Re: GDP: chattiness in @seealso

2007-11-16 Thread Graham Percival

Mats Bengtsson wrote:
I'm not sure what the conclusion is, but maybe we can reformulate the 
original question into:

- Should we introduce some new concepts under @seealso or should this
 always be done in the main text?


Let's go with "main text".  I'm not completely dead-set on this option, 
but I don't think anybody is going to seriously argue against it.



- For issues that are somewhat peripheral and just need a short intro and a
 cross reference, should they be described in running text or in an item 
list?


The problem with an item list is that it raises questions about anything 
which is _not_ on the list.  For example, @ref{Ties} doesn't appear in 
the item list in Durations.  And if we move all the explanations away 
from the @seealso, it would seem really weird to include @ref{Ties} in 
the main text, then again in an item list in the main text, then _again_ 
in the complete reference list in @seealso.


So right now my vote is to describe anything which needs describing in 
the main text.


- Under @seealso, can we find a nice layout that allows for a mix of 
pure links
 which already have self-explanatory names, with links that need a 
sentence of

 explanation?


I can't think of any /nice/ layout.  A list that alternates 
full-sentences and single-word references is going to look weird.  A 
typical comma-separated horizontal list would still look slightly 
awkward even with the parenthetical remarks.



With all that in mind, I'm proposing that we move all sentences into the 
main text, and have a simple

@seealso

Notation Reference: @ref{foo}, @ref{bar}.




- Is there a need to repeat all links from the main text also in @seealso?


Perhaps not a need, but I feel a strong desire to do so.  It's a simple 
enough job for the Formatters.




A side comment: in "Durations" under @refbugs, the term "glyphs"
is used several times. Is this is term that is well-known to all 
readers, or

is it only known to hackers?


I deliberately decided not to spend time agonizing over this.  I just 
dumped Han-Wen's email in that section.  As a general rule, the @refbugs 
will be more technical, harder to understand, and may possibly be a 
simple copy&paste from a hacker's email or bug respose.


Making @refbugs easier to read might be something to schedule in the 
third round of GDP, but certainly not before that.  (first round: making 
GDP presentable again.  Second round: making GDP significantly better 
than 2.11 in all respects.  Third round: perfection)


Cheers,
- Graham


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


Re: GDP: chattiness in @seealso

2007-11-16 Thread Mats Bengtsson



Graham Percival wrote:
At the very least, I want it clear which sentence refer to the 
Notation Reference, and which sentences refer to the other parts of 
the docs.


... I _really_ think this is completely unnecessary, though.  And if 
you want to add full sentences to every single notation reference 
@ref{}, I assume you want to do the same for every @lsr{dir,snippet}, 
every @internalsref{}, etc ?



Mats, you're the yardstick for efficient NR use.  What do you think of 
the compact vs. full sentence form of @seealso ?  I don't want to 
approve any change that makes the NR harder to use for knowledgeable 
users, and IMO this is one such change.
In my opinion, the main issue is not how chatty the @seealso is but 
rather how

much information you include and where you place it. Regarding your current
version of "Durations", I think it's great to mention about and 
reference to all these
issues that are related to the main topic of the section. One question 
is if this information
should be below @seealso or in the main text. I have previously mainly 
thought of
the @seealso as a kind of reference list ("bibliography") as it is used 
in scientific
papers, i.e. a collection of all cross references mentioned in the main 
text.
In "Durations", you instead use @seealso to introduce new 
concepts/issues that
have not been mentioned in the main text and I kind of like the idea of 
putting

these in an item list, where they are easier to spot than in running text.
Still, for cross references that have already been mentioned in the main 
text,

it would seem chatty to repeat part of the information again in @seealso.
For the link to "Proportional duration", for example, it's probably 
necessary

to provide the concept in a full sentence, so the readers realize what the
link is all about. For "Writing rests", on the other hand, it's pretty 
obvious

from the section name itself.

I'm not sure what the conclusion is, but maybe we can reformulate the 
original

question into:
- Should we introduce some new concepts under @seealso or should this
 always be done in the main text?
- For issues that are somewhat peripheral and just need a short intro and a
 cross reference, should they be described in running text or in an 
item list?
- Under @seealso, can we find a nice layout that allows for a mix of 
pure links
 which already have self-explanatory names, with links that need a 
sentence of

 explanation?
- Is there a need to repeat all links from the main text also in @seealso?

A side comment: in "Durations" under @refbugs, the term "glyphs"
is used several times. Is this is term that is well-known to all readers, or
is it only known to hackers?

  /Mats


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


Re: GDP: chattiness in @seealso

2007-11-15 Thread Graham Percival



Eyolf Østrem wrote:

On 15.11.2007 (16:19), Graham Percival wrote:

 ... I _really_ think this is completely unnecessary, though.  And if you
 want to add full sentences to every single notation reference @ref{}, I
 assume you want to do the same for every @lsr{dir,snippet}, every
 @internalsref{}, etc ?


No, not really. My only concern -- since you asked for general principles
-- is that there shouldn't be a rule to preclude explanation where it is
desirable.


That's why I proposed #3 -- short explanations could be added if needed, 
but most of the time you wouldn't need any explanation.  To take the 
current examples, if you're looking at Durations and see a link to 
"writing rests" (or simply "rests", which is what we should have there), 
I don't think there's any question about why it might be useful.


Also, remember that this is the format inside the @seealso -- not about 
having links in the main doc section.


Cheers,
- Graham


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


Re: GDP: chattiness in @seealso

2007-11-15 Thread Eyolf Østrem
On 15.11.2007 (16:19), Graham Percival wrote:
> 
>  At the very least, I want it clear which sentence refer to the Notation
>  Reference, and which sentences refer to the other parts of the docs.

Agreed.

>  ... I _really_ think this is completely unnecessary, though.  And if you
>  want to add full sentences to every single notation reference @ref{}, I
>  assume you want to do the same for every @lsr{dir,snippet}, every
>  @internalsref{}, etc ?

No, not really. My only concern -- since you asked for general principles
-- is that there shouldn't be a rule to preclude explanation where it is
desirable. This will be the case, I imagine, with references to some
complicated function in an altogether general section, or in other cases
where the reference in its barest form is less than obvious. In many cases,
I agree that an extra description will be fluff and should be avoided.

>  Mats, you're the yardstick for efficient NR use.  What do you think of
>  the compact vs. full sentence form of @seealso ?  I don't want to
>  approve any change that makes the NR harder to use for knowledgeable
>  users, and IMO this is one such change.

How do you define a knowledgeable user in this respect? One who is
knowledgeable in using the docs will know to look for links in the seealso
sections, and I can't see how it would make it more difficult to use it
with an extra pointer or two (like "Remember to bring the towel from your
hotel room") -- and one as knowledgeable about LP as Mats probably won't
need those links in any case :)

sorry, the question wasn't for me, so I'll shut up.

Eyolf

-- 
Life, like beer, is merely borrowed.
-- Don Reed


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


Re: GDP: chattiness in @seealso

2007-11-15 Thread Eyolf Østrem
On 14.11.2007 (16:18), Graham Percival wrote:
>  I think we should have a consistent format for the NR; which one do
>  people prefer?


>  I have a slight preference against #2 (sentences everywhere), since IMO

I know you have, and you know this is the one I prefer. Giving a hint at
WHY one should seealso ain't fluff. This isn't dungeons and dragons ("you
are in a dark cave. To the east there is a link to Proportional notation,
to the south is a snippet." etc). In other words ... 

>  in most cases it's obvious why somebody might want to look at other
>  section.

... I'd say that in SOME cases it's obvious, but in many it's not, and if a
general rule is needed, I'd go for 2 (with 3 as a variant). 

eyolf


-- 
Law always chooses sides on the basis of enforcement power.  Morality and
legal niceties have little to do with it when the real question is:  Who has
the clout? 

  -- Bene Gesserit Council Proceedings:  Archives #XOX232


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


Re: GDP: chattiness in @seealso

2007-11-15 Thread Graham Percival

Eyolf Østrem wrote:

On 14.11.2007 (16:18), Graham Percival wrote:

 I have a slight preference against #2 (sentences everywhere), since IMO


I know you have, and you know this is the one I prefer. Giving a hint at
WHY one should seealso ain't fluff. This isn't dungeons and dragons ("you
are in a dark cave. To the east there is a link to Proportional notation,
to the south is a snippet." etc).


I think you mean "this isn't zork", not D&D.  :)


 in most cases it's obvious why somebody might want to look at other
 section.


... I'd say that in SOME cases it's obvious, but in many it's not, and if a
general rule is needed, I'd go for 2 (with 3 as a variant). 


Well, you're the one who'll have to go through and write sentences for 
every single @seealso section, so it's no skin off my back... what about 
the new Durations?  (see tomorrow's GDP; should be online in about two 
hours.  If you're not certain if you're seeing the updated ones or not: 
the updated one stick things in an itemized list)


At the very least, I want it clear which sentence refer to the Notation 
Reference, and which sentences refer to the other parts of the docs.


... I _really_ think this is completely unnecessary, though.  And if you 
want to add full sentences to every single notation reference @ref{}, I 
assume you want to do the same for every @lsr{dir,snippet}, every 
@internalsref{}, etc ?



Mats, you're the yardstick for efficient NR use.  What do you think of 
the compact vs. full sentence form of @seealso ?  I don't want to 
approve any change that makes the NR harder to use for knowledgeable 
users, and IMO this is one such change.


Cheers,
- Graham


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


Re: GDP: chattiness in @seealso

2007-11-15 Thread David Fedoruk
Good questions!
>
> (another overdue discussion question, sorry)

IMHO this isn't a problem. The crucial thing to remember is that
program documentation isn't for programmers (they have comments in the
working code), its for us non-programmers who struggle with programs
like lilypond because it requires us to learn to think differently
about notes and music than we are used to.

Full sentences in the language of the documentation is preferable. A
full sentence engages, or draws the reader into the program in a way
that the terseness of choice 1 or 3 cannot do. It can also prevent the
reader from ducking into blind alleys. This prevents much head banging
(less risk of permanent damage to) and general frustration.

Links and "see also" references are good ways to redirect someone who
is not quite sure where to look for what they are looking for. Many
times I have a question which I do not know the correct words for. I
guess at how the question might be worded. I am right about 50 percent
of the time. When I am not right, those see also references usually
are the ones that save me from much head banging.

I have (on occasion) actually sat in a chair away from the computer
and read computer documentation. I don't believe documentation is
simply on-the-fly reading. It should be a well thought out description
with instruction on how to use the program it accompanies. It is a
literary work of its own. The skill and craft of it is to be on point;
engaging and readable all at the same time.

What I am saying is that presentation matters. It is essentially first
contact for newbies using lilypond. The longer you use lilypond, the
less you depend on documentation. The speed with which that happens is
in large part due to the skill with which the documentation is
written. Without a doubt, number 2 is the best option.

I had not intended to use so many words, but this sums up my thoughts
on documentation and specifically on this question.

>
> Take a look at the "see also" sections in
> NR 1.1.3 Displaying pitches: Instrument transpositions
> and
> NR 1.2.1 Writing rhythms: Durations
>
> In 1.1.3, we have a short, compact format:
> 
> Notation reference: Quoting other voices, Transpose.
> 
>
> In 1.2.1, there's much more explanation of why people might want to look
> at each link:
> 
> For ways of specifying durations for the syllables of lyrics and ways of
> aligning lyrics to notes see Vocal music.
>
> For a description of how to enter rests see Writing rests.
>
> A note with the duration of a quadruple breve may be entered with
> \maxima, but this is supported only within ancient music notation; see
> Ancient notation.
>
> Optionally, notes can be spaced proportionately to their duration. For
> details of this and other settings which control proportional notation
> see Proportional notation.

I have often found the footnotes in a book or article as interesting
or more interesting than the main piece of writing.


Cheers,
David


--
David Fedoruk
B.Mus. UBC,1986
Certificate in Internet Systems Administration, UBC, 2003


http://recordjackethistorian.wordpress.com
"Music is enough for one's life time, but one life time is not enough
for music" Sergei Rachmaninov


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


Re: GDP: chattiness in @seealso

2007-11-15 Thread Valentin Villenave
2007/11/15, Graham Percival <[EMAIL PROTECTED]>:

> I have a slight preference against #2 (sentences everywhere), since IMO
> in most cases it's obvious why somebody might want to look at other
> section.

I don't! I like full sentences :)

> Option #3 is preferring no explanation at all, but allowing short
> phrases in parentheses.

IMHO this is the best choice, by far.

Valentin


___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user


GDP: chattiness in @seealso

2007-11-14 Thread Graham Percival

As always, GDP:
http://web.uvic.ca/~gperciva/

(another overdue discussion question, sorry)


Take a look at the "see also" sections in
NR 1.1.3 Displaying pitches: Instrument transpositions
and
NR 1.2.1 Writing rhythms: Durations

In 1.1.3, we have a short, compact format:

Notation reference: Quoting other voices, Transpose.


In 1.2.1, there's much more explanation of why people might want to look
at each link:

For ways of specifying durations for the syllables of lyrics and ways of
aligning lyrics to notes see Vocal music.

For a description of how to enter rests see Writing rests.

A note with the duration of a quadruple breve may be entered with
\maxima, but this is supported only within ancient music notation; see
Ancient notation.

Optionally, notes can be spaced proportionately to their duration. For
details of this and other settings which control proportional notation
see Proportional notation.
-


I think we should have a consistent format for the NR; which one do
people prefer?

I have a slight preference against #2 (sentences everywhere), since IMO
in most cases it's obvious why somebody might want to look at other
section.  For example, if you're looking at Durations, it should be
pretty obvious if you personally want to look at:
Writing rests, Ancient notation, or Proportional notation.  I'll grant
that the link to Vocal music might need some explanation, though.

Of course I'm famously (and in most of your opinions, unnecessarily :)
opposed to fluff.


Option #3 is preferring no explanation at all, but allowing short
phrases in parentheses.  ie 1.2.1 would turn into this:
--
See also

Notation Reference: Writing rests, Ancient notation (for \maxima),
Proportional notation, Vocal music (for specifying durations with lyrics).



Cheers,
- Graham



___
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user