Software that makes placemats

[jdaw1]

First use of droplets was at the 1966 horizontal on 10th Sept 2019 (placemats and parameters by which made).

In this DropletsCharges, there is a +20 attractive charge on Dow; -2 repelling charge on Offley; and other small charges not relevant for this calculation (+1 Cr; -3 G-2016); +1 GC). So going from Dow to a little past Offley, somewhere in Taylor’s circle, is a balance point (not quite Lagrange, nor quite a termination shock) which — for my palate — gives some intricate structure. Different opinions welcomed.

Where is this balance point? At a distance d past Offley, measured in radii, in Mathematica notation

Solve[{D[20/(d + 2)^2 - 2/d^2, d] == 0 && d > 0}, d]

That’s the real root of 9 d³ − 6 d² − 12 d − 8 = 0   ⇒   d ≈ 1.73, which is in happy optical agreement with the output.

[akzy]

I've never massively been a fan of these turning points, always caused me numerical nightmares. I'd be tempted to shift charges to hide it in the 'T' of Taylor's.

[jdaw1]

I've never massively been a fan of these turning points, always caused me numerical nightmares. I'd be tempted to shift charges to hide it in the 'T' of Taylor's.

But we’re doing different things. You are precisely modelling something difficult. Whereas I am using a simulation of a physical process to make something aesthetically appealing.

So I disagree.

[akzy]

I feel all the years of torment have had a terrible effect on me.

Where is this balance point? At a distance d past Offley, measured in radii, in Mathematica notation

Solve[{D[20/(d + 2)^2 - 2/d^2, d] == 0 && d > 0}, d]

That’s the real root of 9 d³ − 6 d² − 12 d − 8 = 0   ⇒   d ≈ 1.73, which is in happy optical agreement with the output.

Having a quick look at this, I notice that d~1.73~sqrt(3). Do you have any insight as to why this is the case?

[jdaw1]

I feel all the years of torment have had a terrible effect on me.

Where is this balance point? At a distance d past Offley, measured in radii, in Mathematica notation

Solve[{D[20/(d + 2)^2 - 2/d^2, d] == 0 && d > 0}, d]

That’s the real root of 9 d³ − 6 d² − 12 d − 8 = 0   ⇒   d ≈ 1.73, which is in happy optical agreement with the output.

Having a quick look at this, I notice that d~1.73~sqrt(3). Do you have any insight as to why this is the case?

Coincidence: d = (1 + 10^⅓ + 10^⅔) × 2/9; d² = (7 + 4×10^⅓ + 10^⅔) × 4/27 ≈ 3.00138.

[akzy]

Just a thought on a small modifier you could make to this package. Consider a dipole charge structure, (i.e. a closely spaced +ve and -ve charge, perhaps hidden in some text?) and you'd have magnetic dipole field lines!

[jdaw1]

At yesterday’s ’66 horizontal some claimed that the droplets resembled a uniflagellar haploid gamete produced by a male mammal (not quite the original phrasing).

For my taste such a gamete has a very clear boundary between head and tail. This is not true of these droplets, for which the thickness is linear in the distance along central path. So I disagree, but nonetheless do not wish such thing to be falsely perceived.

One possible change would be to increase the value of the parameter DropletsOuterWidthStart.

With DropletsOuterWidthStart = 0.48, as was true for the ’66 placemats

With DropletsOuterWidthStart = 1.44:

With DropletsOuterWidthStart = 2.16:

With DropletsOuterWidthStart = 2.88:

What’s the smallest value of DropletsOuterWidthStart such that the droplets don’t appear to be sperm cells?

Animated:

[DaveRL]

To me the original don't look like sperm, which have a bulbous head and long wiggly tail. I prefer the original, but if there has to be a change would vote for the smallest change tolerable. I guess those that see the shapes as sperm have a wild time with Paisley Pattern.

[PhilW]

Coincidence: d = (1 + 10^⅓ + 10^⅔) × 2/9;

Which can be simplified to d = 2/(10^⅓ - 1)
(I don't think that helps in any way, nor is likely of much interest to anyone, but I noticed it could be simplified so...)

[PhilW]

What’s the smallest value of DropletsOuterWidthStart such that the droplets don’t appear to be sperm cells?

An observation; you have chosen to replace the arrow-headed lines with small linear shapes all heading towards one or more large circles. The impression may not only be a consequence of your linear shapes. However, presumably if we can replace the arrow-headed lines with sperm-like shapes, then we could potentially use many alternatives which might generate different interpretations - ants spring to mind as an example.

[jdaw1]

To me the original don't look like sperm, which have a bulbous head and long wiggly tail. I prefer the original

Fully agree, but significantly many seem not to.

Which can be simplified to d = 2/(10^⅓ - 1)

Is the surdish denominator a simplification? Maybe — tastes differ.

An observation; you have chosen to replace the arrow-headed lines with small linear shapes all heading towards one or more large circles. The impression may not only be a consequence of your linear shapes. However, presumably if we can replace the arrow-headed lines with sperm-like shapes, then we could potentially use many alternatives which might generate different interpretations - ants spring to mind as an example.

The intention was ‘Droplets’, which fit the drinking theme. Surely we don’t want ants on the table, nor sperm.

[jdaw1]

Would shorter droplets work?

[DRT]

I'm beginning to regret mentioning this. What my comment was intended to convey is that the small part of my brain that still thinks like a 15 year old boy^† recognised something in this design that made him snigger. That was it really.



^† I believe it to be a scientific fact that all men above the age of 15 have this part of their brain.

[jdaw1]

Decision:

[DaveRL]

^† I believe it to be a scientific fact that all men above the age of 15 have this part of their brain.

True. My wife sighs, while my sons and I giggle. I'm a bad influence, apparently.

[jdaw1]

True. My wife sighs, while my sons and I giggle. I'm a bad influence, apparently.

You reprobate! Obviously, obviously, this has never happened to me.

[jdaw1]

Just a thought on a small modifier you could make to this package. Consider a dipole charge structure, (i.e. a closely spaced +ve and -ve charge, perhaps hidden in some text?) and you'd have magnetic dipole field lines!

Sorry, forgot to report. The structure of DropletsCharges has been changed to overload the position specification. As before it can be an integer, a WithinPage glass number; now also allowed is an array of the form [x y], in which [0 0] is bottom-left of the page. Which allows positive and negative charges to be very close: example, with four charges arranged as two dipoles at 90° to each other, the charges within each dipole being apart by 12pt = ⅙″ = 4.2⅓mm. (Separately, note the file size is 437kb, which drops to 106kb with /Droplets false def.)

[akzy]

Sorry, forgot to report. The structure of DropletsCharges has been changed to overload the position specification. As before it can be an integer, a WithinPage glass number; now also allowed is an array of the form [x y], in which [0 0] is bottom-left of the page. Which allows positive and negative charges to be very close: example, with four charges arranged as two dipoles at 90° to each other, the charges within each dipole being apart by 12pt = ⅙″ = 4.2⅓mm. (Separately, note the file size is 437kb, which drops to 106kb with /Droplets false def.)

The dipole example looks really sharp. Looking forward to seeing it in some placemats.

[jdaw1]

The dipole example looks really sharp. Looking forward to seeing it in some placemats.

I don’t like it as much as the examples with isolated poles. YMMV.

For those wanting something like that, I did it in a slightly complicated way — numbers could have been used instead of formulae.

Code: Select all

/Droplets true def
/DropletsCharges [  % Length a multiple of 3: SheetNum or /All; position being either, type integer, the centre of the WithinPage circle, or array [x y]; charge being numeric; perhaps repeat.
	/All  {[MgnL 0.75 mul PageWidth MgnR sub 0.25 mul add 6 sub  PageHeight MgnB MgnT sub add 2 div      ]}  -1
	/All  {[MgnL 0.75 mul PageWidth MgnR sub 0.25 mul add 6 add  PageHeight MgnB MgnT sub add 2 div      ]}   1
	/All  {[MgnL 0.25 mul PageWidth MgnR sub 0.75 mul add        PageHeight MgnB MgnT sub add 2 div 6 sub]}  -1
	/All  {[MgnL 0.25 mul PageWidth MgnR sub 0.75 mul add        PageHeight MgnB MgnT sub add 2 div 6 add]}   1
] def  % DropletsCharges

[jdaw1]

A request has come from AHB, which I paraphrase here. He was guillotining a page of decanter labels, and wanted thin grey lines between them to guide his chopping.

I have two sources of reluctance, on which comment is invited.

• The decanter labels are spread over the decanter-label page, usually A4, to be as tightly packed as possible subject to a minimum size (DecanterLabelsMaxSmallerDimension×DecanterLabelsMaxLargerDimension). But unless the page size is a fortuitous multiple of these minima in both dimensions, typically the decanter labels will be slightly further apart. This may well be further apart than the size of the business cards to the backs of which they are to be glued. So the correct place to guillotine depends on the size of the cards, not known at the time of making the PDF (and even if known, cards will typically be further apart).

• If I cut by hand, as is usual, I’m, unlikely to follow the grey lines precisely, and their remnants will be unsightly.

I could add a small grey line at the edge of the page, for those determined to guillotine precisely between. But I’m not sure that even this should be enabled by default.

Please comment.

[jdaw1]

The decanter labels are spread over the decanter-label page, usually A4, to be as tightly packed as possible subject to a minimum size (DecanterLabelsMaxSmallerDimension×DecanterLabelsMaxLargerDimension).

I could show more cut lines, perhaps only short and at the edge, such that each rectangle is DecanterLabelsMaxSmallerDimension × DecanterLabelsMaxLargerDimension, lessening this problem. But not eliminating: if the useful areas of the cards that happen to be used is smaller than DecanterLabelsMaxSmallerDimension × DecanterLabelsMaxLargerDimension, this cutting would be wrong.

[PhilW]

A request has come from AHB, which I paraphrase here. He was guillotining a page of decanter labels, and wanted thin grey lines between them to guide his chopping.

I have two sources of reluctance, on which comment is invited.

<snip ...>

Please comment.

I agree with the second point, though understand AHB's request. Perhaps give the feature a global flag, disabled by default for those who prefer to cut without lines, but which Alex (and anyone else) can enable to use when wanted?

[jdaw1]

Done: Boolean parameter DecanterLabelsGuillotineMarks added to software. Alex: please inspect this example, and also the relevant page of the document for the this year’s visit to the White Horse.

[Alex Bridgeman]

Thank you.

Some observations:
(1) The short lines at the top overlap and confuse with the text of the instructions. I don't think these short lines add anything so I suggest they are removed.
(2) the broken lines around each label are more complex than I had in mind — my thought was simply to put a line halfway between each label rather than creating a gutter which needs to be trimmed twice. The edge of the paper on the left hand side can be the boundary for the left-most label, a series of lines separates this label from its companions to the right with a final line indicating where to cut on the right-hand side of the last label.

Top and bottom cutting lines would continue to be needed.
(3) Is there a reason for the decanter label cutting lines to be different colours and some being bold while others are not?

[jdaw1]

Some confusion.

the broken lines around each label are more complex than I had in mind

Is there a reason for the decanter label cutting lines to be different colours and some being bold while others are not?

These don’t print. Display only. Have you a different PDF viewer?

I don't think these short lines add anything so I suggest they are removed.

These short lines, ≈ 2mm, are the only relevant lines that print.

[akzy]

/DropletsCharges [ % Length a multiple of 3: SheetNum or /All; position being either, type integer, the centre of the WithinPage circle, or array [x y]; charge being numeric; perhaps repeat.
/All {[MgnL 0.75 mul PageWidth MgnR sub 0.25 mul add 6 sub PageHeight MgnB MgnT sub add 2 div ]} -1
/All {[MgnL 0.75 mul PageWidth MgnR sub 0.25 mul add 6 add PageHeight MgnB MgnT sub add 2 div ]} 1
/All {[MgnL 0.25 mul PageWidth MgnR sub 0.75 mul add PageHeight MgnB MgnT sub add 2 div 6 sub]} -1
/All {[MgnL 0.25 mul PageWidth MgnR sub 0.75 mul add PageHeight MgnB MgnT sub add 2 div 6 add]} 1
] def % DropletsCharges[/code]

I've started having a little play with the droplets feature. I'm wanting to play around with the positioning of the charges and ideally don't want to just use raw units (which i presume are mm ?) and instead tie them to geometries as you did in this example. I struggled to find documentation for what these key words mean such as MgnL/R/T/B (which which i can assume is "something" - left/right/etc.). Also, are there key words associated to circle location? I'd appreciate if you could point me in the direction of where to look.

[jdaw1]

Please help me improve the relevant section of the manual.

The array DropletsCharges is of a length that is a multiple of three, as follows. 0 An integer SheetNum for this charge, or /All meaning all sheets. 1 A location: this can be an integer, the centre of the WithinPage circle on that sheet; or a length array of length two, [x y] in which [0 0] is bottom-left of the page. 2 A numeric charge.

So to put charge at the centre of circle i on SheetNum=0, that’s just 0 i charge.

Also, from the manual:

In PostScript sizes, distances, lengths, widths (and, approximately, font sizes) are denominated in points, where 72 pt = 1″ = 1 inch = 25.4 mm, and hence 1 pt = 1⁄72″ = 0.01388″ = 127⁄360 mm = 0.35277 mm, and 1 mm = 360⁄127 pt ≈ 2.83 pt. Even areas are in square points, there being 72×72=5,184 to the inch², and just over 8 to the mm².

Also, from a more scary manual page:

GlassPositions, a triple-depth array holding the positions of the glasses. GlassPositions SheetNum get WithinPage get is an array, [x y], where x and y are the position of the centre of the glass placement.

MgnL/R/T/B

Used margin, so that gap between the edge of the page and non-header non-footer page contents: Left; Right; Top; Bottom. Usually equal to Margin…, unless SideBySideGlassesTastingNotes is true (which it probably isn’t).

[akzy]

Greatly appreciated. I will have a play and report back.

[Doggett]

Personally on the latest iteration for the Rebello Valente tasting, I think the droplets detract from the classic nature of the placemat and is like a very light version of Pollock’s number 14. I am all for pimping up the mats but this one is not for me... thoughts?

[jdaw1]

Personally on the latest iteration for the Rebello Valente tasting, I think the droplets detract from the classic nature of the placemat and is like a very light version of Pollock’s number 14. I am all for pimping up the mats but this one is not for me... thoughts?

I’d like to know what others think.

Some food for thought.

• From the log page of the Rebello Valente placemats: “Droplets will cover approximately 7.3% of non-margin bare page (ignoring overlaps and hollow inners, assuming no sharp turns, etc).”

• Thoughts on other decorative things? E.g.,:

◊ Spirals, Messias tasting 23 Oct 2019:






◊ CrossHatching, 1994 tasting on 11 June 2019:






◊ Flowers, Blind and Informal tasting on 21 May 2019:






◊ Stars, DRT’s Annual Elixir tasting on 26 Mar 2019:
 

[Doggett]

Quite liked the spirals, flowers and crosshatching. Not sold on the stars but the bold defining of the main elements is good to my taste a la Flowers.

[Glenn E.]

I like all of the above for different reasons. But if I'm honest, I probably won't use these features. We're pretty basic with our placemats and are very happy with just the core functionality.

I might try spirals (or anything else that fills the circle) some time. For some reason filling the glasses circle is more appealing to me than filling the empty space between glasses.

[jdaw1]

For some reason filling the glasses circle is more appealing to me than filling the empty space between glasses.

Is this a quirky feature of Simon and our American cousins (and hence perhaps ignorable), or is this a general preference?

Quite liked the spirals, flowers and crosshatching. Not sold on the stars but the bold defining of the main elements is good to my taste a la Flowers.

Flowers not so different to the Stars (compare this use of Hearts). Why the different preference?


I seek guidance.


Reminder: many many placemats shown at the list of placemats.

[Andy Velebil]

Quite liked the spirals, flowers and crosshatching. Not sold on the stars but the bold defining of the main elements is good to my taste a la Flowers.

Ditto

[Glenn E.]

For some reason filling the glasses circle is more appealing to me than filling the empty space between glasses.

Is this a quirky feature of Simon and our American cousins (and hence perhaps ignorable), or is this a general preference?

It is such a mild preference for me - on a feature that I'm not likely to use anyway - that it can be safely ignored, at least as far as I'm concerned.

[jdaw1]

Is it the shape of the flowers versus that of the stars? Or is it that in the examples above the stars are pale on pale, whereas the flowers are pale on dark?

Compare, for example, the 1963 horizontal on 10 April 2018.





Or, similar, the Cockburn vertical on 02 Oct 2018.

[PhilW]

I’d like to know what others think.

I tend to prefer a single pattern on a set of placemats - or at least a coherent one; so:
- outline or solid wording, with pattern filling in circles and clear background
- outline or solid wording, with clear circles and pattern filled background
- pattern-filled wording, with clear circles and clear or same/related pattern background

So, for example, when a pattern is used in the background (rays, diagonals, spirals, sperm and others) I prefer empty circles; or vice versa. Matching background and text-fill can be good (complementary might also work well).

Where a pattern is used to fill the circle, it should not also fill the text, which should then be clear or solid; noting this is as per your Messias example above, but not always per previous placemats.

Am not keen on gray-and-pattern-filled text (such as flowers example, or stars example), nor black-and-pattern filled text.

[jdaw1]

I tend to prefer a single pattern on a set of placemats - or at least a coherent one; so:
- outline or solid wording, with pattern filling in circles and clear background
- outline or solid wording, with clear circles and pattern filled background
- pattern-filled wording, with clear circles and clear or same/related pattern background

Please post some examples (the dates would suffice) of near-fails and only-just-successes.

Reminder: many many placemats shown at the list of placemats.

Thank you.

[jdaw1]

Placemats created.

My bugs, or yours? (E.g., spacing in Circlearrays. E.g., the table of contents has a link to the log page, but there isn’t a log page — was it removed in a post-process rather than by /OutputLogToPage false def? Please send me the code.)


This post moved by jdaw1 from its organisation thread, re Tue 28 Jan 2020.

[flash_uk]

Placemats created.

My bugs, or yours? (E.g., spacing in Circlearrays. E.g., the table of contents has a link to the log page, but there isn’t a log page — was it removed in a post-process rather than by /OutputLogToPage false def? Please send me the code.)

I noticed both the spacing thing, and the lack of log page. Will email the .ps.

The spacing thing I think is my doing. I think I used initials in more than one circlearrays field.


This post moved by jdaw1 from its organisation thread, re Tue 28 Jan 2020.

[jdaw1]

Score one each.

Code: Select all

/Circlearrays [
	[ (NAC) () (NAC) ]
	[ (IDJ) () (IDJ) ]
	[ (MPM) () (MPM) ]
	[ (Dagger) () (Dagger) ]
	[ (Double Dagger) () (Double Dagger) ]
] def

The empty strings explain the weird gaps.

But the absence of a log page was a failure of the author of the code to realise that there was a difference in behaviour between Adobe Distiller and Ghostscript. Fixed. Apologies.

[flash_uk]

The empty strings explain the weird gaps.

Sackcloth and ashes. This requires some adaptation to the wizard coding, to cope with blank fields. I will get round to that, but not at the pace with which you fixed the log page matter.

[jdaw1]

Next question, is about branding, but likely as not the most consistent usage is in the software and its output.

Are we PascalCase (www.ThePortForum.com) or camelCase (www.thePortForum.com)? Current usage is PascalCase; but there would be less need for kerning with camelCase.

Currently we use PascalCase:

Preference? Comment?

[flash_uk]

I prefer PascalCase, as it reinforces the often used acronym TPF. Where would the kerning difference happen with camelCase? Around the first dot?

[jdaw1]

Yes: “.T” needs kerning; “.t” doesnʼt, or at least, it matters much less.

[jdaw1]

During the Emergency at 67 Pall Mall on 25 February 2020, Neil, and perhaps others, asked that the parameter NeckTagsHoleRadius be larger. Currently its default is 42.52pt ≈ 15mm, so a diameter of 3cm. What size is wanted?

Suggestion: 54pt = 19.05mm, so a diameter of 38.1mm. Comment?

[jdaw1]

Suggestion: 54pt = 19.05mm, so a diameter of 38.1mm. Comment?

No comment, so done. If there is later disagreement, it can be undone.

[jdaw1]

I’ve decided that the code should exist in a proper repository, to allow submission and discussion of issues, to allow others to contribute to the PostScript, and for better death-proofing. Hence ∃ github.com/jdaw1/placemat.

Two issues ask a question of GitHub experts (Software licence; Documentation: HTML or Markdown or other?). If you can answer, please do.

[jdaw1]

The move to GitHub is complete. If you have published links to this software, please update them to:
• http://github.com/jdaw1/placemat/
• http://github.com/jdaw1/placemat/blob/m ... lacemat.ps
• http://raw.githubusercontent.com/jdaw1/ ... lacemat.ps
as appropriate.

[jdaw1]

Do any programmers know whether there would be sufficient advantages in moving my code to SourceForge.net? My prior is ‘no’, but I’m willing to be persuaded otherwise.

In perusing this thread for outstanding issues, this was seen.