[Skencil-users] Fwd: Join forces for skencil1/sK1?

[Bill Baxter]

On 6/22/07, Bernhard Reiter <bernhard at intevation.de> wrote:
>
> > My suggestion is using "sK1" name for latest branch, because:
> > 1. This code is highly differ from 0.6/0.7 versions
>
> I believe this is tightly coupled with the question of how many
> branched we keep alive. See below.
>
> > 2. This code name is already announced as in Russian community and in
> west
> > open source world.
> > 3. This name is derived from Sketch
>
> Skencil is quite introduced as well, though I agree that sK1 is a stylish
> name that is derived from it and its file format.



Just an outsider's opinion here, but Skencil is a much nicer name if  you
ask me.  It immediately suggests a program that has something to do with
drawing.  sK1 by itself suggests absolutely nothing.   I could see using sK1
as a subtitle maybe.  Something like "Skencil : sK1" to differentiate it
from "Skencil : Classic", but just branding the thing "sK1" seems like a bad
idea.


The goals of a "mature vector graphics editor" seems to be compatible
> with Skencil's goal, unless somebody comes and want to use it as
> component,
> I do not think we should make provisions for it.
> That the separation of model-view-controler would enable
> something like skconvert to be run in scripts is quite cool, so I would
> like
> to keep or extend this as another goal.


Being available to use as a component would fill a niche that isn't
currently served.  I'm currently writing my own vector drawing program from
scratch because I couldn't find a decent component to build on.    Still, it
is quite a small niche, so it may not be worth the effort to create a good
clean component design just to serve that niche.  Of course creating a
design with clean separation is a good idea anyway, it just takes more time.

> Only  Scribus has similar features, but this application is not
> illustration
> > program.


Right, I think Scribus aims to be more a of a Quark wannabe.

So, my personal view of sK1 future is a competitor of CorelDRAW
> > but on Linux platform. And future users pool should grow from former
> > CorelDRAW users.
>
> Well it would be cool if we could target windows as well as GNU/Linux
> and X11 running plattform and even MacOS-X. Tcl/tk is pretty portable
> and pyGTK was as well in the past, so this should be doable.
> But this is more a personal goal. Having a great application for GNU/Linux
> is a huge plus as well.


I agree.  But I'm a big advocate for doing everything cross-platform,
always.  There's no good reason to arbitrarily cut off 80% of the potential
users in the world by being Linux-only.  Or even cutting off 20% by being
Windows only.  Unless your app is so OS-specific that a port wouldn't make
sense.  Say, something like a Gnome desktop configurator, or an app where
80% of the code is platform-specific and would have to be rewritten
per-platform anyway.  But for everything else, it might as well be written
cross-platform, or at least factored from the beginning with cross-platform
in mind.  (That's why wxWidgets is my GUI of choice.)

> Inkscape Team also declared that it's application is something like
> > CorelDRAW or Illustrator, but on my opinion Inkscape is more similar to
> > specialized editor like Macromedia Flash (at least on current
> development
> > stage).
>
> I cannot really judge, because I did not really use Inkscape a lot.
> My next implementation target with Skencil (or sK1) or whatever would
> be to add the map extension for the Thuban-Map-SVG. (Check out my svg
> export
> extension for thuban.intevation.org ).


Hmm.  I'm a beginner at both, but to me Inkscape is much more usable than
Skencil right now.  And I strongly disagree with the assessment that
Inkscape is a Flash competitor.  It doesn't even support animation right
now, so claiming it competes with Flash is just silly.   It is most
definitely a CorelDRAW/Illustrator competitor.  It may gain animation
features in the future, but right now it is a vector drawing program, plain
and simple.    And it works cross-platform via GTK ports.

The main thing that limits Inkscape's potential in my opinion is that it is
actually an SVG editor, as opposed to a general-purpose vector drawing
program.  If the feature doesn't exist in the SVG spec, it won't be added to
Inkscape.  Of course if your goal is to export SVG this is not a limitation
at all -- it's a huge feature.  And SVG at some level aims to be a general
purpose vector drawing description language.  So if it makes sense for an
application to want to do it, chances are that it will be added to the SVG
spec eventually, even if it isn't there currently.  Some of the Inkscape
devs are actually involved in the W3C SVG group.

Anyway, these are just my two cents.  I tried both Skencil and Inskcape at
the same time.  I had trouble figuring out how to do much of anything with
Skencil (after getting past the troubles I had just installing it), but I
found it very easy to use (and install) Inkscape.  When I look at Skencil
and Inkscape side-by-side, I seriously have to ask myself "what is the point
of skencil?"  Inkscape is cross plaform, well-thought-out UI, lots of
features, pluggable script extensions in multiple languages (but mainly
Python), written in C/C++ (==pretty fast and also easy installation),
pretty many decent tutorials available.  Uses GTK just like Skencil (so
unfortunately both look equally ugly and alien on Windows and Mac).  So the
main thing that seems to differentiate Skencil is that it's in Python
(except the low-level stuff, which seems to be X-Windows-specific C
extension modules).  And maybe also the fact that Skencil doesn't care about
SVG compatibility, so it is less constrained in how it can evolve.

I could definitely see Skencil catching up to Inkscape in a year or so,
thanks to the speed and ease of pure Python development.  But on the other
hand, being Linux-only restricts the pool of potential developers who might
take an interest in the project and decide to contribute.  Quite a few of
the Inkscape contributors come from the Windows world (including me -- I've
already made a couple of contributions to Inkscape now, developed on
Windows).

Anyway, just my outsider's opinion.  Inkscape is most definitely
head-to-head with Skencil, so I don't think you should ignore it.  You
should study it.  See what you can learn from it.  First you must clearly
decide for yourselves what it is that Skencil does, or will do, better than
Inkscape, and then clearly explain to the world on the basis of that why we
should rush to use and support Skencil rather than Inkscape.

I'll just conclude by noting that if all other things were equal about
Skencil and Inkscape (features and cross-platform issues), then I would
*definitely* prefer to work on Skencil because GTK+ is terribly painful to
work with.  Inkscape is slowly converting over to GTKmm/C++ which is a
little better.  But neither can hold a candle to Python for ease of
development.

This post isn't meant to be discouraging.  Quite the opposite.  I think
you're a little behind Inskcape right now, but if you focus on making
Skencil easily available to the widest audience possible (I.e. port it to
Win and Mac, and make sure installation is smooth and simple everywhere), I
think the ease of working with Python will help lure in more developers and
accelerate the pace of Skencil development to the point where it can rocket
past Inkscape.

Regards,
--bb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.wald.intevation.org/pipermail/skencil-users/attachments/20070622/f14823c2/attachment-0001.html