Template talk:col1
Add topicNeeds documentation. For instance what is the matching closing template? — hippietrail (talk) 02:56, 16 March 2015 (UTC)
The following information has failed Wiktionary's deletion process (permalink).
It should not be re-entered without careful consideration.
Now that {{ant2}}
, {{coord2}}
, {{der2}}
, {{hyp2}}
, {{rel2}}
, and {{syn2}}
are all aliases of {{col2}}
, and similarly for other numbers of columns, I suggest eliminating the lesser-used aliases. Following is a table of all the aliases and their uses (note that the #uses for the main template includes uses of the aliases). I suggest replacing and deleting all aliases with #uses < 1000 (the threshold I've been using for the need for deprecation). By this measure, we should keep {{der2}}
, {{der3}}
, {{der4}}
, {{rel2}}
, {{rel3}}
, {{rel4}}
and rename/delete the rest.
Benwing2 (talk) 04:30, 3 April 2019 (UTC)
- I support merging all the column templates into a single one. There is no rule or even guideline for choosing how many columns to use, it's left entirely to the personal preference of the editor, which results in different entries having different numbers of columns for no apparent reason. We should pick a number and make everything use that. —Rua (mew) 15:18, 6 April 2019 (UTC)
- @Rua OK. For now I think I'll just orphan the less-used templates and we can figure out later how to choose the number of columns automatically. Benwing2 (talk) 15:30, 6 April 2019 (UTC)
- I guess we should delete
{{col1}}
because it doesn't make the 1000-uses threshold. DCDuring (talk) 16:01, 6 April 2019 (UTC)
- I guess we should delete
- @Rua OK. For now I think I'll just orphan the less-used templates and we can figure out later how to choose the number of columns automatically. Benwing2 (talk) 15:30, 6 April 2019 (UTC)
- I'm sure there is a lot of arbitrary variation, but you're proposing to take away all flexibility so there's no way to deal with exceptional cases except by hard-coded kludges. There are factors in choosing a format such as the length of the terms (Chinese is very compact, an agglutinative language is going to be more diffuse), the number of the terms, the desired height and width to be occupied, and possibly others that neither you nor I is aware of. Sometimes two columns look fine and three look awful for no easily explainable reason. There's a fine line between a consistent "house style" and procrustean rigidity- "make everything use that" sounds a lot like the latter. Chuck Entz (talk) 18:35, 6 April 2019 (UTC)
- My point is that the need for such flexibility has never been explained, and the way that one should choose the number of columns has never been documented anywhere. Instead, it's arbitrary as you say. It is trivially easy to choose the number of columns based on the length of the terms, so that in itself is not a reason to favour manually choosing. It merely codifies and automates what we already did by hand before. But perhaps we are going a step too far, and should really codify the guidelines first. I invite you to participate. —Rua (mew) 18:46, 6 April 2019 (UTC)
- I think we should provide the capability of auto-choosing the number of columns (and probably encourage people to do so) but not force all users to use it. Benwing2 (talk) 02:47, 8 April 2019 (UTC)
- Isn't the "right" number of columns substantially dependent on the effective width of the browser frame. I don't get the impression that we are very good at that kind of thing. I have a fairly wide monitor, but I like having at least two browser windows visible and I like to be able to easily read our content without spectacles. Sometime the high-column-count tables are unusable. Until such time as we can make the number of columns auto-adjust to browser frame width, this exercise could be a waste of time. DCDuring (talk) 03:32, 8 April 2019 (UTC)
- Much of the unusability or ugliness in the high-column-count templates is due to the length of the longest word in a given table entry. If that length exceeds the available space, then the long word runs over into the next column, possibly overwriting format elements (eg, "*") or content. DCDuring (talk) 03:44, 8 April 2019 (UTC)
- @DCDuring Are you saying it's a waste of time to implement auto-choosing the number of columns? I guess I agree with you if we're doing it independently of the frame width, but maybe there's a way to make it dependent on the frame width using CSS. I know that other sites are able to do this. Maybe we could do something like create CSS classes "multicol-2-3-4" (which means "two columns with narrow frames, 3 columns with mid-width frames, 4 columns with wide frames"), "multicol-3-4-6" (similar), etc., and then have the module code compute the length of the longest word and choose the appropriate CSS class. Benwing2 (talk) 18:24, 13 April 2019 (UTC)
- You have obviously taken my concern on board. I appreciate that, but cannot be any help on technical aspects of implementation. Other sites, though not all of them, do seem to have resolved this kind of thing satisfactorily. The good news about solving this is that the solution will appear in many entries. The bad news is that it probably will take a while to address implementation issues, including interaction with things like the right-hand-side table of contents and other right-hand side elements. I expect that folks will have some different preferences about spacing between columns. DCDuring (talk) 20:40, 13 April 2019 (UTC)
- Perhaps it would be a good idea to use the
column-width
CSS properties instead of thecolumn-number
properties, and choose a column width that displays most of the terms to good advantage. It would probably be better to do it by hand rather than by module, since optimum width depends on the fonts that browsers happen to choose (though they may be nudged towards certain fonts by MediaWiki:Common.css), and a module can't adequately determine the widths of characters. If this option were taken, all the numbered column templates could be deprecated in favor of a single one with a width parameter, or maybe they could be replaced by column templates for narrow, medium, and wide columns. I use column width in the table of contents of my possibly incorrect header page. (It may not display correctly in all browsers.) — Eru·tuon 21:11, 13 April 2019 (UTC)
- @DCDuring Are you saying it's a waste of time to implement auto-choosing the number of columns? I guess I agree with you if we're doing it independently of the frame width, but maybe there's a way to make it dependent on the frame width using CSS. I know that other sites are able to do this. Maybe we could do something like create CSS classes "multicol-2-3-4" (which means "two columns with narrow frames, 3 columns with mid-width frames, 4 columns with wide frames"), "multicol-3-4-6" (similar), etc., and then have the module code compute the length of the longest word and choose the appropriate CSS class. Benwing2 (talk) 18:24, 13 April 2019 (UTC)
- Much of the unusability or ugliness in the high-column-count templates is due to the length of the longest word in a given table entry. If that length exceeds the available space, then the long word runs over into the next column, possibly overwriting format elements (eg, "*") or content. DCDuring (talk) 03:44, 8 April 2019 (UTC)
- Isn't the "right" number of columns substantially dependent on the effective width of the browser frame. I don't get the impression that we are very good at that kind of thing. I have a fairly wide monitor, but I like having at least two browser windows visible and I like to be able to easily read our content without spectacles. Sometime the high-column-count tables are unusable. Until such time as we can make the number of columns auto-adjust to browser frame width, this exercise could be a waste of time. DCDuring (talk) 03:32, 8 April 2019 (UTC)
- I think we should provide the capability of auto-choosing the number of columns (and probably encourage people to do so) but not force all users to use it. Benwing2 (talk) 02:47, 8 April 2019 (UTC)
- My point is that the need for such flexibility has never been explained, and the way that one should choose the number of columns has never been documented anywhere. Instead, it's arbitrary as you say. It is trivially easy to choose the number of columns based on the length of the terms, so that in itself is not a reason to favour manually choosing. It merely codifies and automates what we already did by hand before. But perhaps we are going a step too far, and should really codify the guidelines first. I invite you to participate. —Rua (mew) 18:46, 6 April 2019 (UTC)
- I'm sure there is a lot of arbitrary variation, but you're proposing to take away all flexibility so there's no way to deal with exceptional cases except by hard-coded kludges. There are factors in choosing a format such as the length of the terms (Chinese is very compact, an agglutinative language is going to be more diffuse), the number of the terms, the desired height and width to be occupied, and possibly others that neither you nor I is aware of. Sometimes two columns look fine and three look awful for no easily explainable reason. There's a fine line between a consistent "house style" and procrustean rigidity- "make everything use that" sounds a lot like the latter. Chuck Entz (talk) 18:35, 6 April 2019 (UTC)
- Delete - this can easily be solved with using CSS in the
{{col}}
template. This is not 1999. Theknightwho (talk) 10:59, 27 May 2022 (UTC)
- RFD-deleted: All originally proposed deletions have been performed. Discussion of further action (such as merging all related templates into
{{col}}
) has grown stale, and can be revived elsewhere. — excarnateSojourner (talk · contrib) 18:54, 16 November 2022 (UTC)