I'm torn between wanting to shamelessly pander to Illustrator users or decide if doing it differently would be better enought to mitigate the inconsistancey!
Illustrator is far from being the standard. Xara, CorelDraw, Freehand, Canvas all have sizeable user communities, and as a rule they don't imitate Illustrator keybindings but do most things their own way. Besides, there's GIMP and other linux graphic soft that we must take into account. So I think in each case, we should review the existing approaches and either select one of them (not necessarily Illustrator) as the best, or devise our own.
And by the way, at least global keybindings should ultimately be user configurable (it's difficult to do that for all keybindings, but at least those in shortcuts.c must be taken from the config file).
On Sun, 23 Nov 2003, bulia byak wrote:
Date: Sun, 23 Nov 2003 10:56:22 -0500 From: bulia byak <bulia@...23...> To: inkscape-devel@lists.sourceforge.net Subject: [Inkscape-devel] (no subject)
I'm torn between wanting to shamelessly pander to Illustrator users or decide if doing it differently would be better enought to mitigate the inconsistancey!
Illustrator is far from being the standard.
It is hard to say any application is the standard but I'd venture to say it is the market leader and as such is a de facto standard.
What I want to know is how the hell they get away with charging ~ $500 when CorelDraw 11 is a mere ~ $150 (prices based on quickly looking at adobe.com and corel.com)!?
I do believe that Adobe Illustrator is much less dominant in the Vector drawing space than Adobe Photoshop is in Raster Graphics and I look forward to Inkscape forcing them even harder to justify their prices.
Xara,
when you said Xara last time i could only think of Xara 3D but i did a bit of research and it seems they have a bit more than just 3D http://www.xara.com/
Xara X does not even support SVG, I'm not impressed at all.
CorelDraw,
I should think CorelDraw when I think vector graphics but I dont. I just not familiar with it at all. Lauris Kaplinski said older versions of CorelDraw helped inspired him with Sodipodi.
(I read a recent review of CorelPainter and they have made its interface much more like Adobe photoshop in a bid to sway Adobe users. I've no idea what the latest version of CorelDraw is like though).
Freehand,
I'd have claimed Freehand was the second biggest commercial vector graphics progam largely due to the strength of Flash and Macromedia's other products.
Canvas all have sizeable user communities,
I can only guess you mean this Canvas http://www.deneba.com/products/canvas8/ and it is a technical graphics program more than an artistic vector graphics program.
While we are enumerating Vector Graphics programs I may as well mention that Jasc Web Draw 1.0 has a reasonably nice and clean interface and the SVG source editing looks useful. By the time it gets to version 2 or 3 I expect it will be more substantial and a really good program. I'm quite impressed by the progress they have made with Jasc Paint shop Pro 8, so I expect they will also turn WebDraw into something good enough at a reasonable enough price.
and as a rule they don't imitate Illustrator keybindings but do most things their own way.
Besides, there's GIMP and other linux graphic soft that we must take into account. So I think in each case, we should review the existing approaches and either select one of them (not necessarily Illustrator) as the best, or devise our own.
And by the way, at least global keybindings should ultimately be user configurable (it's difficult to do that for all keybindings, but at least those in shortcuts.c must be taken from the config file).
All I am trying to say is that Inkscape cannot please everyone and that it makes most sense to look first at Illustrator (and Freehand and maybe CorelDraw in that order of precedence).
I am not saying the Inkscape developers should disregard other software and other programs certainly should provide inspiration just to follow the leader unless there is a significant benefit to doing it differently. By following Adobe Illustrator we make it much easier for people to leverage existing knowledge (training, documentation, and other support) and we make migration away from Illustrator much easier.
I think it makes more sense to look at vector graphics programs rather than raster graphics programs. Some projects make a point of being different and good luck with that but if you want to maximize your audience then the default shortcuts should take the path of least resistance. I really hope that Inkscape look first to the leading vector graphics programs and also carefully consider the Gnome Human Interface Guidelines (push for changes if you strongly disagree) and KDE Usability Guidelines.
It would be nice if Inkscape allowed users to rebind menu items to other keys the way the GIMP does (off by default in GIMP 1.3/2.0 to prevent accidental rebinding). It would also be quite nice to be able to easily switch between keybinding profiles, the GIMP has a photoshop-menurc just for this purpose but unfortunately has no user visable way of changing it (by which I mean you can only change it by removing the standard menurc and replacing it with the photoshop-menurc) but even then the default keybindings should try to appeal to the largest possible audience.
Sorry to go on so long, I do take this a little too seriously sometimes.
Sincerely
Alan Horkan http://advogato.org/person/AlanHorkan/
PS Opinions, disclaimer, etc
PPS Another factor is the behaviour of Adobe (think Sklyarov) and their ridiculous prices which metaphorically makes me want to slap them.
Alan Horkan wrote:
All I am trying to say is that Inkscape cannot please everyone and that it makes most sense to look first at Illustrator (and Freehand and maybe CorelDraw in that order of precedence).
Ah yes, because free software can only chase the tail lights of commercial software. Here was me thinking we'd evaluate ideas based on merit...
I think that it is most important to match the behaviour of other gnome apps.
njh
On Mon, 24 Nov 2003, Nathan Hurst wrote:
Alan Horkan wrote:
All I am trying to say is that Inkscape cannot please everyone and that it makes most sense to look first at Illustrator (and Freehand and maybe CorelDraw in that order of precedence).
Ah yes, because free software can only chase the tail lights of commercial software. Here was me thinking we'd evaluate ideas based on merit...
I think that it is most important to match the behaviour of other gnome apps.
njh
This is surely a topic that could be discussed heatedly and vigorously, but for the sake of keeping good company it would be best if we didn't!
However the issue of Illustrator compatibility is an important one and I think we need to have a position as a project, since the topic is certainly going to be asked again and again if we don't.
It is a very good point that Illustrator is a dominant application and that many users will be familiar with it already. We could do much worse than to clone Illustrator! However, has not been a goal of either Sodipodi nor Inkscape, and I'd daresay it probably shouldn't be. It's important that our developers feel they have the freedom to make unique design decisions if they feel the need, because we want to encourage contributors to be creative (and hopefully prolific!) and not feel constrained to (boring?) reimplementation work.
How does this sound as a compromise position:
"Where a better solution cannot be found, default to the way Illustrator does it."
I think this emphasizes that we want to encourage developers to be creative and try new ideas, but gives them a standard of measure to compare against. And more importantly, for developers who are unsure what to do it gives a reference they can start from.
What do folks think?
Bryce
On Sun, 23 Nov 2003, Bryce Harrington wrote:
<snip>
This is surely a topic that could be discussed heatedly and vigorously, but for the sake of keeping good company it would be best if we didn't!
Couldn't resist responding already.
I hope my previous post was suitabley respectful, I want to encourage and the last thing I want to do is cause flames.
However the issue of Illustrator compatibility is an important one and I think we need to have a position as a project, since the topic is certainly going to be asked again and again if we don't.
It is a very good point that Illustrator is a dominant application and that many users will be familiar with it already. We could do much worse than to clone Illustrator! However, has not been a goal of either Sodipodi nor Inkscape, and I'd daresay it probably shouldn't be. It's
I agree. I'll try not to labour the point so much.
important that our developers feel they have the freedom to make unique design decisions if they feel the need, because we want to encourage
How does this sound as a compromise position:
"Where a better solution cannot be found, default to the way Illustrator does it."
That is brilliant, just the sentiment I was trying to convey. Abiword works by a similar rule of thumb and I think it has worked very well for them and they often do change things for the better.
I think this emphasizes that we want to encourage developers to be creative and try new ideas, but gives them a standard of measure to compare against. And more importantly, for developers who are unsure what to do it gives a reference they can start from.
What do folks think?
I think Inkscape has attracted a great crowd of people and enthusiasm and will go far!
Bryce
Sincerely
Alan Horkan http://advogato.org/person/AlanHorkan/
On Mon, 24 Nov 2003, Alan Horkan wrote:
This is surely a topic that could be discussed heatedly and vigorously, but for the sake of keeping good company it would be best if we didn't!
Couldn't resist responding already.
I hope my previous post was suitabley respectful, I want to encourage and the last thing I want to do is cause flames.
No prob, it brought up a good point that hadn't been dealt with for Inkscape. I think since it was a topic that we'd run into when we were working on Sodipodi that there was some sensitivity, but it sounds like underneath we're in violent agreement.
It is a very good point that Illustrator is a dominant application and that many users will be familiar with it already. We could do much worse than to clone Illustrator! However, has not been a goal of either Sodipodi nor Inkscape, and I'd daresay it probably shouldn't be. It's
I agree. I'll try not to labour the point so much.
It's actually good to bring it up as we discuss things, from the perspective that some of us (like me) don't own Illustrator (nor wish to). So having this sort of input from those who are familiar with it may be the best way to ensure that knowledge gets fed in.
important that our developers feel they have the freedom to make unique design decisions if they feel the need, because we want to encourage
How does this sound as a compromise position:
"Where a better solution cannot be found, default to the way Illustrator does it."That is brilliant, just the sentiment I was trying to convey. Abiword works by a similar rule of thumb and I think it has worked very well for them and they often do change things for the better.
Great! I actually can claim no credit for it - I cribbed it entirely from Nathan, and also thought it felt it summarized things succinctly. :-)
Bryce
On Mon, 24 Nov 2003, Nathan Hurst wrote:
Date: Mon, 24 Nov 2003 11:22:06 +1100 From: Nathan Hurst <Nathan.Hurst@...38...> To: Alan Horkan <horkana@...44...> Cc: inkscape-devel@lists.sourceforge.net Subject: Re: Keybindings in general [was Re: [Inkscape-devel] (no subject)]
Alan Horkan wrote:
All I am trying to say is that Inkscape cannot please everyone and that it makes most sense to look first at Illustrator (and Freehand and maybe CorelDraw in that order of precedence).
Ah yes, because free software can only chase the tail lights of commercial software. Here was me thinking we'd evaluate ideas based on merit...
I think that it is most important to match the behaviour of other gnome apps.
I mentioned the Gnome and KDE guidelines too, I guess I should have mentioned them first but they are only guidelines and they dont contain all the answers, in fact they dont even contain that much explanation of why they choose to do things certain ways, what compromises and tradeoffs where made.
Think of the users, think what is likely to be the best possible answer for the most possible people.
Those applications are by no means perfect but consistancy is very helpful for most users, there are known benifits to keeping things similar, that is the merit of this idea.
Show me something that is clearly better - not just different - then of course it is worth making things a little bit more difficult to learn at first if it it is significantly better.
I'm only pointing out that it makes sense not to do things differently unless there is a very good reasons to do otherwise.
Inkscape will still be a very different application, have full source code available, an unbeatable price, (faster, smaller?) and lots of other distinguishing features I'm sure.
Inkscape will likely do things that commercial software will never do! With the right infrastructure Inkscape can encourage and make it easy for others to help out with plugins and extensions. It would be fantastic if Inkscape suppoted even half of the XML vector graphics formats such OpenOffice Draw (libgsf might help make this easier), Kivio, Karbon14, Killustrator/Kontour, Dia amongst others. The very openness of Inkscape will make it easier to eventually support lots of different scripting langauges. I'm sure it wont take long before people the world over to translate Inkscape into their local langauage something that commercial software rarely manages. (Not sure best how to make a quick comment about spellchecking but that too is something Inkscape could get cheaply from the underlying desktop, a feature i dont believe Illustrator has).
I'm sure there will be many other way for Inkscape to differentiate itself. I just want to help as best I can to improve Inkscape.
I am glad you have a healthy amount of skepticism. Thanks.
Sincerely
Alan Horkan http://advogato.org/person/AlanHorkan/
On Mon, 24 Nov 2003, Alan Horkan wrote:
Think of the users, think what is likely to be the best possible answer for the most possible people.
I'm only pointing out that it makes sense not to do things differently unless there is a very good reasons to do otherwise.
Inkscape will still be a very different application, have full source code available, an unbeatable price, (faster, smaller?) and lots of other distinguishing features I'm sure.
I'd like to inject a couple comments into this thread regarding goals, since we've touched on it here in relation to userbase acquisition.
I think often people assume the goal of every software application is to get as many users as possible. For proprietary software, sure, that translates into more money since each user (in theory!) pays for their copy of the software. Or it gets you bought out by Microsoft, or whatnot. ;-)
But for an open source project, is it the case that more users equals more goodness? Are users the ultimate customers?
My feeling is that no, users are more of a means to an end. Lots of users can translate into a larger pool to draw developers from, and of course potential developers will consider the (potential or actual) size of the userbase when deciding whether to bother submitting their changes. But of course in these cases the real goal is getting more developer activity.
Sure there can be some personal satisfaction in knowing that the software is very widely used. Many times the desire to gain more users is what drives developers to work hard. But careful what you wish for! With many users comes many demands for changes. Your open source project can feel more like an job you can't get away from and don't get paid for! ;-)
From my experience there's three motivations for open source development
that can easily be translated into goals; one is practical, one idealogical, one educational, and one social:
First, you want good software available at a good price. For example, this is why I got involved with Sodipodi originally. I wanted a good drawing tool that I could tinker with to improve for my own d*mn needs. I suspect this holds true for a lot of us. We have a job or a hobby that needs a good tool. We want to make a good tool that'll last us and do well by us for the remainder of our lives. If few others use it, well then we can be special. If lots of users use it, and in doing so, it results in a better tool for me, then well that's great too.
Second is the desire to make the world, or more specifically the open source community, better. This is why adhering to and improving open standards is important, for example. For this motivation, a large userbase is kind of an objective, but again it's slanted in favor of developers since by definition the "open source community" is people who contribute to open source projects.
Third is to learn new things. Many open source projects started or gain contributions from people who are doing it out of curiousity, interest, or need to learn new skills, either for career purposes or personal fulfilment. A large userbase (implying a widely known application) can *maybe* translate into recognition when interviewing, but otherwise is not very relevant.
Fourth is a desire to work with good folk and have fun working together. Because, really, for the developers this is a hobby, and hobbies should be fun, especially if they're going to take up a lot of your time. For this, the personality, friendliness, and dedication of the users and developers is much more important than their sheer quantity.
So, for all these reasons, in a way other developers are the real customers of the open source developers, and users are just lucky sods that get all their benefits as a side effect! ;-)
For Inkscape, I think that we still need to grow the userbase. However I feel that we can achieve that easily enough by just focusing on making the code better; this is best achieved by encouraging developers to develop.
Bryce
P.S. I should add to the above that I use the verb 'develop' to mean more than just programming. Develop could mean a variety of activities that improve the tool, including especially documentation, design, application art, testing, website, and administrative work. Even just providing feedback can be considered development work if it is done with consideration and in an organized fashion.
participants (4)
-
Alan Horkan -
Bryce Harrington -
bulia byak -
Nathan Hurst