to and stimulate the retinal cells? Can the animal... see itself thinking?
Actually also being actively ported to other languages, including javascript, ruby and python.
Not familiar with computational design? Check out some of the examples on the processing gallery.
If you are planning on doing a lot of complex curved work, you might want to look at Rhino -- it's a highly capable surface modeler. I often describe it as "like autoCAD but inherently 3d and without the suck." The basics can be learned very quickly and you can do relatively sophisticated work with it shortly there after. It's actually kind of a pleasure to work with. Version 5 is just out and I haven't seen it, but I can't imagine it makes a huge dent in some of the UI and paradigm limitations of that package, though. The big, and ultimately frustrating limitation of Rhino is that it's a surface modeller. It's great for sketching ideas out and exploring potential and complex geometries, but if you want to 3d print something, you need to make sure you have a water-tight model, which can be a pain. Making changes to models after the fact can also be a pain.
If you are planning to send drawings out to contractors for fabrication work, or ever want to work in an architecture office, autoCAD is (god help you) still a viable route to do that sort of work. It's industry standard, the only problem is that it's paradigm for drawing is one that was bad even by the standard of R&D prototypes done in the 60s -- it's basically the equivalent of an electronic drawing board. It brings some level of intelligence to the problem of drawing, but it fundamentally doesn't understand that two orthogonal views of the same object should be linked.
Solidwork is a great option if you're willing to put in a lot of time to understand it's paradigm: everything is about defining geometrical constraints and relationships. It's all number and sketch driven. The joy of it is that if you set things up to be reconfigurable, you can change tweak fundamental elements of your model. The horror of it is when you need to make a change that you didn't anticipate, the complex network of relationships that you've created can totally break. This is often not actually as bad as it sounds -- the relationship networks can be repaired and there are best practices and design patterns for setting up assemblies and parts, but there's definitely a bit of a feeling of trepidation on opening a complex solidworks model and thinking about making a change to it after weeks or months of work. After more than a year and a half of working with this package, I went out and bought one of those mega Bible books on it, and it's been very rewarding to page through. As I said, a great option *if* you want to invest the time in learning a deep deep tool. The other thing to understand is that Solidworks is a solid-modeler: it inherently understands that the world is composed of closed, water-tight objects. It's associative and parametric, which are technical terms related to the fact that you can build associations via sketch geometry and drive dimensions and other elements via numbers (extending all the way to creating a link with a spreadsheet, if you want). It's also fairly easy to get drawings out of it, if you ever do want to communicate with a fabricator or someone else: it'll do the projections for you. It's definitely not as aggressively NURBSy as Rhino (NURBs are the name for the splines that construct curvy surfaces) but you can do curved and organic shapes with it. Just be prepared to go down another rabbit hole -- there's an entire solidworks bible devoted just to surfacing techniques.
I know some people who prep laser cutting files in either Illustrator or InkScape. Personally, the though of doing serious cad work in Illustrator makes my teeth hurt. And I'm a proficient Illustrator user.
I've been lucky to generally have access to commercial grade cad packages. My very limited experience with free and open-source CAD suggests it's skill a bit of a ghetto. This may have changed in the past year or so -- if people can recommend a genuinely good, O.S. package, by all means jump into it. But if it hasn't, I would suggest that learning any CAD package, even the relatively user-friendly Rhino, is a pretty big