Talk:Tree traversal

Wiki markup problem :Category v. Category references

Note: This documents a markup glitch or documentation error or non-intuitive feature in the markup. Not sure if this is where this should live.

Problems observed in the following:

...
== Icon and Unicon ==
==={{header|Icon}}===
<lang Icon>procedure main()
...
end</lang>

...
Note: A [[Category:Unicon]] specific example has not been provided.    <=== SIDE EFFECT THIS CAUSES the task to show up in tasks completed Unicon

I diagnosed the side effect - Tree Traversal shows up as a completed task under Unicon.

• There is no {{header|Unicon}} tag anywhere on the page.
• The [[Category:Unicon link]]produces nothing visible unlike other [[..]] tags. This also causes the problem.
• Replacing the reference above with [http://rosettacode.org/wiki/Category:Unicon Unicon] doesn't cause the problem.

When you want to make a visible link to a category you need to put a : in front like this: [[:Category:Icon]] or [[:Category:Icon|custom text]]. These will show this: Category:Icon or custom text. Is that what you were trying to do? --Mwn3d 02:51, 13 April 2010 (UTC)

Yes I was but that what's odd is that Template:Example-needs-review and Category_talk:Icon w/o the :

Also, thanks with the : the Side effect above seems to be gone --Dgamey 03:48, 13 April 2010 (UTC)

"Don't masquerade as being Pythonic"

Some slightly silly suggestions being made here in an oddly threatening tone. 'Masquerade' ? and on my user page "underhanded and subterfuge on your part that is aimed at confusing the readership as to what constitutes idiomatic Python" What ?

These eccentric complaints do leave me just a little puzzled ... (and, incidentally, also tempted to suggest that you apply a good linter to the slight opacities of your English, but that is neither here nor there) Hout (talk) 03:18, 8 March 2019 (UTC)

I do understand the huge importance which you attach to the 'Pythonic' subset of Python, but I have to say that beyond being grateful for the PEP linters, which I always apply, I am not over-interested in it per se, and I have certainly never claimed to be modelling or demonstrating it a purist version of it. Nor am I remotely interested in steering any future developments in the 'Pythonic' dialect of Python, the core traditions of which are predominantly procedural.

I do like Python, and, like everyone, I find its libraries very useful indeed.

I also like functional programming, by which I mean constructing code, wherever possible, by the composition of pure functions, minimising the use of mutation.

I like to code in this style (the core discovery of which is that 'immutability changes everything', and that math works better as an ally than as an opponent – the rest is commentary. Go and study it) because I find that it helps.

In my context, writing and composing pure functions enables:

1. Faster and simpler coding and refactoring
2. Higher levels of code reuse
3. A little bit more reliability

I am, however, aware that this style of coding requires the acquisition of a few more concepts than the imperative mode, and can look unfamiliar (and even alarming or discouraging) to those whose habits and preferences are more solidly procedural.

(It is well known that there is sometimes a bit of tension (and some feelings of de-skilling) in work-places during transitions towards more functional styles)

I am also aware that your own interest in Rosetta code is very much built around your oft-repeated proposition that it is all about being idiomatic.

My own interest in it has a slightly different focus. I don't think of RC as a parade-ground or a museum for modelling spotless uniforms. (I feel that automatic linters can take care of major lint).

Instead my personal view is that presenting solutions to the same task (in as many different languages and idioms as possible) provides insight into how languages are similar as well as different, and helps learners with a grounding in one approach to a problem to learn another. (See the RC landing page formulations).

We can disagree on that core focus, and there is absolutely no need for us to persuade each other. We can just submit different code.

I am not quite sure why you feel that I am aiming to 'flood' RC with examples in order to mislead readers about exactly what the Pythonic subset is like. To be honest I am not terribly interested in why you might think that, just as I was not terribly interested when you were (equally eccentrically, in my view) trying to persuade people, on these pages and on Reddit, that my code was automatically generated by a secret transpiler project :-) (A Haskell to Python transpiler ? Good grief ... whatever for ?)

I am writing Python code as composition of pure functions, because I find that quicker, more reliable, and more interesting. I always use the PEP linter, but I am certainly not claiming to be modelling or restricting myself to the narrowly 'Pythonic' subset of that language, which is not intended to be optimised for functional programming anyway, and PEP8 sternly warns us against the trap of 'foolish consistency' (the rigid confusion of quality with compliance).

I dare say that you may partially believe your conspiracy theories (secret transpilers, 'masquerades' and 'underhand subterfuges'), and I have no particular interest in weaning you off them, but I must ask you to refrain from using them to authorise campaigns of bullying and harassment. I am not, oddly enough, a demon, and by somehow persuading yourself that I am, you are, I fear tempting and authorising yourself to behave more than a little disgracefully.

Give us all, and yourself, a break

If you don't like my code (for example for tree traversals on this page) then just write different code. Mine works well for me, and I hope it contains some abstractions which you and others might find useful or interesting too, but if not, that's fine.

I'm sure that your code works well for you.

Quality is a function of optimisation for context, and context varies.

Go and write good code for your context. Ranting does Python no good, and even the most rigid style-sheet compliance doesn't fix bugs or make bad code good.

Enough already ! Hout (talk) 03:18, 8 March 2019 (UTC)

Aigh ... more of this bullying grafitti and strange non-sensical labels.

'Multi-language' ? What does that even mean ?

Stop it ! Hout (talk) 03:23, 8 March 2019 (UTC)