xarino project
Gemini
member of the Ubuntu forums called the "xarino" project.
I have sent a pm to the creator of the project to see if we might be
able to merge the two projects to get things rolling along again.
twright
I am Tom Wright, and I did most of the work on the xarino project. I
am more than happy for you to use any and all of the work we did to
help revive what might be the world's most awesome vector editor.
Most of the work which initially got done with was branding
definitions, looking through the source code and compiling notes on
the most important bugs. We also got it to build properly with the
latest version of wxwidgets and had started looking at modernizing
some bits of UI related code (e.g. to get rid of all of the deprecated
warning on compilation) but not many of these tasks were actually
finished because of other constraints we faced.
All in all I must admit I made a pretty poor job of reviving the
project but now I have more coding experience I would love to return
to it and make some more progress. One other thing is that I am not
sure about the legal issues around keeping the name xaralx, if you
want it xarino is yours for the taking along with all of the code left
in our svn repo and admin rights to our google code site.
Gemini
Thanks for posting in.
I will have more time to look things over next weekend.
I know one thing for sure... Carl Worth had explained to me that there
is a ton of work involved with rebuilding for the Cairo ren.engine. He
is still willing to offer advice and tech support to any Dev who is
interested in taking on the task. Only thing is, I am not sure only
one person can handle it?
This is the biggest part of the work which needs to be done. But.. if
we could get this portion under way, and get some really good ground
covered, I am sure I could then approach Charles, who may then be able
to talk the Linux dev he knows to jump in and, Charles might have more
of something to approach Magix with if enough headway is made. to
support the project. (even if only from a legal standpoint)
If not, I will discuss with Charles the prospect of branching off as
Xarino. The reason I say this is, it would be nice to see if we can
get some support behind the project, attracting more contributing
devs, and that may be actually possible if both Charles and Carl are
involved. (even if only on the sidelines)
Let me know what you want to do.... :)
Tom Wright
I think that there is a lot of good we could do on the project before
getting more developer interest but a lot of it is dependant on how we
approach it.
Probably the biggest issue is whether we really want to rip out cdraw
when that is such a monumental task - I personally wonder whether it is
worth it for a change which would add no new features and possibly take
away some of Xara's biggest benefits.
On the other hand I think that we would have serious problems in the
long term with a binary blob in the centre of the program and it may
impact out ability to get new developers. Perhaps with the work that has
already been done it might be a manageable task but seeing as it killed
the original project when it had many experienced coders behind it,
serious planning would be needed before any attempt to replace it.
One other idea I have been thinking about is whether currently Inkscape
is closer to what we hope to achieve than the legacy XaraLX code
especially if we intend to base it on Cairo. It might just be a lot less
work to fork Inkscape and rewrite the UI in the style of XaraLX, it all
depends on what everyone wants out of the program (and how intertwined
Inkscape's engine and UI are).
Naming wise I think that launching with a new name would be the best I
idea as it does not force us to be tied to Xara Xtreame for Windows and
ensure that we are OK legally. Really this depends on how Charles and
Magix feel about the project so I will leave this with you.
Finally in terms of code my specialities are GUI stuff and maintenance
so I do not know how much use I will be in reworking the core of the
program. I don't have that much free time at the moment but I would love
to do as much as I can to help with the project. Once we decide on the
path of the project, I think the best thing to do is to do one release
which is essentially code cleanup and modernisation then start on the
big things; depending on what we want to do, posting some GUI mockups
and using the forums might help gather interest.
On Thu, 2009-10-01 at 20:26 -0700, Gemini wrote:
> I will have more time to look things over next weekend.
> I know one thing for sure... Carl Worth had explained to me that there
> is a ton of work involved with rebuilding for the Cairo ren.engine. He
> is still willing to offer advice and tech support to any Dev who is
> interested in taking on the task. Only thing is, I am not sure only
> one person can handle it?
> This is the biggest part of the work which needs to be done. But.. if
> we could get this portion under way, and get some really good ground
> covered, I am sure I could then approach Charles, who may then be able
> to talk the Linux dev he knows to jump in and, Charles might have more
> of something to approach Magix with if enough headway is made. to
> support the project. (even if only from a legal standpoint)
> If not, I will discuss with Charles the prospect of branching off as
> Xarino. The reason I say this is, it would be nice to see if we can
> get some support behind the project, attracting more contributing
> devs, and that may be actually possible if both Charles and Carl are
> involved. (even if only on the sidelines)
>
> Let me know what you want to do.... :)
--
Thanks,
Tom Wright
Scott
keep Xara and not fork to Inkscape, for one thing we would loose what
makes Xara so good the rendering speed and the nice way the tools work.
If you kept Cdraw would that put a life span on Xara/xarino?
Some of my thoughts.
Gemini
beginning.
The engine xaralx currently uses is CDraw and it is not open-sourced.
It will be extremely difficult to move forward and gain any interest
in the project, if that very subject is not addressed.
This is also where the renaming / forking may run into a snag.
To try to branch off, using the current rendering engine probably
won't go unnoticed and is sure to cause some trouble.
You have probably seen this before but, (just in case) it sums up the
issue: and this is basically where both Charles and Carl still stand
to this day.
http://www.linux.com/archive/feature/60491
Tom Wright
> The binary rendering engine is the biggest issue, and was so from the
> beginning.
> The engine xaralx currently uses is CDraw and it is not open-sourced.
> It will be extremely difficult to move forward and gain any interest
> in the project, if that very subject is not addressed.
> This is also where the renaming / forking may run into a snag.
> To try to branch off, using the current rendering engine probably
> won't go unnoticed and is sure to cause some trouble.
As far as I remember from my work on Xarino, the library comes with a
quite permissive license which allows the project to use it as long as
it is not changed but of course I am no lawyer so I do not know all of
the issues involved.
> I'm not a dev or know anything about coding. IMHO I'd like to see us
> keep Xara and not fork to Inkscape, for one thing we would loose what
> makes Xara so good the rendering speed and the nice way the tools
> work.
> If you kept Cdraw would that put a life span on Xara/xarino?
> Some of my thoughts.
I agree that we should strive to make it as much like the official Xara
but if we are left with is a slightly out of date codebase designed
around another rendering engine and an old version of Xara's interface
(in wxwidgets which whilst multiplatform is not nice and native like GTK
or QT). We almost certainly will not be able to replace the engine
without more developers.
If we do decide to keep Cdraw then we will be able to get some instant
results which would be cool but I long-term future of the program might
be jeopardised.
--
Thanks,
Tom Wright
Scott
the speed in which it works?
Inkscape from 4.7 and up is very slow, making it hard to use at least
for me.
thanks for clarifying things some what.
Tom Wright
> Won't using a different engine defeat the reason why Xara is so good,
> the speed in which it works?
> Inkscape from 4.7 and up is very slow, making it hard to use at least
> for me.
> thanks for clarifying things some what.
I am afraid to say that there is a high chance that much of Xara's
awesome speed will probably will be lost with switching from CDraw but
Cairo and Inkscape are constantly improving and I can think of some
other features from Xara which could help like a flexible render detail
slider.
In any case, even whilst CDraw is awesome now, we only have a binary
only library based on a snapshot of an old version which does not get
updated and will not still be the fastest thing around for ever. It was
my original aim with Xarino to stick with CDraw and there are still
benefits to keeping it so it is definitely worth talking about.
--
Thanks,
Tom Wright
Scott
taken to try and save Xaralx.
I now have a better handle on what needs to be done and which I had the
skills to help. Keep up the work.
Gemini
some of the more important deps.
There are other rendering engines out there but, going in with a clean
build of cairo may not be as bad (slow) as one may think.
Remember, Inkscape has become bogged over time with patches, etc. so,
this may be worth asking Carl about. Would we notice the same
reduction in speed as the current build of Inkscape if we are starting
from scratch?
We could also start taking a look at some other rendering engines as
well? Maybe there is something more comparable to CDraw.
I think someone from Ubuntu recently updated wxwidgets ? If so, it
would be nice to grab that source.
eventually we will also have to start thinking about how this will be
packaged, So we are able to share with the communities?
AutoPackage works fairly well
I also have "some" experience with building rpm's, but not debs (and
alien is not always the best solution)
I think I will ask Charles to peek in and give some input on this
current conversation (when he has time) to know exactly where we will
stand, and then we can figure where to go from there, because I am
really not sure how plausible it will be to carry on with CDraw. I
know that really stinks but, think about the fact of the potential of
gaining newer features from newer releases of Xara if we can replace
CDraw. I am not sure if that possibility still stands now with Magix
at the helm but, this was Charle's thoughts in the past.
I gotta try to get him to look in asap so we can get our bearings.
I'll send him an email this weekend
Gemini
I cannot imagine a day without xaralx, I love Inkscape too but, I work
too fast for it and it crashes on me alot :P
There are also some great features within xaralx which Inkscape does
not have. So, to me this project is well worth any effort put into it.
There are droves of people who want to see this happen, it is just a
matter of doing what we can to get it moving again. We have options
And one does not necessarily need to posses programing skills to help.
I am not the most technically skilled person in the world, and I often
am not very educated on subjects I speak on but, I have the drive to
see this happen... I have been at it somewhat for a couple years now
(starting over at TalkGraphics.com, and actually when they closed the
xaralx board, I offered up LGU as a place where anyone interested in
xaralx can go.
The interest has not begun yet, there are only a few of us there who
would die if xaralx ceased to be (or some derivative of it)
So... stick around Scott, we're more than glad to have anyone
interested, involved in any way they can be ;)
Tom Wright
> Remember, Inkscape has become bogged over time with patches, etc. so,
> this may be worth asking Carl about. Would we notice the same
> reduction in speed as the current build of Inkscape if we are starting
> from scratch?
different versions etc. might have problems depending on the complexity
of drawing. Inkscape have also developed their own custom canvas widget
so if anything is slowing it down outside the library it will be that.
> We could also start taking a look at some other rendering engines as
> well? Maybe there is something more comparable to CDraw.
I think for Linux Cairo is the fastest and best supported 2d drawing
layer - I am not sure there is any decent alternative but we could look
into it.
> I think someone from Ubuntu recently updated wxwidgets ? If so, it
> would be nice to grab that source.
> eventually we will also have to start thinking about how this will be
> packaged, So we are able to share with the communities?
> AutoPackage works fairly well
> I also have "some" experience with building rpm's, but not debs (and
> alien is not always the best solution)
Part of what we did on Xarino was getting it to build against the newer
wxwidgets (2.8, I think they are on 3 now though). It should just be a
matter of going through the compiler errors.
As for packaging I think we should use launchpad for everything and
create a ppa - I think we need to stick to native packaging in order to
keep things simple. There has also been a lot of buzz about Quickly on
Ubuntu planet later so we might want to try that later.
> I think I will ask Charles to peek in and give some input on this
> current conversation (when he has time) to know exactly where we will
> stand, and then we can figure where to go from there, because I am
> really not sure how plausible it will be to carry on with CDraw. I
> know that really stinks but, think about the fact of the potential of
> gaining newer features from newer releases of Xara if we can replace
> CDraw. I am not sure if that possibility still stands now with Magix
> at the helm but, this was Charle's thoughts in the past.
>
> I gotta try to get him to look in asap so we can get our bearings.
> I'll send him an email this weekend
Great, once we have a bit more info then we will be able look at how
much there is to do.
--
Thanks,
Tom Wright
Scott
Gemini
I am hoping he has the time to have a quick look at our discussion
here, so we can really get down to setting up a plan.
We will need to be patient, as he is quite a busy guy, but I assure
you, he has always been good at responding fairly quickly.
Haha, I always get excited when the potential of beginning further
work on xaralx is being discussed.
I will say this... Should "anything" be done regarding xaralx
development, leading to a new version release... it will be noticed
and attract more attention than most realize at this moment, and has
the potential to attract people with coding experience. It will be a
kinda big "Linux news" item throughout the communities ;)
Tom Wright
> Haha, I always get excited when the potential of beginning further
> work on xaralx is being discussed.
> I will say this... Should "anything" be done regarding xaralx
> development, leading to a new version release... it will be noticed
> and attract more attention than most realize at this moment, and has
> the potential to attract people with coding experience. It will be a
> kinda big "Linux news" item throughout the communities ;)
along again. When we get the that stage it would be great if we could
set up a basic web presence then reach out to the community via Slashdot
and various forums - if by then we also have an easy to install package
which is less buggy, uses the newer wxwidgets and maybe has a bit of UI
cleanup (maybe some new icons keeping with the Xara style) we might be
able to attract some real intrest.
--
Thanks,
Tom Wright
cmo
It’s open source so you can do anything you like to it, but I would
suggest that it makes more sense to submit any changes to the Xara
repository since that does get followed and will likely raise more
interest amongst others if they see it actively being worked on. I
can’t promise anything though. Similarly it would seem to make sense
to keep the name the same not least because that’s the name people
will recognise. But of course you’re free to do what you want.
On the issue of replacing CDraw, I’ll just say these points;
a) While some people are fanatical about not having any closed source
parts, and I can understand that, a larger section of the Linux
community, I believe, is perfectly willing to have closed source
binary elements, if that’s what’s required in order to get the
software they want working. Witness the number of closed source
device drivers, and even things like Flash. So the first question to
ask is whether it’s necessary, or a priority, to replace the CDraw
rendering engine.
My personal view is, like Tom’s comment earlier, it probably makes
sense to focus on other priorities to make it a more stable, usable
application, (e.g. to work with the latest WxWidgets and recent
distros) than to replace the closed source CDraw part.
b) But assuming you do decide to replace the CDraw rendering engine,
then replacing CDraw with Cairo should not be that difficult for a
reasonably skilled developer. Cairo and CDraw are very similar - they
are both low-level bezier ‘postscript’ based anti-aliasing vector
rendering engines with a broadly similar capability set.
It’s my opinion, confirmed by the Inkscape devs that I know, that this
work should be very considerably more easy than, for example, making
Inkscape use the Cairo rendering engine. I would also suggest it’s a
lot less work than trying to morph Inkscape into a Xara-like clone.
As to whether using a different engine will affect Xara’s speed? The
answer is that yes it will BUT a significant part of Xara’s speed is
due to the internal data structures and efficiency of the algorithms
used - it’s not all just down to CDraw. So I suspect for day-to-day
use, Cairo it will be perfectly fast enough, and quite likely faster
than Inkscape. Also remember that there are hardware accelerated
Cairo versions - Carl can tell you a lot more about that.
But as always with such open source projects, it comes down to finding
developers willing to put the time and energy into moving the project
forward. I can’t help you on that score.
Gemini
exactly the content I was looking for.
So guys, there we have some vital input, which clears up any questions
any of us may have had (me especially)
Now, I am one of those many, who could care less that CDraw is a
binary element... as long as xaralx works :P
And I know many others who feel the same. So, with that in mind, and
now knowing where continuing the use of CDraw stands, I am in full
agreement with Tom that replacing the ren. engine is not the biggest
issue.
It is not that I doubted you Tom, but my fears where more based around
the difficulty of attracting any independant developers, as long as
CDraw remains a binary element. It will be difficult, but not
impossible.
I guess it is now a matter of continuing the work set in the xarino
project?
We also see that using the name of xaralx is not an issue, and is in
fact, in a way suggested for good points. But we can talk on this more
to decide which name to use.
And, I think Charles is right, that once we begin to share any
progress upstream, this project will gain some attention. Let's pray
it attracts, especially experienced/ skilled programmers and such. ;)
One last thing I will touch on, regarding changing out the ren engine
is... such work would definitely be more likely to attract developers
from many corners of the Linux community. But, again... that is
something we can discuss later down the road, after we have made some
progress and if we see that work does not seem to be attracting these
specific types of people.
But, a website is a very good idea, and we can all take part in
promoting it and getting the word out to the masses...
Tom Wright
a important step closer to having a plan.
On Mon, 2009-10-05 at 04:52 -0700, cmo wrote:
> It’s open source so you can do anything you like to it, but I would
> suggest that it makes more sense to submit any changes to the Xara
> repository since that does get followed and will likely raise more
> interest amongst others if they see it actively being worked on. I
> can’t promise anything though. Similarly it would seem to make sense
> to keep the name the same not least because that’s the name people
> will recognise. But of course you’re free to do what you want.
I think that submitting changes back to the XaraLX repository would be a
good I idea but seeing as we do not have write access and I am unsure
any of that infrastructure is still maintained this might be a problem.
Perhaps the best course of action is to create a new repository in
Launchpad etc. and set it to automatically submit our changes upstream.
As for the name I think that just depends on what users would be most
comfortable with and any legal issues.
> On the issue of replacing CDraw, I’ll just say these points;
> a) While some people are fanatical about not having any closed source
> parts, and I can understand that, a larger section of the Linux
> community, I believe, is perfectly willing to have closed source
> binary elements, if that’s what’s required in order to get the
> software they want working. Witness the number of closed source
> device drivers, and even things like Flash. So the first question to
> ask is whether it’s necessary, or a priority, to replace the CDraw
> rendering engine.
>
> My personal view is, like Tom’s comment earlier, it probably makes
> sense to focus on other priorities to make it a more stable, usable
> application, (e.g. to work with the latest WxWidgets and recent
> distros) than to replace the closed source CDraw part.
Whilst for now I think it would be perfectly fine to keep using CDraw,
it will almost certainly be an issue later so I think we should start
work on important, user visible tasks first whilst keeping replacing
CDraw on our roadmap with an eye to rip it out later (which should not
be any harder than doing it now).
> As to whether using a different engine will affect Xara’s speed? The
> answer is that yes it will BUT a significant part of Xara’s speed is
> due to the internal data structures and efficiency of the algorithms
> used - it’s not all just down to CDraw. So I suspect for day-to-day
> use, Cairo it will be perfectly fast enough, and quite likely faster
> than Inkscape. Also remember that there are hardware accelerated
> Cairo versions - Carl can tell you a lot more about that.
Hm, that would make us the first hardware accelerated vector editor for
Linux I think - a very nice place to be :-).
> But as always with such open source projects, it comes down to finding
> developers willing to put the time and energy into moving the project
> forward. I can’t help you on that score.
Well I will do my best to help the project when I have the time/energy -
hopefully if enough people care about it for it to be worth doing then
they would not mind getting involved.
On Mon, 2009-10-05 at 07:00 -0700, Gemini wrote:
> It is not that I doubted you Tom, but my fears where more based around
> the difficulty of attracting any independent developers, as long as
> CDraw remains a binary element. It will be difficult, but not
> impossible.
Feel free to doubt me; I am no expert in graphics so I am not really
aware of how large a task switching to Cairo will be.
> I guess it is now a matter of continuing the work set in the xarino
> project?
> We also see that using the name of xaralx is not an issue, and is in
> fact, in a way suggested for good points. But we can talk on this more
> to decide which name to use.
Well as I have never used Xara for windows I would not be the best
person to do the naming - as far as I am concerned either is good. In
any case it would save a lot of hassle with the makefiles.
Coding wise I think once we have got a good repository for the code then
I at least would be ready to start doing some more coding.
--
Thanks,
Tom Wright
Gemini
should be back on Thursday Oct. 8
I am hoping Tom, that you and I can email so I can get a better idea
of what you are able & willing to contribute, and what types of
skilled individuals we need to look for from there. ;)