Report (Wub)

Wub utility for … reports, presumably.

CMcC no. a general utility for generating HTML tables which happens to be distributed along with Wub - now you can get it directly. Yay.

(The paragraphs CMcC originally posted on the wiki was the same, links and everything, as for color (Wub).)

CMcC That was an oversight.

I moved it back to Report, adding some verbiage as to what the original content described (a tcllib utility) and disambiguating the Wub analogue by using the word also.

FWIW:

  • I don't think category suffixes in titles by means of trailing parentheses is good form.
  • I also don't think splitting pages so each page can be neatly categorised is necessarily a good thing.

Lars H: OK, then — it seems I'll have to spell this out.

(It would perhaps have been slightly more in tradition to use the author's surname than "Wub", but that was the first thing that came to my mind.)
  • This practice has nothing to do with categorizing pages — on the contrary, many such pairs of pages belong to the same sets of categories. The reason for having several pages is instead the need to keep the topics apart! Had a single page been used to discuss both mysql binding (Smith) and mysql binding (Klaren) then little but confusion would follow, as they probably weren't interchangable.

Tcllib's report and report (Wub) obviously aren't interchangable, but how would you expect people to keep their respective advices apart if you insist on keeping both on the same page? There's little problem as long as one is just a "there's also" note, but presumably you expect more to follow.

CMcC it seems that you wouldn't have to spell out a truly established practice. It seems that one man's established practice is another man's mono-structual monomania. Don't get me wrong, I think wiki gardening is excellent and useful, but when it leads to camping on well-known and evocative nouns (such as color or report) I think it vivisects the discourse and doesn't foster it.

By insisting that titles unambiguously and uniquely identify their content, you (a) reduce the possibility of productive/creative/accidental linkages, (b) duplicate information which Categories facility is supposed (or should I say alleged) to provide.

I see your subscripted-() examples, I wonder how many of them you are personally responsible for, and I raise you XML. HTML and many other pages which are topics but not yet (and perhaps not ever) categories. It is fitting, in my opinion, that a utility to handle color is on or around the color page, fitting that a utility for web report generation is on the report page. As it would be fitting if report contained something for generating pdf reports, and/or a series of musings on the nature of reports in general.

I propose a distinction between topics and categories and suggest that titles represent topics and not categories.

LV Actually, I expect that I am probably responsible for a number of the pages mentioned.

My opinion on the topic, is that if I were writing about two or more divergent entities today that happen to share the same name, then I would likely start out with a page mentioning each of the meanings, and generate a link to a seperate page for each for specific details. That way, the average user who searches for just the word will find a page that briefly mentions where to find the detail about what s/he is researching, and may provide a serendipous moment of discovery upon reading the other meanings on the same page.

Lars H: Calm down, CMcC — your jumps to conclusions and fixations on claims noone has made are starting to remind me of the Unperson! (Come to think of it, this streak of creating a series of pages with almost the same content does so too, although one can infer from the brevity of the titles that it wasn't Him in diguise.) No need to get all manical about the matter.

The wiki is fairly large, and some practices are less common than others, so I wouldn't be surprised if there are some that I haven't noticed, but that's no reason to get all cranky when someone else points out that the practice exist. I knew page-pairs such as those listed above existed, but I still had to search around for quite some time before I could exhibit examples, and no, I didn't create any of them. It still struck me as the natural way of handling the content that you added.

As for the utility of a disambiguating (not necessarily categorising) suffix, one may also take a look in Wikipedia. There we have, as separate pages:

By the same reasoning as you put forth above, it may be argued about each and every one of these things that they should be on the main http://en.wikipedia.org/wiki/Meridian page, from which would however follow that all of the material on these things should be on the same page. As that is a patently absurd conclusion, the argument is flawed, even if it may seem "creative" to link Meridian, California with Meridian (train).

CMcC I don't believe I jumped to any conclusions, nor fixated on anything.

This wiki is not Wikipedia. Nor does it want to be. Wikipedia is not paradigmatic of a Tcler's Wiki.

There is no sound reason that practices found necessary and useful on WP would necessarily be useful or beneficial here. You have given examples from WP, but have not shown that similar circumstances exist here, nor have you given any substantive argument that WP's methodology is beneficial here. You have given no examples here of situations where we could all agree the principle you promulgate (and had claimed was standard practice here) has been soundly applied or deserving of extension to other cases.

Argument by analogy is inherently a weak form of argument.

If I may be indulged, to prosecute this argument more fully: one of your examples, Lars, is Miscellaneous Tcl procs (Dillinger) vs. Miscellaneous Tcl procs (Corey) vs. Miscellaneous Tcl procs (Kirsch).

I would like to illustrate the folly of this by reference to some synonyms of 'Miscellaneous': assorted, mixed, motley, sundry, consisting of a haphazard assortment of different kinds. So what your methodology suggests we do is take a motley, haphazard collection of sundry Tcl procs, and break them into smaller motley haphazard collections by name. Is entropy really thereby decreased, or have we just taken one haphazard assortment and replaced it by three haphazard assortments classified by an unimportant piece of information?

Additionally, no methodology would justify camping on color.


AMG, waaaaay off topic: It appears that the Wiki will include an opening parens in the URL but not a closing parens. Hence I have replaced the closing parentheses with %29's. How odd. Look at https://wiki.tcl-lang.org/_revision/22417?V=7 to see what I mean.