On Sat, May 2, 2009 at 3:18 PM, jidanni@jidanni.org wrote:
OK, I have reverse engineered the site notice. A non-Javascript user must apparently do the following to see it, else a better method would be mentioned in a <noscript>.
$ w3m -dump_source \ http://upload.wikimedia.org/centralnotice/wikipedia/en/centralnotice.js?207x... | ascii2uni -qa 7 | perl -pwle 's/\n/\n/g' | perl -nlwe 'next unless /^<table/../^</table>/;print' | w3m -dump -T text/html
Please participate in a vote to determine the future copyright [Hide] terms of Wikimedia projects (vote ends May 3, 2009). Vote now! [Help us with translations!]
Please participate in a vote to determine the future copyright [Hide] terms of Wikimedia projects (vote ends May 3, 2009). Vote now! [Help us with Scholarship applications for Wikimania 2009 are now open. translations!] Apply now!
Maybe monitoring the URL of the source file, if any, would be less painful. Is there a general pattern for where it is kept for all the WMF sites?
Actually, I think there is a seed a good idea here. It should be straightforward to create a special page that showed all current site notices without using javascript. That could then be linked to, via noscript or similar option, to provide people lacking javascript support with a place to be able to read the notices.
To partially answer your question, central notices are managed through [1]. It isn't the world's most useful interface if all you want to do is read notices, but it is probably better than parsing the javascript.
-Robert Rohde