<?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 for techtangents</title>
	<atom:link href="http://www.techtangents.com/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.techtangents.com</link>
	<description></description>
	<lastBuildDate>Wed, 25 Jan 2012 14:44:19 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.2.1</generator>
	<item>
		<title>Comment on Purity and Testability by techtangents</title>
		<link>http://www.techtangents.com/purity-and-testability/comment-page-1/#comment-1987</link>
		<dc:creator>techtangents</dc:creator>
		<pubDate>Wed, 25 Jan 2012 14:44:19 +0000</pubDate>
		<guid isPermaLink="false">http://www.techtangents.com/?p=419#comment-1987</guid>
		<description>Good point. Thanks.

Recursive types may not be exhaustively testable due to infinite domain, but properties of them can still be proved inductively and tested over large sample sets a la QuickCheck. While your continuum is correct, I think the testability penalty incurred in introducing recursive types is much lower than when introducing impurity.</description>
		<content:encoded><![CDATA[<p>Good point. Thanks.</p>
<p>Recursive types may not be exhaustively testable due to infinite domain, but properties of them can still be proved inductively and tested over large sample sets a la QuickCheck. While your continuum is correct, I think the testability penalty incurred in introducing recursive types is much lower than when introducing impurity.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Purity and Testability by Daniel Waterworth</title>
		<link>http://www.techtangents.com/purity-and-testability/comment-page-1/#comment-1985</link>
		<dc:creator>Daniel Waterworth</dc:creator>
		<pubDate>Sun, 13 Nov 2011 09:40:10 +0000</pubDate>
		<guid isPermaLink="false">http://www.techtangents.com/?p=419#comment-1985</guid>
		<description>Functional purity is certainly important to testability, but it&#039;s not the whole story. You said, &quot;It means that I can potentially write a test which runs the function over each element of the domain&quot;, but the domain can be - and in fact usually is - infinite, because recursive types are more common than non-recursive types.

So there&#039;s a continuum of testability:

impure code &lt; pure code with recursive types &lt; pure code with no recursive types</description>
		<content:encoded><![CDATA[<p>Functional purity is certainly important to testability, but it&#8217;s not the whole story. You said, &#8220;It means that I can potentially write a test which runs the function over each element of the domain&#8221;, but the domain can be &#8211; and in fact usually is &#8211; infinite, because recursive types are more common than non-recursive types.</p>
<p>So there&#8217;s a continuum of testability:</p>
<p>impure code &lt; pure code with recursive types &lt; pure code with no recursive types</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Teaching Theory by ddoctor</title>
		<link>http://www.techtangents.com/teaching-theory/comment-page-1/#comment-233</link>
		<dc:creator>ddoctor</dc:creator>
		<pubDate>Thu, 04 Feb 2010 21:20:52 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=289#comment-233</guid>
		<description>&gt; Teaching tends to be about remaining flexible and responding to the reactions of your audience.
Ah. That&#039;s a very good point.</description>
		<content:encoded><![CDATA[<p>&gt; Teaching tends to be about remaining flexible and responding to the reactions of your audience.<br />
Ah. That&#8217;s a very good point.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Teaching Theory by Spyder</title>
		<link>http://www.techtangents.com/teaching-theory/comment-page-1/#comment-232</link>
		<dc:creator>Spyder</dc:creator>
		<pubDate>Thu, 04 Feb 2010 13:42:50 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=289#comment-232</guid>
		<description>It&#039;s not about having one method that works for everyone - we all learn at different rates and often via different methods.  Me for example, I&#039;m fine with abstract concepts as long as I understand and accept the detail first :)

Teaching tends to be about remaining flexible and responding to the reactions of your audience.  Don&#039;t stick to a pre-set plan just because you have one, if people understand a topic quickly then pace through and slow down when they start to struggle.

Not that you did a particlarly bad job - nor am I claiming to be an awesome teacher myself. Just offering some thoughts :)</description>
		<content:encoded><![CDATA[<p>It&#8217;s not about having one method that works for everyone &#8211; we all learn at different rates and often via different methods.  Me for example, I&#8217;m fine with abstract concepts as long as I understand and accept the detail first <img src='http://www.techtangents.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
<p>Teaching tends to be about remaining flexible and responding to the reactions of your audience.  Don&#8217;t stick to a pre-set plan just because you have one, if people understand a topic quickly then pace through and slow down when they start to struggle.</p>
<p>Not that you did a particlarly bad job &#8211; nor am I claiming to be an awesome teacher myself. Just offering some thoughts <img src='http://www.techtangents.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Filesystems by Symphonious &#187; Better File System APIs</title>
		<link>http://www.techtangents.com/filesystems/comment-page-1/#comment-231</link>
		<dc:creator>Symphonious &#187; Better File System APIs</dc:creator>
		<pubDate>Thu, 21 Jan 2010 08:30:09 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=276#comment-231</guid>
		<description>[...] Dylan complained about the lack of testability with file system APIs, which reminded me of a neat looking library I stumbled across while looking at build systems: EntityFS. Among a bunch of other useful sounding things it provides:   File system APIs are backend-independent. File systems can be created in RAM memory, on Zip files, on file system directories, etcetera [...]</description>
		<content:encoded><![CDATA[<p>[...] Dylan complained about the lack of testability with file system APIs, which reminded me of a neat looking library I stumbled across while looking at build systems: EntityFS. Among a bunch of other useful sounding things it provides:   File system APIs are backend-independent. File systems can be created in RAM memory, on Zip files, on file system directories, etcetera [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on The Desk by Suneth</title>
		<link>http://www.techtangents.com/the-desk/comment-page-1/#comment-230</link>
		<dc:creator>Suneth</dc:creator>
		<pubDate>Mon, 19 Oct 2009 20:16:14 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=255#comment-230</guid>
		<description>Strange! I have no issues with both seats... I&#039;ve been to IKEA million times and packed the car with useless furniture (don&#039;t tell that to my lady...)
Is there a lock at the back of the seat?</description>
		<content:encoded><![CDATA[<p>Strange! I have no issues with both seats&#8230; I&#8217;ve been to IKEA million times and packed the car with useless furniture (don&#8217;t tell that to my lady&#8230;)<br />
Is there a lock at the back of the seat?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Coding in more than just text by ddoctor</title>
		<link>http://www.techtangents.com/coding-in-more-than-just-text/comment-page-1/#comment-229</link>
		<dc:creator>ddoctor</dc:creator>
		<pubDate>Tue, 01 Sep 2009 22:58:42 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=229#comment-229</guid>
		<description>Yeah, that&#039;s the kicker, isn&#039;t it? There&#039;s a few approaches that could help:

a) Keep the format as text, and link to external objects, rather than directly embedding. Then it&#039;s up to the IDE to provide the unified editing experience.

b) Just use text formats, but use graphical tools... a lot of this is happening.

c) Research new ways of diffing.

d) Maybe we don&#039;t need diffs for everything. Sacrifice them as the trade-off. There&#039;s already behaviors in some systems (e.g. user-configurable behaviors) that aren&#039;t diffable. Maybe that&#039;s a problem, but maybe we can come up with other ways of solving the same problems that diffs do.

It&#039;s hard to match text&#039;s strengths, for sure, but some tradeoffs along these lines could improve the development experience.

I just can&#039;t see us coding in code forever.</description>
		<content:encoded><![CDATA[<p>Yeah, that&#8217;s the kicker, isn&#8217;t it? There&#8217;s a few approaches that could help:</p>
<p>a) Keep the format as text, and link to external objects, rather than directly embedding. Then it&#8217;s up to the IDE to provide the unified editing experience.</p>
<p>b) Just use text formats, but use graphical tools&#8230; a lot of this is happening.</p>
<p>c) Research new ways of diffing.</p>
<p>d) Maybe we don&#8217;t need diffs for everything. Sacrifice them as the trade-off. There&#8217;s already behaviors in some systems (e.g. user-configurable behaviors) that aren&#8217;t diffable. Maybe that&#8217;s a problem, but maybe we can come up with other ways of solving the same problems that diffs do.</p>
<p>It&#8217;s hard to match text&#8217;s strengths, for sure, but some tradeoffs along these lines could improve the development experience.</p>
<p>I just can&#8217;t see us coding in code forever.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Coding in more than just text by ajsutton</title>
		<link>http://www.techtangents.com/coding-in-more-than-just-text/comment-page-1/#comment-228</link>
		<dc:creator>ajsutton</dc:creator>
		<pubDate>Tue, 01 Sep 2009 17:38:34 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=229#comment-228</guid>
		<description>Ah, that old chestnut.  It seems appealing until you try and put it in a version control system when it just up and dies usually.  That or try running it through a diff tool.  Text is just so powerful, flexible and universal it&#039;s benefits far outweigh it&#039;s weaknesses.</description>
		<content:encoded><![CDATA[<p>Ah, that old chestnut.  It seems appealing until you try and put it in a version control system when it just up and dies usually.  That or try running it through a diff tool.  Text is just so powerful, flexible and universal it&#8217;s benefits far outweigh it&#8217;s weaknesses.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Noise by ddoctor</title>
		<link>http://www.techtangents.com/noise/comment-page-1/#comment-227</link>
		<dc:creator>ddoctor</dc:creator>
		<pubDate>Sun, 23 Aug 2009 22:32:31 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=215#comment-227</guid>
		<description>Oh, a system fan header! Good thinking. I&#039;ll give it a shot. My BIOS has the same setting, but it hasn&#039;t done a whole lot to date.

As for your boot drive - can the BIOS boot off a USB stick? Or, you could network boot it.</description>
		<content:encoded><![CDATA[<p>Oh, a system fan header! Good thinking. I&#8217;ll give it a shot. My BIOS has the same setting, but it hasn&#8217;t done a whole lot to date.</p>
<p>As for your boot drive &#8211; can the BIOS boot off a USB stick? Or, you could network boot it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Noise by Spyder</title>
		<link>http://www.techtangents.com/noise/comment-page-1/#comment-226</link>
		<dc:creator>Spyder</dc:creator>
		<pubDate>Sun, 23 Aug 2009 14:19:08 +0000</pubDate>
		<guid isPermaLink="false">http://techtangents.wordpress.com/?p=215#comment-226</guid>
		<description>Assuming the CPU isn&#039;t at 100%, I can only think of two reasons for the fan to be going flat out; either you accidentally plugged it into a system fan header or something in the windows software is messing with it.  My server&#039;s BIOS has this handy &quot;target CPU temp&quot; and when I stopped using it as a desktop I upped that to 60.  Can barely tell it&#039;s running (although it is in the lounge room, not usually a quiet place).

I&#039;ve just finished ripping out five HDDs (total 1.3tb) from my MythTV server and replacing them with a WD green power 1tb (which is super quiet, and only $110 atm).  It&#039;s a little less storage space but after cleaning out some of the recordings I only have 500gb used.  In the last year or so I have realised that I don&#039;t need any of these storage drives online 24/7 and while USB self-powered drives are my storage medium of choice I picked up a desktop SATA caddy so I can access the data on any of those 5 drives if I need to (and I&#039;m planning to print out an index of the contents :) ).  The drives won&#039;t wear out as fast this way, and it means less power usage (admittedly not much, but the WD uses 5w under load and each of my old drives are 7w idle).

The only question now is what to do with the old 80gb IDE boot drive that only has 4gb used... I&#039;m thinking about replacing it with a CF card (via one of those nifty IDE adapters).</description>
		<content:encoded><![CDATA[<p>Assuming the CPU isn&#8217;t at 100%, I can only think of two reasons for the fan to be going flat out; either you accidentally plugged it into a system fan header or something in the windows software is messing with it.  My server&#8217;s BIOS has this handy &#8220;target CPU temp&#8221; and when I stopped using it as a desktop I upped that to 60.  Can barely tell it&#8217;s running (although it is in the lounge room, not usually a quiet place).</p>
<p>I&#8217;ve just finished ripping out five HDDs (total 1.3tb) from my MythTV server and replacing them with a WD green power 1tb (which is super quiet, and only $110 atm).  It&#8217;s a little less storage space but after cleaning out some of the recordings I only have 500gb used.  In the last year or so I have realised that I don&#8217;t need any of these storage drives online 24/7 and while USB self-powered drives are my storage medium of choice I picked up a desktop SATA caddy so I can access the data on any of those 5 drives if I need to (and I&#8217;m planning to print out an index of the contents <img src='http://www.techtangents.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' />  ).  The drives won&#8217;t wear out as fast this way, and it means less power usage (admittedly not much, but the WD uses 5w under load and each of my old drives are 7w idle).</p>
<p>The only question now is what to do with the old 80gb IDE boot drive that only has 4gb used&#8230; I&#8217;m thinking about replacing it with a CF card (via one of those nifty IDE adapters).</p>
]]></content:encoded>
	</item>
</channel>
</rss>

