Talk:Main Page/Archive 20070630
Talk ('discussion' tab) pages -- signatures
Appending your signature when you add a comment, particularly when replying to something already written, makes it easier to identify who's saying what, a la web forums. --John Stumbles 11:59, 16 December 2006 (GMT)
Here's a reply to the above --John Stumbles 12:11, 16 December 2006 (GMT)
It would be nice if the wiki kept track of threads and increased the indent for each reply but it doesn't seem to do that (although some users on wikipedia make it seem to do so using <blockquote>tags around their text</blockquote>). --John Stumbles 12:02, 16 December 2006 (GMT)
This quickly gets messy if you try to hand-craft multiple layers of indentation as you have to put your own <blockquote>tags</blockquote> before the closing </blockquote> you're replying to.
Kids: don't try this at home!
Trouble with category pages
I added a [[Category:Plumbing]] tag to the Plumbing page but when I followed the Category: link at the bottom of the page I got:
Editing Category:Plumbing From DIYWiki Jump to: navigation, search You've followed a link to a page that doesn't exist yet. To create the page, start typing in the box below (see the help page for more info). If you are here by mistake, just click your browser's back button. Preview Sorry! We could not process your edit due to a loss of session data. Please try again. If it still doesn't work, try logging out and logging back in.
If I just 'Save page' with an empty page and again try to follow the link I get the same error. However if I create a page with some random text and save that I get correctly directed to the newly created Category page. I can then edit it and delete the random text, save again and I still get to the new page (which is what I want). Odd.
I find that what turn out to be the useful category headings are often not what I epxected them to be ahead of time. (I manage a few info libraries elsewhere.)
I uploaded a bunch of image files for an article then realised I didn't need some of them, but there doesn't seem to be any way to delete them.
They're shown in Special:Unusedimages
--John Stumbles 18:21, 3 January 2007 (GMT)
There are a few redundant articles could be deleted too. Whatever the method, it may be best to delete the how to after its been done! NT 19:14, 3 January 2007 (GMT)
How do you do a 'revert' then? Could be handy to know. NT 06:07, 4 February 2007 (GMT)
Go to history and find the version you want to go back to, go into edit and save it without making any changes to the text (but do fill in the Summary to say you're reverting it, and why, of course!)
--John Stumbles 12:15, 4 February 2007 (GMT)
Ohh - easy! Thanks. NT 22:08, 4 February 2007 (GMT)
DIY pages on wikipedia
There are some middling to bloody-awful DIY related pages on wikipedia which need some work:
- Do_it_yourself could do with UK-ifying and putting in some decent links (like to here!)
- Handyperson <groan>
Naming & Capitalisation conventions
Some time ago I put in the Naming section:
When referring to one article from another it helps to have a consistent naming convention. Perhaps we should follow Wikipedia's naming convention [] especially regarding capitalisation of article names and whether they should be in the singular or plural []
However no-one (least of all me :-)) seems to be doing that so maybe we should change this to reflect actual practice which seems to be to Capitalise All Initials and use plurals for subjects which are a class of things e.g. Round Tuits
I agree, for one reason or another it hasnt been followed, and its less confusing to stick with what we've got, so thats what I do for now. NT 05:27, 15 February 2007 (GMT)
OK I've changed it to reflect more-or-less what we've got now, although what we've got isn't consistent as to whether we capitalise all words in an article name or just the main ones, but wtf, life's too short :-)
--John Stumbles 11:25, 15 February 2007 (GMT)
Taking a local copy of the Wiki content
Don't let's everyone do this all the time for the sake of the server but this saves copies of most of the useful & interesting pages. It actually fetches all of the pages then throws away a lot which makes it not very server-friendly.
As well as HTML-formatted article pages it also keeps edit and version pages:
- from edit pages you can re-create an article directly from the markup contained in the page's edit box.
- version pages show the evolution of articles
wget -w -r --convert-links http://wiki.diyfaq.org.uk --reject '*§ion=*','*=Special:*','*&diff=*'
The --convert-links option is supposed to make links within pages work as local copies but that doesn't seem to work for me: urls are of the form
Maybe that has something to do with it?
The --reject option prevents wget keeping various pages . The following arguments are regular expressions matching file names:
- edit pages for sections of articles (only the edit page for the whole article is kept)
- [[Special:Specialpages|Special] pages
- diffs between versions
My wget version is
GNU Wget 1.10.2
Import or Link to Wikipedia content?
Where Wikipedia has an article on something of interest to us here the question arises whether to link to it or copy it.
- I suggest that where the Wikipedia article covers everything we'd want to cover (or could be expanded to do so within Wikipedia policies and guidelines) we should link to it (and contribute to it if necessary). The problem if we were to copy it is that as the Wikipedia article is developed, expanded, corrected etc our copy gets out of sync, left behind and stagnant (or vice versa if it's our copy that gets the attention). It seems reasonable to copy an extract of the Wikipedia article as a taster or sample of the target of the link since people (at least ones like me) tend to think twice about following a link unless they're really interested, and/or find an article consisting solely of a link to something else a bit unfinished.
- Where we should be importing content is as a basis for developing an independent article of our own, where it would be inappropriate in terms of Wikipedia's policies to develop their article in the direction we want to go. For example there's a Wikipedia article on DIY which discusses the rise of DIY in the USA and suggests that it is a return to once-common practice: "since the beginning of time, people have used their own abilities and available tools and technologies to take care of their own needs, make their own clothing, and so on". We might be interested in developing this idea with reference to Barry Bucknell, Nicholas Saunders, John Seymour, uk.d-i-y and Uncle Tom Cobbleigh and all. However Wikipedia's Guidelines for writing high quality articles state:
- Articles should contain only material that has been published by reputable sources. Editors adding new information into an article should cite a reputable source for that information
- Articles may not contain any previously unpublished arguments, concepts, data, ideas, statements, or theories
so this sort of material (including the Wikipedia DIY article as it stands, IMHO!) is not suitable for inclusion in Wikipedia since it expounds the authors' own ideas rather than reporting cite-able external sources.
- In many cases there's a Wikipedia article telling us more than we probably want to know about something which we've got an article about (where our article is unsuitable for Wikipedia). A link to the Wikipedia article can be useful for those interested: see e.g. our Limescale article.
--John Stumbles 23:35, 21 March 2007 (GMT)
That sounds like a good summary to me John. Import as I see it for articles where we want to start with what wikipedia has, but develop it in a different direction, and/or cut it down.
The only problem with this approach is that we currently have so many large holes in our article repertoire precisely because various subjects have been covered well elsewhere. If in such cases we never import, this will probably never become anything like a complete diy wiki.
What is the solution to this issue? I dont know. All the following options have problems afaics:
- Redirect to wikipedia pages or other external links in such cases
- Import wikipedia articles (lack of updating from wikipedia)
- Write an article from scratch (who would rewrite things and why)
- Have near-empty articles with just an intro and a link to wikipedia or other site (works, but doesnt look good)
Cant think of other options offhand. NT 02:59, 22 March 2007 (GMT)
I suspect we'll find that wikipedia articles are more general and theoretical and that we'll probably want to start with a reference to their article but then get into the practicalities in an article of our own. I think the Limescale article was a good example of that: their article covers the chemistry of it in great depth, ours covers how it affects us and what we do about it.
Of more concern to me in terms of overlap is the DIY Wikibook which is setting out to do pretty much what we are doing (though naturally with a more global - read USAnian! - perspective). However theirs seems to be more wide-ranging (fish breeding, gauss guns ...) and I see us as having a much narrower but deeper concentration on serious UK-based 'domestic' DIY.
--John Stumbles 11:57, 23 March 2007 (GMT)
With a total of just 1 article with useful content, I'm not sure we need worry too much about wikibooks.
I suppose there is no clear universally applicable answer to the question of what to do when something has already been covered well elsewhere. Maybe I'll import one good wikipedia article partly as a practical experiment & see how it works. Hopefully the 2 will develop in different directions, as our wiki and wikipedia do have different audiences and aims. NT 15:16, 23 March 2007 (GMT)
Wiki articles suffer from an age old bugbear: the pulls in both directions as to what level of information to include in the article. More info is useful for some, yet makes the artcle much less accessible for many.
Since at least 2 of our articles have reached this point, I'll try splitting one into 2 articles, one that speaks to the main points, and the other for more on the subject, in the hope of achieving both aims rather than only one. If I get the time! NT 15:19, 23 March 2007 (GMT)
Yes I find that's what happens e.g. with the Central Heating Design article. As it's filled out I've split off one section and will do others (Round Tuits permitting). Then the main article can be an arm-waving introduction to the main principles and the sub-articles can get into the details.
--John Stumbles 11:40, 24 March 2007 (GMT)
I've just moved a link from this page to one more specific to the subject (Energy Efficiency, in this case).
I suggest that the links on this page should be to general DIY subjects which don't fit into a narrower, more specific category.
Does this seem reasonable?
--John Stumbles 18:11, 29 March 2007 (BST)
Sounds good to me, else we'd grow an enormous page here.
To some extent it seems we're feeling our way on this wiki on some points, with a few different things tried along the way.
I'm happy to say importing a wiki page seems to have been successful so far. The 2 were tweaked quite a lot to fit the purpose of this site better. NT 23:37, 29 March 2007 (BST)
--John Stumbles 01:55, 4 April 2007 (BST)
Categories (and sub-categories?)
(Moved from Talk:Shelving Units)
Shelving units are more or less always made from wood or metal, and are thus woodwork & metalwork. I'm sure more articles in said categories will turn up later on. So I dont see a reason to remove these categories.
Why the terms metal & wood instead of metalwork & woodwork? Because there is more to each subject than -work, there is also sourcing and choosing the materials, and asstsd other matters relevant to the materials that would not really come under the headings woodwork & metalwork NT 09:54, 22 April 2007 (BST)
I would be inclinded to categorise materials (wood, metal, plastics etc) separately from techniques for working. There's probably more in common between working with soft metals, hard plastics and medium-hard woods than there is between working with two different woods such as oak and balsa. So maybe we should have categories for 'wood' and 'woodwork[ing]' etc, and articles on, say, drills could be in categories woodwork[ing] metalwork[ing] etc. I think we might also be able to have a heirarchy of categories so, say, 'woods' and 'metals' could both be sub-categories of 'materials'. Wikimedia commons uses such a scheme for categorising pictures (and other media) so a picture of a penny-farthing might be in category 'antique bicycles', and that category is itself in categories 'antiques' and 'bicyles', and so on ...
What's the point of this all? Well I was looking at list of Special:Categories: it already scrolls off the bottom of my fairly hi-res screen and so I was wondering if it could be made more compact by reducing or grouping together categories. What do you think? Is it worthwhile even trying to do this?
How this relates to shelving units I don't know: since you discuss choice of woods I was probably wrong to remove that category (so I've put it back). Since the article isn't about choosing types of metal to work with in the same was as for wood I'd be inclined to leave that out but I'll leave that call to you.
--John Stumbles 13:11, 22 April 2007 (BST)
I cant think of any problem with there being many categories, if and only if the articles in the categories are relevant to their categories. I know this place is small now compared to 'pedia, but it probably will grow, and I daresay many of the now busy categories had only one entry for a while.
Categories are just one of the ways users can find information on here, and will grow, so personally I wouldn't want to remove valid cats.
You raise a fair point about plastic, metals and woods, yet we need to categorise them all somehow, and I think we need end users not to have to do work to undertand our category system, just so they can use it to find things. Hence I've been trying to stick with a simple, visible and immediately understood system.
I guess there will always be some amount of differing ideas on what categories to have, but I dont see a big problem at this earlyish time in having semi-free-range categories, and article contributors can choose to use them or not.
"What's the point of this all? Well I was looking at list of Special:Categories: it already scrolls off the bottom of my fairly hi-res screen and so I was wondering if it could be made more compact by reducing or grouping together categories. What do you think? Is it worthwhile even trying to do this?"
Personally I'd rather fill the useful categories up with links to relevant articles. In fact I can already think of another article that qualifies for the noise reduction category.
A busy categories page could be added if it is thought it would help. NT 16:41, 22 April 2007 (BST)
What about the heirarchical categories idea? Worth trying? --John Stumbles 17:36, 22 April 2007 (BST)
Very logical, but doesnt it hide some of them from view when you look at the Special:Categories page? If so, single layer cats might make it easier for users to find things, both in a specific search and much more so in a 'lets see what looks interesting' search.
Not that it makes a great difference either way. NT 19:37, 22 April 2007 (BST)
OK I've tried it. It doesn't hide sub-categories from the main Special:Categories which I had hoped it would :-( It's a pity that page doesn't show a tree-like representation of categories e.g.
Tuits + Round Tuits + Square Tuits
etc, but since a sub-category can be a member of multiple higher-level categories that wouldn't necessarily be very useful.
And maybe we should rename Category:Adhesives to Adhesives & Sealants - ?
--John Stumbles 11:09, 23 April 2007 (BST)
"Question: do we want to put articles in the parent category as well as the child one? E.g. Silicone Sealant and Putty & Mastic are in both Materials and Adhesives."
Well, if we dont, clicking the main category fails to show up many articles in it. If we do, the subcats just narrow the search down, so I suppose we ought to put em in both. What do you think?
- I like the Commons way: putting each article into only the most specific sub-[sub-[sub-]]]categories rather than having a string of overlapping categories in each article. However this doesn't always work: I just moved some articles that were in category:Domestic Hot Water out of category:Heating, but things like solar heating and heat banks can be DHW or non-DHW heating so need to be in both. In the end I decided to get rid of DHW and just have heating: the category's not so big it needs sub-dividing (yet!) --John Stumbles
"And maybe we should rename Category:Adhesives to Adhesives & Sealants - ?"
I sense a slippery slope though. Where is the dividing line between sealant and paint? AFAIK there isnt one, paint is handy as a sealant. Yet glue and paint are very different - even though some paints can be used as glues.
Would separate cats/subcats of adhesive & sealant work? With many items being in both cats. Not that its something I'm too worried over :) NT 13:51, 23 April 2007 (BST)
I'm inclined to go for adhesives and sealants as one cat. Partly because that's how Screwfix and Toolstation categorise them! And there are many products which advertise themselves as both.
It probably all seems like a fuss over nothing, but looking at the dog's breakfast of categories WP's got I'd like to try to keep ours sane!
--John Stumbles 23:56, 23 April 2007 (BST)
I dont honestly think its going to work John. Article contributors / editors will do their own thing - this wont be our corner forever. And every person that comes here has their own idea about categories they will use.
With article content its posible to check and verify or refute information, but your or my idea of categories is no more or less valid than any other reasonable idea.
I think the best option is to accept the free aka dogs dinner approach, and build up those categories we think of value by tagging all relevant articles with those cats. To do otherwise is only going to cause trouble, and wont lead to an agreed structure for long, if at all.
Wikipedia has been through all this, and I think the present situation with it is a lesson we could learn from. There are reasons it is as it is. And it does work.
If you want a neat clean categories page you could create one with only those cats you think valuable. Maybe call it Popular Categories or something. NT 11:56, 27 April 2007 (BST)
Yes, I can't see any brilliant way of categorising things, I guess we should let it evolve --John Stumbles 20:43, 27 April 2007 (BST)
If there are cats we think are good ones, we can create them and tag all appropriate articles with them, but I'm more wary of deleting cats others create. That should swing the balance some, toward less unnecessary (in your/my/our view) extra cats. NT 18:28, 28 April 2007 (BST)
Might there be any mileage in protecting this page? NT 18:57, 28 May 2007 (BST)
Not much point, they would only go scribble on something else. At least its easy to keep track of the twats here.
--John Rumm 20:18, 28 May 2007 (BST)
Reformat glossary pages?
The glossary pages currently do not have the word entries configured as headings. This seems to make it impossible to link to them such that the entry you are linking to is automatically positioned at the top of the page. Anyone know if there is a way round this, or do we need to make every entry a heading? (or do an alphabet heading for each section perhaps)
--John Rumm 20:23, 28 May 2007 (BST)
If we did headise each word it'd double the article length. That's why I initially did it similarish to how it is now.
Do we really want to link to words in the glossary? I never have as I suspect people would get fed up following links to simple basic minimal info stuff. I try to make each link go to something with more info than that.
We could list the appropriate subject glossary as a 'see also' link for each article. We've only got one subject glossary so far, but over 40 articles in it already. I reckon that would work, what do you think? NT 23:19, 28 May 2007 (BST)
Am tempted to start using 3 month blocks for spam first offence. What do you think? NT 23:47, 17 June 2007 (BST)
There's only been one spam from an address which was blocked and the block expired, and lots of blocks on addresses have expired and no further spam's been had from them, so I think 1 month is probably OK. --John Stumbles 18:48, 19 June 2007 (BST)
OK, didnt know that. Cheers. NT 23:21, 19 June 2007 (BST)