The Ultimate Dogfooding Story

This guy is my personal hero, but -- alas -- this is also a story about <i>worse is better</i>:<p><a href="http://www.taunton.com/finehomebuilding/how-to/articles/sawstop-revisited.aspx" rel="nofollow">http://www.taunton.com/finehomebuilding/how-to/articles/saws...</a><p><i>"Having heard all of the various reasons put forth by the manufacturers for not doing anything, I firmly believe it simply comes down to money," says Gass. "They cannot figure out how to make more money by adding SawStop. They are not paying for the injuries that occur now, so why should they spend money to change their product to eliminate a cost they aren't bearing?"</i><p>So, although this guy has offered his safety technology to all tool manufacturers, SawStop is only available on SawStop-branded saws, which cost a bit more money, which causes people to bitch and moan, because who wants to spend money on a feature that merely prevents injuries?<p>The appropriate software-platform analogy is left as an exercise for the reader. ;)

Lest you think he started with the finger - it was hot dogs first...<p>"Ultimately, it seemed like the most reliable technique would have to involve contact detection, Gass recalls. And because your body has an electrical capacitance, it offers the potential for that. I figured that if you put a voltage into the saw blade, your body could absorb some of that signal. Then the voltage in the saw blade would drop.<p>Within a week, Gass formulated most of the idea’s details in his mind. Thirty days later, he completed his first working prototype. Initially, he tested the prototype by touching the side of the saw blade with a finger. And while that proved the saw could stop in a fraction of a second, he still didn’t know if it was quick enough to prevent serious injury. Gass wondered how deep a 4,000-rpm saw blade would cut into human flesh during the microseconds that it took to stop the blade. To answer that, he needed to touch the blade teeth.<p>"I came up with the idea of using hot dogs, and it worked pretty well," Gass says. "It’s cheap, readily available, and you don’t get any protesters coming to your door."<p>From a Design News article - he was chosen as DN's Engineer of The Year 2007 (aside: DN is one of the <i>best</i> pure applied engineering publications that I actually read - apply for a subscription)<p><a href="http://www.designnews.com/article/print/5897-Man_on_a_Mission.php" rel="nofollow">http://www.designnews.com/article/print/5897-Man_on_a_Missio...</a>

That's not what "eating your own dog food" means. It means using your product to do your own work. As in: if Gass were a woodworker and used the saw on his own projects, he'd get better ideas about saw design.<p>I like the saw story too, but its applicability to software projects is...? Here's the moral of the story offered by Atwood:<p><i>Nothing exudes confidence like software developers who are willing to stick their own extremities into the spinning blades of software they've written.</i><p>That's inane even by his standards.

The original video of Steve Gass sticking his own finger into the table-saw: <a href="http://www.dumpert.nl/mediabase/325071/6bb3c29d/time_warp_cirkelzaag.html" rel="nofollow">http://www.dumpert.nl/mediabase/325071/6bb3c29d/time_warp_ci...</a><p>(the one on CH is of a hotdog... this one is of a real, live human digit!!!)

It is a good story. I don't know if you can call the risk of a finger "ultimate," though; when Nikola Tesla bet his life on the skin effect of very high frequency alternating current as a PR counter Edison's animal electrocution demonstrations.

I've heard of the Sawstop technology before and what I heard was very depressing if true.<p>Basically none of the table saw manufacturers want to deal with this guy because they are afraid or lawsuits.<p>The logic goes like this, right now everybody knows a table saw will maim you, there's no expectation that it won't.<p>But if they implement that technology, and it then fails, <i>then</i> there will be a BIG lawsuit.<p>Meanwhile everybody who uses a table saw and sees Sawstop, wants it.<p>But the guy is having a very difficult time setting a major manufacturing operation to ship Sawstop table saws himself.<p>Again, I don't know if this is true, but it's very sad if it is.<p>-- EDIT:<p>I missed this:<p><a href="http://news.ycombinator.com/item?id=453822" rel="nofollow">http://news.ycombinator.com/item?id=453822</a>

A lot of medical researchers apparently also experimented on themselves.

I would argue that this was a horrible decision, even if everything turned out in the end. It reminds me of when a kid makes a really stupid decision that by chance turns out well, and the parents (correctly) yell at him despite the result.<p>If it had turned out that there was a minor flaw/"bug" in his implementation, he'd lose his finger for no reason.<p>What would be much better is to make lots of test saws with his technology and give them to carpenters for free. The worst thing that could happen is that someone who would have lost their finger anyway would lose their finger. The best would be free publicity.

Is what Jeff Atwood doing even legal?!?! You can't just "quote" 80% of someone's work and then publish it as your own! In the article I counted only 6 sentences that Jess Atwood wrote himself. Can I just, say, "quote" a best seller, add a few words of commentary and then publish it? I think not.

I seem to remember the founder of American Body Armor performing demonstrations on himself by firing a .38 revolver at himself while wearing armor ... or something. heavily paraphrased.

Here is the video of the creator testing it with his own finger: <a href="http://www.youtube.com/watch?v=E3mzhvMgrLE" rel="nofollow">http://www.youtube.com/watch?v=E3mzhvMgrLE</a>

This may be the first application for table saws, but this has been around for awhile. It's basically the kind of saw that doctors use to remove casts.

A guy I once worked with aptly referred to the "dog-fooding" as "feel the pain" approach.

I don't like the principle behind the dog food. It means that you create a product good for dogs but not humans. Something not made for you. Then, when you start eating it, you improve it to a level being acceptable to you, a human.<p>I like to think that I create human food, a product that I need and care about if it's good or not at the start.