csh
/
y.st.
forked from y.st./y.st.


			
				
					
						
						
							123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146
							<?xml version="1.0" encoding="utf-8"?>
<!--
                                                                                     
 h       t     t                ::       /     /                     t             / 
 h       t     t                ::      //    //                     t            // 
 h     ttttt ttttt ppppp sssss         //    //  y   y       sssss ttttt         //  
 hhhh    t     t   p   p s            //    //   y   y       s       t          //   
 h  hh   t     t   ppppp sssss       //    //    yyyyy       sssss   t         //    
 h   h   t     t   p         s  ::   /     /         y  ..       s   t    ..   /     
 h   h   t     t   p     sssss  ::   /     /     yyyyy  ..   sssss   t    ..   /     
                                                                                     
	<https://y.st./>
	Copyright © 2015 Alex Yst <mailto:copyright@y.st>

	This program is free software: you can redistribute it and/or modify
	it under the terms of the GNU General Public License as published by
	the Free Software Foundation, either version 3 of the License, or
	(at your option) any later version.

	This program is distributed in the hope that it will be useful,
	but WITHOUT ANY WARRANTY; without even the implied warranty of
	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	GNU General Public License for more details.

	You should have received a copy of the GNU General Public License
	along with this program. If not, see <https://www.gnu.org./licenses/>.
-->
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml">
	<head>
		<base href="https://y.st./en/weblog/2015/10-October/30.xhtml" />
		<title>glib is messing up the file-sorting order. &lt;https://y.st./en/weblog/2015/10-October/30.xhtml&gt;</title>
		<link rel="icon" type="image/png" href="/link/CC_BY-SA_4.0/y.st./icon.png" />
		<link rel="stylesheet" type="text/css" href="/link/basic.css" />
		<link rel="stylesheet" type="text/css" href="/link/site-specific.css" />
		<script type="text/javascript" src="/script/javascript.js" />
		<meta name="viewport" content="width=device-width" />
	</head>
	<body>
		<nav>
			<p>
				<a href="/en/">Home</a> |
				<a href="/en/a/about.xhtml">About</a> |
				<a href="/en/a/contact.xhtml">Contact</a> |
				<a href="/a/canary.txt">Canary</a> |
				<a href="/en/URI_research/"><abbr title="Uniform Resource Identifier">URI</abbr> research</a> |
				<a href="/en/opinion/">Opinions</a> |
				<a href="/en/coursework/">Coursework</a> |
				<a href="/en/law/">Law</a> |
				<a href="/en/a/links.xhtml">Links</a> |
				<a href="/en/weblog/2015/10-October/30.xhtml.asc">{this page}.asc</a>
			</p>
			<hr/>
			<p>
				Weblog index:
				<a href="/en/weblog/"><abbr title="American Standard Code for Information Interchange">ASCII</abbr> calendars</a> |
				<a href="/en/weblog/index_ol_ascending.xhtml">Ascending list</a> |
				<a href="/en/weblog/index_ol_descending.xhtml">Descending list</a>
			</p>
			<hr/>
			<p>
				Jump to entry:
				<a href="/en/weblog/2015/03-March/07.xhtml">&lt;&lt;First</a>
				<a rel="prev" href="/en/weblog/2015/10-October/29.xhtml">&lt;Previous</a>
				<a rel="next" href="/en/weblog/2015/10-October/31.xhtml">Next&gt;</a>
				<a href="/en/weblog/latest.xhtml">Latest&gt;&gt;</a>
			</p>
			<hr/>
		</nav>
		<header>
			<h1>glib is messing up the file-sorting order.</h1>
			<p>Day 00237: Friday, 2015 October 30</p>
		</header>
<p>
	There is an issue present in the file managers associated with many of the common desktop environments that causes files to be sorted in a bizarre and unintuitive manner.
	Instead of sorting files is such a way that every character is treated as just a character, they are sorted instead treating digits as somehow special.
	All consecutive digits are treated as a single character and sorted from least to greatest.
	This makes it very difficult to find files with names that begin with large strings of digits if the beginning digits are known but the number of digits is not.
	This is far worse than the issue of sorting files case-insensitively.
	With the help of a <a href="https://forum.xfce.org/viewtopic.php?pid=31785#p31785">forum post</a>, I&apos;ve tracked down the problem.
	The problem is glib, which is probably used by most <abbr title="GIMP Toolkit">GTK+</abbr>-based desktops.
	This means that I cannot escape the issue by switching to another desktop, as <abbr title="K Desktop Environment">KDE</abbr> is out of the question.
	Furthermore, I can&apos;t find any information on disabling this &quot;feature&quot; in glib.
	When I get the chance, I need to try out some other window managers and see if I can find one that does not have this annoying issue while still being usable.
</p>
<p>
	I tried out a new email client today called Sylpheed.
	It seems alright so far, though does have one issue during setup.
	Sylpheed handles proxies on a per-mail-account basis, so you cannot set up your proxy before setting up your accounts.
	Furthermore, the proxy cannot be set up during the initial setup of a given account.
	Instead, an account must be set up to run over clearnet, then edited to run through the proxy instead.
	To prevent any data leaks, I set the account up to use <code>//localhost.</code> as the mail server, then fixed it to use the real mail server once the proxy was set.
	However, updating the <abbr title="Internet Message Access Protocol">IMAP</abbr> and <abbr title="Simple Mail Transfer Protocol">SMTP</abbr> servers did not seem to actually allow me to access my mail.
	Eventually I found the problem, which is that Sylpheed does not update the server&apos;s directory tree upon editing the account or upon restarting the email client.
	If you use this method to prevent data leaks over clearnet, you&apos;ll need to tell Sylpheed to refresh the directory tree once the email account settings are in place.
</p>
<p>
	In the <abbr title="Cascading Style Sheets">CSS</abbr> course I&apos;m taking, the professor talked about creating <abbr title="Cascading Style Sheets">CSS</abbr> rules that only apply to hyperlinks to websites that use specific <abbr title="Top Level Domain">TLD</abbr>s.
	The professor&apos;s method does not function outside the toy website we are working with, but it was an interesting concept none the less.
	Specifically, the <abbr title="Cascading Style Sheets">CSS</abbr> selectors she used checked the last few characters of the <code>href</code> attribute to see if they, for example, were &quot;.edu&quot;.
	However, this method requires that no hyperlinks specify a file path.
	At the very least, the <code>/</code> path should be used, but even if a trailing slash was added to the selector, it would not select hyperlinks that used a different path.
	Using a different selector, one could select hyperlinks pointing to <abbr title="Uniform Resource Identifier">URI</abbr>s containing &quot;.edu&quot; (or even &quot;.edu/&quot;), but even that would select hyperlinks pointing to <abbr title="Uniform Resource Identifier">URI</abbr>s containing this text anywhere in their path, not just those that actually use the &quot;edu.&quot; <abbr title="Top Level Domain">TLD</abbr>.
	I&apos;m not sure it is possible to achieve the desired effect without making assumptions that will not always hold true.
	Speaking of <abbr title="Cascading Style Sheets">CSS</abbr>, when I have time, I should set my stylesheet to wipe as much of browsers&apos; default style sheets to achieve a more consistent look.
</p>
<p>
	I think that I might move my weblog into onion land.
	My current setup, along with my job hunt, is putting a rift in online identity.
	I don&apos;t want potential employers reading my journal and discriminating against me for its contents, so I&apos;ve had to set up a shallow mask for use only when job hunting.
	But if I torify the personal parts, only people that use Tor or know about the Tor-to-Web gateways will see my more private thoughts.
	I won&apos;t have to be &quot;{legal name}@clovermail.net&quot;, I can be &quot;{legal name}@{my domain}&quot;.
	This applies to more than just email though.
	I plan to get more into software development on my own time, which will of course be released under my real name, not my legal name, using accounts that are tied to my clearnet website.
	With my weblog hosted there, I would have no choice but to hide everything I am behind that shallow mask.
	By moving the personal part of my website to onion land, I insure that most of the people that see it will either be like-minded or technically savvy.
	I will no longer be as much of a blatant target for the nosy.
	When I do this, I&apos;ll use <abbr title="Cascading Style Sheets">CSS</abbr> to hide links to the onion address, then link to a style sheet on the onion that will reveal them.
	It&apos;s not the perfect solution, but it should be pretty decent.
</p>
<p>
	This entry will probably be the last to go live before I set up the onion website and move the weblog to it.
	That means I&apos;m going to have to go dark for a bit, as I do not even have a server machine available right now.
	I&apos;ll spend that time preparing the style sheets and rewriting the links of this website to begin in &quot;//y.st/&quot; or &quot;//quystystxtvdgyst.onion/&quot;, depending on the link.
	I&apos;ll probably continue to update my canary daily and will definitely write weblog entries in the mean time to make available when the onion comes back up.
</p>
<p>
	My <a href="/a/canary.txt">canary</a> still sings the tune of freedom and transparency.
</p>
		<hr/>
		<p>
			Copyright © 2015 Alex Yst;
			You may modify and/or redistribute this document under the terms of the <a rel="license" href="/license/gpl-3.0-standalone.xhtml"><abbr title="GNU&apos;s Not Unix">GNU</abbr> <abbr title="General Public License version Three or later">GPLv3+</abbr></a>.
			If for some reason you would prefer to modify and/or distribute this document under other free copyleft terms, please ask me via email.
			My address is in the source comments near the top of this document.
			This license also applies to embedded content such as images.
			For more information on that, see <a href="/en/a/licensing.xhtml">licensing</a>.
		</p>
		<p>
			<abbr title="World Wide Web Consortium">W3C</abbr> standards are important.
			This document conforms to the <a href="https://validator.w3.org./nu/?doc=https%3A%2F%2Fy.st.%2Fen%2Fweblog%2F2015%2F10-October%2F30.xhtml"><abbr title="Extensible Hypertext Markup Language">XHTML</abbr> 5.1</a> specification and uses style sheets that conform to the <a href="http://jigsaw.w3.org./css-validator/validator?uri=https%3A%2F%2Fy.st.%2Fen%2Fweblog%2F2015%2F10-October%2F30.xhtml"><abbr title="Cascading Style Sheets">CSS</abbr>3</a> specification.
		</p>
	</body>
</html>