Cloned Tiles interface rework proposal
Congratulations on another terrific release! I hope you’re all taking a well-deserved break.
In the meantime, I wanted to discuss whether it’s possible to rework the Tiled Clones interface, mainly to address the following issues: - Location: Tiled Clones is too powerful a tool to be hidden in Edit. Proposal: move Tiled Clones to Object, and have it open a dialogue like Fill, Align etc rather than a floating box (Ctrl + Shift + C). - The interface: the current interface is very powerful, but it feels like staring down a math/geometry class. :P Separating Tiles and Rotate and adding a preview area using simple examples would be nice. - “How do I make it rotate around This point?” Have the option to select an on-screen guideline, like how it works with filters and LPEs. For rotations, you select an angled line.
Now, I’m assuming that most uses of tiled clones would fall into a few major categories: - Tiles: including simple symmetry and various tiles options. - Radial Clone: flowers, laces, round borders, clocks, gears... - Objects Along Path: Like Pattern Upon Path, except for objects. Example: a line of trees bordering a curving river and fading into the background.
So, long-story short, my proposal for more user-friendly Tiled Clones interface:
http://a.imagehost.org/view/0611/Tiles http://a.imagehost.org/view/0362/Radial
Options that I’ve missed will of course have to be added back in. Thoughts?
I use tiled clones extensively in my work, and it just feels too fundamental a tool to be kept over in the edit menu, so I agree with that suggestion. It could use at least a default keyboard shortcut to invoke it!
I personally disagree somewhat about the reclassification into three tabs, as the tabs as they exist now correspond precisely to the various fundamental linear transformations (but where's skew?). I've always appreciated the precision they afford me this way. If you'll allow me to augment Valerie's suggestion about the create tiled clones, I would like to see two changes: First, that the scaling values the user enters should be more mathematically intuitive. For instance, downscaling should be achieved by entering a percentage 0 < p < 100, not a percentage p<0, as it stands now. This follows from the mathematical definition of size change percentage: PercentSize = (SizeTransformed/SizeOriginal) * 100, which will never be negative. This always catches me, because I think "Well, I want every clone to be 95% as large as the original," and Inkscape then proceeds to enlarge each clone by 195%! Secondly, when creating tiled clones the first clone always overlays the original, and that always gets in my way of adjusting the original. Why does it do that?
I have to believe there are reasons for Inkscape operating as it does in regard to these two suggestions, but I don't know them. Of course I can and do still make use of the tools as they are in Inkscape; they aren't broken or buggy and this isn't urgent.
Thanks, Mike
On Fri, Aug 27, 2010 at 7:33 AM, Valerie <valerie_vk@...36...> wrote:
Congratulations on another terrific release! I hope you’re all taking a well-deserved break.
In the meantime, I wanted to discuss whether it’s possible to rework the Tiled Clones interface, mainly to address the following issues:
- Location: Tiled Clones is too powerful a tool to be hidden in Edit.
Proposal: move Tiled Clones to Object, and have it open a dialogue like Fill, Align etc rather than a floating box (Ctrl + Shift + C).
- The interface: the current interface is very powerful, but it feels like
staring down a math/geometry class. :P Separating Tiles and Rotate and adding a preview area using simple examples would be nice.
- “How do I make it rotate around This point?” Have the option to select an
on-screen guideline, like how it works with filters and LPEs. For rotations, you select an angled line.
Now, I’m assuming that most uses of tiled clones would fall into a few major categories:
- Tiles: including simple symmetry and various tiles options.
- Radial Clone: flowers, laces, round borders, clocks, gears...
- Objects Along Path: Like Pattern Upon Path, except for objects. Example:
a line of trees bordering a curving river and fading into the background.
So, long-story short, my proposal for more user-friendly Tiled Clones interface:
http://a.imagehost.org/view/0611/Tiles http://a.imagehost.org/view/0362/Radial
Options that I’ve missed will of course have to be added back in. Thoughts?
Sell apps to millions through the Intel(R) Atom(Tm) Developer Program Be part of this innovative community and reach millions of netbook users worldwide. Take advantage of special opportunities to increase revenue and speed time-to-market. Join now, and jumpstart your future. http://p.sf.net/sfu/intel-atom-d2d _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
On Fri, Aug 27, 2010 at 10:27 AM, Michael Nachtigal <chainsawchihuahua@...400...> wrote:
see two changes: First, that the scaling values the user enters should be more mathematically intuitive. For instance, downscaling should be achieved by entering a percentage 0 < p < 100, not a percentage p<0, as it stands now.
The problem is, if you allow it to specify 90%, that would mean 90% of the previous row/column size, not the original size. This way for example you will never be able to go down to zero size, because you will only approach zero asymptotically. The current display from 0% is not ideal also, but I think it makes this more clear.
to enlarge each clone by 195%! Secondly, when creating tiled clones the first clone always overlays the original, and that always gets in my way of adjusting the original. Why does it do that?
Not always. If you randomize positioning, it will not overlay the original.
I catch the opportunity to ask a related question: I couldn't find a (simple) way to achieve a tiling effect of an arbitrary object given x and y spacing in whatever absolute unit. This would be more useful when you want a spacing that makes the clones overlap (as in an Escher tessellation) because you can't use the trick of adding a bigger borderless rectangle to obtain the desired offset and deleting it after cloning. Now shift is specified as % of the original object: is that really useful to anybody (I ask to know, not to be provocatory)? Also, I couldn't find a way to ignore border thickness, i.e. using geometric bounding box.
Of course both results can be achieved with some %-math looking at the object dimensions and border thickness but this is not the human way: the computer should work for me, not the opposite!
Thanks. Luca
On Fri, 2010-08-27 at 10:07 -0700, LucaDC wrote:
Now shift is specified as % of the original object: is that really useful to anybody (I ask to know, not to be provocatory)?
In the cases I had, absolute values (usually px) would have been of advantage.
When I had to create 38 concentric circles with constant stroke width and constant gaps, Cloned Tiles didn't offer a solution. I had to wrestle with the Interpolate extension, instead.
That's why I think the first step to a better Cloned Tiles interface should be about defining the desired capabilities, perhaps by looking at what can and can't be done with it, currently.
On Fri, Aug 27, 2010 at 8:33 AM, Valerie <valerie_vk@...36...> wrote:
Congratulations on another terrific release! I hope you’re all taking a well-deserved break.
In the meantime, I wanted to discuss whether it’s possible to rework the Tiled Clones interface, mainly to address the following issues:
- Location: Tiled Clones is too powerful a tool to be hidden in Edit. Proposal: move Tiled Clones to Object, and have it open a dialogue like Fill, Align etc rather than a floating box (Ctrl + Shift + C).
Maybe we should just move all the clone commands to Object, for consistency.
Dockable dialog: certainly, this was the original plan. It just needs to be coded.
- The interface: the current interface is very powerful, but it feels like staring down a math/geometry class. :P Separating Tiles and Rotate and adding a preview area using simple examples would be nice.
Preview: yes of course, it would be great to add it.
Separate Rotation: yes, that would be good too, as a shortcut. For selecting rotation axis, we need to offer several options: not only by dragging a handle on screen but via shortcuts such as "object center", "top left corner", etc.
http://a.imagehost.org/view/0611/Tiles http://a.imagehost.org/view/0362/Radial
I'm afraid there won't be enough space for all the shift/angle/opacity etc parameters in the space you provided. They are now in their own tabs for a reason - they simply won't fit otherwise.
Dockable dialog: certainly, this was the original plan. It just needs to be coded.
Great!
Separate Rotation: yes, that would be good too, as a shortcut. For selecting rotation axis, we need to offer several options: not only by dragging a handle on screen but via shortcuts such as "object center", "top left corner", etc.
Ah, like that proposal for another part of Inkscape long ago... the one with the 3x3 square that allowed you to choose a corner. So, the possible choices would be: - the afore-mentioned 3x3 square - selecting a handle on-screen - any other options to speak of?
http://a.imagehost.org/view/0611/Tiles http://a.imagehost.org/view/0362/Radial
I'm afraid there won't be enough space for all the shift/angle/opacity etc parameters in the space you provided. They are now in their own tabs for a reason
- they simply won't fit otherwise.
Quite true. I left them this way because I already separated Tiles and Rotation, and I was hesitating about adding sub-tabs within each. So, there are several possibilities: - All the tabs remain as they are, though there is a shortcut to start out as Rotation - Tiles and Rotation are separate tabs of Tiled Clones, and each has sub-tabs for the remaining options - Tiles are under Tiled Clones, and Rotation is under a separate dock called Rotated Clones, where everything is expressed in terms of normal and orthogonal coordinates. Some duplicate functions, to be sure, but probably more intuitive for the user. - Any other options?
I welcome more input so that I can try to come up with more mock-ups and eventually a blueprint.
2010/8/28 Valerie <valerie_vk@...36...>:
Ah, like that proposal for another part of Inkscape long ago... the one with the 3x3 square that allowed you to choose a corner. So, the possible choices would be:
- the afore-mentioned 3x3 square
- selecting a handle on-screen
- any other options to speak of?
There should be a Tile tool that would work like the selector tool, but only transform the rotation centers of objects; the objects themselves would only be translated. There could also be control points in the upper left and bottom right that would expand the "coverage area" of the tiling by creating new clones; they would snap to an integer number of clones. Extra control points would appear on the first clone in the second row and the second clone in the first row to edit per-row or per-column rotation and scaling. Editing the shifts and rotations as numbers in a dialog is just wrong.
We could also look into creating a new SP object which defines a tiling (e.g. <svg:g inkscape:type="tiling" ... >). This could make it easier to address some usability issues with the current dialog.
Regards, Krzysztof
Thanks for all the suggestions! Here are some points so far:
1. Location - Tiled clones should be moved to Objects for greater visibility - Shortcut - Open a side dock instead of a floating box
Any objections I've missed?
2. Preview for symmetry types
Here are my preview proposals: http://i.imagehost.org/view/0688/Symmetry_list
Sorry, when I did this, I didn't notice Thorsten's proposal yet: http://thorwil.files.wordpress.com/2010/08/wallpaper_groups.png?w=176&h=...
(it certainly does a better job of explaining the rotational transformations, that I frankly didn't quite understand)
Thoughts? And although a visual selector may take up too much space, could there be a link to a full list of possible transformations, maybe a tutorial page with further explanations besides each type or each group of transformations.
3. No distinction between original and clones
I'm personally in favour of this, but I'm not sure what the rest of you think. I can see advantages to being able to grab any random clone and start changing it, most people eventually have to delete the original anyway. However, I probably missed the reasons against this. Thoughts?
4. Tile boundaries
http://i.imagehost.org/view/0177/Symmetry
I've included 3 choices of tile boundaries: - “of object”, which corresponds to the current possibility - absolute bounds, with a choice of object alignment (to either corners or to the center, with a 3x3 box. Because of space constraints, I didn't add “Object alignment” before the box picture, but that explanation will appear on mouse-over) - from selected. A drop down appears of the paths you have selected along with the object. You have to choose a single path with 2 straight segments. Inkscape makes a parallelogram out of this for the tiling, choosing one as the x axis and the other as the y axis. You can choose to invert the two if needed for the symmetries.
5. Choice of units
Small change in the interface to allow units besides %:
http://j.imagehost.org/view/0856/Shift
I've noticed that I forgot to add skew in the tabs, so just keep that in mind. I didn't quite get some of the suggestions that have been proposed, so if I've missed something, could you do a mock-up to explain?
Is it too early for me to start a blueprint draft page?
I'll do something separate for radial transformation, as they use polar coordinates.
On Tue, 2010-08-31 at 09:38 -0700, Valerie wrote:
Thanks for all the suggestions! Here are some points so far:
- Location
- Tiled clones should be moved to Objects for greater visibility
- Shortcut
- Open a side dock instead of a floating box
Any objections I've missed?
The way rows and columns apply to groups like P4G or P6M is strange. Like I said elsewhere, I think the numbers should be relative to full pattern repetitions (P3M1 for example would need a 1/6 resolution for the number of repetitions).
Thoughts? And although a visual selector may take up too much space, could there be a link to a full list of possible transformations, maybe a tutorial page with further explanations besides each type or each group of transformations.
If the list gets a label "Wallpaper group:", I wonder if that could be turned into a link to documentation.
- No distinction between original and clones
I'm personally in favour of this, but I'm not sure what the rest of you think. I can see advantages to being able to grab any random clone and start changing it, most people eventually have to delete the original anyway. However, I probably missed the reasons against this. Thoughts?
Original and clones are treated differently: - You have to select the original for editing - You must take care to not delete the original, as that turns all clones into independent objects
VS
There are only clones: - Choose any clone to edit them all - Delete any clone (no need to check the statusbar or to try editing to keep clones and original apart).
But maybe things are as they are because of the structure of SVG?
Exponent has no unit. It's the shift value to the power of this exponent, where 1 is thus neutral.
BTW, I'm not sure how Randomize works. Would 5% mean that the range of possible deviations is from -2.5% to 2.5%? if it's something else, it might not make sense to offer absolute units there.
A) Clones without Original?
On 1/9/10 09:24, Thorsten Wilms wrote:
On Tue, 2010-08-31 at 09:38 -0700, Valerie wrote:
- No distinction between original and clones
I'm personally in favour of this, but I'm not sure what the rest of you think. I can see advantages to being able to grab any random clone and start changing it, most people eventually have to delete the original anyway. However, I probably missed the reasons against this. Thoughts?
Original and clones are treated differently:
- You have to select the original for editing
- You must take care to not delete the original, as that turns all
clones into independent objects
VS
There are only clones:
- Choose any clone to edit them all
- Delete any clone (no need to check the statusbar or to try editing to
keep clones and original apart).
But maybe things are as they are because of the structure of SVG?
AFAIU clones don't exist without linking to an original. Deleting the original turns all clones into originals too (i.e. it unlinks them).
SVG 1.1 > Document Structure > The ‘use’ element: http://www.w3.org/TR/SVG11/struct.html#UseElement
Maybe the original could be 'hidden' in the <defs> section (like <symbol>'s, with a toggle in the dialog to 'unhide' it for editing).
Alternatively the first clone tile could be skipped, as proposed in Bug #309873 in Inkscape: “skip first tile clone” https://bugs.launchpad.net/inkscape/+bug/309873
B) List of feature requests for tiled clones in the bug tracker:
579617 13 of 17 Wallpaper groups at symmetry tab of clones behaves erratically ” https://bugs.launchpad.net/inkscape/+bug/579617
553912 Cannot create offset parallel tiled clones of vertical or horizontal lines https://bugs.launchpad.net/inkscape/+bug/553912
412768 Clone tiling needs "Array" between points option https://bugs.launchpad.net/inkscape/+bug/412768
319648 Symmetry tab needs help on modes https://bugs.launchpad.net/inkscape/+bug/319648
309873 skip first tile clone https://bugs.launchpad.net/inkscape/+bug/309873
281719 clone dialog improvement https://bugs.launchpad.net/inkscape/+bug/281719
171900 Tile Clones: User defined transformation order https://bugs.launchpad.net/inkscape/+bug/171900
171804 Tiled Clone Trace Enhancements https://bugs.launchpad.net/inkscape/+bug/171804
170835 Absolute units for shift in Tile Clones https://bugs.launchpad.net/inkscape/+bug/170835
170823 Selecting all (tiled) clones of an object https://bugs.launchpad.net/inkscape/+bug/170823
170745 cancel button for for tile cloning https://bugs.launchpad.net/inkscape/+bug/170745
C) most often reported bug with 'Tiled Clones'
168651 "Tiled Clones" inside transformed groups broken https://bugs.launchpad.net/inkscape/+bug/168651
~suv
Thank you very much for the list of bugs!
Maybe the original could be 'hidden' in the <defs> section (like <symbol>'s, with a toggle in the dialog to 'unhide' it for editing).
Alternatively the first clone tile could be skipped, as proposed in Bug #309873 in Inkscape: “skip first tile clone” https://bugs.launchpad.net/inkscape/+bug/309873
I see. Has there been any discussion on which solution is preferred? I would personally prefer to see the second solution, and possibly some sort of border appear around the original when one of the clones is selected with the Tile Clones box active. Does that seem acceptable?
579617 13 of 17 Wallpaper groups at symmetry tab of clones behaves erratically https://bugs.launchpad.net/inkscape/+bug/579617
Okay, that's a programming issue, I can't help there unfortunately...
553912 Cannot create offset parallel tiled clones of vertical or horizontal lines https://bugs.launchpad.net/inkscape/+bug/553912
This can also get solved with the two additional tile bound methods proposed.
412768 Clone tiling needs "Array" between points option https://bugs.launchpad.net/inkscape/+bug/412768
This sounds a bit like alignment or interpolation? With the option to define fixed boundaries, it should be easier to achieve what the user wants by using a boundary box whose size is X/n and Y/m (X and Y being the page dimensions, and n/m the number of copies).
319648 Symmetry tab needs help on modes https://bugs.launchpad.net/inkscape/+bug/319648
281719 clone dialog improvement https://bugs.launchpad.net/inkscape/+bug/281719
Basically, the currently proposed symmetry preview area. :) It shouldn't hurt to make it bigger like suggested here though.
171900 Tile Clones: User defined transformation order https://bugs.launchpad.net/inkscape/+bug/171900
Unless the user has examples in mind that are not limited to rotation, I'm actually thinking of an option to address this in a separate dock for rotations only. Basically, a checkbox in Scale for "Scale from Object center" or "Scale from Rotation Center."
171804 Tiled Clone Trace Enhancements https://bugs.launchpad.net/inkscape/+bug/171804
This is both a programming and interface issue... I'm not sure how this could be addressed...
170835 Absolute units for shift in Tile Clones https://bugs.launchpad.net/inkscape/+bug/170835
Covered with the tile boundary selection. :)
170823 Selecting all (tiled) clones of an object https://bugs.launchpad.net/inkscape/+bug/170823
An extra button then. I could add that to the interface, as well as adding back "select original" (since there needs to be an original).
170745 cancel button for for tile cloning https://bugs.launchpad.net/inkscape/+bug/170745
Sounds like something not limited to tile cloning. Some filters are also guilty of big RAM consumption. It'd be better to have a button next to some progress bar to cancel the operation.
C) most often reported bug with 'Tiled Clones'
168651 "Tiled Clones" inside transformed groups broken https://bugs.launchpad.net/inkscape/+bug/168651
Ah, another programming bug.
I can't help with programming unfortunately, I don't have the skills... But it seems our current discussions cover a big part of the interface requests. Thanks for the bug links! I'll start a wiki page later this week with a blueprint proposal.
On 1/9/10 17:30, Valerie wrote:
Thank you very much for the list of bugs!
Maybe the original could be 'hidden' in the <defs> section (like <symbol>'s, with a toggle in the dialog to 'unhide' it for editing).
Alternatively the first clone tile could be skipped, as proposed in Bug #309873 in Inkscape: “skip first tile clone” https://bugs.launchpad.net/inkscape/+bug/309873
I see. Has there been any discussion on which solution is preferred? I would personally prefer to see the second solution, and possibly some sort of border appear around the original when one of the clones is selected with the Tile Clones box active. Does that seem acceptable?
Skipping the first clone tile and using the original in place breaks 'color' mode for the first tile:
Modifying the color of clones only works when the color of the original is unset (the unset fill of the original is rendered black, its unset stroke is not rendered at all; and the clones use the individually assigned colors).
~suv
While looking at the list of bugs ~suv sent, I realized how much space there is in the Symmetry tab, much bigger than is needed for a simple preview of symmetry types. So, since the Tiled Clones interface is so confusing, why not simply add instructions into them? After all, I've noticed that they don't have to be long, they just have to be easily found by users.
I've made a wiki page gathering everything so far (though we may need a new list of symmetry previews since Thorsten is experimenting with the preview styles)
http://wiki.inkscape.org/wiki/index.php/Tiled-Clones
As you can see, Simple translations now has tips for rotation, for example. This way, I think there's no need to have something separate for radial transformations, since users can just follow instructions. For anything more complicated, they can look up details.
They will also be useful for P3 and P6 type transformations. I still haven't managed to get them working myself, so if there needs to be value input into the Shift or Rotation tabs, now's the time to say what is needed? Thanks! :)
Feel free to mess with the wiki page.
2010/8/31 Valerie <valerie_vk@...36...>:
- Choice of units
Small change in the interface to allow units besides %:
In the long run, minor corrections to the tiling dialog are pointless. The geometry of the tiling should be editable on the canvas using the node tool.
That's what the information contained in <svg:g type="tiling" ...> would be for.
Regards, Krzysztof
Hi!
I had another look at the dialog.
Inkscape defaults to shifting the clones by 100% width of the selected object for columns and 100% height for rows. So the parameters on the Shift page are actually about the deviation from that, but the interface doesn’t make that clear at all.
Shift and Scale only take %, but should allow absolute values with a unit of the user’s choice.
"Rows, columns" wouldn’t make sense for radial arrangements that should also be possible.
"Width, height": it could be made clearer that this option will fill the specified area and in what directions (original defines top left).
Use saved size and position of the tile checkbox: what is the use case for this option? The tool-tip shows that this needs a lot of explanation: “Pretend that the size and position of the tile are the same as the last time you tiled it (if any), instead of using the current size”.
The Exponent parameter on the Shift and Scale pages depends on the tool-tip to explain that it defines whether rows will be spaced evenly (1), converge (1).
Trace page: Well, non of my tests produced anything sensible or useful.
The Symmetry page only contains a pop-up list (what GTK+ erroneously calls a combo box) full of mysterious items: http://thorwil.files.wordpress.com/2010/08/screenshot_create_tiled_clones_sy...
These are the 17 wallpaper groups, all possible tilings with translational symmetry. http://en.wikipedia.org/wiki/Wallpaper_group I don’t think knowledge of these should be expected. At the very least, the term wallpaper groups should be mentioned. Even once you know what this is about, the descriptions might not help you much with recalling the patterns or with predicting the outcome based on the selected object.
The Wikipedia article includes diagrams, but I didn’t find them to help much. http://tavmjong.free.fr/INKSCAPE/MANUAL/html/Tiles-Symmetries.html is much better. There are 11 groups based on rectangles (2 of them can be parallelograms), 1 on right-angled rectangles (rectangles cut apart diagonally) and 5 on hexagon subdivisions.
Here’s an attempt at creating the most simple schematics, leaving out points of rotation and mirror axes to just depict orientation. The place taken by the selected object is darkened: http://thorwil.files.wordpress.com/2010/08/wallpaper_groups.png?w=176&h=...
These could be added to the descriptions given now.
On Mon, Aug 30, 2010 at 8:38 AM, Thorsten Wilms <t_w_@...123...> wrote:
Hi!
I had another look at the dialog.
Inkscape defaults to shifting the clones by 100% width of the selected object for columns and 100% height for rows. So the parameters on the Shift page are actually about the deviation from that, but the interface doesn’t make that clear at all.
Well, it only makes sense to say that base shift is 100% for P1. If the shift is accompanied by various rotations and flips, it's not clear how to count percentages and whether they make sense at all. The best formulation is: these values move clones from the position they would have after the symmetry transform, in the units of the tile's width/height. This is why I'm against displaying 105% instead of 5% there, because that 105 would be a sum of unsummable values: the 100% move is different from the 5% move (except when we're using P1).
Shift and Scale only take %, but should allow absolute values with a unit of the user’s choice.
Yes, this just needs to be coded.
"Rows, columns" wouldn’t make sense for radial arrangements that should also be possible.
Agreed, we need some kind of shortcut for radial placement, which would disable irrelevant values.
"Width, height": it could be made clearer that this option will fill the specified area and in what directions (original defines top left).
Feel free to edit the tooltip or label!
Use saved size and position of the tile checkbox: what is the use case for this option?
You made a tiling, then scaled the tile, and want to repeat or slightly change the tiling, but without scaling it - leaving it the same frequency (even if with overlapping or gaps). Feel free to edit the tooltip to make it clearer.
The Exponent parameter on the Shift and Scale pages depends on the tool-tip to explain that it defines whether rows will be spaced evenly (1), converge (1).
You can suggest a way to make this clear without a tooltip?
Trace page: Well, non of my tests produced anything sensible or useful.
So what were you tracing?
The Symmetry page only contains a pop-up list (what GTK+ erroneously calls a combo box) full of mysterious items:
The empty space below is reserved for explanatory illustrations, which would make this a lot less mysterious.
Here’s an attempt at creating the most simple schematics, leaving out points of rotation and mirror axes to just depict orientation. The place taken by the selected object is darkened: http://thorwil.files.wordpress.com/2010/08/wallpaper_groups.png?w=176&h=...
Nice start, but I think they can be made even more clear. Why the skew? How will this fill a larger space, not just the first bunch of tiles?
These could be added to the descriptions given now.
Definitely yes.
Trace page: Well, non of my tests produced anything sensible or useful.
This is (imoo) one of the super best excellent features of tiled clones!
http://www.inkscape.org/screenshots/?version=0.42 http://tavmjong.free.fr/INKSCAPE/MANUAL/html/Tiles-Trace.html
-Rob A>
On Mon, 2010-08-30 at 13:15 -0300, bulia byak wrote:
On Mon, Aug 30, 2010 at 8:38 AM, Thorsten Wilms <t_w_@...123...> wrote:
Hi!
I had another look at the dialog.
Inkscape defaults to shifting the clones by 100% width of the selected object for columns and 100% height for rows. So the parameters on the Shift page are actually about the deviation from that, but the interface doesn’t make that clear at all.
Well, it only makes sense to say that base shift is 100% for P1. If the shift is accompanied by various rotations and flips, it's not clear how to count percentages and whether they make sense at all. The best formulation is: these values move clones from the position they would have after the symmetry transform, in the units of the tile's width/height. This is why I'm against displaying 105% instead of 5% there, because that 105 would be a sum of unsummable values: the 100% move is different from the 5% move (except when we're using P1).
Ah, right!
Let's see if I get things right: The P1 scheme creates a clone and translates it and will treat the clone as the new original for the possible next operation (all links refer to the true original, though, I guess).
The operation will be repeated until the number of clones +1 reaches the specified number of steps for one direction. Afterwards the entire result is used as selected object for the other direction.
But all other wallpaper groups have a cell size > 1. To get to see one complete P4G pattern, you need to specify 2 rows, 8 columns, even though the resulting arrangement is 4 x 4. I wasn't able to deduce those numbers, I had to try.
This makes me think that for all but P1, using the original object as unit for rows and columns is the wrong approach. What is really of interest are repetitions of the full pattern (though depending on the wallpaper group, 1/2, 1/3, 1/4, 1/6 or 1/12 could be the resolution).
Then I guess for all wallpaper groups (except P1), once one pattern is full, additional clones will be used to create translated duplicates of that pattern. So an implicit P1 as follow-up, no recursion.
"Width, height": it could be made clearer that this option will fill the specified area and in what directions (original defines top left).
Feel free to edit the tooltip or label!
How about:
--- Fill width and height, starting from the top left: [ ]x[ ] [px] --- or: --- Fill area, starting from the top left: Width: [ ] Height: [ ] [px] ---
Use saved size and position of the tile checkbox: what is the use case for this option?
You made a tiling, then scaled the tile, and want to repeat or slightly change the tiling, but without scaling it - leaving it the same frequency (even if with overlapping or gaps). Feel free to edit the tooltip to make it clearer.
Sounds to me like a user having such needs could well just copy the tile and scale it back, making the cost of such an odd option taking space and diverting user attention too high. I mean, it's like an undo that doesn't happen, except for the result of the next operation, so you work with one state, while you see another.
The Exponent parameter on the Shift and Scale pages depends on the tool-tip to explain that it defines whether rows will be spaced evenly (1), converge (1).
You can suggest a way to make this clear without a tooltip?
* Add an info text * Use sliders with 3 marks for converging, even and diverging
Trace page: Well, non of my tests produced anything sensible or useful.
So what were you tracing?
Tried with a photo and various shapes in a new drawing without using layers. I can get some transparency and size variations, but no color changes (I did check "Color" in "3. Apply ...".
The Symmetry page only contains a pop-up list (what GTK+ erroneously calls a combo box) full of mysterious items:
The empty space below is reserved for explanatory illustrations, which would make this a lot less mysterious.
I think a list box with integrated illustrations could work.
Here’s an attempt at creating the most simple schematics, leaving out points of rotation and mirror axes to just depict orientation. The place taken by the selected object is darkened: http://thorwil.files.wordpress.com/2010/08/wallpaper_groups.png?w=176&h=...
Nice start, but I think they can be made even more clear. Why the skew? How will this fill a larger space, not just the first bunch of tiles?
You mean the first 2, left side? just to mark the ones that work with parallelograms. That should be left out anyway, as long as there is no parameter to assume a skewed bounding box.
I think showing more than 1 cell/repetition only makes it harder to understand the patterns. Maybe faded out to the right and bottom, but even that takes precious space.
Oh, and should there be an option to rotate the wallpaper patterns by 90° in either direction, original tile staying top left? Especially looking at PM, PG, CMM.
These could be added to the descriptions given now.
Definitely yes.
I'm very much willing to bring the SVG into shape so it can be dropped in, just tell me how you want it.
On 30/8/10 20:41, Thorsten Wilms wrote:
On Mon, 2010-08-30 at 13:15 -0300, bulia byak wrote:
On Mon, Aug 30, 2010 at 8:38 AM, Thorsten Wilms <t_w_@...123...> wrote:
Use saved size and position of the tile checkbox: what is the use case for this option?
You made a tiling, then scaled the tile, and want to repeat or slightly change the tiling, but without scaling it - leaving it the same frequency (even if with overlapping or gaps). Feel free to edit the tooltip to make it clearer.
Sounds to me like a user having such needs could well just copy the tile and scale it back, making the cost of such an odd option taking space and diverting user attention too high. I mean, it's like an undo that doesn't happen, except for the result of the next operation, so you work with one state, while you see another.
Some examples illustrating how this option works and why it's not as 'odd' as it might seem to you: http://tavmjong.free.fr/INKSCAPE/MANUAL/html/TilePattern.html http://www.inkscape.org/screenshots/gallery/inkscape-0.41-CVS-linux-tiles1.png http://linuxformat.co.uk/wiki/index.php/Inkscape_-_cloning_and_tiling#Tessellation
Hopefully it will not get dropped from a reworked 'Tiled Clones' dialog, as it allows creating seamless tiles or tesselations: editing the contents of the base tile (a group) & recreating the tiled clones even if the dimensions of the bbox of the base tile have changed (because the saved dimensions and rotation center of the group are used).
~suv
On Tue, 2010-08-31 at 02:52 +0200, ~suv wrote:
Some examples illustrating how this option works and why it's not as 'odd' as it might seem to you: http://tavmjong.free.fr/INKSCAPE/MANUAL/html/TilePattern.html http://www.inkscape.org/screenshots/gallery/inkscape-0.41-CVS-linux-tiles1.png http://linuxformat.co.uk/wiki/index.php/Inkscape_-_cloning_and_tiling#Tessellation
Thanks! I now understand that it is about allowing overlap, making it easier to edit graphics on the tile edges.
I would propose working within a clipped group, but as noted in the first linked article, this can result in drawing artifacts on the boundaries.
Specifying the area of overlap would be cumbersome, as it will vary during editing.
The solution has to be about defining an area in absolute terms that differs from the current bounding box. That's what the current solution does, but in an obscure way.
How about:
[] Assume tile size of [width] * [height]
combined with a "Pick dimensions from selection" button?
No need to tile once in advance, no hidden values and I think less of a mental workout to see the usefulness in allowing overlap.
On Mon, 2010-08-30 at 20:41 +0200, Thorsten Wilms wrote:
The operation will be repeated until the number of clones +1 reaches the specified number of steps for one direction. Afterwards the entire result is used as selected object for the other direction.
Not "+1", as one clone will be placed right on top of the prototype, so rows * columns is the number of clones, without subtracting 1.
It's easy to overlook this, as without transparency, the prototype below the first clone will be invisible (ignoring that it sometimes shows on the edges, if you look very closely). Is this really the right thing to do? If I as user want a 3 x 3 tiling, I want 9 tiles, not 10. Why make me deal with one extra?
Note that there is the insecurity of not knowing what happens if you delete the original. Other applications take the approach of not differentiating between original and clones.
I am having trouble keeping the whole thread in my head, so I will respond to the discussion below:
Thorsten: "Here's an attempt at creating the most simple schematics, leaving out points of rotation and mirror axes to just depict orientation. The place taken by the selected object is darkened: http://thorwil.files.wordpress.com/2010/08/wallpaper_groups.png"
Bulia: "Nice start, but I think they can be made even more clear. Why the skew? How will this fill a larger space, not just the first bunch of tiles?"
Thorsten: "You mean the first 2, left side? just to mark the ones that work with parallelograms. That should be left out anyway, as long as there is no parameter to assume a skewed bounding box. I think showing more than 1 cell/repetition only makes it harder to understand the patterns. Maybe faded out to the right and bottom, but even that takes precious space. Oh, and should there be an option to rotate the wallpaper patterns by 90° in either direction, original tile staying top left? Especially looking at PM, PG, CMM."
Just to get the math right (from what I understand of it). - For P1, the skew is essential, because P1 does not allow rotation symmetries, nor (glide) reflections. When you make it a square, it becomes a member of p4m. (note that square or rectangle makes a difference) - I think for P2, the skew is not essential, but I am not sure. - The drawing for PG is wrong, it should not have rotation symmetry. - The drawing for PMM is wrong, the bottom two squares should be upside down.
(have not looked at the others in more detail)
Ciao, Johan
On Tue, 2010-08-31 at 12:32 +0200, J.B.C.Engelen@...1578... wrote:
Just to get the math right (from what I understand of it).
- For P1, the skew is essential, because P1 does not allow rotation symmetries, nor (glide) reflections.
When you make it a square, it becomes a member of p4m. (note that square or rectangle makes a difference)
I don't understand how P1 with a square becomes a member of P4M at all.
The following only fill the plane with square prototypes: P4, P4G. Not sure if I should make the others non-square because of that.
- I think for P2, the skew is not essential, but I am not sure.
- The drawing for PG is wrong, it should not have rotation symmetry.
- The drawing for PMM is wrong, the bottom two squares should be upside down.
Fixed those, thank you! http://thorwil.files.wordpress.com/2010/08/wallpaper_groups1.png
-----Original Message----- From: Thorsten Wilms [mailto:t_w_@...123...] Sent: Tuesday, August 31, 2010 15:01
On Tue, 2010-08-31 at 12:32 +0200, J.B.C.Engelen@...1578... wrote:
Just to get the math right (from what I understand of it).
- For P1, the skew is essential, because P1 does not allow
rotation symmetries, nor (glide) reflections.
When you make it a square, it becomes a member of p4m. (note that square or rectangle makes a difference)
I don't understand how P1 with a square becomes a member of P4M at all.
The following only fill the plane with square prototypes: P4, P4G. Not sure if I should make the others non-square because of that.
Sorry for the confusion. The confusion arose because I was thinking about a featureless square; because the other squares have a feature drawn in them to make them assymetric for rotations and reflections. You are right, P1 does not require the skew. (now I would almost write, "of course")
-Johan
On Tue, Aug 31, 2010 at 10:01 AM, Thorsten Wilms <t_w_@...123...> wrote:
http://thorwil.files.wordpress.com/2010/08/wallpaper_groups1.png
I think it would be less confusing to use a tile that is obviously a single object. For example, a square with one rounded corner. As it is now, it's a bit motley with that darker triangle.
Also, I would like to see how the entire plane is filled. So I think we need three levels of opacity used: opaque original tile, more transparent bunch of its immediate adjacent transformations, and still more transparent repeatings of these transformations to the end of the preview window.
On Tue, 2010-08-31 at 12:56 -0300, bulia byak wrote:
On Tue, Aug 31, 2010 at 10:01 AM, Thorsten Wilms <t_w_@...123...> wrote:
http://thorwil.files.wordpress.com/2010/08/wallpaper_groups1.png
I think it would be less confusing to use a tile that is obviously a single object. For example, a square with one rounded corner. As it is now, it's a bit motley with that darker triangle.
Marking one corner isn't enough, as it doesn't show a difference between rotation and reflection.
Tried a few things: http://www.foopics.com/showfull/9c37a7ecb9b96d1499428c0adfc7cc2c
I think the black outlines help, as heavy-handed as they are (could be just dark instead of pitch black).
participants (9)
-
unknown＠example.com -
bulia byak -
Krzysztof Kosiński -
LucaDC -
Michael Nachtigal -
Rob Antonishen -
Thorsten Wilms -
Valerie -
~suv