As far as I can tell, all GUI toolkits are basically the same.
They all have some sort of base Widget that everything else that can be drawn inherits from.
They all have basically the same widgets - Window, Scrollbar, Button, Dialog, FileSelector, DrawingArea, Menu, Container, etc.
They all use event driven architecture with a "main loop" that responds to user events through application registered callbacks.
Most have some sort of "GUI-builder" program (ala Glade for GTK+).
As far as I can tell, most language bindings to each individual toolkit are more or less a literal translation of the API. This makes it seem to me like any programming language would be just about as productive as any other.
Some toolkits bill themselves as not just a GUI toolkit, but an "application framework", for example wxWidgets. They add on APIs for other stuff like networking, data structures, logging, threading, and database access. Considering that most of this other stuff usually has better libraries to access the functionality you need, it seems like it wouldn't be particularly important in deciding between toolkits. In fact, if you know you already have this stuff covered, it would be beneficial to choose a toolkit that is simple and know it is just a GUI toolkit, like GTK+ or FLTK.
Are there GUI libraries out there that are radically different from this mold?
As someone trying to break into GUI programming, how would you suggest to choose between a GUI toolkit - or does it really even matter which one? What programming language tends to be easiest for developing GUI applications - or should I just stick with what I know?
You're writing this question as if you've never used the these toolkits. I'm not really sure what the actual question is here. Have you ever used Swing? Does that seem as the same level of productivity as the .NET WinForms API? Really, saying anything about any of these will just lead to a flamewar or series of downvotes. There are differences. Of course, there are many similarities. A lot of that has to do with many of the fundamental constructs you need in an event driven GUI programming environment, such as a message loop. Of course, there are probably many other ways to do it but it is a proven method.
I don't know of any that "radically" break from this. Probably the most different GUI library for general purpose GUI applications I know of is from REBOL with its VIEW:
http://www.rebol.com/docs/view-guide.html
However, in reality its not "radically" different.
One big difference in productivity is the tools for these different toolkits have very different levels of maturity. You're kind of asking two questions. One question is more theoretical: "Are any of these toolkits fundamentally different?" and the other is about productivity and that is "Are there differing levels of productivity for the various toolkits?" The first question is very debatable. The second question has the clear answer: Yes.
There are many questions to ask. First what programming language or platform will you be using? Do you need portability to different platforms like Windows, Mac OS X, and Linux? Will it be just Windows or just Linux? Obviously, you wouldn't want to use WinForms on Linux (or Gnome/KDE if anyone wants to be pedantic). You could use GTK+ on Windows, but the "widgets" won't have same look and feel as those offered by native Windows toolkits. You could choose Swing, but it is really only available in Java and has its own issues. I don't think Swing has a great level of productivity, personally. I think having to add all these "adapters" and connections all this "layout" gook is not that productive. Some people do though.
Another aspect is the selection of built-in widgets your chosen toolkit holds and what the selection is among third-party free and commerical widgets. If you choose a more obscure toolkit, you won't have as many exotic sort of widgets to choose from. For instance, you would find a Mac OS X style "dock" widget for .NET WinForms and probably a few others but maybe not in GTK+ (actually, it probably exists there, but its just example).
So, in short, I don't think you can really say that all the GUI toolkits are just interchangeable. There are many other issues to consider than whether they have base widgets and message loops.
Related
I have been playing around with php-gtk recently and in the past I have experimented with Java to make GUI 'hello world' apps.
However both these types of applications have had a bit of a clunky (almost childish) look and feel to them. I cannot deny that they are handy for making apps for in-house use (and I totally respect the amount of community effort that goes into these projects). But I would not necessarily be proud to sell it as a commercial application with a price tag of, say, £450 or £1,000.
If I wanted to make an application that had the look and feel of, say, Firefox for Windows, or Adobe xyz, what GUI/language should I use?
Is the 'professional sheen' or smart look and feel 100% down to the designers or is it the case that, no matter how good a designer is, picking the right GUI framework is essential to get that look?
There are a few aspects to having a polished UX for a piece of software.
Using the most native framework for the platform. Win32/WPF for Windows, Cocoa for Mac etc.
Application's visual artefacts are coherent - this includes images, graphics, toolbar icons etc.
Following the platforms guidelines and best practices.
It's overlapping a bit with Igor's reply, but here's my take:
Native Control Look - UI controls today have a rather complex appearance. There are many visual cues we instinctively derive from them, and even if it's a white rectangle with some frame, with the wong shadow it looks strangely out of place. A context menu often doesn't just open today, it slides in from some direction, or fades in.
Native Control Behavior - Even more complex than UI, there's a lot of detail to behavior: different context menus depending on click position, different "hot" areas when selecting or dragging items, keyboard shortcuts, etc.
Attention to detail - There's a lot of consistent UI behavior to discover on any platform. Just the way arrow keys work in a tree control WRT selecting, opening and closing nodes.
Just look at Windows: Most non-native toolkits get the basic keyboard navigation wrong - Arrow keys, Home, End, PgUp and PgDown, behavior modified with Ctrl, extending selection with Shift gives up to 32 behaviors. Copy & Paste is traditionally with Ctrl+C/Ctrl+X/Ctrl+V and Shift+INS,Shift+DEL, and missing. Mouse double click often selects a word, mouse triple click sometimes a sentence, line or paragraph.
Response time and Muscle Memory - There are, basically, two UI operation modes:
act-look loop, where you wait for the response before deciding the next step,
playback from muscle memory, which is much faster and requires less mental processing ressources.
There are, however, two requirements for that: response must be uniform and "instant", and the next action must be registered correctly immediately (at least within 10 ms)
Often enough, with non-native toolkits, this gets hard by the response lagging behind one or two actions (the mind locks on the discrepancy), and by toolkits that take 50ms or more to show a menu, in which time a click isn't registered as intended.
A polished UI takes long to get right - A good control library can solve most of the per-control issues, but there's some final 10% taking 90% of the time, and you have control interactions. You have to try different approaches, you have to expect users with FPS-trained reflexes, you have to try all kinds of workflows.
Cross-Platform toolkits can't get it perfectly right - they are stuck between a rock and a hard place: They can opt for internal consistency independent of the platform, or being consistent with platform they currently run on. To get it right, the latter often requires platform-dependent code in the calling code, the actual thing you are trying to avoid.
Always try to use the GUI framework that's used by your desktop environment. .NET's libraries are probably the best for creating Windows apps. GTK+ is always the best on GNOME, while Qt works well on KDE - even though all three work on each other's systems, their visual appeal decreases with their lack of visual integration.
The GUI API/language used is utterly irrelevant to UI design, although some APIs make it easier or fasterto implement.
Good UI is about:
Good graphic design/artwork (visual balance and symmetry, complementary colours, visually 'pleasing' shapes and layouts, visual consistency within your app and with the other apps around it - consistent positioning, sizes, gaps, colours, etc)
Understanding the user's workflows and making what they want to do easy and intuitive. This often means implementing 3 or 4 ways of achieving the same action (e.g. "Copy and paste" can usually be achieved by: main-menu->copy/paste, context-menu->copy/paste, ctrl+c/v, button:copy/paste, drag-and-drop)
Keeping everything simple. Remove as much as possible to cut the UI back to just what the user needs and no more.
Being intuitive and not surprising the user. Controls should look like the controls the user knows, work like the controls the user knows, and be located in the places a user expects given their previous experience with the computer.
Following the conventions (position OK/Cancel buttons in the standard locations, use the OS-defined colours for highlighting selected objects etc)
To get this, you need to look at lots of "good" applications and dissect what makes them good. Get a good artist/graphic designer to draw you good icons etc. And spend a lot of time thinking about the user's workflows.
Make sure you separate the business logic from the UI - this will allow you to re-skin the app easily to improve the UI. And usually the data the program needs is not related to the way the user needs to use the application - don't be tempted to just expose your x,y,z variables in editable fields! UI is the layer that hides your implementation and makes it usable!
I don't care for the way Java application GUIs tend to look using the usual toolkits. If you like the way Firefox looks, you might look into XUL and the GUI framework shared by most Mozilla applications. Komodo Editor/IDE use the same tools (along with several other applications). GTK is very powerful, and I really doubt it is what is preventing your applications from having that professional sheen. Keep exploring it's features, and rethinking the best way to display your components, and I'm sure you'll stumble upon an arrangement that feels better.
On the other hand, it isn't all about the toolkit. Good interface design is an art, and interfaces like Firefox have evolved through endless amounts of feedback. The best thing to do is talk to users and find out what will make them more comfortable using your application. I've found that software tends to look good when it is also functional.
I would recommend spending a lot of time in software that you find pleasant to use. Make notes of the way things are done, and look for commonalities among interface elements. Most software sticks to a pretty common set of principles that make using the software easier, and the more you explore the software you find appealing, the sooner patterns will start to emerge.
Good luck!
Currently I am preparing exercises about networks and mobile communications for students. I was thinking about creating an interactive user-interface which enables the user to drag&drop predefined elements and then implement a logic based upon element distances etc.
An example would be to place two base stations (a predefined element with several properties), set the scale in the interface and then check the signal interferrence in the environment (user-interface).
The first part might be too abstract whereas the example might be too specific, but I was wondering whether there already exists any friendly framework or language which enables developers to create interactive interfaces (for teaching/learning purpouses) in short ammount of time. Usually I write applications for PC environment in .NET but in this case it would take too much time to create a specific interface for every exercise.
I would appreciate if anyone could suggest any way to create interactive user-interface in short ammount of time. Are there any special programming languages or development tools for this kind of applications or are there any useful frameworks for .NET, Java or any other language to speed up the development of user-interfaces?
Thank you!
Perhaps sketchflow could be of interest
although more intended as a tool for mocking up GUI's. It does also have the ability to hook up behaviours and other short snippets of code to your testgui.
Take a look at Mockingbird. I'm not sure if it's exactly what you are looking for, but it is a pretty sweet web application and works well for the development of UIs (at least the wireframes).
There are lots of tools for this purpose -- but most of them only work well if your user interface follows one of a few fairly well-defined (and widely used) patterns. Just for example, if you're creating a database front-end, there are quite a few tools to make the job quite easy.
One often-overlooked tool for this kind of job is Visio. Visio is normally thought of primarily for drawing and diagramming, but you can also attach behaviors to objects so double-clicking an object can run a macro of your choice, opening forms, doing calculations, etc. The macros are (at least normally) written in VBA, which probably isn't the greatest choice for big projects, but as you said, for a situation like this you're mostly interested in short snippets.
There are, of course, a number of downsides to this, such as the inability (at least AFAIK) to deploy such a diagram as a web-based application. Like Henry Ford's "they can have any color they want as long as it's black", you can respond to any user action you want as long as it's a double-click.
Microsoft expression blend might bee something cool too
http://www.microsoft.com/france/expression/default.aspx
I'm now with a idea to start the development of a bare bones Qt/GTK+-like framework, but I want to know some things before I start the creation of this project:
What is the structure of GTK+ and Qt?
Do I need to develop a window manager to build my own framework?
Some resources to start?
Developing a GUI/Application framework is a significant undertaking. You might want to be very clear about why you need to write yet an other framework.
Both projects you mention are open source. Why not start there?
GTK: git clone git://git.gnome.org/gtk+
Qt: git clone git://gitorious.org/qt/qt.git
Ed You ask what the structure of GTK and Qt are, whether you need to write your own widow manager (answer: no) and how to get started. Answers to at least the first two are in the source code. Don't forget, great practitioners in any field learn by watching others. Reading code is no different.
Writing a GUI/app framework would be a great learning experience, but even a fairly small app framework would be a very big job, and not something you really should tackle until you're fairly expert in writing applications using several other frameworks and widget toolkits.
I did something like this once, back in the early years of this decade. That was after I'd been programming for the Mac for over 15 years, Windows over 10, and had programmed both directly to their native graphics, event, and widget APIs, as well as various object-oriented toolkits for them including PowerPlant, MFC, and MacApp. When I started working on a PalmOS application, I spent a couple of weeks writing a very small app framework modeled on PowerPlant. But I could not have succeeded at all without those decades of broad and deep experience with so many GUI systems.
Doing this for Linux/X11 is even more work. That's because, unlike Mac OS and Windows, neither X11 nor Linux supply built-in user interface widgets, or much in the way of graphics primitives or text layout capabilities. GTK+ is part of the GNOME ecosystem; it provides the widgets, gets its message queue and internal communications from GObject, relies on GDK to abstract and simplify its graphics and event communications with X11, and uses Pango and Cairo for text rendering and layout. I work all through that system, and it probably represents many dozens of person-years of hard work by a lot of really smart people. And I'm sure Qt is very similar.
So if you really want to do this, I would recommend you:
Write programs with a lot of different app and widget toolkits, on multiple operating systems. That will help you learn not just how such systems work, but why they are designed as they are. And it will give you some feeling for what works well, and what works poorly.
Contribute bug fixes or new features to one or more of the various open-source frameworks. GTK+ has a list of tasks for beginners to work on. Another great open-source framework is wxWidgets.
Become an expert-level C/C++ programmer.
When you've done that for a few years, you will have the expertise suitable for tackling your own framework.
That sounds like a major undertaking, at least as a starting project.
Not sure what you mean by "the structure" of e.g. GTK+. You can see the object hierarchy for GTK+, that tells you at least how the implemented objects (GTK+ is an object-oriented API) relate to each other. You can guess how the code can be structured, from that information.
And no, you don't need to write your own window manager; the toolkits mainly concern themselves with what happens inside windows, not with the window management itself. Of course you could decide that your "platform" should have a wider scope, and include a WM.
I think some of the answers here might exaggerate a bit. Obviously making something of the same quality, width and depth as Qt and Gtk is a huge untertaking. But you can make simpler stuff and still learn a lot about how it works. I suggest doing like I did in university. Use OpenGL with Glut. Then you got basic drawing functionality and event system in place already. You then need to create classes for buttons, text fields etc.
If you want to make it really simple then each component just needs to know where it is drawn and have some sort of bounding box where you check whether mouse click are inside or not. You also needs to create some system which makes it possible for buttons, check boxes etc to tell the rest of your code that they were clicked.
This isn't really the rocket science people here make it out to be. Games have made their own very simple GUI toolkits for years. You can try that approach as well. I have modeled a simple GUI tookit on top of a game engine before. Your buttons and textfield could be simply be sprites.
But yeah, if you want to make something that will compete with Gtk+ and Qt, forget about it. That is a team effort over many years.
There are many more technologies and tools available to build the front end for an application.
Which is the best technology/tool/platform available using which I can build a better GUI, by which I'll be able to build a nice looking as well as an efficient GUI?
Definition of "better" includes factors such as efficiency,user friendliness,better content control mechanism, navigation and many more.
I know this is a question about which GUI toolkit you should use, but your first technology for producing a user-friendly UI is pen and paper. Sketch out some mock-ups. Draw buttons and menus on construction paper, cut them out, and glue them together. Then try your mockups on about a half-dozen people. You'll quickly find out what makes a good UI.
It doesn't matter how good the UI looks or whether it uses the latest snazzy effects -- if your users can't figure out how to use it, they'll go elsewhere. You need to learn what works for your target audience before you write a single line of code.
Read Don't Make Me Think to learn how to make mock-ups and do user testing.
If' you're tracking an IP address, you'll definitely want to create your GUI Interface in Visual Basic.
Use the .NET framework in Visual Studio 2005/2008/2010 Studio. I haven't developed in any other environment, but I have been able to create nice looking apps in this IDE / Framework.
"Best" depends on what your evaluation function is.
For ease of development, and high quality UI, in a non-web based app it's hard to beat C#/VB or any other .NET language and environment for a windows-based app. Depending on the quality of the UI, MPF will give you greater flexibility and control, whereas windows form will make it easier to develop.
Having used Windows Presentation Foundation for a while now i would highly recommend it. There is a pretty big learning curve and, to be honest, MSFT should have included some controls (the datagrid being the biggest one) that were not included by default (but will be in .NET 4.0). Where WPF and XAML exceed is providing a foundation from which you can build just about anything. You can style ANY part of ANY control and build your own composite controls from scratch. A lot of thought went into binding and value converters and once you get used to the declarative nature of XAML you wont want to turn back. The company I work for has been using it for a couple of years now and the difference between the GUIs we used to develop (mainly winforms and asp.net) and what we develop now are night and day in terms of both look-and-feel and functionality. My two cents anyway...
It depends.
What device will the GUI be used on, hand held, PC, Mac?
What platform Windows, Linux, Web?
What kind of application will it be, accounting, email client, web application?
What audience will be using the application (a GUI aimed at a child may be different than one aimed at an adult)?
All of these things must be taken into account before even starting to formulate an answer to your question.
You have several choices for developing a GUI.
first, if cross platform is an issue consider using Java or Python.
you can also use Adobe AIR and develop the gui in Flex.
If you direct the product to windows only users .NET WPF is the best solution, with a very rich set of control and examples.
You can also use .NET with mono for cross platform compatibility, but WPF isn't currently supported.
Desktop, Mobile, Windows, Linux, Database, OpenGL: Nokia Qt. Wiht Python - PyQt development process is shortest and easy. Application containing all required python & qt libraries and modules is around 30MB with Inno Setup installation is 8MB and will work on Windows 2000 and newer for Python 2.6.x, Python 2.5.x based application will run from Windows 95 to Windows 7.
I think all the attributes you list -- efficiency, user friendliness, etc -- are attributes of a good design rather than a good toolkit. Just about any toolkit can be used to meet those goals. I think the question might be different if you were asking about eye candy, fancy multi-media, etc. There are definitely some toolkits that do that better than others.
If you're interested in usability first (and it sounds like you are), focus on the design then pick whatever toolkit meets your current abilities and can handle your design. For example, if you require 3D images that might narrow your choices; likewise if you need to show videos, that will influence which toolkits you can choose from.
So, start with a good design. From that, create a list of requires for the toolkit -- rich editing controls, video, 3D, etc. And then look for a toolkit that provides what you need.
The best toolkit in the world won't make up for poor design.
Personally after having used Win32, Forms and WPF then going to Mac/iPhone GUI development, I very much prefer the flexibility and high quality of visuals in the Mac/iPhone GUI.
One of the most useful examples is the fact that in NSTable/UITable controls (ListViews or similar in Windows), every cell is a fully customisable View (a Control in Windows).
Where in a ListView you have very little customisation for each cell/item in the view since you only provide details, not an actual control, an NSTable/UITable asks you for a table cell which you can add anything to, such as buttons, switches and image views.
Mac OS GUI development to me is a LOT more flexible and more consistently flexible in that regard. Everything is a View so I can my own contents to anything.
Have you considered Silverlight?
It can be used to create internet applications, but it can also be run out of browser to create desktop applications. It's has significant overlap with WPF though there are differences which might catch you out when swapping from one to the other.
Expression Blend 3 is a very good visual designer and the code it produces is quite efficient.
Is it worth to try to keep your GUI within the system looks ?
Every major program have their own anyways...
(visual studio, iexplorer, firefox, symantec utilities, adobe ...)
Or just the frame and dialogs should be left in the system look 'n feel range ?
update:
One easy exemple, if you want to add a close button to your tab, usually you make it against your current desktop theme. But if the user has a different theme, your close button is out of place, it doesn't fit the system look anymore.
I played with the uxtheme api, but there is nothing much you can do, and some themes i've seen are incomplete sets.
So to address this issue, the best way i see, is to do like visual studio/firefox/chrome roolup your own tab control with your theme...
I think, that unless your program becomes a very major part of the users life, you should strive to minimize "surprises" and maximimze recognizability (is that even a word?).
So, if you are making something that is used by 1.000 people for 10 minutes a day, go with system looks, and mechanisms.
If, on the other hand, you are making something that 100 people are using for 6 hours a day, I would start exploring what UI improvements and shortcuts I could cram in to make those 6 hours easier to deal with.
Notice however, that UI fixes must not come at the expense of performance. This is almost always the case in the beginning when someone thinks that simply overriding the OnPaint event in .Net will be sufficient.
Before you know it you are once again intercepting NC_PAINT and NC_BACKGROUNDERASE and all those little tricks to make it go as fast as the built-in controls.
I tend to agree with others here- especially Soraz and Smaci.
One thing I'll add, though. If you do feel that the OS L&F is too constraining, and you have good grounds for going beyond it, I'd strive to follow the priciple of "Pacing and leading" (which I'm borrowing here from an NLP context).
The idea is that you still want to capitalise as much as possible on your intended audidences familiarity with the host OS (there will be rare exceptions to this, as Smaci has already covered). So you use as much as possible of the "standard" controls and behaviours (this is the "pacing") - but extend it where necessary in ways that still "fit in" as much as possible (leading).
You've already mentioned some good examples of this principle at work - Visual Studio, even Office to some extend (Office is "special" as new UI styles that cut their teeth here often find their way back into future OS versions - or de-facto standards).
I'm bringing this up to contrast the type of apps that just "do it their way" - usually because they've been ported from another platform, or have been written to be cross-platform in GUI as well as core. Java apps often fall into this category, but they're not the only ones. It's not as bad as it used to be, but even today most pro audio apps have mongrel UIs, showing their lineage as they have been ported from one platform to another through the years. While there might be good business reasons for these examples, it remains that their UIs tend to suck and going this route should be avoided if in any way possible!
The overriding principle is still to follow the path of least surprise, and take account of your user's familiarity with the OS, and ratio of their time using your app to others on the OS.
Yes, if only because it enables the OS to use any accessability features that are built in like text-to-speech. There is nothing more annoying for someone who needs accessability features to have yet another UI that breaks all the tools they are used to.
I'd say it depends on the users, the application and the platform. The interface should be intuitive to the users, which is only the same as following system UI standards if they are appropriate for those users. For example, in the past I have been involved in developing hand held systems for dairy and bread delivery on Windows CE hand helds. The users in this case typically were not computer literate, and had a weak educational backround. The user interface focussed on ease of use through simple language and was modelled on a pre-existing paper form system. It made no attempt to follow the Windows look and feel as this would not have been appropriate.
Currently, I develop very graphical software for a user group that is typically 3rd level educated and very computer literate. The expectation here is that the software will adhere to and extend the Windows look and feel.
Software should be easy and intuitive where possible, and how to achieve this is entirely context dependent.
I'd like to reply with another question (Not really Stackoverflow protocol, but I think that, in this case, it's justified)
The question is 'Is it worth breaking the OS look and feel?'
In other words,
Do you have justification for doing so? (In order to present data in some way that's not possible within normal L&F)
What do you gain from doing so? (Improvinging usability?)
What do you lose from doing so? (Intuitiveness & familiarity?)
Don't simply do it 'To be different'
It depends on how wide you would define system look'n feel... But in general, you should keep it.
Do not surprise the user with differentiating from what he is used to. That's one of the reasons why we call him user ;-)
Firefox and Adobe products usually don't because they are targeting several plattforms which all have their own L&F. But Visual Studio keeps the typical Windows L&F. And, as long as you are developing only for Windows, so should you.
Apart from the fact that there is no well-defined look-n-feel on Windows, you should always try to follow the host platform native L&F. Note however that look-n-feel is just as much about how a program behaves as how it looks. Programs which behave in a counter-intuitive way is just as annoying as programs sporting their own ugly widgets.
Fraps is a good example (IMHO) of a program which is actually very useful, but breaks several user interface guidelines and looks really ugly.
If you're developing for Apple's Mac OS X or Microsoft Windows, the vendors supply interface guidelines which should be followed for any application to be "native".
See Are there any standards to follow in determining where to place menu items? for more information.
If you are on (or develop for) a Mac, then definitely YES!
And this should be true for Windows also.
In general, yes. But there's the occassional program that does well despite being not formatted for all the OSes it runs on. For example, emacs runs pretty much contrary to every interface guideline on OS X or Windows (and probably even gnome/KDE) and it's not going away any time soon.
I strongly recommend making your application look native.
A common mistake that developers who are porting an application to a new platform seem to make is that the new application should look-and-feel like it does on the old platform.
No, the new application should look-and-feel like all the other application that the user is used to on the new platform.
Otherwise, you get abominations like iTunes on Windows. The same UI design may be exactly right on one platform and very wrong on the next.
You will find that your users may not be able to pin-point why they dislike your application, but they just feel it hard to use.
Yes, there are valid exceptions, but they are rare (and sure enough, they tend to be the major applications like Office and Firefox, rather than the little ones). If you are unsure enough to have to ask on StackOverflow, your application isn't one of them.