<?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"
	>
<channel>
	<title>Comments on: Development in Variable Width</title>
	<atom:link href="http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/</link>
	<description>readers are plentiful, thinkers are rare</description>
	<pubDate>Fri, 09 Jan 2009 23:13:52 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: Daniel</title>
		<link>http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/#comment-140</link>
		<dc:creator>Daniel</dc:creator>
		<pubDate>Tue, 16 May 2006 19:50:23 +0000</pubDate>
		<guid isPermaLink="false">http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/#comment-140</guid>
		<description>Just as I thought.  Most people waste so much time on aligning things to the right of other text.  It's such a waste to bother with this regardless of the font you use.  Same thing with asterisks around a comment.  It's a shame to see how much people waste their time when formatting comments.  The visual studio.net IDE is now legitimizing that practice by automatically beginning each new line of a multi-line comment with a star.  I just find it annoying.  I wonder what's the point.


Other strange coding practices I don't understand: 

Using vi or emacs.  I hesitate to criticize this outright because in theory having an all keyboard interface would be faster once you got past the very steep learning curve.  But do the proponents of these systems realize how many shortcuts are available to users of modern windows-based text editors?  I got to see a real seasoned unix guy using vi the other day.  He's a very fast typist besides.  He looked very slow and clumsy using it.

Indenting with spaces instead of tabs.  I've seen many editors offer the option of using spaces.  Why?  Tabs seem much smarter to me.  The key is pressed once, both to insert the tab and then to delete it if necessary, so it's easier.  Most editors have tab width settings besides, so if someone really likes code indented far, they can choose to display tabs wider than I would need to.  Tabs conform better to personal preference.  Besides, it's much easier to convert a file from being tab-indented to space-indented if desired.  You can pretty much always indiscriminately search-and-replace in that direction, whereas multiple spaces can't always be replaced by a tab in code.</description>
		<content:encoded><![CDATA[<p>Just as I thought.  Most people waste so much time on aligning things to the right of other text.  It&#8217;s such a waste to bother with this regardless of the font you use.  Same thing with asterisks around a comment.  It&#8217;s a shame to see how much people waste their time when formatting comments.  The visual studio.net IDE is now legitimizing that practice by automatically beginning each new line of a multi-line comment with a star.  I just find it annoying.  I wonder what&#8217;s the point.</p>
<p>Other strange coding practices I don&#8217;t understand: </p>
<p>Using vi or emacs.  I hesitate to criticize this outright because in theory having an all keyboard interface would be faster once you got past the very steep learning curve.  But do the proponents of these systems realize how many shortcuts are available to users of modern windows-based text editors?  I got to see a real seasoned unix guy using vi the other day.  He&#8217;s a very fast typist besides.  He looked very slow and clumsy using it.</p>
<p>Indenting with spaces instead of tabs.  I&#8217;ve seen many editors offer the option of using spaces.  Why?  Tabs seem much smarter to me.  The key is pressed once, both to insert the tab and then to delete it if necessary, so it&#8217;s easier.  Most editors have tab width settings besides, so if someone really likes code indented far, they can choose to display tabs wider than I would need to.  Tabs conform better to personal preference.  Besides, it&#8217;s much easier to convert a file from being tab-indented to space-indented if desired.  You can pretty much always indiscriminately search-and-replace in that direction, whereas multiple spaces can&#8217;t always be replaced by a tab in code.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Warren Falk</title>
		<link>http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/#comment-139</link>
		<dc:creator>Warren Falk</dc:creator>
		<pubDate>Tue, 16 May 2006 14:25:27 +0000</pubDate>
		<guid isPermaLink="false">http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/#comment-139</guid>
		<description>I think programmers didn't switch to variable length fonts because the switch to variable length fonts didn't happen in their favorite development environments.  Ironically development environments are often the last things to see certain innovations because developers get used to doing things a certain way, and also tend to become self-righteous zealots that preach that their way to programming nirvana is the only way. How else can you explain emacs, and even vi's continued use?  Neither of those can yet do variable width fonts.

I'm using Tahoma right now, but I'm sure there are better ones out there.  I'd like to mix and match them, actually.  I'd like to take some curly braces and angle brackets from monospace and insert them in my variable width.  I wouldn't mind getting my slash or dot back in my zero, even though I can easily tell a zero from the letter 'O' because it's narrower.

Tabs are not quite sufficient in their typical implementation.  The only tabs guaranteed to line up are those off the left margin.  Once a tab is based off of entered text, the width of that text may span more tabstops in one environment than another causing the tab to be inconsistent.

There are very few cases where code that was intended to line up did not so far.  The only ones I've seen are when the developer went crazy lining up declarations, or formatted a comment by putting a square border of asterisks all the way around it.</description>
		<content:encoded><![CDATA[<p>I think programmers didn&#8217;t switch to variable length fonts because the switch to variable length fonts didn&#8217;t happen in their favorite development environments.  Ironically development environments are often the last things to see certain innovations because developers get used to doing things a certain way, and also tend to become self-righteous zealots that preach that their way to programming nirvana is the only way. How else can you explain emacs, and even vi&#8217;s continued use?  Neither of those can yet do variable width fonts.</p>
<p>I&#8217;m using Tahoma right now, but I&#8217;m sure there are better ones out there.  I&#8217;d like to mix and match them, actually.  I&#8217;d like to take some curly braces and angle brackets from monospace and insert them in my variable width.  I wouldn&#8217;t mind getting my slash or dot back in my zero, even though I can easily tell a zero from the letter &#8216;O&#8217; because it&#8217;s narrower.</p>
<p>Tabs are not quite sufficient in their typical implementation.  The only tabs guaranteed to line up are those off the left margin.  Once a tab is based off of entered text, the width of that text may span more tabstops in one environment than another causing the tab to be inconsistent.</p>
<p>There are very few cases where code that was intended to line up did not so far.  The only ones I&#8217;ve seen are when the developer went crazy lining up declarations, or formatted a comment by putting a square border of asterisks all the way around it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Daniel</title>
		<link>http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/#comment-138</link>
		<dc:creator>Daniel</dc:creator>
		<pubDate>Tue, 16 May 2006 14:12:04 +0000</pubDate>
		<guid isPermaLink="false">http://www.warrenfalk.com/blog/2006/05/10/development-in-variable-width/#comment-138</guid>
		<description>I've often thought about this too.  Why is fixed width so necessary to the programmer?

I can only think of two reasons which I think you mentioned already: to make certain significant characters more visible (like curly braces and parentheses), and to make the spacing standard so other developers can choose a different monospace font and have no trouble.

Other than that, I can see no good reason.  I do have a few questions though:

Why do you think that programmers use fixed fonts when the rest of the world has made the transition away from them as soon as variable width fonts were available?  Maybe it's because I can't see why tabs don't do a good job of providing all the "lining up" that any code needs.

What font do you like to use?

What are some examples where you intended to line up code but it doesn't line up?</description>
		<content:encoded><![CDATA[<p>I&#8217;ve often thought about this too.  Why is fixed width so necessary to the programmer?</p>
<p>I can only think of two reasons which I think you mentioned already: to make certain significant characters more visible (like curly braces and parentheses), and to make the spacing standard so other developers can choose a different monospace font and have no trouble.</p>
<p>Other than that, I can see no good reason.  I do have a few questions though:</p>
<p>Why do you think that programmers use fixed fonts when the rest of the world has made the transition away from them as soon as variable width fonts were available?  Maybe it&#8217;s because I can&#8217;t see why tabs don&#8217;t do a good job of providing all the &#8220;lining up&#8221; that any code needs.</p>
<p>What font do you like to use?</p>
<p>What are some examples where you intended to line up code but it doesn&#8217;t line up?</p>
]]></content:encoded>
	</item>
</channel>
</rss>
