Fw: Re: [Passepartout] PPT + Inkscape
--------- Original Message -------- Da: passepartout@...273... To: "passepartout@...273..." <passepartout@...273...> Cc: "Dylan VanHerpen" <dvanherpen@...210...> Oggetto: Re: [Passepartout] PPT + Inkscape Data: 23/05/04 22:32
Dylan VanHerpen <dvanherpen@...210...> writes:
Fredrik,
Hey, I'm just half the development team :-)
I'm impressed with the functionality already present in Passepartout, and anxiously waiting for the next update. Like others, I'd like SVG support, truetype font support and better type handling.
TrueType support is being worked on. Type handling?
I've also been using Inkscape, which already is very useable for single page layouts (especially when the Pango conversion is completed with the next release). It seems to me that Inkscape and PPT complement each other rather well. What would you think about combining the strengths of both efforts, and developing PPT as a layout module for Inkscape? Inkscape seems to be further along in terms of font handling (Pango), vector graphics and object layering (with alpha-blending). Combining the two would make for a kick-ass DTP solution with a great universal document format.
Collaboration with Inkscape has been suggested before and it certainly makes a lot of sense, though I don't think merging the programs is a good idea.
Inkscape seems to have a very nice version of the gnome canvas, more capable than the vanilla version, which I'm not even sure has a maintainer.
I don't know if Pango has much to do with the font support in Inkscape. Inkscape can convert fonts to Beziér curves. I don't think it could do that without reading the font files. We'd like to know how it manages to find out what font files correspond to which fonts in X.
We haven't been in touch with any of the Inkscape devs though.
Passepartout-dev wrote:
Collaboration with Inkscape has been suggested before and it certainly makes a lot of sense, though I don't think merging the programs is a good idea.
Would you like to share your reasons why not?
We(a bunch of inkscape devels) think that by building multiple frontends on the inkscape code base for different application types we can make a more unified application set, and that it can be done with less programmer-hours. Having a look through the passepartout source, I think you would benefit greatly from our expertise with several things - much better canvas system, a powerful document tree handling system, superior maths core, better h-n-j and of course support for import and export of vector graphics.
On the cons side, we don't handle any dtp type file formats yet, don't support multiple pages, have poor support for linking to multiple documents (no interface in particular), terrible printing support (being addressed) and clearly no direct interface for working with dtp documents.
I think your team could easily provide that expertise for inkscape - we have talked about doing it ourselves, but you have obviously battled and overcome several difficult problems in your work. A possible route would be to start sharing the common subsystems, and leaving the two applications separate. Another idea would be to start adding passepartout's tools to inkscape and merging the code base.
I think the latter is probably more efficient, and would be a significantly innovative step in the world of DTP and illustration.
Inkscape seems to have a very nice version of the gnome canvas, more capable than the vanilla version, which I'm not even sure has a maintainer.
We do, and we've been slowly working on building a separatly buildable library for use by other projects such as passepartout.
I don't know if Pango has much to do with the font support in Inkscape. Inkscape can convert fonts to Beziér curves. I don't think it could do that without reading the font files. We'd like to know how it manages to find out what font files correspond to which fonts in X.
A few of our developers have been working on using pango and found it is not entirely suitable for drawing programs (and for similar reasons, DTP programs).
We haven't been in touch with any of the Inkscape devs though.
Hop onto the jabber channel and have a chat! We love to chat!
njh
Nathan Hurst <Nathan.Hurst@...38...> writes:
Passepartout-dev wrote:
Collaboration with Inkscape has been suggested before and it certainly makes a lot of sense, though I don't think merging the programs is a good idea.Would you like to share your reasons why not?
Well, for one thing it would mean a lot of work. There are only two of us. We'd have to get to know the Inkscape code and do a feasability study first.
Secondly, I am a bit worried we'd be making a Frankenstein's app. A lot of compromises would have to be made. As Bulia Byak pointed out, as long as Inkscape uses SVG as its document format, it would be a bad idea to add things that are not supported by SVG.
We(a bunch of inkscape devels) think that by building multiple frontends on the inkscape code base for different application types we can make a more unified application set, and that it can be done with less programmer-hours.
Well, I encourage that. But Passepartout is almost two years old and probably has a framework that is radically different from Inkscape's. The only large-scale "merge" that makes sense to me is if you fork Inkscape and transform it into a DTP program by adding bits and pieces from Passepartout.
better h-n-j
What is h-n-j?
Inkscape seems to have a very nice version of the gnome canvas, more capable than the vanilla version, which I'm not even sure has a maintainer.We do, and we've been slowly working on building a separatly buildable library for use by other projects such as passepartout.
The gnome canvas does not seem to support rotated text, but you do. Do you convert the text to curves for that to work?
I don't know if Pango has much to do with the font support in Inkscape. Inkscape can convert fonts to Beziér curves. I don't think it could do that without reading the font files. We'd like to know how it manages to find out what font files correspond to which fonts in X.A few of our developers have been working on using pango and found it is not entirely suitable for drawing programs (and for similar reasons, DTP programs).
It isn't, no. There is a great divide between fonts on the screen and fonts on paper in Unix/X. Perhaps Fontconfig can help with that, but I'm not sure how to use it.
We solved the problem by letting ghostscript render text. The problem is, it is not fast and it won't discover TrueType fonts automatically.
Hop onto the jabber channel and have a chat! We love to chat!
Ok, maybe we will.
On Tue, 25 May 2004, Fredrik Arnerup wrote:
Nathan Hurst <Nathan.Hurst@...38...> writes:
Passepartout-dev wrote:
Collaboration with Inkscape has been suggested before and it certainly makes a lot of sense, though I don't think merging the programs is a good idea.Would you like to share your reasons why not?
Well, for one thing it would mean a lot of work. There are only two of us. We'd have to get to know the Inkscape code and do a feasability study first.
Secondly, I am a bit worried we'd be making a Frankenstein's app. A lot of compromises would have to be made. As Bulia Byak pointed out, as long as Inkscape uses SVG as its document format, it would be a bad idea to add things that are not supported by SVG.
Here are my thoughts on this. I've been on a few projects where other projects join up - maybe about a dozen instances. It's generally quite amicable and seems to be in everyone's best interests. But in each case the second group ended up just kind of dying out, without any code or ideas getting transferred across. I'm not sure if those issues were time issues, cultural incompatibility, technical problems, or just "too many cooks in the kitchen", but in any case, evidence suggests it's much trickier to make work than you'd think.
So... I definitely love the idea of combining forces, but realistically I don't think it'd be in Passepartout's interests to wholesale join Inkscape. I am much more in favor of the "sister projects" approach like we have with Scribus, where we just agree to help each other out and look for ways to enhance interoperability, but remain distinct projects with distinct priorities and goals.
Bryce
participants (4)
-
Bluefuture -
Bryce Harrington -
Fredrik Arnerup -
Nathan Hurst