Slashdot Mirror
Instrumented GIMP To Identify Usability Flaws
Mike writes "New users of the GIMP often become frustrated at the application's unwieldy user interface. Now Prof. Michael Terry and a group of researchers at the University of Waterloo have created ingimp, a modified version of the GIMP that collects real-time usability data in order to help the GIMP developers find and fix its usability problems. Terry recently gave a lecture about ingimp and the data it collects. During each session, ingimp records events such as document creation, window manipulation, and tool use. A log of these events is sent to the ingimp server for analysis. The project hopes to answer questions such as 'What is the typical monitor resolution of a GIMP user?' and 'Is the GIMP used primarily for photo editing or drawing?'"
I like the idea, but will the folks who use ingimp be at all representative of the user population at large? ... Especially of the user population that would complain about accessibility / usability. Is it worth it or is anyone talking about making such a thing an integral part of any project?
take it easy, but take it.
At the beginning it is hard - just like many programs. But my experience is, that you get used to it pretty fast.
Or they could just call it GIMPshop....
So, if I invent a version that gives data on why the name sucks (the otehr main problem with the program), will the developers pay attention to that too?
"As God is my witness, I thought turkeys could fly." A. Carlson
I don't remember ever having a problem figuring out GIMP. But it would drive me insane if they start changing things around on me.
I already see one potential problem with this approach, and that is that it collects usability statistics from ingimp users, not GIMP users. How would it be guaranteed that the two groups are statistically equivalent?
(No, I have not RTFA yet.)
There are far too many anomalies of usability, lack of features and intricacies required for Gimp. Today, Photoshop is the industry leader, and anyone doing serious editing is using it. To be successful, Gimp must surpass it in more than one way (the one way being free). Kind of like what Firefox did to IE. Unfortunately, Gimp is no where ready for that. And I get a feeling that it is (heading to) nowhere.
... may be even Blender. The problem is that no one wants to be in the middle. Utilities need to rise to the top, or they face the fate of XMMS. I hope there will be a replacement in GTK too, just to show Gimp how to use the toolkit :)
I have been using Gimp for a long time. When I first installed Linux it was the only program everyone used to talk about. KDE's kolourpaint was not yet there for general purpose paint-brush replacement. I have used it for years under the hood of open-source fanboyism. And I think that is the reason why it has suffered. It had no competition, and now it is just a software which you don't want to open, again.
Now, I know it is not a paint-brush replacement. But it is neither a Photoshop replacement... and the middle land is already full of other utilities. Inkscape, Krita,
PS: posted this on journal before... this is shameless re-posting.
is in getting others to use the program because of its name. Lets have a contest to rename the GIMP.
You can put lipstick on a pig but in the end, you still have a pig.
With cameras and microphones and other things:
----
"Our performance traces indicate large amounts of cussing when images are resized."
---
"Wow. During that file open, three hundred users gave the finger to the camera."
"And that one guy --"
"I don't want to talk about that guy. Wahwahwahwahwah I-can't-hear-yoooo. Don't remind me of what he did."
---
"Nine hundred instances of users hitting the computer with a hammer while cropping. At least, that what we think the accelerometers were saying."
--
"The rapid rise in temperature was probably caused by the users pouring gasoline on the system and lighting a match. We'll try to address that issue in the next release."
Any sufficiently advanced technology is insufficiently documented.
For a while, I actually believed the folks that repeated ad nauseum the mantra that GIMP's user interface was difficult compared to their beloved Photoshop. Then one day I sat down to try to do some quick photo edits on a Photoshop box. Two hours later, I gave up on its bizarre layer model and just installed the GIMP so I could get some work done.
The ease-of-use of a graphic user interface, in general, correlates far more with the user's pre-existing familiarity with the interface than it does with any design decisions of the interface itself. There are certainly areas where GIMP's user interface could be improved, but let's not pretend like it's some kind of embarrassment -- because it's not.
If they want to know how GIMP is typically used, that's easy. GIMP's typical and most popular use is for people to say, "Hey, you can edit your photos under Linux with GIMP, and you don't have to use Wine and Photoshop."
But professionals using GIMP for doing real work? That's atypical. Hopefully that will change.
You know, how do you recognize a project is someone's pet project? It's an overcomplicated solution for a problem with a trivial solution.
Want to find out what makes the GIMP ui suck? Ask the damn users! They won't exactly shy away from telling you.
I'm a Photoshop user and I have GIMP installed here to use the occasional esoteric plugin functionality. Let me tell you few things you can immediately get busy fixing:
1. for some reason GIMP developers decided every single thing needs its own window and its own menu bar. It's weird as f*ck: put the entire layout in a single window with integrated panel layout (similar to how Eclipse does it, for example).
2. each plugin is its own modeless exe dialog that takes arbitrary amount to start after it was called (at which time you can modify the processed image.. sometimes, and sometimes GIMP crashes because of it): create a proper lean plugin API and modal plugin dialog.
3. the menus and options are all over the place: there seems to be no strategy at all about what goes where
4. GIMP has really bad startup time, and performance, compared to commercial graphics editors (such as Photoshop)
5. There's no way at all to organize your layers in a more complex setup: there are no layer groups, layer folder, or anything like that. It's just a big sack of flat layers, that you can select one at a time, and link them together. This is Photoshop 4 level functionality, and most graphics editors are waaaay past that by now.
6. There are no proper drawing tools in Gimp at all. For a graphics package that claims to be targeted at geeks making icons and web devs making web designs, this is ridiculous. We're forced to fake our ways with selection tools and scripts, which covers only a fraction of what we need.
7. A personal issue I have with Gimp: no proper grid. I use the grid in Photoshop all the time, set on unobtrusive "pixel" mode, and usually at 8, 16, 32 pixels with subdivisors. In Gimp, no subdivisors, no pixel mode, and for some reason the *mere fast of displaying* the grid, makes everything slow down to a crawl.
The project hopes to answer questions such as 'What is the typical monitor resolution of a GIMP user?' and 'Is the GIMP used primarily for photo editing or drawing?'"
Looks to me like they're about to fall into the fallacy that caused Daimler-Chrysler to do a redesign of the Jeep line that killed their market.
The marketing department looked at what fraction of SUVs were actually used off-road. They came to the conclusion that it was small. So they redesigned their line to be more comfortable on-road at great cost to its off-road performance.
Turns out that a significant fraction of their market was people who NEEDED the off-road capability - and had the resources to pay for it, reliably buying cars, year after year, through all economic cycles.
Jeep stopped being the car they needed and became another clone of the rest of the market: "Mall Terrain Vehicles" that LOOK like an off-road car but are really just a funny-looking small/high van that qualifies as a "truck" to escape the fleet mileage regulations. Their guaranteed market went elsewhere and they were in head-to-head competition with a slew of vehicles over which they had no advantage.
Similarly, Coke looked at all the people buying Pepsi, saw that they were younger and that Pepsi's main difference was that it was sweeter, and replaced Coke with New Coke, which was sweeter yet. Result: People who drank Coke because they liked a less-sweet drink switched to Pepsi.
And then there was the high-ranking officer in WW II who spent months counting all the bullet holes on the returning bombers, then did a big presentation on how those areas should have armor added. At the end of his presentation a lower-ranking officer asked "Shouldn't we, instead, add more armor to those areas that are only lightly holed? After all, this sample represents only the planes that came back."
= = =
I think the same thing could happen here: Paying attention to what people do a lot of just focuses on what you're already doing right - at the cost of ignoring the things that people do occasionally, or only some people do, but which they need to have. Further, the things they do rarely may be used rarely specifically BECAUSE they're hard to use and the interface needs improvement.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
That is an awful mistake for F/OSS fanboys. "Oh, it's free, so we shouldn't complain". This is like being blind to the problem. If it's free and it works, why isn't EVERYBODY using it? (In other words, why is Mozilla Firefox MUCH MORE popular than the GIMP? Think about it).
Sometimes we can forget that graphical applications are meant to be used by designers who use most of their time retouching photographs and stuff. Here, time is money. And if the lack of usability in the GIMP makes me spend 5 times more the time than I would with Photoshop (and i'm being considerate), it's just not worth switching. To put it another way, Photoshop's user interface _IS_ worth the price. I still can't believe the GIMP guys CANNOT make something as user friendly (or don't want to, which is worse). It shocks me and frustrates me.
A quote from a designer's blog:
Ooooh... what a bold statement! The GIMP is *NOT* user-focused. Don't tell me.
See, professionals don't want just "a better pile of poo" to do their imaging work. They (and I, too) want something that IS EASY TO HANDLE. Because in graphical applications, form is function. And this is something that many programmers (at least many of those that I've discussed with) simply fail to understand.
Is GIMP still being developed? This is a serious question.
I've been a big GIMP fan for years. Years ago, I was excited about the 2.0 release of GIMP. It brought many new features and the UI got a serious revamp. But now it has been several years, and it seems that GIMP development has slowed down. They're still releasing newer versions with bug fixes, but no new features. For example: I recently bought a Wacom tablet, and while GIMP has Wacom support, I miss some of the things that Photoshop has, such as support for variable brush width based on tablet pressure. The long-awaited GEGL, which was introduced years ago and will supposedly add CMYK and 16-bit support, is still not ready, and to my knowledge is still pre-alpha. (Not that I need CMYK and 16-bit, but at least that silence all the complainers.)
A year or two ago I also read an article about someone wanting to sponsor GIMP development. But that effort went nowhere, as his request was eventually ignored.
What is going on? Is GIMP still being actively developed? Are the GIMP developers still interested in adding new features?
Yes, there seriously is. The big areas where I love Photoshop and hate Gimp revolve around layer properties, layer blending, transparent backgrounds, and grouping and copying layers.
From the presentation slides, it seems like 200 people have installed it (netting "over 100,000 commands" in the log files). Obviously more will do so in response to the Slashdot article (and appropriate web pollination)... but aren't these self-selected geeks already? How are you going to get non-geeks to install this instead of the regular GIMP (assuming you convinc them to take a look at it)?
Furthermore, how does this help determine what GIMP isn't doing properly? I mean, if you have various tools at your disposal, and GIMP sucks at doing X, then you might do half your work in GIMP and the other half in another app. So all the usability problems around X won't show up in the logs -- almost a kind of self-denial.
I use Photoshop on a nearly daily basis. Last time I tried GIMP it was not ready for professional print design, to be sure, and only probably good enough for desktop publishing or Web graphics. How about Pantone or CMYK support? Non-destructive layer effects? Variable-sized brushes? Actually useful text formatting?
$nice = $webHosting + $domainNames + $sslCerts
That said, there are some basic problems that surpass the complexity.
However, most of this pales to the limitations that are inherent in the functionality. One of my biggest gripes is that the anti-aliasing code is sloppy in non-uniformly implemented. Try this: select a circle, and then use Edit->Stroke Selection. Select a 2 pixel stroke line and go. You will get absolutely HORRID aliasing. The same thing happens (though not quite as bad) with the paint tool stroking.
Overall, the GIMP is an amazing and powerful tool, but it has some serious warts.
Then there was the time IBM instrumented a mainframe to determine what instructions were heavily used so they could focus their optimize-the-microcode effort on them.
They found one particular instruction that accounted for some exceedingly large fraction of the execution time. So they went to work on the microcode and doubled its speed. Then they deployed the new microcode and measured the application performance, expecting to see a big improvement.
It didn't change a bit.
After a little more research they discovered they'd optimized the idle task's wait loop.
= = = =
Collecting data can be useful. But making good decisions based on it requires wisdom and insight.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
The biggest problem I have with GIMP is it's interface. It's clear the application was designed by programmers and not designers. I feel like they've tried to cram too much onto the screen and they suffer from a similar problem I was with Microsoft applications. They try to offer too many ways to do things and get too technical with details. I don't need 10 different sliders for customizing a brush. If I want a custom brush I should be able to just create the graphic as I would anything else then just drag it into a custom brush box and be done with it.
Photoshop is getting progressively more bloated but I still find it more fluid than anything else. I'm not constantly hindered by the application.
The solution isn't to do more coding. The data they gather may result in solutions that only complicate the issue. What they should do is sit down with a small team of designers. Include people with experience in photo-editing, website layouts and interface design. Ideally, find people that have little to no experience with GIMP. Work with this team to develop an interface. And most importantly, keep things simple.
Inevitably, most applications end up being overly complex because of some overwhelming desire to cram in every last feature the developer can dream up. There also seems to be little planning. Build a set of guidelines and adhere to them. And one last thing, be sure that all essential function can be activated via the keyboard. When I'm doing time-consuming production I don't want to hunt around for small icons, or be constantly switching between the mouse and keyboard.
Maybe there's a really good reason that Photoshop does things the way it does. It's designed for professionals, and hasn't changed a whole lot in at least a decade, so I gotta figure that for professionals, Photoshop is made to work the way most graphics professionals need it to work. The layer thing IS complicated, but from what little I know about graphic design, all of that stuff is very useful (necessary?).
I don't respond to AC's.
Because those people as usual either don't need a complex graphics applications nd therefore won't learn how to use it or have used unlicensed copies of PhotoShop for years and hate Gimp because it's interface isn't the same ("waaah, it's not the same menu, I have to use the second button of the mouse, the horror")...
The last time I used PhotoShop was on a MacII, before porting it to Windows was even considered. I've been using Gimp since the first release and never found its interface to be unusable. The Gimp is well documented and fairly easy to learn. I currently use it for my photo correction and retouching needs and it works just fine.
Like a poster mentioned above, I'm fairly sure I never could do as much with PhotoShop without spending a significant amount of time learning a new complicated piece of software. Which I'll never do because PS costs around 1000 €s in Europe and because it doesn't run in Linux anyway (which is what I happen to do my work in).
So Gimp and Krita suit me just fine (along with BibblePro and digiKam).
May contain traces of nut.
Made from the freshest electrons.
I watched the video, and the only thing that stuck in my mind, is that I think you're not qualified to study usability if you have to Alt-Tab through a bunch of firefox instances because you haven't discovered tabs yet.
"Follow me" the wise man said, but he walked behind.
1. for some reason GIMP developers decided every single thing needs its own window and its own menu bar. It's weird as f*ck: put the entire layout in a single window with integrated panel layout (similar to how Eclipse does it, for example).
2. each plugin is its own modeless exe dialog that takes arbitrary amount to start after it was called (at which time you can modify the processed image.. sometimes, and sometimes GIMP crashes because of it): create a proper lean plugin API and modal plugin dialog.
3. the menus and options are all over the place: there seems to be no strategy at all about what goes where
I find these issues to make GIMP nearly unusable. I'm always fighting with it. How do you get around these problems?
Ok, some questions, then. What if an interface offered approaches more consistent than Photoshop's or GIMP's? For instance, in the main toolbox, both Photoshop and GIMP mix area selection tools (eg, rect, ellipse) with tools that actually modify the image using areas (eg, smudge, fill.) Because of this, the only way to know what a tool in the tool box will do is to become familiar with it; the location (in the toolbox) doesn't define the type of functionality. What if the area selection tools were in one toolbox, and tools that modified the image were able to be placed in another - perhaps just the ones you think you'll need today? In terms of usability, this type of approach associates physical location with function; this *should*, theoretically, enhance usability.
The same thing applies to layers. Photoshop's interface treats layers like they were not images, rather, as if they were only components of images. But essentially, they are images, as demonstrated by the ability to select one and edit it as if it was the image. What if a four-layer image allowed you to see, and edit, all four layers at once, just as if they were normal images, while changes to the sum of all the layers, let's call that the "master" image, are visible in yet another window? Wouldn't that be more consistent than treating a layer as if it were something other than an image? It provides direct, and simultaneous, access to everything at once (many layers begin to bring window management into the equation, but those skills are even more basic than anything inside an image editor.)
Before you answer, I would like to point out to you how many complaints that couch themselves as usability complaints refer to an application not working "like" Photoshop, and how often the phrase "industry standard" is brought up; it seems to me that when complaints of this type are voiced, they refer to learning the person has already done, and they want *compliance*, because they already have (a set of) muscle memory that they work with. They actually don't want better or easier, because better and easier is different, and different will impede their progress while they learn (if they are even willing to learn!)
I make decisions about these precise things as part of my job; I'd be very interested in specific opinions from anyone on these issues.
I've fallen off your lawn, and I can't get up.
If I have nothing to hide, you have no reason to search me
("waaah, it's not the same menu, I have to use the second button of the mouse, the horror")...
My main complaint about the Gimp UI is the multiple windows. It's cliche, but it's still true.
2.0 helped a great deal because they group things together so there are less, but it's still a pain. Now instead of raising just one window (Photoshop) I have to raise three (the photo, the toolkit, and the navigator/layers/other stuff panel). On Linux this isn't horrible because I can dedicate a virtual desktop to it, but (1) on Windows it's a downright pain in the ass and (2) I shouldn't have to dedicate a virtual desktop to a program to make it usable!
I mostly agree, but I find item #1 to be misguided: " put the entire layout in a single window with integrated panel layout"
8 153&pid=19817169#19818683
If you want better window management, use a better window manager. Putting the window management features into GIMP would actually cripple the program for many of us. Photoshop's MDI is a great way to work around the limitations of the window manager in MS Windows. But it's still a kludge. A better window manager is a far better solution, and there are plenty of good solutions already available.
More details are in another comment:
http://developers.slashdot.org/comments.pl?sid=24
Lack of user feedback is not the problem with the gimp. Users have being telling the Gimp programmers for years what's wrong with their UI.
The problem is that gimp programmers ignore all critism of their UI and likewise they will ignore this ridiculously complicated solution to gather user feedback.
They (KDE) tried, but the gnome fundamentalists didn't let them. This was in the middle of the gnome/kde wars, when Qt wasn't pure enough for some.
You can still find references on the web:
http://dot.kde.org/1096230607/1096270511/
I always think, what if they did it ? It would be better than gimp for sure, seeing the quality of other kde apps at the time.
A better url:
http://lwn.net/Articles/101709/
Open any book about usability. On the first pages you'd come across the axiom to avoid creating a menu structure which is more than two levels deep. Gimp's menu structure on the other hand is deeply nested. There are so many basic usability problems which could be easily corrected without using any special software.
..at least for those of us using Windows XP or (shudder) Vista:
Paint.NET is getting better and better (and has an active user community creating plugins, etc). I tried it about 2 years ago and wasn't all that impressed, but as of my latest inspection, it's pretty useful software. Just make sure to check out the forums for effects and tutorials.
"Murphy was an optimist" - O'Toole's commentary on Murphy's Law