Wikipedia:Phase II bug reports: Difference between revisions
Content deleted Content added
No edit summary |
Confirming XHTML bug |
||
Line 3:
== UNCONFIRMED ==
''Newly submitted bugs which noone has been able to duplicate yet.''
I'm not sure if this is a bug: when you search for something, the old logo shows up on the search page, as in http://wikipedia.com/search.fcgi?request=happy▼
'''All's Well That Ends Well'''
Line 18 ⟶ 20:
This isn't really a bug, but I thought I should pass it along. There are some footnotes in the [[geologic Timescale]] 11/26/01 that are superscripted using HTML. They don't look all that bad in IE, but they do somewhat alter the spacing of the vertical bars following them on the same line in this ASCII table. They look worse in a text mode browser like LYNX than in IE or Netscape. Same will be true of other HTML that alters text sizes. That is the way things are I guess. Also, I suspect that ASCII tables and ASCII art (if any) need to be checked with LYNX. I trimmed the [[geologic Timescale]] to get a one to one of intended lines to displayed lines when using Lynx. Others might want to do the same if they have similar situations.
<XHTML bug moved to NEW>
'''Pages claim to be XHTML'''▼
Wikipedia pages include a DOCTYPE that claims the content is XHTML Basic. Given just how far from the truth this is, and how difficult it will be to ensure correctness when anyone can enter a range of HTML tags, no DOCTYPE should be included at all.▼
Examples of HTML used that isn't XHTML Basic below.▼
See [http://validator.w3.org/check?uri=www.wikipedia.com] for an even more picky analysis.▼
* img tag isn't closed - should be <img ... />.▼
* html lang, body bgcolor, img align and img border attributes aren't in XHMTL Basic.▼
* hr and font tags aren't in XHTML Basic.▼
* list items and paragraphs aren't closed: < li> ... '''< /li>'''.▼
* Some attribute values aren't quoted, e.g. type=text must be type="text".▼
* Some inline elements aren't contained in block-level elements, like the toolbar at the bottom of the page.▼
This does have actual effects on the pages: in Mozilla, the top hr element overlaps the logo, and nested indents (like on [[Carey Evans/Talk]]) don't work.▼
The pages also claim to be UTF-8 encoded XML (<?xml ...?> PI at the top of the page) while the HTTP headers say ISO-8859-1.▼
:--[[Carey Evans]], 17 November 2001
== NEW ==
Line 86 ⟶ 70:
----
▲'''Pages claim to be XHTML''' 17 November 2001
▲I'm not sure if this is a bug: when you search for something, the old logo shows up on the search page, as in http://wikipedia.com/search.fcgi?request=happy
▲Wikipedia pages include a DOCTYPE that claims the content is XHTML Basic. Given just how far from the truth this is, and how difficult it will be to ensure correctness when anyone can enter a range of HTML tags, no DOCTYPE should be included at all.
▲Examples of HTML used that isn't XHTML Basic below.
▲See [http://validator.w3.org/check?uri=www.wikipedia.com] for an even more picky analysis.
▲* img tag isn't closed - should be <img ... />.
▲* html lang, body bgcolor, img align and img border attributes aren't in XHMTL Basic.
▲* hr and font tags aren't in XHTML Basic.
▲* list items and paragraphs aren't closed: < li> ... '''< /li>'''.
▲* Some attribute values aren't quoted, e.g. type=text must be type="text".
▲* Some inline elements aren't contained in block-level elements, like the toolbar at the bottom of the page.
▲This does have actual effects on the pages: in Mozilla, the top hr element overlaps the logo, and nested indents (like on [[Carey Evans/Talk]]) don't work.
▲The pages also claim to be UTF-8 encoded XML (<?xml ...?> PI at the top of the page) while the HTTP headers say ISO-8859-1.
''I can confirm this. The problem with nested indents is particularly bad, as it makes some pages more difficult to read. -- [[Taral]]''
| |||