by someone
Wed Mar 14th, 2007 at 11:44:30 AM EST
This story begins a few weeks back, when I was sitting here in front of this very same computer, reading this very same Eurotrib. There were some comments in various diaries that I though added up to something interesting. As I began my work of copy-paste I quickly grew annoyed. First, copy some text, then slap it in a block quote and provide a link with the author name. These several steps ought to be done by the computer, not me! This is true as well when getting material from other web sites, except then we'd like to link to the main site, and not to as specific content marker as in a comment. Well, we are here, the Word Wide Internet 2.0 is here, Firefox is here, and all the infrastructure to define ones own functionality is in place. Thus my choice was made, and I began what I can only describe as a thrilling fall into the Black Hole that is xul, javascript, and the Document Object Model.
Let me thus introduce you to what I wrested from this bottomless pit of despair.(It's not pretty, this code interface, but it does get the job done once one takes the time to learn how.)
I call it the Update [2007-3-15 5:45:45 by someone]:(new version)TribExt, and it is intended to grow into a site-specific browser extension for Firefox.
Features:
When text is highlighted the right-click context menu has an option to "Copy Html, URL, title" of the page, a command also available using Ctrl-Shift-C. If the selection is made wholly within one comment on the Eurotrib, we copy instead the selected html, name of the comment author, and link to the comment. These are wrapped up in a blockquote, with the author or title providing the text of the link.
Under the "View" - "Sidebar" menu there is a new option called the "Copy Sidebar". It can also be toggled on and off with Ctrl-Shift-S. It contains all the things copied in this manner, and saves its info in a file for when you close the browser. It has a "clear" button, and each copied item has a small red "x" by which it can be discarded. Items are copied to this sidebar regardless of whether it is open or not, they will show when it is opened. The last copied item is also placed on the system clipboard for easy access. The items in the sidebar can be pasted within Firefox with a simple drag-and-drop action into a text area. Oh, and if you hover the mouse over an item it will popup as it would display when pasted.
Having developed thus version 0.1 I am now hoping someone would like to give it a try. So, are you running Firefox? (If not, you really should get it here) If so, would you consider downloading and installing the Update [2007-3-15 5:45:45 by someone]:(new version) TribExt? To do this, download the file to your computer. Then open it with Firefox ("File" - "Open File" == Ctrl-O) which will install it in the appropriate place.
Note that version 0.1 means it is a bit under tested. It might not always copy what you want it to if the html in the selected text is a bit weird. I do filter it right now, and transform into Html allowed by ET comments. (minus div, they gave me some trouble... And images, as they would otherwise be hot-linked.) If you copy a portion of a table you will likely have to slap the html table tags around the <tr>'s and <td>'s. But it should mostly work. I hope.
Two known minor bugs: You have to click the "Clear" button twice for it to have any effect. The ampersand character is turned into its html character code equivalent for display in the copy popup window since it otherwise causes a bit of a mess. This doesn't work right, and the character code (&) is displayed instead. The pasted html doesn't have this problem. The blue text in the sidebar are links, though the pointer doesn't change when you move the mouse over them. They will open the linked page in the currently selected tag in the main browser area if you click on them. Tell me if you find other problems or have ideas of how things should work.
So, if you have some time, please consider giving this a test. I am hoping to do some more interesting Eurotrib content specific extension work later, and some interest from the community would be nice. Use the comments below for testing if you wish. (It is possible to display content from a website any way one wishes! One can in fact assemble content. Thus, client side development will allow for those features I have always wished for the Eurotrib: A page of responses to my comments, sorted by whether I have seen them or not, and responded or not. The possibility to "read" only part of a comment thread and leave the rest of the comments as unread, etc. This can be done entirely on the client side, keeping track of the information locally, and presenting content differently than served. This is all very exciting!)
Examples:
Copied from the Guardian:
Comment is free: A year to change the world
The Guardian's Comment is free website is celebrating its first birthday today with a special series of blogs from dozens of its regular contributors - all giving their own answer to a single question: What is the one thing you would most like to see happen by this time next year?
(What would I like to see from "Comment is free"?? Some intelligent content, perhaps?)
Copied from an ET comment:
Sven Triloqvist:
That book was largely superceded by the classic work of D. Gital, 'Principals 0f Regressive Enumeration' which unified the Cascading Theories of the 17th century Dutch mathematician Count Backvoord.
Screen shot, with mouse hovering over one of the copied items, displaying its popup. (or "tooltip" as it is called...) Click for larger picture.
I'm looking forward to reading the Eurotrib again now, since I can apparently not both learn a new code interface and read at the same time. And with that, I'm off for a drink! Cheers!
