<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Archimedes: A Big Enough Lever</title>
	<atom:link href="http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/</link>
	<description>Come for the words, stay for the...HEY! Come back!</description>
	<lastBuildDate>Sat, 13 Mar 2010 15:28:57 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.5</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: SeanMcTex</title>
		<link>http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/comment-page-1/#comment-48951</link>
		<dc:creator>SeanMcTex</dc:creator>
		<pubDate>Tue, 20 May 2008 15:55:32 +0000</pubDate>
		<guid isPermaLink="false">http://www.mcmains.net/?p=1017#comment-48951</guid>
		<description>@Patrick: Good to see you! 

MT is a somewhat similar concept, though more generalized in execution. In fact, given that MT has some good tools for creating HITs, one might be able to implement Archimedes on top of it.

The things that make Archimedes especially attractive to programmers (or at least to this one) is the ability to create requests for work (HITs in Amazon&#039;s parlance) without breaking programming flow. Even if it were build on the substrate of MT, one would still need a piece to watch source code checkins, and to mediate between the system where that lives and Amazon&#039;s system for doling out the work. (And even then, I&#039;m not sure one could verify that the submitted code passed the unit tests and the human acceptance review before giving the work a thumbs-up on MT.)</description>
		<content:encoded><![CDATA[<p>@Patrick: Good to see you! </p>
<p>MT is a somewhat similar concept, though more generalized in execution. In fact, given that MT has some good tools for creating HITs, one might be able to implement Archimedes on top of it.</p>
<p>The things that make Archimedes especially attractive to programmers (or at least to this one) is the ability to create requests for work (HITs in Amazon&#8217;s parlance) without breaking programming flow. Even if it were build on the substrate of MT, one would still need a piece to watch source code checkins, and to mediate between the system where that lives and Amazon&#8217;s system for doling out the work. (And even then, I&#8217;m not sure one could verify that the submitted code passed the unit tests and the human acceptance review before giving the work a thumbs-up on MT.)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Patrick</title>
		<link>http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/comment-page-1/#comment-48950</link>
		<dc:creator>Patrick</dc:creator>
		<pubDate>Tue, 20 May 2008 15:24:34 +0000</pubDate>
		<guid isPermaLink="false">http://www.mcmains.net/?p=1017#comment-48950</guid>
		<description>Other than the hardware part, how does this differ from &lt;a href=&quot;http://www.mturk.com/mturk/welcome&quot; rel=&quot;nofollow&quot;&gt;Amazon&#039;s Mechanical Turk&lt;/a&gt;.</description>
		<content:encoded><![CDATA[<p>Other than the hardware part, how does this differ from <a href="http://www.mturk.com/mturk/welcome" onclick="javascript:pageTracker._trackPageview('/outbound/comment/www.mturk.com');" rel="nofollow">Amazon&#8217;s Mechanical Turk</a>.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: david</title>
		<link>http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/comment-page-1/#comment-47091</link>
		<dc:creator>david</dc:creator>
		<pubDate>Sat, 10 May 2008 02:11:01 +0000</pubDate>
		<guid isPermaLink="false">http://www.mcmains.net/?p=1017#comment-47091</guid>
		<description>Now that I&#039;ve actually read your post, your idea makes a lot more sense to me.  What I was trying to say at lunch is that programming is definitely a great skill to learn, but I was wondering out loud when we might hit a saturation point where encouraging 3 world countries to amass armies of independent programmers might actually do more harm than good.  We have so much productive capacity in the world today, we really need to find a way to steer it in a way that is profitable for the individual, beneficial to society, and sustainable in the long term.

At some point there could be too few projects and so many reasonably skilled programmers that the price would go down, and/or these people who invested months and years of learning would find their skill gradually less marketable.  I really liked your idea and was just trying to think of other skills that could be incorporated into this sort of system, so that a variety skills would be developed and a variety of opportunities would available.

The thing is, those projects can develop in parallel and don&#039;t have to be a part of Archimedes.  For this specific niche, your idea is very innovative and seems worth looking into.</description>
		<content:encoded><![CDATA[<p>Now that I&#8217;ve actually read your post, your idea makes a lot more sense to me.  What I was trying to say at lunch is that programming is definitely a great skill to learn, but I was wondering out loud when we might hit a saturation point where encouraging 3 world countries to amass armies of independent programmers might actually do more harm than good.  We have so much productive capacity in the world today, we really need to find a way to steer it in a way that is profitable for the individual, beneficial to society, and sustainable in the long term.</p>
<p>At some point there could be too few projects and so many reasonably skilled programmers that the price would go down, and/or these people who invested months and years of learning would find their skill gradually less marketable.  I really liked your idea and was just trying to think of other skills that could be incorporated into this sort of system, so that a variety skills would be developed and a variety of opportunities would available.</p>
<p>The thing is, those projects can develop in parallel and don&#8217;t have to be a part of Archimedes.  For this specific niche, your idea is very innovative and seems worth looking into.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: SeanMcTex</title>
		<link>http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/comment-page-1/#comment-47020</link>
		<dc:creator>SeanMcTex</dc:creator>
		<pubDate>Fri, 09 May 2008 16:04:16 +0000</pubDate>
		<guid isPermaLink="false">http://www.mcmains.net/?p=1017#comment-47020</guid>
		<description>Thanks for the comments! I love that I can think of something on the way to work and have great, intelligent feedback (from great, intelligent people) within an hour or two.

@JasonY: Yep, it&#039;s an original idea, though a bit of judicious googling appears that others have tried somewhat similar things. I haven&#039;t seen that any of them provided the hooks into a source code management system, though.

Your MIT scenario is exactly the kind of thing that I had in mind! People around the world don&#039;t lack time or intelligence, just good ways to put those resources to us.

I&#039;m not sure offhand where one would take this to get it going; I would guess that grants might be a good approach, but I&#039;ve no experience with such things.

@Steve Ivy: The virus-checking is a great idea! Thanks for that!

I like the unit-testing too -- lets the computers do more of the work for us! I also thought that as part of a Phase 2, it might be cool to build tools to make generating acceptance tests easier for non-programmers, so that managers or people with product ideas could specify them without learning a unit-testing framework. That&#039;s considerably more ambitious, though.</description>
		<content:encoded><![CDATA[<p>Thanks for the comments! I love that I can think of something on the way to work and have great, intelligent feedback (from great, intelligent people) within an hour or two.</p>
<p>@JasonY: Yep, it&#8217;s an original idea, though a bit of judicious googling appears that others have tried somewhat similar things. I haven&#8217;t seen that any of them provided the hooks into a source code management system, though.</p>
<p>Your MIT scenario is exactly the kind of thing that I had in mind! People around the world don&#8217;t lack time or intelligence, just good ways to put those resources to us.</p>
<p>I&#8217;m not sure offhand where one would take this to get it going; I would guess that grants might be a good approach, but I&#8217;ve no experience with such things.</p>
<p>@Steve Ivy: The virus-checking is a great idea! Thanks for that!</p>
<p>I like the unit-testing too &#8212; lets the computers do more of the work for us! I also thought that as part of a Phase 2, it might be cool to build tools to make generating acceptance tests easier for non-programmers, so that managers or people with product ideas could specify them without learning a unit-testing framework. That&#8217;s considerably more ambitious, though.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steve Ivy</title>
		<link>http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/comment-page-1/#comment-47019</link>
		<dc:creator>Steve Ivy</dc:creator>
		<pubDate>Fri, 09 May 2008 15:49:54 +0000</pubDate>
		<guid isPermaLink="false">http://www.mcmains.net/?p=1017#comment-47019</guid>
		<description>I love it! The unit tests in particular are a great idea for speccing work. A realy &quot;duh!&quot; moment. I&#039;ve heard TDD described as contract-driven development - this just makes it a &quot;real&quot; contract!

You might, in addition, provide hooks in the service so that a user could ask, say McAfee or Symantec to pre-check submitted code for virus-y bits. Just a thought. 

I still love it! Get that man an angel investor, quick!</description>
		<content:encoded><![CDATA[<p>I love it! The unit tests in particular are a great idea for speccing work. A realy &#8220;duh!&#8221; moment. I&#8217;ve heard TDD described as contract-driven development &#8211; this just makes it a &#8220;real&#8221; contract!</p>
<p>You might, in addition, provide hooks in the service so that a user could ask, say McAfee or Symantec to pre-check submitted code for virus-y bits. Just a thought. </p>
<p>I still love it! Get that man an angel investor, quick!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: JasonY</title>
		<link>http://www.mcmains.net/2008/05/09/archimedes-a-big-enough-lever/comment-page-1/#comment-47018</link>
		<dc:creator>JasonY</dc:creator>
		<pubDate>Fri, 09 May 2008 15:49:21 +0000</pubDate>
		<guid isPermaLink="false">http://www.mcmains.net/?p=1017#comment-47018</guid>
		<description>Brilliant. I think you should copyright the idea and the sue the pants off of anyone who tries to use it. Who needs to change the world when you can get rich off of IP piracy?

Ha. No, but seriously, I think this is has the potential to be a real breakthrough idea. Is it original? If so, color me impressed.

In light of the fact that MIT has put some of their standard programming classes online for free, there is no limit to where this could go. It is now possible for a poor child in a destitute country to get the equivalent of a fairly good (self-taught) education in programming, and a system like Archimedes (inspired name) could leverage (ouch!) that into a life-changing prospect, while at the same time speeding up the rate of progress. My only concern is that it would devalue the education of those unfortunate folks who have put themselves into usurious debt to attend the MIT classes in meatspace. But hey, the wheels of the future crush the slow, I guess.

Is there anyone who you could talk to about this? I&#039;ll link it from my blog, but I doubt any of my 30 daily readers is in a position to get this rolling.

Great brainstorm, Sean. Glad I knows ya.</description>
		<content:encoded><![CDATA[<p>Brilliant. I think you should copyright the idea and the sue the pants off of anyone who tries to use it. Who needs to change the world when you can get rich off of IP piracy?</p>
<p>Ha. No, but seriously, I think this is has the potential to be a real breakthrough idea. Is it original? If so, color me impressed.</p>
<p>In light of the fact that MIT has put some of their standard programming classes online for free, there is no limit to where this could go. It is now possible for a poor child in a destitute country to get the equivalent of a fairly good (self-taught) education in programming, and a system like Archimedes (inspired name) could leverage (ouch!) that into a life-changing prospect, while at the same time speeding up the rate of progress. My only concern is that it would devalue the education of those unfortunate folks who have put themselves into usurious debt to attend the MIT classes in meatspace. But hey, the wheels of the future crush the slow, I guess.</p>
<p>Is there anyone who you could talk to about this? I&#8217;ll link it from my blog, but I doubt any of my 30 daily readers is in a position to get this rolling.</p>
<p>Great brainstorm, Sean. Glad I knows ya.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
