Tuning-Math archive - first upload

HI there,

Here is the first complete upload of the Tuning-Math
archive.

I'm posting this to tuning-math and also to the
main tuning forum. Did tuning-math first because
it is a smaller forum, so a smaller archive
to upload. But it will give an idea of how
the main tuning archive will be presented as well.

Here it is:

http://www.robertinventor.com/tuning-math/index.html

Or to open it up showing the contents list right away:

http://www.robertinventor.com/tuning-math/frame.html?index.html

Comments welcome. It is to be considered a beta of the
archive format and subject to possible change.

I thought I'd say something about how you link
to the content - as it is at present. This is also
subject to change, right now I think consider links
you make to the archvie as experimental. They may
not all work when I re-upload it in a week or two
from now. But once the format is decided I will
make it so that they will work from then on no
matter whether messages get added or more people
opt in - at least, for links to individual
messages within the archive.

So anyway here it is as it is now:

To link to an individual message,
you need to hide the contents list and then navigate to it
and you will see a url like this in the address bar:
http://www.robertinventor.com/tuning-math/s___6/msg_5750-5774.html#5767

You can also link to an individual page and show it within
the contents list. Just navigate to the page,
hide the contents, then show it again, and you will now
see a url such as:

http://www.robertinventor.com/tuning-math/frame.html?s___6/msg_5750-5774.html

Unfortunately, it seems you can't show
contents list + link to an individual message.
This url, which one might think would work,
only gets you the page and not the individual
message:

http://www.robertinventor.com/tuning-math/frame.html?/s___6/msg_5750-5774.html#5767

I don't know if there is any work around for it yet.

Also in case you wonder about it - this can't be used
as a forum script at all.

The only way I can add messages to it is by uploading
the entire archive. That is for technical reasons
to do with the whole approach used to make it.
Add a single message and you will need to re-upload
MBs of html files to update the archive, and I don't
think it can be adapted to be interactive.
So, it is only usable as an archive maker after the event.

Questions like that however should be discussed
over in metatuning rather than here if anyone
wants to follow it up and discuss it more
- I've been discussing it with Kurt
already.

I'll leave it up for a while so we can discuss anything
that arises, & probably do the final upload
in a week or two. If you want to opt in be
sure to let me know as it is a large upload
and will only be done rarely.

Look out for the colouring of quoted posts
(idea of Kurt's) and the
option to look up links in
the wayback machine :-).

Robert

HI there,

Here is the first complete upload of the Tuning-Math
archive.

I'm posting this to tuning-math and also to the
main tuning forum. Did tuning-math first because
it is a smaller forum, so a smaller archive
to upload. But it will give an idea of how
the main tuning archive will be presented as well.

Here it is:

http://www.robertinventor.com/tuning-math/index.html

Or to open it up showing the contents list right away:

http://www.robertinventor.com/tuning-math/frame.html?index.html

Comments welcome. It is to be considered a beta of the
archive format and subject to possible change.

I thought I'd say something about how you link
to the content - as it is at present. This is also
subject to change, right now I think consider links
you make to the archvie as experimental. They may
not all work when I re-upload it in a week or two
from now. But once the format is decided I will
make it so that they will work from then on no
matter whether messages get added or more people
opt in - at least, for links to individual
messages within the archive.

So anyway here it is as it is now:

To link to an individual message,
you need to hide the contents list and then navigate to it
and you will see a url like this in the address bar:
http://www.robertinventor.com/tuning-math/s___6/msg_5750-5774.html#5767

You can also link to an individual page and show it within
the contents list. Just navigate to the page,
hide the contents, then show it again, and you will now
see a url such as:

http://www.robertinventor.com/tuning-math/frame.html?s___6/msg_5750-5774.html

Unfortunately, it seems you can't show
contents list + link to an individual message.
This url, which one might think would work,
only gets you the page and not the individual
message:

http://www.robertinventor.com/tuning-math/frame.html?/s___6/msg_5750-5774.html#5767

I don't know if there is any work around for it yet.

Also in case you wonder about it - this can't be used
as a forum script at all.

The only way I can add messages to it is by uploading
the entire archive. That is for technical reasons
to do with the whole approach used to make it.
Add a single message and you will need to re-upload
MBs of html files to update the archive, and I don't
think it can be adapted to be interactive.
So, it is only usable as an archive maker after the event.

Questions like that however should be discussed
over in metatuning rather than here if anyone
wants to follow it up and discuss it more
- I've been discussing it with Kurt
already.

I'll leave it up for a while so we can discuss anything
that arises, & probably do the final upload
in a week or two. If you want to opt in be
sure to let me know as it is a large upload
and will only be done rarely.

Look out for the colouring of quoted posts
(idea of Kurt's) and the
option to look up links in
the wayback machine :-).

I know the subject matter is of course
rather technical for tuning list
members. After leaving it up long enough
to get initial comments, I'll perhaps
do the last month or some such
for the main tuning list and
ask for comments on that too.

However, it is also a good one
to start with apart from size,
as that very technicalilty does mean tht
it may raise more issues such as
presentation and formatting#
and special characters
with fewer messages :-).

Robert

Hi there,

Forgot to say:

For help see the help section in
http://www.robertinventor.com/tuning-math/about_this_archive.html

Also there is no zip yet. I'll leave that until later
- but can upload one right away if anyone particularly
wants it even for this beta version.

Robert

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...> wrote:
> HI there,
>
> Here is the first complete upload of the Tuning-Math
> archive.(snip)>
> Robert
....space....Jumpin' Dave writes: I uploaded this group. I found
myself at the mercy of the encyclopedia of terms. I spent well over a
hour reading terms.
some terms are starting to make sense. Wish I got a hold of microtonal
instruments first and learned by ear, So I could say,"oh, I knew
that!" Although, some of my music does resemble this style. I am
making all the mistakes I can.

on 7/21/04 5:06 PM, Robert Walker <robertwalker@ntlworld.com> wrote:

> HI there,
>
> Here is the first complete upload of the Tuning-Math
> archive.
>
> I'm posting this to tuning-math and also to the
> main tuning forum. Did tuning-math first because
> it is a smaller forum, so a smaller archive
> to upload. But it will give an idea of how
> the main tuning archive will be presented as well.
>
> Here it is:
>
> http://www.robertinventor.com/tuning-math/index.html

Fantastic!

> Look out for the colouring of quoted posts
> (idea of Kurt's) and the
> option to look up links in
> the wayback machine :-).

I looked at that and curiously I found that I liked the *aesthetic* a little
better than what I am used to in my email client, but actually it didn't
function as well for me in terms of helping me quickly be able to see the
non-quoted material as distinct from the quoted. I'm pretty sure this is
because the approach used of coloring the background on quoted lines but
leaving the text black creates less contrast than I am used to between
quoted and non-quoted text. In my email client the non-quoted text is the
only thing in black and as it turns out this means it is a lot easier for it
to pop out from the quoted text which is in various colors (blue, green,
red, purple, all on the dark side but a lot less dark than black).

So I'd like to suggest trying the approach of colorizing the text itself.

Also because of how badly yahoo (or whatever) is constantly messing up
quoted lines (causing extra line wraps that cause quote levels to be lost),
I think that adding a blank line between different quoting levels tends to
work against rather than for overall readability, because those mess-ups are
quite frequent.

Even better would be to write an algorithm that would recognize the common
mess-up pattern and fix it. It always involves extra line wraps causing
alternate lines to be very short and at a lower quoting level. I think this
could be fixed in the middle of paragraphs but perhaps not at the end, with
high reliability. It would be great to get rid of that nonsense if it can
be done!

The only other comment so far is that I think for some reason consecutive
quotes without spaces between them look better, are less visually
distracting and are more clearly quoting rather than something else. That
is I like

>>>

better than

> > >

Everything else looks great so far. For myself I set my browsers text size
to 75% for better viewing and less scrolling. It might be good to get
people's input about what size works best for them. I think yahoo's is a
little smaller, but also too small (and certainly too narrow and not
monospace as needed) yet with too much linespace so wasting a lot of screen
unnecessarily.

Thanks.

-Kurt

Hi Kurt,

> So I'd like to suggest trying the approach of colorizing the text itself.

Okay fine I've done that.

> Also because of how badly yahoo (or whatever) is constantly messing up
> quoted lines (causing extra line wraps that cause quote levels to be lost),
> I think that adding a blank line between different quoting levels tends to
> work against rather than for overall readability, because those mess-ups are
> quite frequent.

> Even better would be to write an algorithm that would recognize the common
> mess-up pattern and fix it. It always involves extra line wraps causing
> alternate lines to be very short and at a lower quoting level. I think this
> could be fixed in the middle of paragraphs but perhaps not at the end, with
> high reliability. It would be great to get rid of that nonsense if it can
> be done!

Thanks for the suggestion. I've removed the gaps and made an algorithm
to deal with the word wrapping giltches.
The thing there is that usually it is a word wrap by only one or
two words - the quoted text was wrapped just to the Yahoo
boundary - then when quoted again some of the lines go just
over so they spill over by just a single word or sometimes
two words at least that is one guess about how it works.

So I just did it to deal with that.

I wouldn't want to try and actualy fix the wrapping
because just occasionally someone might actually
reply with a single word or two words in the
middle of all the quoted text. But that will be
much rarer than the other way round and will
leave it to the reader to pick up on it.

> The only other comment so far is that I think for some reason consecutive
> quotes without spaces between them look better, are less visually
> distracting and are more clearly quoting rather than something else. That
> is I like

> >>>
>
> better than
>
> >

Okay I've done that too.

One thought - occasionally > may be used for some other meaning
at the start of the text (as a symbol). Don't know if
> > is likely to be sued too. If it does then
it will replace it by >> which mat not be a matter
of concern - we can meet that if / when we encounter it.

I've re-uploaded section 1 and 2 with these fixes (and will
gradually upload the others during the day).

http://www.robertinventor.com/tuning-math/frame.html?s___1\index.html

Also fixed a bug in the frame - it wasn't working
in an old Netscape 4.7 and is now - so if anyone found it didn't
work for them, try again.

On the matter of the text size - I've
made it for next upload so that the entire archive will
use a single style sheet. So at least
with the downloaded archive you will be
able to chagne the text sizes by editing
that style sheet. Also you will be able
to change the text and background
colours for the various levels of quoting
too.

Robert

Hi Kurt,

> So I'd like to suggest trying the approach of colorizing the text itself.

Okay fine I've done that.

> Also because of how badly yahoo (or whatever) is constantly messing up
> quoted lines (causing extra line wraps that cause quote levels to be lost),
> I think that adding a blank line between different quoting levels tends to
> work against rather than for overall readability, because those mess-ups are
> quite frequent.

> Even better would be to write an algorithm that would recognize the common
> mess-up pattern and fix it. It always involves extra line wraps causing
> alternate lines to be very short and at a lower quoting level. I think this
> could be fixed in the middle of paragraphs but perhaps not at the end, with
> high reliability. It would be great to get rid of that nonsense if it can
> be done!

Thanks for the suggestion. I've removed the gaps and made an algorithm
to deal with the word wrapping giltches.
The thing there is that usually it is a word wrap by only one or
two words - the quoted text was wrapped just to the Yahoo
boundary - then when quoted again some of the lines go just
over so they spill over by just a single word or sometimes
two words at least that is one guess about how it works.

So I just did it to deal with that.

I wouldn't want to try and actualy fix the wrapping
because just occasionally someone might actually
reply with a single word or two words in the
middle of all the quoted text. But that will be
much rarer than the other way round and will
leave it to the reader to pick up on it.

> The only other comment so far is that I think for some reason consecutive
> quotes without spaces between them look better, are less visually
> distracting and are more clearly quoting rather than something else. That
> is I like

> >>>
>
> better than
>
> >

Okay I've done that too.

One thought - occasionally > may be used for some other meaning
at the start of the text (as a symbol). Don't know if
> > is likely to be sued too. If it does then
it will replace it by >> which mat not be a matter
of concern - we can meet that if / when we encounter it.

I've re-uploaded section 1 with these fixes.

http://www.robertinventor.com/tuning-math/frame.html?s___1\index.html

Also fixed a bug in the frame - it wasn't working
in an old Netscape 4.7 and is now - so if anyone found it didn't
work for them, try again.

On the matter of the text size - I've
made it for next upload so that the entire archive will
use a single style sheet. So at least
with the downloaded archive you will be
able to chagne the text sizes by editing
that style sheet. Also you will be able
to change the text and background
colours for the various levels of quoting
too.

Robert

Hi there,

I've fixed the thing about linking to an individual message
within the frame.

I've uploaded sections 1 and 2 so far with the fix.

Now to link to an individual message within
the archive, just navigate to that message
and use the Contents link after the
message number.

This will put a url into the browser
address bar which you can then use
to link to that individual message .

Or if you want to link to the message
without the contents listing, click the
"Hide Contents" link next to the
message and use the url that brings
up.

Techy note:

The solution was to replace the
# in the link by an &.
It seems that you can't have a #
after a ? in the url
for some reason. Then
my script for writing out
the frame dynamically just changes
the & back into a # again.

Fixed :-).

Robert

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>
wrote:
> HI there,
>
> Here is the first complete upload of the Tuning-Math
> archive.

Very nice!

> I'll leave it up for a while so we can discuss anything
> that arises ...
>
> Robert

The numbering of the messages does not follow the original, e.g.,
#3946 in the original is #4505 in the archive.

If the messages in the archive were all numbered consecutively, and
if the archive contained *fewer* messages than the original, then I
would expect that the number in the archive would be *lower* than in
the original.

However, since I have extensive records of past messages on certain
subjects *by number*, I would want to see the original message
numbering retained, if possible.

--George

--- In tuning@yahoogroups.com, "David A. Lovatto" <slipindave@y...>
wrote:
> --- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>
wrote:
> > HI there,
> >
> > Here is the first complete upload of the Tuning-Math
> > archive.(snip)>
> > Robert
> ....space....Jumpin' Dave writes: I uploaded this group. I found
> myself at the mercy of the encyclopedia of terms. I spent well over
a
> hour reading terms.
> some terms are starting to make sense. Wish I got a hold of
microtonal
> instruments first and learned by ear, So I could say,"oh, I knew
> that!" Although, some of my music does resemble this style. I am
> making all the mistakes I can.

David, if you need more firsthand experience alternate tunings, you
need to download Scala and try some of these out for yourself:

http://www.xs4all.nl/~huygensf/scala

--George

Hi there,

Just to say I added a search to the tuning-math archive
after a suggestion by Gene on the tuning-math list.

http://www.robertinventor.com/tuning-math/search_tuning-math.html

Robert

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>

/tuning/topicId_54749.html#54761
> On the matter of the text size - I've
> made it for next upload so that the entire archive will
> use a single style sheet. So at least
> with the downloaded archive you will be
> able to chagne the text sizes by editing
> that style sheet. Also you will be able
> to change the text and background
> colours for the various levels of quoting
> too.
>
> Robert

***This is really amazing...! All done with XML??

Joseph

hi Robert,

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>
wrote:
> Hi there,
>
> Just to say I added a search to the tuning-math archive
> after a suggestion by Gene on the tuning-math list.
>
> http://www.robertinventor.com/tuning-math/search_tuning-math.html
>
> Robert

AWESOME!

is there any way you can integrate this with the rest
of the archive? instead of having to use a separate link.

also, would you please change the links to any
Encyclopaedia terms to the current (and permanent) address?

http://tonalsoft.com/enc

expired domain names are:

http://ixpres.com/interval/dict
http://sonic-arts.org/dict

here's an example page i found in a search which has
a broken link to "limma" in my message #4186:

http://www.robertinventor.com/tuning-math/s___5/msg_4175-4199.
html#udpm13

if the links are to terms which are in the Encyclopaedia,
they should have the new link.

thanks.

-monz

on 7/21/04 8:00 PM, David A. Lovatto <slipindave@yahoo.com> wrote:

> --- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...> wrote:
>> HI there,
>>
>> Here is the first complete upload of the Tuning-Math
>> archive.(snip)>
>> Robert
> ....space....Jumpin' Dave writes: I uploaded this group. I found
> myself at the mercy of the encyclopedia of terms. I spent well over a
> hour reading terms.

Speaking of this, and of the whole terminology problem that was brought up
several days back. How about *this* idea Robert and Monz:

(1) Monz periodically generates a list of all indexable tuning terms, and
gives this list to Robert
(2) Robert adds a frame (maybe) or something that allows an alphebetized
list of all tuning terms used in a given message thread (or whatever appears
on a page) to be displayed somewhere off to the side, and also maybe (why
not) as hyperlinks within the text. I can imagine the hyperlinks might
interfere with readability, in which case a subtle color change of the
indexed terms could still be used so the reader gets a heads up that they
can go click somewhere for a definition of that term.
(3) [Technical.] Given this structure, it might be good if Monz created a
cgi that takes a tuning term as an argument so that a simple url could be
created for any tuning term. Otherwise trying to do the above might be
unwieldy.

Whadya think?

This would help to ease the terminology problem somewhat.

(The next trick would be to start a migration towards tunings lists that
provide the same services via email, i.e. embedded hyperlinks for tunings
terms, and/or a list of alphabetized hyperlinks at the end of each message
for terms used in that message. But one step at a time, so the value of the
idea can be determined from experience.)

-Kurt

on 7/22/04 9:08 PM, Kurt Bigler <kkb@breathsense.com> wrote:

> on 7/21/04 8:00 PM, David A. Lovatto <slipindave@yahoo.com> wrote:
>
>> --- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...> wrote:
>>> HI there,
>>>
>>> Here is the first complete upload of the Tuning-Math
>>> archive.(snip)>
>>> Robert
>> ....space....Jumpin' Dave writes: I uploaded this group. I found
>> myself at the mercy of the encyclopedia of terms. I spent well over a
>> hour reading terms.
>
> Speaking of this, and of the whole terminology problem that was brought up
> several days back. How about *this* idea Robert and Monz:
>
> (1) Monz periodically generates a list of all indexable tuning terms, and
> gives this list to Robert
> (2) Robert adds a frame (maybe) or something that allows an alphebetized
> list of all tuning terms used in a given message thread (or whatever appears
> on a page) to be displayed somewhere off to the side, and also maybe (why
> not) as hyperlinks within the text. I can imagine the hyperlinks might
> interfere with readability, in which case a subtle color change of the
> indexed terms could still be used so the reader gets a heads up that they
> can go click somewhere for a definition of that term.
> (3) [Technical.] Given this structure, it might be good if Monz created a
> cgi that takes a tuning term as an argument so that a simple url could be
> created for any tuning term. Otherwise trying to do the above might be
> unwieldy.

On second thought, probably a lot could be done just by Robert fairly easily
(?) without any help from Monz just by generating a term to url index from
Monz's dictionary index frame.

-Kurt

>
> Whadya think?
>
> This would help to ease the terminology problem somewhat.
>
> (The next trick would be to start a migration towards tunings lists that
> provide the same services via email, i.e. embedded hyperlinks for tunings
> terms, and/or a list of alphabetized hyperlinks at the end of each message
> for terms used in that message. But one step at a time, so the value of the
> idea can be determined from experience.)
>
> -Kurt

--- In tuning@yahoogroups.com, Kurt Bigler <kkb@b...> wrote:

> Speaking of this, and of the whole terminology problem
> that was brought up several days back. How about *this*
> idea Robert and Monz:
>
> (1) Monz periodically generates a list of all indexable
> tuning terms, and gives this list to Robert
> (2) Robert adds a frame (maybe) or something that allows
> an alphebetized list of all tuning terms used in a given
> message thread (or whatever appears on a page) to be
> displayed somewhere off to the side, and also maybe
> (why not) as hyperlinks within the text. I can imagine
> the hyperlinks might interfere with readability, in which
> case a subtle color change of the indexed terms could still
> be used so the reader gets a heads up that they can go
> click somewhere for a definition of that term.
> (3) [Technical.] Given this structure, it might be good
> if Monz created a cgi that takes a tuning term as an
> argument so that a simple url could be created for any
> tuning term. Otherwise trying to do the above might be
> unwieldy.
>
> Whadya think?

i don't know anything about cgi.

but these ideas illustrate why we would like to host the
tuning lists at Tonalsoft. the tuning lists could be
closely integrated with the online version of the
Encyclopaedia.

-monz

hi Robert,

i see that your archive also preserves the blank spaces
in our diagrams and tables! excellent!!

-monz

on 7/22/04 7:22 AM, Robert Walker <robertwalker@ntlworld.com> wrote:

> Now to link to an individual message within the archive, just navigate to that
> message and use the Contents link after the message number.
>
> This will put a url into the browser address bar which you can then use to
> link to that individual message .
>
> Or if you want to link to the message without the contents listing, click the
> "Hide Contents" link next to the message and use the url that brings up.

This works fine, but seems like you are taking advantage of a side-effect to
get a feature, which people can get used to but is indirect. It may not
seem worth adding another link for directness, but perhaps just clicking on
the message number itself is an intuitive way to get a url for that message
number. Just a thought.

> Techy note:
>
> The solution was to replace the # in the link by an &. It seems that you can't
> have a # after a ? in the url for some reason. Then my script for writing out
> the frame dynamically just changes the & back into a # again.

I don't get this. At a single level of url, the # comes after the ?, i.e.
the fragment comes after the query string. If you want your query string to
contain an embedded url (or partial url) then you do need to escape the
characters that the top-level url parser recognizes or else map it in some
way. My reference says url parsing works like this (with extra spaces for
readability):

scheme :// host : port / path ? query # fragment

> Fixed :-).

So the above is sort of moot but I'll leave it FWIW.

-Kurt

on 7/22/04 7:12 AM, Robert Walker <robertwalker@ntlworld.com> wrote:

> Hi Kurt,
>
>> So I'd like to suggest trying the approach of colorizing the text itself.
>
> Okay fine I've done that.

That did the trick. Now the new text stands out just fine. Too bad you're
not working for Apple. I'd just report bugs to you and they'd be fixed the
next day instead of being filed in perfect bureaucratic form for posterity.
;)

The color schemes are quite good but might be refined just a little--perhaps
to use the least clear color schemes for the oldest material? Mind you this
is a fine point, but I think we're already getting to the fine points fast!

When looking at this, here's a message that contains 6 consecutive levels
levels of quoting (7 if you include unquoted, and so is useful to allow
people to look at the color scheme:

http://www.robertinventor.com/tuning-math/s__12/msg_11000-11024.html#11005

Maybe you have an even better one handy.

As it hits me the only thing I really don't like is that the olive/brown on
yellow is used for level 1 quoting, because that is the most important level
of quoting since it is what the new text is directly responding to. To me
the olive/brown on yellow is the least clear (perhaps least contrast between
foreground and background) and might be used for all quoting levels above
your maximum limit if you have such a limit.

The rest of my comments here are too technical and so I'll post them on
metatuning.

>> Even better would be to write an algorithm that would recognize the common
>> mess-up pattern and fix it. It always involves extra line wraps causing
>> alternate lines to be very short and at a lower quoting level. I think this
>> could be fixed in the middle of paragraphs but perhaps not at the end, with
>> high reliability. It would be great to get rid of that nonsense if it can
>> be done!
>
> Thanks for the suggestion. I've removed the gaps and made an algorithm
> to deal with the word wrapping giltches.

Wow it's really amazing. I haven't seen one of those ugly glitches
anywhere. Only thing is it looks like the last line in each colorized block
often has one less quote than the lines before it.

> I wouldn't want to try and actualy fix the wrapping because just occasionally
> someone might actually reply with a single word or two words in the middle of
> all the quoted text. But that will be much rarer than the other way round and
> will leave it to the reader to pick up on it.

Hmm. Funny I didn't see any straggling words though. Maybe I was on the
wrong pages this time to see glitches.

>> The only other comment so far is that I think for some reason consecutive
>> quotes without spaces between them look better, are less visually
>> distracting and are more clearly quoting rather than something else. That
>> is I like
>>>>>
>> better than
>>>

> Okay I've done that too.

Yes, I think it does look better.

> One thought - occasionally > may be used for some other meaning
> at the start of the text (as a symbol). Don't know if
>>> is likely to be sued too. If it does then
> it will replace it by >> which mat not be a matter
> of concern - we can meet that if / when we encounter it.

I think you mean if > is not the first character, i.e. it is preceeded by
whitespace? Otherwise I don't get it. It is essentially ambiguous, but
that ambiguity comes with the medium.

> On the matter of the text size - I've made it for next upload so that the
> entire archive will use a single style sheet. So at least with the downloaded
> archive you will be able to chagne the text sizes by editing that style sheet.

No kidding? I didn't know style sheets could be edited by the end user (the
one doing the browsing). How? Maybe this is only supported by specific
browsers, maybe only on Windows?

-Kurt

on tuning, on 7/22/04 7:12 AM, Robert Walker wrote:

> Hi Kurt,
>
>> So I'd like to suggest trying the approach of colorizing the text itself.
>
> Okay fine I've done that.

So regarding the color scheme, here are the more technical comments that I
wanted to keep off the main tuning list. Again here is a good url for
seeing the colors of 7 levels of quoting:

http://www.robertinventor.com/tuning-math/s__12/msg_11000-11024.html#11005

Continuing where I left off on tuning... Regarding the progression of
colors, I probably prefer a fixed "wall" limit to rotating through a fixed
list of color pairs, for hopefully obvious reasons. On the other hand it is
a moot point if the limit is high enough.

So I think I minor tweak will do it, regarding the colors of quoting level 1
as I said in tuning. But if you want to have an actual method/algorithm
behind the color generation here is one algorithm that comes to mind
(untested mind you):

* Increasing quoting levels would have decreasing contrast between text and
background.

* However even level 1 should have a much lighter text than level 0.

* Perhaps level 1 should be distinct in some other way since it is the text
being replied to. This "other way" should not be too dramatic, but perhaps
just by level 0 being highest contras, level 1 being next highest but with a
gap after that that distinguishes other levels more significantly. Or it
just might fall out of the progression...

* With increasing level, text then might continue to get lighter while
backgrund gets darker but with an asymptote to prevent fading out and to
prevent the background becomong so dark that it is distracting. The
background on level 6 is perhaps a little dark and does distract.

* Then within that structure if the hue were to rotate by a constant angle
that does not divide into 360 then you have an algorithm for a pattern that
doesn't repeat at any level. I actually used such an algorithm for hue
succesfully in a charting program to generate the colors for successive
bars. IIRC I used approximatly 1 radian as the hue delta, but I'm really
not sure, but when you hit a good delta you will know it is right. It looks
like you have a good eye for color.

These are just untested thoughts and I'm not attached to them. You may have
actually done something a little like this either by algorithm or manually,
but the looks of it.

Thanks for all your work and your fantasticly active receptivity to
feedback!

-Kurt

on 7/23/04 4:07 PM, Kurt Bigler <kkb@breathsense.com> wrote:

> on tuning, on 7/22/04 7:12 AM, Robert Walker wrote:
>
>> Hi Kurt,
>>
>>> So I'd like to suggest trying the approach of colorizing the text itself.
>>
>> Okay fine I've done that.
>
> So regarding the color scheme, here are the more technical comments that I
> wanted to keep off the main tuning list.

OOPS. Sorry. I meant to post this to metatuning but posted it here by
mistake! I guess I'll just let it go now rather than reposting it to
metatuning.

-Kurt

Hi Monz,

> > Just to say I added a search to the tuning-math archive
> > after a suggestion by Gene on the tuning-math list.
> >
> > http://www.robertinventor.com/tuning-math/search_tuning-math.html

> AWESOME!

> is there any way you can integrate this with the rest
> of the archive? instead of having to use a separate link.

Yes, for next upload whenever it is, I've added
a link to the search window at the head of
every message - where it has the option to show
or hide the contents, you can also now search
the archive.

This is just a protem solution. The atomz
search engine is free up to 500 pages per
site. We will reach that limit with a few
months for Tuning-math (unless we
increase the number of entries per page
e.g. to 50 instead of 25). We are
well over it with the main Tuning
list.

I just e-mailed them to find out how
much it costs for larger sites. Was
expecting it to be high as it usually
is if you ahve to contact them first
to find out the prices - seems to usually
mean they are catering for big businesses
only. It was - teh starting price is
well outside our range at $10,000
per year!

However, luckily there are scripts one
can use as well. So maybe later
we can use one of those.

The scripts that crawl your site are best
as they will be faster for a large number
of docuents like this and it changes only
when I upload it so you can just get it to
crawl again every time.
.
Some examples here to try out:
http://cgi.resourceindex.com/Programs_and_Scripts/Perl/Searching/Searching_Your_Web_Site/

Anyway more later.

Also thanks for the link update - I've
set it to fix those for the next upload.
If anyone else has a domain change
to the same content on another site,
let me know and I'll add it to my
list of links to fix - easy to do
now I've done these, can do any
others the same way.

Robert

Hi Kurt,

> > ....space....Jumpin' Dave writes: I uploaded this group. I found
> > myself at the mercy of the encyclopedia of terms. I spent well over a
> > hour reading terms.

> Speaking of this, and of the whole terminology problem that was brought up
> several days back. How about *this* idea Robert and Monz:

...

> On second thought, probably a lot could be done just by Robert fairly easily
> (?) without any help from Monz just by generating a term to url index from
> Monz's dictionary index frame.

Yes, there is something that can be done. Well
one simple thing I've done already is to add a link to Monz's
encyclopedia in the search window which you will find if
you do a search again now.

With the next upload each message has a link to the
search page, so his dictionary will be just two clicks away
from any message. If someone feels like doing
a search by keyword of the encyclopedia index
so that you enter a word into a form and
get the best matching encyclopedia entry,
then I could add that to the search page too.

Best approach,though it requires loads more work for someone,
would be to have a dictionary as well.

The dictionary just gives a couple of sentences
for each term by way of definition, then a link to
the encyclopedia.

Here is an example I did:

http://www.robertinventor.com/test_show_defs/index.htm
Note - it doesn't yet work for the Opera browser
at least, not version 7 of it. Hopefully that can be
fixed somehow.

I've added links there to Joe Monzo's encyclopedia
now to show how it could be done.

However, it is a big task for a single person to
do. I reckon it would take weeks if not months
to do dictionary definitions for every term
in Monz's encyclopedia - weeks if you can do
one every ten minutes, and months if it takes
more like an hour each - and you usually
need more words in a dictionary than an encyclopedia.

So I'm a bit unsure about the praciticality of
it but it would be great if it could be done.

Actually I had planned to do a dictionary for
the help for Fractal Tune Smithy in a similar way
- which would only need to define terms actually used
in the help for FTS - but after doing this estimate
here I'm wondering how practical that is too.

Maybe it could be a collaborative effort?
I'm happy to contribute a few definitions
certainly - those ones on that page
and a couple more I did today as an
experiment to get an idea of how long
it would take to do the entire dictionary.
I'd be happy to contribute e.g. maybe
half dozen terms a week - it is quite fun
to do a few of them like that. If we had
a dozen or so of us on it then we'd get
it done pretty quickly.

Well if Joe wanted to do it all of course,
but it is such a big undertaking...

I'm inclined to think that a collaborative
effort would be better, and less strain
on any single person, and more fun too.

> Whadya think?

Great idea!

> This would help to ease the terminology problem somewhat.

Yes, indeed!

One thought is that if we can make such a dictionary,
the javascript archive could be included in any
web page and used by web masters, and software authors
and anyone to add dictionary definitions of higlighted words.

> (The next trick would be to start a migration towards tunings lists that
> provide the same services via email, i.e. embedded hyperlinks for tunings
> terms, and/or a list of alphabetized hyperlinks at the end of each message
> for terms used in that message. But one step at a time, so the value of the
> idea can be determined from experience.)

Well it would be someone else programming that, not my
area at all to write a tuning list script, or hack
an existing cgi script to do this.

But if we had such a dictionary available, then
perhaps it could be used in just this way
at some later date by someone with the necessary
skills in Perl or PhP or whatever.

Robert

> So regarding the color scheme, here are the more technical
> comments that I wanted to keep off the main tuning list.

Sorry to see your hopes dashed. :)

> Again here is a good url for seeing the colors of 7 levels
> of quoting:
>
>http://www.robertinventor.com/tuning-math/s__12/
>msg_11000-11024.html#11005

Beautiful, Robert!! Three cheers!

>Continuing where I left off on tuning... Regarding the
>progression of colors, I probably prefer a fixed "wall"
>limit to rotating through a fixed list of color pairs,
>for hopefully obvious reasons.

Yes, that does sound like a good idea.

>* Increasing quoting levels would have decreasing contrast
>between text and background.

I thought about this but it probably wouldn't work very well
in practice. Probably just 5 highly-contrasting colors is
best, with the wall at 5 levels. Or perhaps the gradient
could be applied to the wall only. For example the 5th color
is fairly light, and gets darker with every subsquent level,
while levels 1-4 are highly contrasted.

-Carl

Hi Kurt,

> http://www.robertinventor.com/tuning-math/s__12/msg_11000-11024.html#11005

> Maybe you have an even better one handy.

Thanks,that's useful. No I don't have a better one handy
as I just made one up in order to figure out the
colours in sequence.

> Wow it's really amazing. I haven't seen one of those ugly glitches
> anywhere. Only thing is it looks like the last line in each colorized block
> often has one less quote than the lines before it.

Yes that is intentional. It's because often the quote level
varies by a number of indentations with blank lines
To avoid stripes consisting of just blank lines
it keeps to the same colour until it meets a
reason to change it. A blank line never counts
as a reason for changing colour, whether indented
or not.

The algorithm as it is now just writes out each line
as it comes to it without look back. To fix it
I would need to redo it so that it can do a look
back. Not hard at all - maybe could do it sone time.

> > I wouldn't want to try and actualy fix the wrapping because just occasionally
> > someone might actually reply with a single word or two words in the middle of
> > all the quoted text. But that will be much rarer than the other way round and
> > will leave it to the reader to pick up on it.

> Hmm. Funny I didn't see any straggling words though. Maybe I was on the
> wrong pages this time to see glitches.

Yes there are a few. Case where the response is
a short sentence of a few words. You may not
even notice them easily as they blend in
with the flow of the rest of the post.

However I found a method that could help with that
- with next upload it will colour a line differently
even if just two words if it starts with a
capitalissed word and ends with punctuation.
Hopefully that will deal with a few of those.

> > On the matter of the text size - I've made it for next upload so that the
> > entire archive will use a single style sheet. So at least with the downloaded
> > archive you will be able to chagne the text sizes by editing that style sheet.

> No kidding? I didn't know style sheets could be edited by the end user (the
> one doing the browsing). How? Maybe this is only supported by specific
browsers, maybe only on Windows?

Sorry, I meant that after you download the archive,
and unzip it onto your local machine, you can edit
your local style sheet. I don't know if you can let the
user edit the style sheet on-line as well - would need some
way for it to persist from one page to another.

One thing you can certainly do is to have
a selection of different style sheets
and let the user choose which one they
want for the entire site.

See this demo:
http://tech.irt.org/articles/js065/custom.htm

I don't know if you can also
let them chose individual
colours and store those in
a cookie - maybe somehow by
using javascript I sort of feel
maybe you might be able to.
Or you could do it
if serving up the style sheet
to the user - using cookies perhaps.

Thanks for the suggestions about colour.
Yes I see what you mean about level
1 not standing out particularly
when you have multiple levels
of replies. Maybe just changing
the colour there from olive to
something else more distinctive for the
first layer would do the trick,
some more intense colour as the
background is already one of the
lightest, just that the text is also
light. I suppose I wanted it to
contrast well with the current text by
being a lot lighter but it
doesn't work so well once you
see it with the other layers.

I can't reslly see it with a fresh
eye right now. I'll try again
next week before the next upload
and see if I can do something better
for the first quoting level.

Yes I did try something a bit like
your idea - my thought was to vary
the colour between successive levels
of quoting, and I applied it rather
the other way around - to get gradually
more intense in colour as they went back.

Then just chose them individually by eye
and didn't keep to it rigidly. That
lets you choose pairs of colours
that go nicely together.

The other thought I had in mind
is that many users are colour blind.
So I try to make sure that colours
are ones that colour blind users
will be able to distinguish. So \I
tried to vary the blue / red component
with each level of quoting
as everyone can distinguish blue
from red if they have any colour
discrimination at all - and also
to vary the intensity of blue
and red which again everyone
should be able to distinguish
if they can distinguish colours.
For the few that can see but can't
distinguish colour, I try to vary
the overall brightness, but that is
less effective here as there are so
many distinctions to do.

Robert

Hi Kurt,

> > Now to link to an individual message within the archive, just navigate to that
> > message and use the Contents link after the message number.
> >
> > This will put a url into the browser address bar which you can then use to
> > link to that individual message .
> >
> > Or if you want to link to the message without the contents listing, click the
> This works fine, but seems like you are taking advantage of a side-effect to
> get a feature, which people can get used to but is indirect. It may not
> seem worth adding another link for directness, but perhaps just clicking on
> the message number itself is an intuitive way to get a url for that message
> number. Just a thought.

Yes I could do. The thing is - if it is done that way usually user would
want the page to stay as it is. So maybe you need a bit of javascript
to check first to see whether the message is showing within a frame
or not. So - if inside a contents list, it jsut shows the contents
list url, and otherwise shows the page url.

So the number would have another meaning too, and not quite
duplicating
<number> - shows url and doesn't change anything.
<contents> - shows it and chagnes to contents if needed.
<hide contents> - shows it and hides contents if needed.

> > Techy note:
> >
> > The solution was to replace the # in the link by an &. It seems that you can't
> > have a # after a ? in the url for some reason. Then my script for writing out
> the frame dynamically just changes the & back into a # again.

> I don't get this. At a single level of url, the # comes after the ?, i.e.
> the fragment comes after the query string. If you want your query string to
> contain an embedded url (or partial url) then you do need to escape the
> characters that the top-level url parser recognizes or else map it in some
> way. My reference says url parsing works like this (with extra spaces for
> readability):

> scheme :// host : port / path ? query # fragment

Oh right. That is sort of what I'd expect too - but it just didn't work.
Maybe I was doing something else wrong?

If I try it like that, then the page doesn't load at all.
Nothing happens.

I tried putting in an alert to alert
me of the input string when the page loaded
and it didn't show up - the page doesn't even load
as far as I can tell (though hard to say as
there is nothing in it before it writes out
what it should write out).

As soon as I remove the # from the url string then it pops up the message
with the input string after the ? so somehow the
# is preventing my page from loading, but don't
know why.

I only tried it out with IE - which I try first
because if a script doesn't work in IE thn
really there is no point at all in considering
it as most visitors use that browser and
aren't likely to swap browser to view the
page.

You can experiment if you like with a demo
page and see if you can get it to work
- easy enough to do.

> > Fixed :-).

> So the above is sort of moot but I'll leave it FWIW.

Yes as it works with this substitituion
trick, and with all the browsers, no
great concern - however of
technical interest so I too
would be interested to know why if anyone finds out.

Robert

on 7/23/04 5:05 PM, Robert Walker <robertwalker@ntlworld.com> wrote:

> Hi Kurt,
>>> ....space....Jumpin' Dave writes: I uploaded this group. I found
>>> myself at the mercy of the encyclopedia of terms. I spent well over a
>>> hour reading terms.
>> Speaking of this, and of the whole terminology problem that was brought up
>> several days back. How about *this* idea Robert and Monz:
> ...
>> On second thought, probably a lot could be done just by Robert fairly easily
>> (?) without any help from Monz just by generating a term to url index from
>> Monz's dictionary index frame.
>
> Yes, there is something that can be done. Well one simple thing I've done
> already is to add a link to Monz's encyclopedia in the search window which you
> will find if you do a search again now.
>
> With the next upload each message has a link to the search page, so his
> dictionary will be just two clicks away from any message. If someone feels
> like doing a search by keyword of the encyclopedia index so that you enter a
> word into a form and get the best matching encyclopedia entry, then I could
> add that to the search page too.

I don't think I'll take on the search thing, but...

The encyclopedia index frame can be run through a filter that will produce
term/url pairs. Have you already done this or was your example below
hard-coded? Some of the terms will be multi-word. Can you handle that?
This is not a 100% solution but will be pretty good.

I need to be able to get the contents of the index frame - can lynx do this?
Would be good to do this in a way that will run on your machine, if
possible. But my remote server is FreeBSD and at home I have OSX, and I
think you have Windows. But I'd like to find some way to avoid yet another
periodic-update bottleneck. I might be able to give you a url that will
deliver monz'z current index page run through the filter.

> Best approach,though it requires loads more work for someone,
> would be to have a dictionary as well.

I'd just get it off the web dynamically. No?

> The dictionary just gives a couple of sentences for each term by way of
> definition, then a link to the encyclopedia.

I see. I'd just skip the short definition part for now. Its very nice but
not essential. One-click access to the encyclopedia page is already a great
value added. Fantastic in fact!

> Here is an example I did:
> http://www.robertinventor.com/test_show_defs/index.htm

This is great. Even without the short definition it would be great. The
way the words are hilited is great, tho I'm not sure how that would apply in
the color-coded case.

> I've added links there to Joe Monzo's encyclopedia
> now to show how it could be done.
>
> However, it is a big task for a single person to
> do. I reckon it would take weeks if not months
> to do dictionary definitions for every term
> in Monz's encyclopedia - weeks if you can do
> one every ten minutes, and months if it takes
> more like an hour each - and you usually
> need more words in a dictionary than an encyclopedia.

Yes, leave that for another time.

> Well if Joe wanted to do it all of course,
> but it is such a big undertaking...

The encyclopedia is already close to having a short definition at the top
for many words, and perhaps that case could be recognized and the content
incorporated in the filtering process.

>> (The next trick would be to start a migration towards tunings lists that
>> provide the same services via email, i.e. embedded hyperlinks for tunings
>> terms, and/or a list of alphabetized hyperlinks at the end of each message
>> for terms used in that message. But one step at a time, so the value of the
>> idea can be determined from experience.)
> Well it would be someone else programming that, not my
> area at all to write a tuning list script, or hack
> an existing cgi script to do this.

Yes, but it is doable, and as Monz said a good reason to put the lists on
his server. The potential just can't be beat. I would want some
assurances/insurances about how catastrophic events can be dealt with, but
I'll bet once people see what is possible here, even google groups won't be
good enough anymore. And at that point I think tonalsoft is currently our
best bet for a fantastic drop-dead integrated tunings list solution.

> But if we had such a dictionary available, then
> perhaps it could be used in just this way
> at some later date by someone with the necessary
> skills in Perl or PhP or whatever.

Yes, these skills are definitely not out of reach. I can do it but I'm not
an absolute whiz who knows exactly the best and most efficient and
up-to-date and maintainable way to do it.

-Kurt

hi Robert,

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>
wrote:

> Yes, there is something that can be done. Well
> one simple thing I've done already is to add a link to Monz's
> encyclopedia in the search window which you will find if
> you do a search again now.
>
> <snip>
>
> Here is an example I did:
>
> http://www.robertinventor.com/test_show_defs/index.htm
> Note - it doesn't yet work for the Opera browser
> at least, not version 7 of it. Hopefully that can be
> fixed somehow.
>
> I've added links there to Joe Monzo's encyclopedia
> now to show how it could be done.

this is absolutely fantastic! i really want to thank
you for all the work you've done on archiving the lists!

one small thing ... instead of calling it "Joe Monzo's
Dictionary" in the links, i would appreciate it if you
used "Tonalsoft Encyclopaedia" instead. thanks.

(as for me taking on any new jobs ... no, sorry,
too busy, can't do it right now.)

-monz

Hi Monz,

> one small thing ... instead of calling it "Joe Monzo's
> Dictionary" in the links, i would appreciate it if you
> used "Tonalsoft Encyclopaedia" instead. thanks.

Yes of course. Done.

> (as for me taking on any new jobs ... no, sorry,
> too busy, can't do it right now.)

No surprise :-).

Robert

on 7/23/04 6:17 PM, Robert Walker <robertwalker@ntlworld.com> wrote:

> Hi Kurt,
>
>> http://www.robertinventor.com/tuning-math/s__12/msg_11000-11024.html#11005
>
>> Only thing is it looks like the last line in each colorized block
>> often has one less quote than the lines before it.
>
> Yes that is intentional. It's because often the quote level varies by a number
> of indentations with blank lines To avoid stripes consisting of just blank
> lines it keeps to the same colour until it meets a reason to change it. A
> blank line never counts as a reason for changing colour, whether indented or
> not.
>
> The algorithm as it is now just writes out each line as it comes to it without
> look back. To fix it I would need to redo it so that it can do a look back.
> Not hard at all - maybe could do it sone time.

Yes, the algorithm I was thinking of that would require no additional text
buffering would be to do defer output of and simply count the number of
consecutive "blank" lines that occur at the same quote level. If a
different quote level is encountered, you flush this virtual buffer. If a
non-blank line at the same quote level is encountered then you output the
color change and then flush the buffer.

More later on the other points. Gotta go.

-Kurt

Hi Monz,

> > Anyway - do you see now why I think a dictionary is a
> > different beast.

> hmmm ... this big thing started out once as a simple
> "Dictionary of Tuning Terms". i guess i should have kept
> that and made the Encyclopaedia a different thing,
> instead of making the one grow into the other. oh well ...
> too late to stop me now! ;-)

Well one person can't do everthing :-).

I'm just like that too, start on what should be a simple
project and it grows and grows quite out of hand
- not that a dictionary would be any simpler a project
if one let that grow as well as much as it could
- each entry would need less time but it would probably
have more words in it than an encyclopedia eventually
one would think, if the comparison with normal dictionaries
is any guide.

Anyway I've just had a thought about it, which is
that even a partial dictionary is better than nothing.
Even if it only has say fifty definitions or
something, if they are commonly used ones,
it could be useful.

Some terms occur over and over again such as
"unision vector" in tuning math. When I
get down to working on this again i
can get it to do a frequency count
for all the terms in your encyclopedia.
So we know which are the most important
ones to do at least in terms of frequency
and can do those ones first.

Once it is underway and people see
how useful it is (I hope) then
maybe it will grow naturally
afer that somehow, so it will
be like a small seed that could
grow into a large dictionary
later on, you never know.
If not found useful then it can just
be removed at that point before
very much work has been done on it.

Does anyone know a good way of
doing content management easily.
I just looked up wiki's
and it seems they are very complex
to install, I wonder if there is
any wiki space around that one
can just use for free?
Or just some way of communally
editing a text file would do.

I've just done a first shot
at a dictionary definition of
unison vector:

unison vector

By way of example fourteen 3/2s are similar
in pitch to a 7/4 (up to octave equivalence)
- the difference in pitch is the septimal schisma of
about 4 cents. If you put these pitches into the 3/2, 7/4
lattice and treat them as equivalent,
this makes (14,1) into a unison vector.
With enough unison vectors, an infinite
lattice can collapse into a finite periodicity block.
This lets you reduce it to a scale with a finite
number of pitches - you can then temper out the unison
vectors, or use them as xenharmonic bridges.

Surely it can be improved.

Robert

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>

/tuning/topicId_54749.html#54835

>
> I've just done a first shot
> at a dictionary definition of
> unison vector:
>
> unison vector
>
> By way of example fourteen 3/2s are similar
> in pitch to a 7/4 (up to octave equivalence)
> - the difference in pitch is the septimal schisma of
> about 4 cents. If you put these pitches into the 3/2, 7/4
> lattice and treat them as equivalent,
> this makes (14,1) into a unison vector.
> With enough unison vectors, an infinite
> lattice can collapse into a finite periodicity block.
> This lets you reduce it to a scale with a finite
> number of pitches - you can then temper out the unison
> vectors, or use them as xenharmonic bridges.
>
> Surely it can be improved.
>
> Robert

***Personally I don't find this entry any clearer than the one that
Monz already has in his dictionary... Erlich really needs to do
this. His most recent paper is the clearest explication of some of
these concepts of tuning I have yet seen...

JP

Hi Joseph,

> ***Personally I don't find this entry any clearer than the one that
> Monz already has in his dictionary... Erlich really needs to do
> this. His most recent paper is the clearest explication of some of
> these concepts of tuning I have yet seen...

Yes, I agree. It isn't significantly shorter either now that I look
at it again.

I'm not attempting to replace the encyclopedia. Just
to have a quick reference look up like a dictionary
and this one didn't work.

Yes if Paul can do a shorter version that's great.
It's one thing to know something and another thing
to explain it clearly and he has probably had
the most experience of anyone in explaining what
a unison vector is.

I have made another attempt which is perhaps
rather more in the style of a dictionary entry -
I don't know if it is any better - it is a little
bit shorter anyway:

"
Unison vector:

Method to reduce an infinite lattice of
pitches to a finite scale, called a periodicity
block. Five limit just intonation twelve tone scales
are periodicity blocks obtained
using the syntonic comma 81/80 and diesis
125/128 as unison vectors, because you
only have one each of notes a comma or
diesis apart from each other.
"

This is surely one of the harder ones
- well many of the mathemtical ones are
going to be pretty hard to do in
a couple of sentences, but one can try.

A self contained definition is easy enough for pentatonic,
aeolian, and septimal and such like words, and
that will be best if possible.

But if not one can try for a short dictionary
style definition that says enough to serve
as a reminder for someone who has
read about the concept before but has perhaps
forgotten what it means precisely.

Robert

Hi Kurt,

> Well anyway I'm still thinking about delivering you term/url pairs if you
> are interested, so please clarify for me whether you can handle multi-word
> "terms" and whether you think the "raw" approach I suggested above would be
> worth doing.

Yes it could be useful indeed. I can parse the contents file myself
and extract them - but if you want to do that for me then it will save a
little work. The easiest format for my program to read would perhaps be to
start each line with the url and then follow with the word(s) for the term
all on one line.

Yes I can handle multi-word terms. I will just ignore multiple white space and >
characters between words when doing the comparisions - that's very easy to do.

I've just found a way of doing the dictionary definitions
as a "tooltip"s using the overlib tool tip library:

http://www.bosrup.com/web/overlib/
which works with all the main browsers they say.

Here it is with tool tips:

http://www.robertinventor.com/test_overlib_tool_tips/index.htm

So the obvious thing really is to make a complete
dictionary from Joe Monzo's contents list - but
leave all the undefined entries blank.

So the undefined ones will just pop up a tip
saying "Look up in Tonalsoft Encyclopedia"
and that's all they say.

Then we can add short definitions above
that for more and more entries as time
goes on. So that gives both at once :-).

If the dictionary idea isn't felt to be helpful
after I have done an experimental upload using
a few to show what it would be like,
then it will be an easy matter to leave out the definitions
altogether, and have tool tips that pop up and lead
to the encyclopedia and leave it at that.

Robert

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>

/tuning/topicId_54749.html#54879

> Unison vector:
>
> Method to reduce an infinite lattice of
> pitches to a finite scale, called a periodicity
> block. Five limit just intonation twelve tone scales
> are periodicity blocks obtained
> using the syntonic comma 81/80 and diesis
> 125/128 as unison vectors, because you
> only have one each of notes a comma or
> diesis apart from each other.
> "
>

***Hi Robert,

Somehow it seems like the entry would have to show that *one* pitch
is *replacing* or *representing* two or more by temperament...

I think the general idea would have to come across before the
specific examples..

Paul is really the person to do this, I believe...

JP

>> Unison vector:
>>
>> Method to reduce an infinite lattice of
>> pitches to a finite scale, called a periodicity
>> block.

Great, except it isn't a method. Possibly a
means. But definitely, it's an *interval*.

>> Five limit just intonation twelve tone scales
>> are periodicity blocks

Not all of them. Try a specific example.

-Carl

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...> wrote:

> Unison vector:
>
> Method to reduce an infinite lattice of
> pitches to a finite scale, called a periodicity
> block.

A comma can hardly be a method. Why not say "a comma used to define
the bounds of a periodicity block"?

Hi Gene,

> A comma can hardly be a method. Why not say "a comma used to define
> the bounds of a periodicity block"?

Thanks for the suggestion. Sounds a good
basis. I'll give it a go. How about this
as the latest attempt:

unison vector:

A small interval of equivalence used to mark
out a finite scale (a periodicity block) within
a lattice of pitches.

Ex. if you use the syntonic comma 81/80 and diesis
125/128 as unison vectors in the (5/4, 3/2) lattice,
you will get only one each of pairs of pitches
a comma or diesis apart:
16/15.....8/5.....6/5.....9/5...(27/20 ~= 4/3)
\ / \ / \ / \ / \
4/3.....1/1.....3/2.....9/8...(27/16 ~= 5/3)
\ / \ / \ / \ / \
5/3.....5/4.....15/8....45/32..(135/64 ~= 25/24)
\ / \ / \ / \ /
(25/24...25/16...75/64...225/128)
\ / \ / \ / \
(~= 16/15....8/5.....6/5.....9/5)
So in this case, you get a twelve tone scale, with
the choice of pitches varying depending on the shape
of the periodicity block and its position
in the lattice.

Robert

Hi Gene,

I didn't explain the reasoning behind changing the wording
of your definition which may seem simpler and not in need of
change as indeed it probably isn't for those familiar
with the field.

It was all in the interest of helping the reader.
So for instance maybe a newbie
has forgotten that a periodicity block is a scale
even - or that it is a region of a lattice
- or that the lattice is infinite and is a lattice
of pitches. The diagram would help with the example,
but I also worded the definition to help give
hints in various ways of what the definition is
all about - the more hints the better their chance
of quickly understanding the concept (I hope).

At least that was the idea and motivation behind
it, don't know if I succeeded or not of course.

Robert

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...> wrote:
> Hi Gene,
>
> > A comma can hardly be a method. Why not say "a comma used to define
> > the bounds of a periodicity block"?
>
> Thanks for the suggestion. Sounds a good
> basis. I'll give it a go. How about this
> as the latest attempt:
>
> unison vector:
>
> A small interval of equivalence used to mark
> out a finite scale (a periodicity block) within
> a lattice of pitches.

This would make more sense to me if it said "lattice of pitch classes".

>> A small interval of equivalence used to mark
>> out a finite scale (a periodicity block) within
>> a lattice of pitches.
>
>This would make more sense to me if it said
>"lattice of pitch classes".

Why's that? They're pitches in the Tenney lattice.

-Carl

--- In tuning@yahoogroups.com, Carl Lumma <ekin@l...> wrote:

> >> A small interval of equivalence used to mark
> >> out a finite scale (a periodicity block) within
> >> a lattice of pitches.
> >
> >This would make more sense to me if it said
> >"lattice of pitch classes".
>
> Why's that? They're pitches in the Tenney lattice.

You wouldn't normally think 2 is a comma which you are tempering out,
but if you leave it be, you don't get a finite region.

Hi Gene,

> This would make more sense to me if it said "lattice of pitch classes".

Yes I understand, that's what a mathematician would say
- a lattice of pitch classes under octave equivaleence.

The periodicity block itself could also be considered
a finite set of pitch classes under the equivalence
relations induced by the unison vectors too.

So e.g. 5/4, or {5} perhaps, would stand for the class
{25/3^4, 5, 3^4, 3^8/5, 3^12/25,...}
and so on so every twelve tone scale consists
of unique representatives chosen from the
twelve pitch classes in the lattice
under the equivalence relations induced
by the lattice vectors.

For a mathematician, that is a clearer way
of putting it - now you understand exactly
what is going on. But for nearly everyone
else it actually obscures the point
and makes it less clear, and I'm not
addressing a mathematician here.

If you are a mathematician you are used to thinking
of a set as an element of other sets but for
non mathematicians that way of thinking is a bit
abstract, while the idea of things being the
same or similar is more accessible as
an intuitively understood notion which a
mathetmatician would express in mathematical
language as a class of elements equivalent under
an equivalence relation. But it is the same
notion and there isnt' any risk of confusion.
We need more precise language in maths
because sometimes there is great risk
of confusion, but not here.

Robert

Hi Gene,

Surely you will know what I mean of course,
but anyway just to correct a slip of the keyboard
in what I wrote:

So e.g. 5/4, or {5} perhaps, would stand for the class
{25/3^4, 5, 3^4, 3^8/5, 3^12/25,...}
and so on so every twelve tone scale consists
of unique representatives chosen from the
twelve pitch classes in the lattice
under the equivalence relations induced
by the unison vectors.

(unison vectors, not lattice vectors, sorry).

- well actually it's under the two unison vectors +
the octave equivalence which I left
understood there. A mathematician would
perhaps introduce some notation to
show that 5 there for instance
actually stands for {..., 5/4, 5/2, 5, 10, 20,...}
though it is fairly common in maths to use
a representative of an equivalence class to
stand for the whole class in formulae so
long as it is understood what is meant.

It is the sort of thing one would think
over carefully when writing a paper
for a maths journal. But I think
out of place here. Though if someone
did an alternative dictionary
of tuning theory for mathematicians it would of course
be done like that, or something similar.

Robert

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...> wrote:

> It is the sort of thing one would think
> over carefully when writing a paper
> for a maths journal. But I think
> out of place here.

I thought "PC sets", meaning sets of pitch classes, where a big deal
in American "set theory" musical theory and constantly in use.

Hi Gene,

> I thought "PC sets", meaning sets of pitch classes, where a big deal
> in American "set theory" musical theory and constantly in use.

Yes, I've come across that as well - maybe they needed to use
the concept of a set rather than just an intuitive notion of
"sameness" for some reason. Maye there was some ambiguity
it resolved in their theory, so that it was needed.

If that weren't the reason, then I'd think the same way about
it as I do here that it is adding extra mathematical
complexity in a place where it isn't needed and
has potential to impede understanding in non
mathematicians. But I rather expect there must be
some reason why they needed to use it. Here
everyone is getting on perfectly well without
referring to pitch classes and I think it
will only confuse things to introduce
the concept. They just talk about a lattice
and regard the elements as being the
"same" under octave equivalence, which
is a natural thing to do as that is a
heard similarity.

While being in the same pitch class set - I don't
know how it works, but maybe it is less intuitive
and immediate than octave equivalence in some
way??

Robert

hi Joseph,

--- In tuning@yahoogroups.com, "Joseph Pehrson" <jpehrson@r...> wrote:

> [re: Robert Walker's brief definition of "unison-vector"]
>
> ***Personally I don't find this entry any clearer than the
> one that Monz already has in his dictionary... Erlich really
> needs to do this. His most recent paper is the clearest
> explication of some of these concepts of tuning I have yet
> seen...
>
> JP

i agree with you about Paul's paper.

meanwhile, lately i've been doing a lot of upgrading to the
Encyclopaedia, and you should check some terms like "wedgie",
"multimonzo", and "prime-mapping".

i'm finally beginning to understand Gene's concepts and
can start trying to present them in a way that "ordinary"
people will understand, or at least get a glimmer of.

i'll be doing lots more work on all of these new terms
as i have time, especially to add the pretty pictures.

:)

-monz

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>

/tuning/topicId_54749.html#54913

> While being in the same pitch class set - I don't
> know how it works, but maybe it is less intuitive
> and immediate than octave equivalence in some
> way??
>
> Robert

***Most probably I'm just misunderstanding this... but Pitch Class
sets (PC) *imply* octave equivalence... It's the same thing...

JP

--- In tuning@yahoogroups.com, "Joseph Pehrson" <jpehrson@r...> wrote:
> --- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>
>
> /tuning/topicId_54749.html#54913
>
> > While being in the same pitch class set - I don't
> > know how it works, but maybe it is less intuitive
> > and immediate than octave equivalence in some
> > way??
> >
> > Robert
>
>
> ***Most probably I'm just misunderstanding this... but Pitch Class
> sets (PC) *imply* octave equivalence... It's the same thing...

I thought they were just sets of equivalence classes, which would
indeed imply octave equivalence.

hi Gene and Robert,

--- In tuning@yahoogroups.com, "Robert Walker" <robertwalker@n...>
wrote:

> Hi Gene,
>
> > I thought "PC sets", meaning sets of pitch classes,
> > were a big deal in American "set theory" musical theory
> > and constantly in use.
>
> Yes, I've come across that as well - maybe they needed to use
> the concept of a set rather than just an intuitive notion of
> "sameness" for some reason. Maye there was some ambiguity
> it resolved in their theory, so that it was needed.

you guys are high-powered with the math, but i can probably
add some historical background ...

modern PC-set theory really has its origins in Milton
Babbitt's compositions and theory papers from the early
1960s. he invented the term "pitch-class".

Babbitt was primarily a 12-tone-serial composer, stepped
in the method invented by Schoenberg, and adding plenty
of his own new twists and complexities as his career
progressed.

from a mathematical point of view, Babbitt's compositions
and theories are extremely interesting. the good thing
is that, at least for me, his music sounds cool too.

there used to be a webpage around which had MIDIs of
his early _3 Compositions for Piano_, which i love.
can't find it now, but here's one with 2 other pieces:

http://www.bright.net/~dlphilp/dp_music2.html

Babbitt's PC-set theory has indeed been the vehicle of
a tremendous amount of analysis in academic circles and
in their journals.

i've been interested for a long time in extending that
to microtonal tunings, and perhaps even JI. when i
first got the idea to use what Gene now calls "monzos"
(prime-factor-exponent-vectors), my purpose was to create
an 8ve-equivalent mathematical notation for JI which
could be manipulated in ways similar to Babbitt's
12 integers 0 ... 11 mod-12.

-monz

--- In tuning@yahoogroups.com, "monz" <monz@a...> wrote:

> Babbitt was primarily a 12-tone-serial composer, stepped
> in the method invented by Schoenberg,

oops, my bad. that's "... steeped in the methed ..."

-monz

--- In tuning@yahoogroups.com, "Gene Ward Smith" <gwsmith@s...> wrote:

/tuning/topicId_54749.html#54924

> --- In tuning@yahoogroups.com, "Joseph Pehrson" <jpehrson@r...>
wrote:
> > --- In tuning@yahoogroups.com, "Robert Walker"
<robertwalker@n...>
> >
> > /tuning/topicId_54749.html#54913
> >
> > > While being in the same pitch class set - I don't
> > > know how it works, but maybe it is less intuitive
> > > and immediate than octave equivalence in some
> > > way??
> > >
> > > Robert
> >
> >
> > ***Most probably I'm just misunderstanding this... but Pitch
Class
> > sets (PC) *imply* octave equivalence... It's the same thing...
>
> I thought they were just sets of equivalence classes, which would
> indeed imply octave equivalence.

***Hi Gene,

Yes, that's the way they teach it in music school... whether it's
mathematically correct or not... :)

JP

--- In tuning@yahoogroups.com, "monz" <monz@a...> wrote:

/tuning/topicId_54749.html#54927

> --- In tuning@yahoogroups.com, "monz" <monz@a...> wrote:
>
> > Babbitt was primarily a 12-tone-serial composer, stepped
> > in the method invented by Schoenberg,
>
>
> oops, my bad. that's "... steeped in the methed ..."
>
>
> -monz

***Fortunately, he was able to clean his shoes after that happened...

JP

on 7/25/04 6:00 PM, monz <monz@attglobal.net> wrote:

> meanwhile, lately i've been doing a lot of upgrading to the
> Encyclopaedia, and you should check some terms like "wedgie",
> "multimonzo", and "prime-mapping".
>
> i'm finally beginning to understand Gene's concepts and
> can start trying to present them in a way that "ordinary"
> people will understand, or at least get a glimmer of.

That's great! I'm excited to read about wedgie.

> i'll be doing lots more work on all of these new terms
> as i have time, especially to add the pretty pictures.

While you are at it, what do you think about the idea of trying to have a
short definition at the beginning of each page which would serve as a
dictionary definition. Would that work within the current form, such that
you might think about adding such at thing at the top when you are working
on a page? If it followed a standard enough form then a definition
extractor could pull this off the webpage for uses in other contexts. Or
should the dictionary be truly separate?

-Kurt

on 7/25/04 9:40 AM, Robert Walker <robertwalker@ntlworld.com> wrote:

> Hi Kurt,
>
>> Well anyway I'm still thinking about delivering you term/url pairs if you
>> are interested, so please clarify for me whether you can handle multi-word
>> "terms" and whether you think the "raw" approach I suggested above would be
>> worth doing.
>
> Yes it could be useful indeed. I can parse the contents file myself
> and extract them - but if you want to do that for me then it will save a
> little work. The easiest format for my program to read would perhaps be to
> start each line with the url and then follow with the word(s) for the term
> all on one line.

Ok, will discuss remaining details with you off-list.

> Yes I can handle multi-word terms. I will just ignore multiple white space and
> >
> characters between words when doing the comparisions - that's very easy to do.
>
> I've just found a way of doing the dictionary definitions
> as a "tooltip"s using the overlib tool tip library:
>
> http://www.bosrup.com/web/overlib/
> which works with all the main browsers they say.
>
> Here it is with tool tips:
>
> http://www.robertinventor.com/test_overlib_tool_tips/index.htm
>
> So the obvious thing really is to make a complete
> dictionary from Joe Monzo's contents list - but
> leave all the undefined entries blank.
>
> So the undefined ones will just pop up a tip
> saying "Look up in Tonalsoft Encyclopedia"
> and that's all they say.
>
> Then we can add short definitions above
> that for more and more entries as time
> goes on. So that gives both at once :-).
>
> If the dictionary idea isn't felt to be helpful
> after I have done an experimental upload using
> a few to show what it would be like,
> then it will be an easy matter to leave out the definitions
> altogether, and have tool tips that pop up and lead
> to the encyclopedia and leave it at that.

This all sounds good.

I'm not sure I like tool tips though. These work better than some, but
having the separate frame you used before has the advantage of allowing you
to mouse over things without obscuring other things you may also be looking
at. If there is a "cookie" way to make it an option to use tool tip versus
separate frame that'd be good, I think. The separate frame was so
uninvasive. Except of course it takes screen space! So I can see why it is
useful. Options are great, for everybody except the person implementing all
the options!

-Kurt

>
> Robert

--- In tuning@yahoogroups.com, Kurt Bigler <kkb@b...> wrote:

/tuning/topicId_54749.html#54930

> on 7/25/04 6:00 PM, monz <monz@a...> wrote:
>
> > meanwhile, lately i've been doing a lot of upgrading to the
> > Encyclopaedia, and you should check some terms like "wedgie",
> > "multimonzo", and "prime-mapping".
> >
> > i'm finally beginning to understand Gene's concepts and
> > can start trying to present them in a way that "ordinary"
> > people will understand, or at least get a glimmer of.
>
> That's great! I'm excited to read about wedgie.
>
> > i'll be doing lots more work on all of these new terms
> > as i have time, especially to add the pretty pictures.
>
> While you are at it, what do you think about the idea of trying to
have a
> short definition at the beginning of each page which would serve as
a
> dictionary definition. Would that work within the current form,
such that
> you might think about adding such at thing at the top when you are
working
> on a page? If it followed a standard enough form then a definition
> extractor could pull this off the webpage for uses in other
contexts. Or
> should the dictionary be truly separate?
>
> -Kurt

***I agree with Kurt here, too. That would be a terrific idea...

JP

hi Kurt,

--- In tuning@yahoogroups.com, Kurt Bigler <kkb@b...> wrote:
> on 7/25/04 6:00 PM, monz <monz@a...> wrote:
>
> > meanwhile, lately i've been doing a lot of upgrading to the
> > Encyclopaedia, and you should check some terms like "wedgie",
> > "multimonzo", and "prime-mapping".
> >
> > i'm finally beginning to understand Gene's concepts and
> > can start trying to present them in a way that "ordinary"
> > people will understand, or at least get a glimmer of.
>
> That's great! I'm excited to read about wedgie.

all the new links that i posted are related, so "wedgie"
will take you to "multimonzo", which takes you to "val",
etc.

i'm hoping that everyone starts seeing why wedgies are
so important. Gene has taken a lot of bashing by other
members of these lists, but it's been my gut feeling right
from his first appearance here about two years ago
that his ideas are tremendously important to the world
of tuning theory.

i wish i had enrolled in my math courses then ...
by now, i'd be past all the trig and calculus prerequisites
and would be digging into linear algebra, which
is what i need to understand Gene's work and which
unfortunately i won't reach now until 2006. :(

i take it as my task to translate Gene's elegant
mathematical definitions and dry lists of numbers
into pretty pictures that anyone can understand,
or at least enjoy. they'll be especially informative
when we make the interactive 3-D Lattice Viewer available.
the more complex tunings don't give away many secrets
on a 2-D diagram, but rotating a 3-D one can show a lot.

> > i'll be doing lots more work on all of these new terms
> > as i have time, especially to add the pretty pictures.
>
> While you are at it, what do you think about the idea of
> trying to have a short definition at the beginning of
> each page which would serve as a dictionary definition.
>< Would that work within the current form, such that
> you might think about adding such at thing at the top
> when you are working on a page? If it followed a standard
> enough form then a definition extractor could pull this
> off the webpage for uses in other contexts. Or should
> the dictionary be truly separate?
>
> -Kurt

as i said a few posts ago, the Encyclopaedia of Tuning
originally started out as a small Dictionary of Tuning Terms.

many pages still have only very small definitions, many
of them generously donated to the cause by John Chalmers.

generally, i *do* try to put a short, concise dictionary-type
definition at the very beginning, and then the larger pages
flesh that out with lots of math, graphics, and MIDI files.

unfortunately, with the explosion of "jargon" in the last
couple of years, which i've had a hard time keeping up with,
i've had to resort to simply copying tuning list or
tuning-math posts *as* the definition ... so those will
still need a new short dictionary definition at the beginning.

anyway, yes, i'll keep trying to do this. and others are
invited to write their own concise definitions to supplement
a page that may need one at the top.

-monz

Hi David,

> Unison vector: The interval between two points on the infinite lattice that
> are considered equivalent when those equivalences are being used to reduce
> the infinite lattice to a finite periodicity block. For example, on the
> two-dimensional 5-limit lattice, one might wish to consider a move four
> steps along the positive 3 axis (i.e. up 81/64) to be equivalent to one step
> along the positive 5 axis (i.e. up 80/64 = 5/4). The ratio between these two
> pitches (i.e. 81/80) is a unison vector. For some purposes the unison vector
> may be written as an actual vector (say <4,-1> is this example or <-4,4,-1>
> if we wish to include factors of two in the cooridinate system) instead of a
> frequency ratio.

Yes fine. It is a typical mathematical type definition.
But it does put some demands on the reader, they need
to puzzle things out. That's not necessarily a bad
thing at all, but it may be interesting to explain
as this often isn't at all obvious to the author.

So, for instance, it assumes that the user knows that the lsttice
is a lattice of pitches, and that they understand that a periodicity
block is a scale, and that by the 3 axis you mean the axis
along which the pitch increments by multiples of 3, and
that they are reasonably comfortable with the idea
of an interval as a vector, and that they can see that
81/64 is the same as 3^4/64.

Try reading it again and try to forget that you know
that a lattice is a lattice of pitches, and
that the 3 axis means the direction along which pitch
increments by intervals of 3/2, and try to forget
your ability to see readily that 81 is the same
as 3^4 (through familiarity) see if you can
figure it out...

By the end of your definition it is quite possible
that a newbie reader may have no idea yet
for instance that a 5-limit just intonation twelve
tone scale is a periodicity block. They may
think they understand it, but if questioned
they might not yet have got that point.

So it depends who it is targetted towards. My definition
was meant for a complete newbie who isn't very
familiar with the idea of a lattice of pitches
at all. So I take the view that they need
as many hints as one can and if one can
hint at things several times so much
the better. That is why it seems back to front -
because I start with the things most familiar
to a newbie rather than the things most fundamental
to the definition.

So you write it out, look at it,
think whether there are any hints you can
add, and whether it can be made to say
the same thing with less technical language,
write it out again, look at it again,
change the order,...

It can take a long time to reduce
a technical definition to a non technical
one. I think of the two the non technical
one is possibly slightly the harder to do.

But readers don't need to stop and puzzle things
out with the less techy style, and there is something
to be said for requiring them to do so
as things you puzzle out may be remembered
more readily because you put more work into
understanding it.

Also I find mathematicians rarely give
concrete examples of the thing they are
defining, tending to leave it up to
the reader to make one up to test
the definition. That again is fine
if you are doing the puzzle type
definition, because the reader will
learn a lot from making up their own example
to fit the definition, but giving one
well chosen concrete example
is helpful if your aim is rather
to minimize puzzlement.

In short, your definition is probably
a suitable one for typical
denizons of the tuning-math
list or mathematician visitors
- while mine is perhaps
more suitable for a visitor
from the main tuning list
who wants to try and figure
out what they are all talking
about over there :-).

Maybe we need two versions
of the dictionary, perhaps again
set using a cookie, or three
even - one with Gene's style
ultra short definition:

A comma used to define the bounds of a periodicity block.

which does have a kind of minimalist poetry
to it doesn't it :-).

I'm not sure what this means in terms
of organising and collaborating,
but maybe we need three dictionaries.

Or, have a mix of entries in
different styles depending on the
contributor perhaps.

Another idea is to have a multiple
definition dictionary source
and have some way for the web
master (rather than the visitor)
to choose which one they want to
set as the preset one, perhaps with
some way for the visitor to choose between them
using cookies.

Robert

Hi Joseph,

> ***Most probably I'm just misunderstanding this... but Pitch Class
> sets (PC) *imply* octave equivalence... It's the same thing...

Oh right, thanks for explaining.
So Gene is right then, they are just pitch class sets
so for a composer familiar with the idea it would indeed
make sense to call it a lattice of pitch class sets.

I enjoyed the Babbit pieces - thanks for the link to them.

Robert

Hi Monz

> anyway, yes, i'll keep trying to do this. and others are
> invited to write their own concise definitions to supplement
> a page that may need one at the top.

Well you are very welcome to use any of mine :-).
Or - how about making it so that the word is
highlighted and when you hover the mouse over
it you see the definition as a tool tip?

You could use the dictionary like that and
have tips for all the terms in the dictionary
- one could do a search and replace of all
the pages to add tips to the words used
indeed.

To take an example, your Aeolian page
doesn't really need an extra definition
at the top does it - because that
would mean you need to say it
all twice, the second time in a little
more detail. But if the word aeolian
was highlighted and a dictionary
definition popped up when you move
the mouse over it, then that
would be unobtrusive and avoid
repetition.

I'd want the dictionary to be
available for any web master or
software author to use for their
web pages or documentation.
Just include the dictionary,
and highlight the words in it
by a method that is quite easy
to use for anyone used to working
with the raw html.

Maybe indeed, I can do a free tool
to automatically search and replace
the text in web pages to add tips for
the dictionary definitions
- I can do that for Windows users
and indeed will probably write one
anyway to add tips to the help
for Fractal Tune Smithy.

Maybe someone else can write
a similar program for other platforms
too.

Just a suggestion.

Robert

Hi Kurt,

> I'm not sure I like tool tips though. These work better than some, but
> having the separate frame you used before has the advantage of allowing you
> to mouse over things without obscuring other things you may also be looking
> at. If there is a "cookie" way to make it an option to use tool tip versus
> separate frame that'd be good, I think. The separate frame was so
> uninvasive. Except of course it takes screen space! So I can see why it is
> useful. Options are great, for everybody except the person implementing all
> the options!

Yes, I can see that. Can perhaps do that if I can find a way to get it to work
with opera - or anyway and just have a warning that it doesn't
work for opera.

Meanwhile though, and less work :-), the overlib library has an option to offset the
tool tip by any desired amount horizontally or vertically, so
perhaps that will do the trick:

http://www.robertinventor.com/test_overlib_tool_tips/index.htm

You might sometimes need to scroll down the page to see the complete
tool tip - but as it is sticky you can do so after it shows up.

Robert

>the overlib library has an option to
>offset the tool tip by any desired
>amount horizontally or vertically, so
>perhaps that will do the trick:
>
>http://www.robertinventor.com/test_overlib_tool_tips/index.htm

Great stuff, Robert!

-Carl

--- In tuning@yahoogroups.com, Carl Lumma <ekin@l...> wrote:
> >the overlib library has an option to
> >offset the tool tip by any desired
> >amount horizontally or vertically, so
> >perhaps that will do the trick:
> >
> >http://www.robertinventor.com/test_overlib_tool_tips/index.htm
>
> Great stuff, Robert!
>
> -Carl

yes, i agree! i like the tool-tip method of showing
the short definitions.

-monz

--- In tuning@yahoogroups.com, Carl Lumma <ekin@l...> wrote:

/tuning/topicId_54749.html#54959

> >the overlib library has an option to
> >offset the tool tip by any desired
> >amount horizontally or vertically, so
> >perhaps that will do the trick:
> >
> >http://www.robertinventor.com/test_overlib_tool_tips/index.htm
>
> Great stuff, Robert!
>
> -Carl

***Robert's "mouseovers" are really great here...

JP