Fwd: Re: Any chance we can make some docs material? (targeting the moon)
I have made videos of workarounds, and anyone making a workaround docs can use my video without my permission. I will upload them soon. Those docs can be deleted once Inkscape gets certain features, and I will delete those videos the minute Inkscape supports certain features.
For now I have this - https://www.youtube.com/watch?v=k5TkIETfQa4
The point of the video is to show users how they can make a easily accessible editable mask in Inkscape without ever having to release mask. It also shows how to get some fade-out effect usually made with raster programs. There's the issue of the mask showing over the target object, and if I hide the editable mask layer, so does the mask goes to black.
The one I will upload after are
2 Gradient Stroke workarounds videos - With Profile LIne, and without profile line
1 tangency line workaround video utilizing spiro, b-spline, and regular bezier path
Way later I will do a shoe render, and how to use clip and masks in order to generate a extremely realistic-looking shoe that is 100 percent vector, but some of those results can easily be achieved with raster programs. I will also try to generate a warped pattern, and show how to achieve that.
I know I'm not the best person to make a tutorial.
On 4/29/2017 6:36 PM, Victor Westmann wrote: Hi Miguel,
I think this is a GREAT advantage for the entire Inkscape project. It makes all the sense of the world to help users do all the needed workarounds and, if they want, help send funds to the project ($$) and/or a bug request on Gitlab(is it at Gitlab really?).
Amazing. The more transparent we are (and we really are) the better.
--Victor Westmann
2017-04-29 10:39 GMT-07:00 Miguel Lopez <reptillia39@...3425...mailto:reptillia39@...3425...>: I want to put out something regarding documentations. What about workarounds to Inkscape limitations as part of the documentation? Some users really need some answers to the limitation of Inkscape. There's another idea I have in mind. For making tutorials shorter, we can use existing tutorials of our own and linking to them as part of advanced section of Inkscape tutorials. Advanced users should already know the program in and out to know how to apply them for advanced rendering. Some of those might need workarounds though.
My only issue with those workaround is the amount of time they require, and the tax on limited resources. Otherwise, if those weren't a issue, they would be fantastic though they're still not a replacement to what other programs has to offer. I can offer making workaround docs if anyone wants me to show how to work around Inkscape limits.
For example :
1) Gradient Stroke -
1a) Method 1 (If profile line aren't needed) - Duplicate and then adjust stroke width, and then convert to path. Every strokes must be converted to path. You copy and paste the path that is going to get removed by applying the path difference. You repeat the process, and then you get every individual stroke which can be colored as however you please.
1b) Method 2 (If profile line is needed) You apply the same thing to the above, but you are not going to use stroke width. Instead, you are going to use power stroke to emulate profile line with gradient stroke.
2) Realistic Rendering of convulated objects like a shoe.
2a) This involves series of clipping, blurring, gradient mesh and brushing. Right now, Inkscape users can only use gradient mesh for basic overall lighting, and some bit of coloring using filtering since gradient mesh is underdeveloped in Inkscape. You apply blur to brushes in order to emulate shading. Doing this a lot can give very sastifying result within Inkscape, but also drains so much of rendering speed within Inkscape even with a powerful computer. But results are literally comparable to raster programs, and in some way even better because well, it's scalable.
3) The lack of warp tool for textures workaround
3a) As of now, we do have lattice tool, and the tweak liquifying tool. Those two can be used as a workaround to the lack of warp tool. Lattice deformation tool should had the option of allowing users to tweak the tangency line though instead of the tangency line being tangent to the horizontal and vertical direction. This is not something that can have a decent workaround since it requires duplicating so many objects and then tweaking it would be a major pain. For complicated textures, it can't be done within Inkscape unless you have a infinite amount of time and resource, and as we all know that's not possible.
4) For people who have trouble with tangency snapping (I'm one of those), and no that snapping option does not help one bit.
4a) The obvious workaround involves using the show handles. You can make tangent line utilizing the result of the show handles LPE. Also, this enables users to be able to create perpendicular, and tangent spiro path.
5) The lack of ease regarding manipulating mask/clip
5a) The workaround to this is well, using the clone as the masking/clipping source while retaining the original for clipping/masking at another location. This workaround works because you can always manipulate the source object, and hide it. It is almost exactly like as if you were manipulating transparency mask within Krita or Photoshop or GIMP. After testing it, it's beautiful really when you change between layer/group.
6) PDF export limitation
6a) The obvious workaround here is well, export to png and then convert to pdf. Of course, some rasterization would be needed if one has to convert to pdf, literally at times.
I think those are the 6 issues that could be addressed via docs tutorial for those who are desperate to find a solution to those. They can always resort to Krita for those (except 6 because pdf export is not planned), but that's not a option if they have to create a vector render. If there's anything I miss regarding workarounds, lemme know. I can probably add more workarounds if I miss anything as I know the program in and out from a user perspective.
On 4/29/2017 11:54 AM, C R wrote:
I don't mean to slow anyone's roll here. But wouldn't it make more sense to put any kind of energy towards documentation into the much discussed, direly needed, user-focused, step by step manual? Rather than starting from scratch on a whole different kind of project?
Yea, this actually makes a lot of sense as a first step.
There are many books out there already, which amount to a series of tutorials. It's not a bad thing. I just think this kind of project is better suited for a single author, or maybe a small team. And I think the project needs the manual much, more more than the community needs another book of tutorials.
I agree. I think the book could be a lot of things in one. But I agree with finishing what we already have before starting something new.
As far as I understand, all that's needed is an English translation of...well can't find a link to the French version. Here's a link to whatever has been translated already: https://fr.flossmanuals.net/start-with-inkscape/introduction/
I can't help with translation, unfortunately. But I'd like to see this finished. So +1 for the suggestion.
Once we have the translation, we'll be off and running to update and finish it! By the way, is there anything those of us who can't translate, can do, to help the translators?
I volunteer to help this effort in what ways are needed.
And won't such a new book of tutorials have to be published? A big obstacle to writing any book is getting it published. You almost have to have an invitation from a publisher to be certain a book will get published. Or publish it yourself, which is not easy eitiher.
Books done in Scribus can be "published" in a variety of ways, opened in browsers, laptops, eReaders, or just printed out. We could sell printed copies along with other Inkscape stuff. Maybe copies signed by members of the project would be kinda cool. No idea what the market is for it, but the idea that we could do all of these at once is attractive, and why I recommend Scribus.
-C
Just my opinion :-)
brynn
-----Original Message----- From: Maren Hachmann Sent: Friday, April 28, 2017 4:01 PM To: Inkscape-Docs ; Inkscape Devel List Subject: Re: [Inkscape-devel] Any chance we can make some docs material? (targeting the moon)
Would it make sense to use gitlab's new subgroups feature for this?
The inkscape-docs team could be a sub-team of Inkscape, that way. There are only 4 members as of now, so changing wouldn't be so difficult as it might be later on.
Maren
Am 28.04.2017 um 16:14 schrieb Martin Owens:
On Fri, 2017-04-28 at 12:39 +0100, C R wrote:
I'd love to quit my job and just do docs. :) Unfortunately, that's what it would probably take to get docs going to the extent we'd like. It's been discussed before, but never gone anywhere because of lack of time/hands involved.
Yes, we should use Scribus to do it. In fact, it should probably be a github project to attract contributors. This way we can patch what needs to be patched when stuff changes in subsequent releases.
Sounds like you have a solid step one Chris.
Here's the inkscape-docs group on gitlab, EVERYONE should join, there should be a button to join:
https://gitlab.com/inkscape-docs
And here's the new book/manual/docs project where files can be put:
https://gitlab.com/inkscape/manuals
I recommend using the wiki attached to the project to plan the adventure slowly. Add a bit at a time and don't rush to have something "complete" but have something small produced.
Best Regards, Martin Owens
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.netmailto:Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.netmailto:Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.netmailto:Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.netmailto:Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.netmailto:Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Am 30.04.2017 um 02:43 schrieb Miguel Lopez:
I have made videos of workarounds, and anyone making a workaround docs can use my video without my permission. I will upload them soon. Those docs can be deleted once Inkscape gets certain features, and I will delete those videos the minute Inkscape supports certain features.
For now I have this - https://www.youtube.com/watch?v=k5TkIETfQa4
The point of the video is to show users how they can make a easily accessible editable mask in Inkscape without ever having to release mask. It also shows how to get some fade-out effect usually made with raster programs. There's the issue of the mask showing over the target object, and if I hide the editable mask layer, so does the mask goes to black.
- Yes, using clones for that is a workaround that is often suggested in the forums, too.
(just a suggestion for your next videos: could you use a screen recording software that also records the mouse pointer? The video is also rather lengthy, and it takes long to get to the point of cloning the layer)
Even if you don't want to finalize an official tutorial, giving ideas is a great way to contribute, too. I think there are other people who like to do the polishing, so you'd just be sharing the work.
Please, when the question of the how/licence/means to use for making a manual is settled, don't forget to come back and add your ideas, Miguel.
Kind Regards, Maren
The one I will upload after are
2 Gradient Stroke workarounds videos - With Profile LIne, and without profile line
1 tangency line workaround video utilizing spiro, b-spline, and regular bezier path
Way later I will do a shoe render, and how to use clip and masks in order to generate a extremely realistic-looking shoe that is 100 percent vector, but some of those results can easily be achieved with raster programs. I will also try to generate a warped pattern, and show how to achieve that.
I know I'm not the best person to make a tutorial.
On 4/29/2017 6:36 PM, Victor Westmann wrote:
Hi Miguel,
I think this is a GREAT advantage for the entire Inkscape project. It makes all the sense of the world to help users do all the needed workarounds and, if they want, help send funds to the project ($$) and/or a bug request on Gitlab(is it at Gitlab really?).
Amazing. The more transparent we are (and we really are) the better.
--Victor Westmann
2017-04-29 10:39 GMT-07:00 Miguel Lopez <reptillia39@...3425... mailto:reptillia39@...3425...>:
I want to put out something regarding documentations. What about workarounds to Inkscape limitations as part of the documentation? Some users really need some answers to the limitation of Inkscape. There's another idea I have in mind. For making tutorials shorter, we can use existing tutorials of our own and linking to them as part of advanced section of Inkscape tutorials. Advanced users should already know the program in and out to know how to apply them for advanced rendering. Some of those might need workarounds though. My only issue with those workaround is the amount of time they require, and the tax on limited resources. Otherwise, if those weren't a issue, they would be fantastic though they're still not a replacement to what other programs has to offer. I can offer making workaround docs if anyone wants me to show how to work around Inkscape limits. For example : 1) Gradient Stroke - 1a) Method 1 (If profile line aren't needed) - Duplicate and then adjust stroke width, and then convert to path. Every strokes must be converted to path. You copy and paste the path that is going to get removed by applying the path difference. You repeat the process, and then you get every individual stroke which can be colored as however you please. 1b) Method 2 (If profile line is needed) You apply the same thing to the above, but you are not going to use stroke width. Instead, you are going to use power stroke to emulate profile line with gradient stroke. 2) Realistic Rendering of convulated objects like a shoe. 2a) This involves series of clipping, blurring, gradient mesh and brushing. Right now, Inkscape users can only use gradient mesh for basic overall lighting, and some bit of coloring using filtering since gradient mesh is underdeveloped in Inkscape. You apply blur to brushes in order to emulate shading. Doing this a lot can give very sastifying result within Inkscape, but also drains so much of rendering speed within Inkscape even with a powerful computer. But results are literally comparable to raster programs, and in some way even better because well, it's scalable. 3) The lack of warp tool for textures workaround 3a) As of now, we do have lattice tool, and the tweak liquifying tool. Those two can be used as a workaround to the lack of warp tool. Lattice deformation tool should had the option of allowing users to tweak the tangency line though instead of the tangency line being tangent to the horizontal and vertical direction. This is not something that can have a decent workaround since it requires duplicating so many objects and then tweaking it would be a major pain. For complicated textures, it can't be done within Inkscape unless you have a infinite amount of time and resource, and as we all know that's not possible. 4) For people who have trouble with tangency snapping (I'm one of those), and no that snapping option does not help one bit. 4a) The obvious workaround involves using the show handles. You can make tangent line utilizing the result of the show handles LPE. Also, this enables users to be able to create perpendicular, and tangent spiro path. 5) The lack of ease regarding manipulating mask/clip 5a) The workaround to this is well, using the clone as the masking/clipping source while retaining the original for clipping/masking at another location. This workaround works because you can always manipulate the source object, and hide it. It is almost exactly like as if you were manipulating transparency mask within Krita or Photoshop or GIMP. After testing it, it's beautiful really when you change between layer/group. 6) PDF export limitation 6a) The obvious workaround here is well, export to png and then convert to pdf. Of course, some rasterization would be needed if one has to convert to pdf, literally at times. I think those are the 6 issues that could be addressed via docs tutorial for those who are desperate to find a solution to those. They can always resort to Krita for those (except 6 because pdf export is not planned), but that's not a option if they have to create a vector render. If there's anything I miss regarding workarounds, lemme know. I can probably add more workarounds if I miss anything as I know the program in and out from a user perspective. On 4/29/2017 11:54 AM, C R wrote: >> I don't mean to slow anyone's roll here. But wouldn't it make more sense to put >> any kind of energy towards documentation into the much discussed, direly needed, >> user-focused, step by step manual? Rather than starting from scratch on a whole >> different kind of project? > Yea, this actually makes a lot of sense as a first step. > >> There are many books out there already, which amount to a series of tutorials. >> It's not a bad thing. I just think this kind of project is better suited for a >> single author, or maybe a small team. And I think the project needs the manual >> much, more more than the community needs another book of tutorials. > I agree. I think the book could be a lot of things in one. But I agree > with finishing what we already have before starting something new. > > >> As far as I understand, all that's needed is an English translation of...well >> can't find a link to the French version. Here's a link to whatever has been >> translated already: >> https://fr.flossmanuals.net/start-with-inkscape/introduction/ <https://fr.flossmanuals.net/start-with-inkscape/introduction/> > I can't help with translation, unfortunately. But I'd like to see this > finished. So +1 for the suggestion. > >> Once we have the translation, we'll be off and running to update and finish it! >> By the way, is there anything those of us who can't translate, can do, to help >> the translators? > I volunteer to help this effort in what ways are needed. > >> And won't such a new book of tutorials have to be published? A big obstacle to >> writing any book is getting it published. You almost have to have an invitation >> from a publisher to be certain a book will get published. Or publish it >> yourself, which is not easy eitiher. > Books done in Scribus can be "published" in a variety of ways, opened > in browsers, laptops, eReaders, or just printed out. We could sell > printed copies along with other Inkscape stuff. Maybe copies signed by > members of the project would be kinda cool. No idea what the market is > for it, but the idea that we could do all of these at once is > attractive, and why I recommend Scribus. > > -C > >> Just my opinion :-) >> >> brynn >> >> -----Original Message----- >> From: Maren Hachmann >> Sent: Friday, April 28, 2017 4:01 PM >> To: Inkscape-Docs ; Inkscape Devel List >> Subject: Re: [Inkscape-devel] Any chance we can make some docs material? >> (targeting the moon) >> >> Would it make sense to use gitlab's new subgroups feature for this? >> >> The inkscape-docs team could be a sub-team of Inkscape, that way. There >> are only 4 members as of now, so changing wouldn't be so difficult as it >> might be later on. >> >> Maren >> >>> Am 28.04.2017 um 16:14 schrieb Martin Owens: >>>> On Fri, 2017-04-28 at 12:39 +0100, C R wrote: >>>>> I'd love to quit my job and just do docs. :) Unfortunately, that's >>>>> what it would probably take to get docs going to the extent we'd >>>>> like. >>>>> It's been discussed before, but never gone anywhere because of lack >>>>> of time/hands involved. >>>>> >>>>> Yes, we should use Scribus to do it. In fact, it should probably be a >>>>> github project to attract contributors. This way we can patch what >>>>> needs to be patched when stuff changes in subsequent releases. >>>> Sounds like you have a solid step one Chris. >>>> >>>> Here's the inkscape-docs group on gitlab, EVERYONE should join, there >>>> should be a button to join: >>>> >>>> https://gitlab.com/inkscape-docs <https://gitlab.com/inkscape-docs> >>>> >>>> And here's the new book/manual/docs project where files can be put: >>>> >>>> https://gitlab.com/inkscape/manuals <https://gitlab.com/inkscape/manuals> >>>> >>>> I recommend using the wiki attached to the project to plan the >>>> adventure slowly. Add a bit at a time and don't rush to have something >>>> "complete" but have something small produced. >>>> >>>> Best Regards, Martin Owens >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> Check out the vibrant tech community on one of the world's most >>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >>>> _______________________________________________ >>>> Inkscape-devel mailing list >>>> Inkscape-devel@lists.sourceforge.net <mailto:Inkscape-devel@lists.sourceforge.net> >>>> https://lists.sourceforge.net/lists/listinfo/inkscape-devel <https://lists.sourceforge.net/lists/listinfo/inkscape-devel> >>>> >> >> ------------------------------------------------------------------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >> _______________________________________________ >> Inkscape-devel mailing list >> Inkscape-devel@lists.sourceforge.net <mailto:Inkscape-devel@lists.sourceforge.net> >> https://lists.sourceforge.net/lists/listinfo/inkscape-devel <https://lists.sourceforge.net/lists/listinfo/inkscape-devel> >> >> >> ------------------------------------------------------------------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >> _______________________________________________ >> Inkscape-devel mailing list >> Inkscape-devel@lists.sourceforge.net <mailto:Inkscape-devel@lists.sourceforge.net> >> https://lists.sourceforge.net/lists/listinfo/inkscape-devel <https://lists.sourceforge.net/lists/listinfo/inkscape-devel> > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Inkscape-devel mailing list > Inkscape-devel@lists.sourceforge.net <mailto:Inkscape-devel@lists.sourceforge.net> > https://lists.sourceforge.net/lists/listinfo/inkscape-devel <https://lists.sourceforge.net/lists/listinfo/inkscape-devel> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net <mailto:Inkscape-devel@lists.sourceforge.net> https://lists.sourceforge.net/lists/listinfo/inkscape-devel <https://lists.sourceforge.net/lists/listinfo/inkscape-devel>
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Can Inkscape get benefits to reduce banding on gradients appliing smoothing from mesh gradients as optional value in the gradient toolbar as a temporary workarround while swich to a higer color data?
Good god, I hope so! That would be amazing! :D If it works, why make it optional? Just make it the default. It's really terrible the way it is. :P
-C
On Sun, Apr 30, 2017 at 10:02 PM, Jabier Arraiza <jabier.arraiza@...2893...> wrote:
Can Inkscape get benefits to reduce banding on gradients appliing smoothing from mesh gradients as optional value in the gradient toolbar as a temporary workarround while swich to a higer color data?
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Easiest way to remove the banding on gradients would be to have dithering in pixman (used by cairo).
See also earlier discussion in [1]
You can find some traces of people having done that (even in old mozilla code[2] with a custom pixman patch[3])
From looking at the pixman repo[4], it does not seem there has been
anything done to do it there. I did not investigate the pixman ml to find if there has been discussions about dithering there.
Perhaps that should also remove banding in blur and derivative steps in filter effects…
Le 01/05/17 à 00:25, Marc Jeanmougin a écrit :
Easiest way to remove the banding on gradients would be to have dithering in pixman (used by cairo).
See also earlier discussion in [1]
You can find some traces of people having done that (even in old mozilla code[2] with a custom pixman patch[3])
From looking at the pixman repo[4], it does not seem there has been
anything done to do it there. I did not investigate the pixman ml to find if there has been discussions about dithering there.
On Mon, 2017-05-01 at 00:25 +0200, Marc Jeanmougin wrote:
Easiest way to remove the banding on gradients would be to have dithering in pixman (used by cairo).
Agreed
See also earlier discussion in [1]
You can find some traces of people having done that (even in old mozilla code[2] with a custom pixman patch[3])
From looking at the pixman repo[4], it does not seem there has been
This is dithering from 8 bit per channel to 5/6/5 bits per channel.
anything done to do it there. I did not investigate the pixman ml to find if there has been discussions about dithering there.
I looked into this some time ago. The biggest problem is that pixman does not support greater than 32 bit color. One would want to generate the gradient at, say 10 bit per channel color and then dither to 8 bit per channel (which including alpha means a minimum of 40 bit color).
BTW, Nvidia has an custom extension to OpenGL that can render SVG without all these problems but they can't get others interested.
https://developer.nvidia.com/nv-path-rendering
Tav
Thanks Mark.
On Mon, 2017-05-01 at 00:25 +0200, Marc Jeanmougin wrote:
Easiest way to remove the banding on gradients would be to have dithering in pixman (used by cairo).
See also earlier discussion in [1]
You can find some traces of people having done that (even in old mozilla code[2] with a custom pixman patch[3])
From looking at the pixman repo[4], it does not seem there has been
anything done to do it there. I did not investigate the pixman ml to find if there has been discussions about dithering there.
participants (7)
-
C R -
ivan louette -
Jabier Arraiza -
Marc Jeanmougin -
Maren Hachmann -
Miguel Lopez -
Tavmjong Bah