<?xml version="1.0" encoding="utf-8"?><!-- generator="b2evolution/1.9.3" -->
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:admin="http://webns.net/mvcb/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:content="http://purl.org/rss/1.0/modules/content/">
	<channel>
		<title>Coup De Des - GUI for concrete poem</title>
						<link>http://hcmc.uvic.ca/blogs/index.php?blog=35</link>
				<description></description>
				<language>en-CA</language>
				<docs>http://backend.userland.com/rss</docs>
				<admin:generatorAgent rdf:resource="http://b2evolution.net/?v=1.9.3"/>
				<ttl>60</ttl>
								<item>
					<title>minor changes to appearance of help page</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=minor_changes_to_appearance_of_help_page&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Thu, 02 Feb 2012 17:36:18 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">9097@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 15) added some links and twiddled with the size of some of the text elements. EdR OK'd</description>
					<content:encoded><![CDATA[added some links and twiddled with the size of some of the text elements. EdR OK'd]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=9097&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>path to mallarme production site</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=path_to_mallarme_production_site&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Tue, 24 Jan 2012 21:11:23 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">9051@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 15) Took me a little while to track this down, so I'm posting it here. 

The URL mallarme.uvic.ca points to /lang02/www/fall_07/coupde/penultima2/ . The coupde folder also contains a penultima and a penultima2noindents folders, which are older versions of the same code as in penultima, now derelict.
That lang02 account also contains a www/coupdedes folder, which has older dev copies of various versions of the poem and GUIs for it.

We should move the production instance either to a new account on the university's unix box, or to the hcmc web server and then have the mallarme.uvic.ca domain name pointed at the new instance. </description>
					<content:encoded><![CDATA[<p>Took me a little while to track this down, so I'm posting it here. </p>

<p>The URL mallarme.uvic.ca points to /lang02/www/fall_07/coupde/penultima2/ . The coupde folder also contains a penultima and a penultima2noindents folders, which are older versions of the same code as in penultima, now derelict.<br />
That lang02 account also contains a www/coupdedes folder, which has older dev copies of various versions of the poem and GUIs for it.</p>

<p>We should move the production instance either to a new account on the university's unix box, or to the hcmc web server and then have the mallarme.uvic.ca domain name pointed at the new instance. </p>]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=9051&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>create 'how to use' page, include video</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=create_how_to_use_page_include_video&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Tue, 24 Jan 2012 19:00:05 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">9047@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 90) I took the "how to use this website" file which was a word document and 
turned it into a web page (http://mallarme.uvic.ca/how_to_use.php). The 
"How to use this web site" link in the footer now links to that new web 
page.

On the how_to_use page I added a video player after the text of the 
instructions. The video player seems to work in all browsers except 
Internet Explorer. The page (including the player) validates, so I'm not 
too keen to spend a ton of time writing special code to work only with 
IE. Immediately below the video player I've added a link that lets the 
user go to the Youtube site and play the video there.

&#60;!-- Here's the original code generated by Youtube, which isn't valid in XHTML 1.1 --&#62;
&#60;object width="640" height="480"&#62;
    &#60;param name="movie" value="http://www.youtube.com/v/X9-rpwoEjww?version=3&#38;hl=en_US&#38;rel=0"&#62;&#60;/param&#62;
    &#60;param name="allowFullScreen" value="true"&#62;&#60;/param&#62;&#60;param name="allowscriptaccess" value="always"&#62;&#60;/param&#62;
    &#60;embed src="http://www.youtube.com/v/X9-rpwoEjww?version=3&#38;hl=en_US&#38;rel=0" type="application/x-shockwave-flash" width="640" height="480" allowscriptaccess="always" allowfullscreen="true"&#62;&#60;/embed&#62;
&#60;/object&#62;

&#60;!-- And here's the output when the code block above is passed through http://www.tools4noobs.com/online_tools/youtube_xhtml/ which does validate --&#62;
&#60;object type="application/x-shockwave-flash" style="width:640px; height:480px;" data="http://www.youtube.com/v/X9-rpwoEjww?version=3"&#62;
    &#60;param name="movie" value="http://www.youtube.com/v/X9-rpwoEjww?version=3" /&#62;
    &#60;param name="allowFullScreen" value="true" /&#62;
    &#60;param name="allowscriptaccess" value="always" /&#62;
&#60;/object&#62;
</description>
					<content:encoded><![CDATA[<p>I took the "how to use this website" file which was a word document and <br />
turned it into a web page (http://mallarme.uvic.ca/how_to_use.php). The <br />
"How to use this web site" link in the footer now links to that new web <br />
page.</p>

<p>On the how_to_use page I added a video player after the text of the <br />
instructions. The video player seems to work in all browsers except <br />
Internet Explorer. The page (including the player) validates, so I'm not <br />
too keen to spend a ton of time writing special code to work only with <br />
IE. Immediately below the video player I've added a link that lets the <br />
user go to the Youtube site and play the video there.</p>

<p>&lt;!-- Here's the original code generated by Youtube, which isn't valid in XHTML 1.1 --&gt;<br />
&lt;object width="640" height="480"&gt;<br />
    &lt;param name="movie" value="http://www.youtube.com/v/X9-rpwoEjww?version=3&amp;hl=en_US&amp;rel=0"&gt;&lt;/param&gt;<br />
    &lt;param name="allowFullScreen" value="true"&gt;&lt;/param&gt;&lt;param name="allowscriptaccess" value="always"&gt;&lt;/param&gt;<br />
    &lt;embed src="http://www.youtube.com/v/X9-rpwoEjww?version=3&amp;hl=en_US&amp;rel=0" type="application/x-shockwave-flash" width="640" height="480" allowscriptaccess="always" allowfullscreen="true"&gt;&lt;/embed&gt;<br />
&lt;/object&gt;</p>

<p>&lt;!-- And here's the output when the code block above is passed through <a href="http://www.tools4noobs.com/online_tools/youtube_xhtml/">http://www.tools4noobs.com/online_tools/youtube_xhtml/</a> which does validate --&gt;<br />
&lt;object type="application/x-shockwave-flash" style="width:640px; height:480px;" data="http://www.youtube.com/v/X9-rpwoEjww?version=3"&gt;<br />
    &lt;param name="movie" value="http://www.youtube.com/v/X9-rpwoEjww?version=3" /&gt;<br />
    &lt;param name="allowFullScreen" value="true" /&gt;<br />
    &lt;param name="allowscriptaccess" value="always" /&gt;<br />
&lt;/object&gt;</p>
]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=9047&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>initial plan for spring 2012</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=initial_plan_for_spring_2012&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Fri, 06 Jan 2012 23:53:43 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">8968@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 90) Met with EDR. He has students and wants them to produce different "readings" of poems already treated, and to do readings of more poems. 

Goal of segment markup is that each segment a grammatically sound snippet (e.g. article and noun constitute one segment). That kind of judgement will have to be made by human whether they start with plain text and create segments or with each word and each punctuation mark in a separate segment and combine segments. If he wants to start with highly segmented XML, then we'll have to write some code to generate that from the plain text. EDR needs to have clear guidelines he provides to students on how to create "grammatically sound" segments. 

Second phase is assembling segments from a poem in various ways to create "readings". As he has more than one students I suggested he exploit that by have 2 students work on same poem, in any one of a number of ways:
- collaborate
- work individually and have peer review
- two students independently work on same poem and compare markup
That would address the problem of idiosyncratic/whimsical/nonsensical "readings" before they ever get to him. 

Did not discuss incorporating Coup De Des into this structure.

EDR will meet with students next week. EDR, Martin and I will meet next week too. 
</description>
					<content:encoded><![CDATA[Met with EDR. He has students and wants them to produce different "readings" of poems already treated, and to do readings of more poems. 

Goal of segment markup is that each segment a grammatically sound snippet (e.g. article and noun constitute one segment). That kind of judgement will have to be made by human whether they start with plain text and create segments or with each word and each punctuation mark in a separate segment and combine segments. If he wants to start with highly segmented XML, then we'll have to write some code to generate that from the plain text. EDR needs to have clear guidelines he provides to students on how to create "grammatically sound" segments. 

Second phase is assembling segments from a poem in various ways to create "readings". As he has more than one students I suggested he exploit that by have 2 students work on same poem, in any one of a number of ways:
- collaborate
- work individually and have peer review
- two students independently work on same poem and compare markup
That would address the problem of idiosyncratic/whimsical/nonsensical "readings" before they ever get to him. 

Did not discuss incorporating Coup De Des into this structure.

EDR will meet with students next week. EDR, Martin and I will meet next week too. 
]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=8968&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>work out specs for markup of new poems</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=work_out_specs_for_markup_of_new_poems&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Tue, 29 Jun 2010 17:00:36 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">6867@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 75) Met with Annick to discuss how to mark up the poems and the various readings for each. Changes from the conventions for Coup De Des necessary because:
- more than one segment can appear on 1 line
- a segment can extend over 2 lines
- a reading can have segments not in sequential order
The second problem is the most potentially problematic, as it makes it much more difficult to write a hierarchical data structure which includes lines and line groups. We may have to include segments consisting of nothing but an empty line to capture white space between line groups

&#60;segment id="s1"&#62;À&#60;/segment&#62;
&#60;segment id="s2"&#62;la nue accablante tu&#60;/segment&#62;

Here's how two readings might look:

reading[1] = new Array(); // reading
reading[1][1] = new Array(); // line 1 of this reading
reading[1][1][1] = "s1"; //id of the first segment of this line of this reading
reading[1][1][2] = "s2"; //id of the second segment of this line of this reading
reading[1][2] = new Array(); // line 2 of this reading
reading[1][2][1] = "s3"; //id of the first segment of this line of this reading

reading[2] = new Array();
reading[2][1] = new Array();
reading[2][1][1] = "s3";
reading[2][2] = new Array();
reading[2][2][1] = "s4";
</description>
					<content:encoded><![CDATA[<p>Met with Annick to discuss how to mark up the poems and the various readings for each. Changes from the conventions for Coup De Des necessary because:<br />
- more than one segment can appear on 1 line<br />
- a segment can extend over 2 lines<br />
- a reading can have segments not in sequential order<br />
The second problem is the most potentially problematic, as it makes it much more difficult to write a hierarchical data structure which includes lines and line groups. We may have to include segments consisting of nothing but an empty line to capture white space between line groups</p>

<p>&lt;segment id="s1"&gt;À&lt;/segment&gt;<br />
&lt;segment id="s2"&gt;la nue accablante tu&lt;/segment&gt;</p>

<p>Here's how two readings might look:</p>

<p>reading[1] = new Array(); // reading<br />
reading[1][1] = new Array(); // line 1 of this reading<br />
reading[1][1][1] = "s1"; //id of the first segment of this line of this reading<br />
reading[1][1][2] = "s2"; //id of the second segment of this line of this reading<br />
reading[1][2] = new Array(); // line 2 of this reading<br />
reading[1][2][1] = "s3"; //id of the first segment of this line of this reading</p>

<p>reading[2] = new Array();<br />
reading[2][1] = new Array();<br />
reading[2][1][1] = "s3";<br />
reading[2][2] = new Array();<br />
reading[2][2][1] = "s4";</p>
]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=6867&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>hilite next and subsequent nodes with js</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=hilite_next_and_subsequent_nodes_with_js&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Wed, 04 Jun 2008 16:53:09 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">3499@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 360) Spent time over last couple of days creating a couple of visualizations
1) Next Node for target node's readings
Get target segment, find out all readings of which this segment is a member, get next segment (if any) for the reading, highlight current segment and each segment which is a next-segment in at least one reading. That didn't show how many readings each next-segment was a member of, so modified the code to increase saturation of highlighting colour proportional to number of readings that segment is the next-segment of. 
2) All subsequent nodes for target node's readings
Get target segment, find out all readings of which this segment is a member, get all subsequent segments (if any) for the reading. Each reading highlighted in a different colour. Used css:before and css:after attribute to put an explicit "start of reading" and "end of reading" characters into page (by using js to add a class to the relevant segments). Three problems: A) Readings which skip elements (1,3,5) where some of those segments are in another reading (1,2,3,4) can't be distinguished from (1,2,3,4,5 and 1,2,3,4).  B) Smaller problem (conceptually, but still a problem for UI) when one reading is a subset of another (1,2,3 and 1,2,3,4), only segment 4 appears in its colour - segments 1,2,3 appear in the colour for that reading. C) Can't visually display how many readings a given segment is a member of - though maybe that could be done in a way similar to the changing saturation solution I used above. 

In general, the problem is to show:
1) remaining segments in all readings for which target segment is a member
2) which reading(s) each of those segments is a member of
3) how many readings(s) a given segment is a member of
I can't use text size, location or italics as they're features of the data. I can use underline/line-through, text colour, bg colour, border attributes, text weight, css:before and css:after based on manipulating the className attribute of the segment.</description>
					<content:encoded><![CDATA[<p>Spent time over last couple of days creating a couple of visualizations<br />
1) Next Node for target node's readings<br />
Get target segment, find out all readings of which this segment is a member, get next segment (if any) for the reading, highlight current segment and each segment which is a next-segment in at least one reading. That didn't show how many readings each next-segment was a member of, so modified the code to increase saturation of highlighting colour proportional to number of readings that segment is the next-segment of. <br />
2) All subsequent nodes for target node's readings<br />
Get target segment, find out all readings of which this segment is a member, get all subsequent segments (if any) for the reading. Each reading highlighted in a different colour. Used css:before and css:after attribute to put an explicit "start of reading" and "end of reading" characters into page (by using js to add a class to the relevant segments). Three problems: A) Readings which skip elements (1,3,5) where some of those segments are in another reading (1,2,3,4) can't be distinguished from (1,2,3,4,5 and 1,2,3,4).  B) Smaller problem (conceptually, but still a problem for UI) when one reading is a subset of another (1,2,3 and 1,2,3,4), only segment 4 appears in its colour - segments 1,2,3 appear in the colour for that reading. C) Can't visually display how many readings a given segment is a member of - though maybe that could be done in a way similar to the changing saturation solution I used above. </p>

<p>In general, the problem is to show:<br />
1) remaining segments in all readings for which target segment is a member<br />
2) which reading(s) each of those segments is a member of<br />
3) how many readings(s) a given segment is a member of<br />
I can't use text size, location or italics as they're features of the data. I can use underline/line-through, text colour, bg colour, border attributes, text weight, css:before and css:after based on manipulating the className attribute of the segment.</p>]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=3499&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>Low birds-eye view visualization</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=low_birds_eye_view_visualization&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Tue, 27 May 2008 22:56:30 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">3439@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 240) The box and stick visualization looks rather clunky, so discussed other options with Greg. We hit on the the idea of viewing the page at a low angle from the horizon, with the words standing up off the page and coming towards you. The closest word is the largest and the words on the rest of the page fade away (in size and/or colour). The interface will allow you to choose a reading and have the page go through it, or allow you to choose a node, present which nodes you can go to next and a means to allow you to choose which node you want. 

Learning about fiddling with svg and specifically css type attributes in svg using javascript, and the degree of support for various attributes (opacity, fill-opacity, etc.) in FF and Safari.

Emile dropped by and seemed impressed with the ideas, but wondered how to do that while making the poem still look like what's on the printed page - I don't see any way to have one representation be simultaneously completely different and reassuringly similar, so we'll have to have some way to instantly navigate from an unconventional presentation to a conventional one. </description>
					<content:encoded><![CDATA[<p>The box and stick visualization looks rather clunky, so discussed other options with Greg. We hit on the the idea of viewing the page at a low angle from the horizon, with the words standing up off the page and coming towards you. The closest word is the largest and the words on the rest of the page fade away (in size and/or colour). The interface will allow you to choose a reading and have the page go through it, or allow you to choose a node, present which nodes you can go to next and a means to allow you to choose which node you want. </p>

<p>Learning about fiddling with svg and specifically css type attributes in svg using javascript, and the degree of support for various attributes (opacity, fill-opacity, etc.) in FF and Safari.</p>

<p>Emile dropped by and seemed impressed with the ideas, but wondered how to do that while making the poem still look like what's on the printed page - I don't see any way to have one representation be simultaneously completely different and reassuringly similar, so we'll have to have some way to instantly navigate from an unconventional presentation to a conventional one. </p>]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=3439&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>apply indents to poem</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=apply_indents_to_poem&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Tue, 27 May 2008 22:50:18 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">3436@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 60) Took K's indents and applied them to the coup de des poem. Everything seemed to work. It's not up to E and K to tweak the px and/or em settings as they wish. Spent more time fixing problems introduced by MSWord in the file provided by K than actually applying the data. </description>
					<content:encoded><![CDATA[Took K's indents and applied them to the coup de des poem. Everything seemed to work. It's not up to E and K to tweak the px and/or em settings as they wish. Spent more time fixing problems introduced by MSWord in the file provided by K than actually applying the data. ]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=3436&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>Set up stylesheet for indents</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=set_up_stylesheet_for_indents&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Fri, 23 May 2008 23:34:32 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">3420@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 90) Katharine provided a list of all the indents she measured in the poem. I've created a class in the stylesheet for each one (definition includes a px-based specification and an em-based specification both derived mathematically from her mm measurements in such a way as to fit into the display box). 
I sent Katherine the poem file and instructions on assigning each line the correct indent value. When I get that back from her I'll post it and then we can look at tweaking it. </description>
					<content:encoded><![CDATA[Katharine provided a list of all the indents she measured in the poem. I've created a class in the stylesheet for each one (definition includes a px-based specification and an em-based specification both derived mathematically from her mm measurements in such a way as to fit into the display box). 
I sent Katherine the poem file and instructions on assigning each line the correct indent value. When I get that back from her I'll post it and then we can look at tweaking it. ]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=3420&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
								<item>
					<title>Helped Clifton on UTF characters in subject line</title>
					<link>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;title=helped_clifton_on_utf_characters_in_subj&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1</link>
					<pubDate>Sat, 01 Mar 2008 00:21:32 +0000</pubDate>
					<dc:creator>sarneil</dc:creator>
					<category domain="main">Activity Log</category>					<guid isPermaLink="false">3007@http://hcmc.uvic.ca/blogs/</guid>
					<description> (Mins: 60) Helped Clifton with problem documented here.</description>
					<content:encoded><![CDATA[Helped Clifton with problem documented <a href="http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=3002&amp;more=1&amp;c=1&amp;tb=1&amp;pb=1">here</a>.]]></content:encoded>
					<comments>http://hcmc.uvic.ca/blogs/index.php?blog=35&amp;p=3007&amp;c=1&amp;tb=1&amp;pb=1#comments</comments>
				</item>
					</channel>
</rss>
