The Oxford Guide - Differences between Version 55 and Version 54 of Comments
|Version 55||Version 54|
|== Line 2 ==|
* Not that I know of. I actually disagree in any case; it's a useful way of distinguishing between internal and external links, so I'd rather the convention was kept. --[[Dom]]
Square brackets around links to external pages look ugly. Is there any chance of preventing them from being displayed? -- James
I was wondering if the Clubs category should be renamed Nightclubs, or if it's unlikely that any other sort of club will appear. :) --Art
- It's true that Clubs is potentially confusing as it could be used to mean University type clubs (ie societies). --Dom
James: My spelling is bad I know. Is it possible to add some sort of spell check? Even if it's just a list of mispelled words so I can go back and re-edit the entry?
- Not my itch; patches welcome :) --Kake
Dom: Perhaps we need specific metadata entries for URLs to other guides/reviews?
- Could do, yes — I mainly worry about making the edit form overcomplex and huge. But this is something the RDF people are interested in, so I'll probably add it. --Kake
James: How about offering the wanted pages ordered by the number of times they are wanted as well as by alphabetical ordering?
- Oh, nice idea. That'll require a patch to CGI::Wiki, so I'll have a think about it. --Kake
James: Can the pages please be validatable HTML of some form or another? I'm trying to write a suitable stylesheet and it would make life a little easier, it's just a few loose </P> tags I believe. Should the <div id="... be <div class="... ?
- Can you make me a list of the specific places where things need changing? Also, 'id' is a valid attribute for the 'div' element - it's for things that you only expect there to be one of. http://www.htmlhelp.com/reference/css/structure.html might be of help. --Kake
- I think the problem is that <hr /> and <<form> ... </form> aren't valid inside paragraph tags, you can spot this on most of the pages as far as I can tell. I could be wrong, but I think that's it. Thanks for the info about the id attribute, that cleared things up for me. -- James
James: Is it nessecary to provide both a map link and the OS co-ordinates? In the current situation if streetmap etc. make a slight change to the formatting of their URL this means that every entry needs to be manually corrected, whereas if we generated the map URL from the coordinates (streetmap provide instructions at http://www.streetmap.co.uk/linkto.htm) only a single change would be needed to the code.
- The problem is that Streetmap have been known to be arsey in the past about software that automatically generates links to them. However, it was slightly more complicated in this particular case because the software itself was doing lookups to streetmap. The fun starts on the london.pm mailing list. Personally I'd be happy with doing a simple conversion as documented on their own site (I didn't notice that page before) but if Kake isn't happy with it this'll require me or someone else to maintain a patch to OpenGuides and this will need some time spent to familiarise myself with the codebase. --Dom
- I'm not sure how using their advertised method to generate the URL is any different (from streetmap's point of view) than having manually generated URLS linked to them. IMHO, it should be okay. The london.pm case appears to be slightly different as they were actively accessing streetmap systems to generate the information. James
- I've now added this to the Oxford installation. If a map link has been supplied, this will be linked; if not, and complete location data has been supplied, one will automatically be generated using the specification mentioned above. Patch submitted to Kake for inclusion; also available as OpenGuides-automaplink.patch temporarily. --Dom
I've changed the navbar to be horizontally laid-out. Is this better or worse? --Dom
- Better although a little padding between the navigation links and the search box would look a little nice, it seems a bit cramped up there. Perhaps better in the long term would be to define the navbar as something like
<div id="navbar">which should give the designer of a style sheet more flexibility over the layout. --James
<div class="navtool"><a href="http://www.ox.compsoc.net/oxfordguide/">Home</a></div> ...
<div class="navtool"> ... </div>
- I've done what you suggested except that <span> is appropriate here, not <div>. I've left the pipe symbols in as separators for the time being; if someone can suggest a better way of doing this so that it'll fall back to something readable without css, let me know. --Dom
Information about businesses etc needs a "last checked" field. Otherwise people won't know how reliable it is - this will particularly be a problem for the opening hours and (less so) phone numbers, from experience with another system that records this kind of information. --Ganesh
HTML fixes: I've made an & -> & change to fix a validation bug in the main node display (in the link to the backlinks page); the other main bug is that line breaks end up being delimited by <p> tags, which is invalid. This probably needs some special logic in Text::WikiFormat (maybe? haven't delved into the code enough yet). --Dom
- I don't like the box around the front page, or the rendering of the HR's in Mozilla. -- James
- I've fixed the HRs (made them smaller) but I don't see what the different between the box on the home page is. Okay, I have just added a navbar to the home page, is that better? -- Dom
- Yes that looks nicer although I feel the html needs tweaking still. The content DIV's also appear to be identical in scope to the body of each page which feels redundant to me. Possibly there should be different div's to wrap around the content and metadata. I could e-mail you an example of my ideas if it's of any help. -- James
See also Comments Archive.
List all versions