[WikiEN-l] Googley comments

Gwern Branwen gwern0 at gmail.com
Tue Sep 8 09:29:43 UTC 2009


On Mon, Sep 7, 2009 at 9:04 AM, Bod Notbod<bodnotbod at gmail.com> wrote:
> On Sat, Sep 5, 2009 at 11:28 AM, Charles
> Matthews<charles.r.matthews at ntlworld.com> wrote:
>
>> The introduction of Talk pages was, it should not be forgotten, one of
>> the most brilliant innovations of the early days of Wikipedia.
>
> Indeed. A very intelligent friend of mine said he often finds the talk
> page as interesting as the article itself. He described them as a
> 'Talmudic commentary'. I keep meaning to make sure that I always read
> the talk page after scanning an article but I don't seem to have
> implanted that idea in my head with sufficient rigour yet, I tend to
> forget.

In the old days, I used to always open a tab when the talk page link
wasn't red. Thanks to projects etc., I can no longer do this since
half the time I'll just find a bunch of template spam.

What I've done is inveigle a JS coder to write the following for my monobook.js:

// load talk page inline
if(wgNamespaceNumber==0) addOnloadHook(getTalkPage)
function getTalkPage() { var tlink =
document.getElementById('ca-talk'); if(tlink.className == 'new')
return; var url = tlink.getElementsByTagName('a')[0].href; url +=
(url.indexOf('?')==-1) ? '?action=render' : '&action=render' ; var tp
= document.createElement('div'); tp.style.border = '1px solid blue';
tp.style.margin = '.5em 0'; tp.style.padding = '.35em';
tp.style.height = '128em'; tp.style.overflow = 'auto'; tp.id =
'ajax-talkpage'; tp.appendChild(document.createTextNode('fetching talk
page...')); document.getElementById('bodyContent').appendChild(tp);
getXML(url,getTalkPageStateChange);}
function getTalkPageStateChange() { switch (getReq.readyState) { case 4:
if (getReq.status == 200) { var tp =
document.getElementById('ajax-talkpage'); clearNode(tp); var txt =
getReq.responseText; tp.innerHTML = txt;} else {
tp.appendChild(document.createTextNode('** Problem ** ' +
getReq.statusText))
}
break;}
}

At the bottom of every article is a second frame, which loads the talk
page. When I've read to the bottom, a glance tells me whether the talk
page is dross or whether there're things worth reading, and I can
continue scrolling.

It doesn't hurt performance much at all. (Sometimes there are so many
templates on the bottom of an article and the top of the talk page
that even this feature doesn't help much, though...)

--
gwern



More information about the WikiEN-l mailing list