examples mix keyword stems and runic stems

[urbit]

examples mix keyword stems and runic stems

[galenwp]

I agree that these should all be the same. Why not just make them all show both styles?

[chc4]

Maybe something like how the React website does it? Have a text box with both, and be able to switch between them via tabs.

[cgyarvin]

I'd love any of these kinds of solutions! No need for anything super fancy...

Sent from my iPhone

On Jul 30, 2016, at 12:23 PM, chc4 [email protected] wrote:

Maybe something like how the React website does it? Have a text box with both, and be able to switch between them via tabs.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

[galenwp]

I say let's start with examples mirrored in both syntaxes. Doing something mirrored should be as simple as adding a component to tree and just wrapping the two examples in it.

@hidwes-matped, would you be up for working on this?

[hoclun-rigsep]

@galenwp if I knew exactly what you meant by 'mirrored,' I would work on this.

[galenwp]

For example, one instance that uses :door and one that uses |= in successive codeblocks.

~zod:dojo> =foo |=(a/@ +(a))
~zod:dojo> (foo 20)
21

~zod:dojo> =foo :door(a/@ +(a))
~zod:dojo> (foo 20)
21

Or did I misunderstand your original thinking?

[hoclun-rigsep]

No that's just fine, I can do that.

[galenwp]

Great! We're here to help if you need it.

[hoclun-rigsep]

On the supposition that this is the appropriate place for it, here's what I
think about the runes and keywords, based solely on one man's experience
throwing himself into hoon for the past two weeks.

I believe keyword forms should never appear in code blocks or example
code. The keyword forms scream 'forced compromise.' They introduce
another 'sigil,' :, when we already have $word, %word, and
&word (am I missing any?). They affect the appearance of indented
code vis-à-vis runic code because they're—usually—five characters
instead of two.

The stem names are still a good thing, because many of the names are
sexily allusive and when I see |_ I already (sub)vocalize door (not
barcab, sorry). In :talk we are typing 'door' and 'gate' at least as
much as we type |_ and |= (but I have yet to see anyone in :talk use
:door when typing code). It already seems to be the case that some
of the stem names are destined to be memorable and freely referred to by
name in :talk—especially the | family and some in $—and some not and I think
that's fine.

My mental monologue goes something like this:

There's this thing called a door, which is spoken 'door' and
spelled 'door' in conversation and prose, and |_ in code.

I imagine you guys already drank a barrel of Anchor and had it out over
this issue, but if you wanted one newb's perspective, here it is.

PS: Going through the docs and editing them has intensified this view.
There are so many indispensable irregular forms that nobody is ever going to be
writing in pure keywordish. There were several examples that I didn't
translate to keyword because I got the sense that nobody would
ever read them.

[juped]

The truth is, nobody likes them but Curtis. But, might I plug http://urbit.org/fora for discussions? (Sign on at http://[your-ship].urbit.org/ to post under your own name on urbit.org.)

[cgyarvin]

Ha ha, you try playing god-emperor sometime. Everybody thinks for some
reason it must be fun. It's no fun being right when everyone else is
wrong, I can tell you that.

hidwes, everything you say is basically true. (Although people who learned
without the keywords do say, internally and externally, "barcab" instead
of "door.") This is in general a really insightful comment and reminds me
of the reality that basically, anyone involved with Urbit at this point is
going to be someone who could have invented it, but just accidentally
didn't. I also could have invented it and just accidentally did.

I have said that on balance, the keywords make Hoon a slightly worse
language, because I think it is more righteous to say "barcab" than "door."
On the other hand, you can talk about "a door", the core, but you can't
really talk about "a barcab."

I think of keywords as serving more or less the function of pinyin in
Chinese today. China is not about to throw away its characters and switch
to pinyin, for a lot of excellent reasons. On the other hand, children
learning Chinese (even Chinese children) start with pinyin. When teaching
Hoon, keywords look way less hard than runes, even though they're not.
In actual fact, it's not really easier to bind keyword to rune to meaning,
than just rune to meaning.

But it just looks easier, from the outside, before you actually try it.
There's something really quite autistic about the pure rune-to-meaning
binding. Anyone can handle it -- but most people don't think they can
handle it. Sadly, marketing matters.

On Sun, Jul 31, 2016 at 6:40 PM, Raymond Pasco [email protected]
wrote:

The truth is, nobody likes them but Curtis. But, might I plug
http://urbit.org/fora for discussions? (Sign on at
http://your-ship.urbit.org/ to post under your own name on urbit.org.)

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/urbit/docs/issues/83#issuecomment-236471697, or mute
the thread
https://github.com/notifications/unsubscribe-auth/AALyAUchl5RSueh01i5KLj8ZXHf0EeDDks5qbU6ogaJpZM4JY3nj
.