Forget About the Mythical Lone Inventor in the Garage | Hacker News Reader

Forget About the Mythical Lone Inventor in the Garage

slate.com

26 pointspolymathist16 comments

Well, like so many other things, it's field-dependent. :-)

If you're working in software, you don't even need a garage: a dorm room or studio apartment, a laptop, and a few thousand dollars to pay for web hosting, will fill your need for physical equipment. I think that one of the great achievements of the past few decades has been the invention of an entire field of technical endeavor in which the base costs are so low.

But not all innovation is in software, and it's good to get a reminder that most fields have substantial equipment materials and equipment costs. A lab doing electronics, biotechnology, energy research, and many other totally necessary things has much higher requirements.

When I was doing microfluidics work in grad school, I was doing some of the cheapest microfabrication research out there. All I needed was a continuous supply of photoresist ($800/L), some decent-quality silicon wafers ($200/pk of 20), access to a high-resolution transparency printer for masks ($150/print), silicone plastic for making devices ($50/lb), and a good fluorescence microscope ($50,000 or so). Oh, and access to a clean room with about $1,000,000 worth of equipment, but that was shared and I don't know what our "rent" for access was...

Edit: costs estimated from the Internet and 2yr-old memories, but I think they're pretty close.

What I find most exciting is that in the coming years, the technologies that right now only seem possible in big labs might experience a revolution– might acheive the ease and cheapness that software development has today. Some examples: Fab@Home for 3d printing http://www.fabathome.org/, Arduino for hardware hacking http://www.arduino.cc/, and BioBricks for genome hacking http://biobricks.org/.

This article really struck me as a fascinating example of garage-based genome hacking that turned into something really useful. http://www.popsci.com/scitech/article/2008-09/are-genome-hac...

Another advantage of a "lab" is interaction with other bright people in the field. It's a key benefit of YC; Jobs designed the pixar building to enhance this; many great software ideas came out of Bell Labs and Xerox PARC; and I've also experienced it myself at my old compsci dept.

The Internet has, if not replace, then extended that metaphor considerably.

Or at least it does when people with a common interest can reach out and find one another.

One of my huge frustrations with so-called "social media" is that they frequently make just this sort of interaction very difficult.

People who empathize similarly can reach out and like one another. However, people who can actually GSD need to wade through a ton of noise in search of signal. Even old Usenet did better (charters and cancelbots FTW).

This is one of the reasons why topically-organized sites (Reddit, HN, etc.) do far better for communications in my book.

I came across this post yesterday about using Ponoko's, relatively inexepensive, laser cutting service for micro-fluidics: http://nznano.blogspot.co.nz/2012/05/simple-laser-cut-millim...

Hey, that's really cool! Thanks for posting the link. I wonder what kind of resolution Ponoko can get, and what the pricing on the parts in the post were.

I don't have a ton of experience with acrylic microfluidics -- I mostly did soft lithography with PDMS -- but this could be a cool alternative to a institutional cutter for small shops/labs.

While I do agree with a lot of the points here, I wonder if perhaps recent advances in technology (particularly those that might be considered platforms for future innovations) are making the idealistic "Garage Inventor" more and more realistic. It probably still requires a big lab to create new technologies or new branches of science, but nowadays anyone with a laptop and some money for web hosting can create something that reaches millions of people. Although that's entirely different from what Edison, Hewlett, and Packard were doing, I'd argue that it's still within the realm of "innovation". (not to say that every website is particularly innovative, just that the possibility is there.)

The author can't be too happy with the picture of Albert Einstein at the top... For those that don't remember this Albert guy made one of the most important discoveries of all time; working in solitude, in the Swiss patent office, in-between reviewing patent applications. :)

And soon after, a team of experimentalists the world over set about proving him right. People rarely remember the experimentalists, the guys who go through all the hard work of providing the evidence to the theorists.

See: Michelson and Morley, Eddington, and others.

http://en.wikipedia.org/wiki/Tests_of_special_relativity http://en.wikipedia.org/wiki/Tests_of_general_relativity

Man who works in giant lab extols virtues of giant labs. Film at 11.

Good piece, with major implications for public policy. Encouraging more people to go into startups is great, but funding more R&D (especially in energy) is crucial as well. Also, Steven Johnson's Where Good Ideas Come From makes this case very well at book length.

I can't wait until some of the programmable-matter type approaches really take off. The closest we have is 3d printing.

What I am hoping to become popular and inexpensive something like 3d printing but with the ability to either assembly parts into machines automatically or the ability to print functional machines all in one go.

When you have inexpensive machines that can do that type of thing, then you really will be able to program reality in your garage, and print out fully assembled machines from open source components you downloaded from the internet.

On a slightly different topic, I was thinking, in response to this article, in a way, no one who is trying to innovate in technology (especially things like software) is really working alone. Instead we are building off of existing platforms and modules that others have created. For example, I am probably one of the most introverted programmers around, but the platform that I am building, ostensibly working on it completely alone, integrates many, many different pieces of open (and closed) source software/APIs, created relatively recently by hundreds or thousands of different people.

I'm working on a product based on jQuery, jQuery UI, WebKit (V8), CodeMirror, NodeJS, jQuery context menu, noty, NowJS (socket.io), CoffeeScript, FckEditor, jQuery editable, jquery ipweditor, Google Web Fonts (more than a dozen contributed fonts by many different authors), Google API hosting, AjaxUpload, NodeJs (NPM, shellJS, formidable, rimraf, async, request, cssmin, vows, zlib, mongolian, etc.) Rackspace API, PayPal/Stripe API, Ubuntu (Debian), Xen. Just look at the huge number of people that have contributed to Debian/Ubuntu, V8 or WebKit or Node.js recently or the rest of it over the years -- there is no way I could even consider doing this project without that stuff. So in a way everyone is in the lab already which is open source/APIs/github or whatever.

What's happening are CNC machines are starting to be sold as consumer devices. You can attach a lot of tools to a CNC-like machine, be it a laser etcher, a plastic squirters, millers, etc. It's kind of like kitchens.

Solitude is great for creativity. Labs/offices/hacker spaces are great for making that creativity a reality. You don't hire in more people in a startup because you want to stifle creativity... you need to get shit done sometimes.

The "why not both" meme comes to mind. Go sit in a hammock for a month or two, come back to the lab and try to get somewhere practical, then go back to the hammock. You need creativity, but you also need critique from peers out on the same ledge you're on.

Which reminds me of the hammock driven development talk[1]. The talk really resonated with me.

[1]: https://blip.tv/clojure/hammock-driven-development-4475586

The inventor in the garage is a real, re-occurring figure in the rise of industries, although the author is correct that these figures are more commercializers and integrators than scientific trailblazers. The irony is that there's no greater example than the namesake of Bell Labs...

Most real paradigm shifts happen outside the mainstream though.

Reply on news.ycombinator.com