Slashdot Mirror
Jakob Nielsen Talks About Usability in FOSS
dokey writes "In an interview with Builder AU, usability expert Jakob Nielsen gives his opinion of usability in Free and open source software. The article echoed what Jon "Maddog" Hall said earlier this year in a keynote at Linux.conf.au -- "Programmers Are From Mars, Users/Managers/Companies are from Venus". Is it time to pay more attention to end-users?(who aren't geeks)"
The attitude of "It works, don't care if you don't like how it works or if you think it's ugly, I like it, if you don't like it than don't use it" is not just in FOSS, it's the attitude of many, if not most, programmers. Despite what it may look like, this isn't flamebait, I'm one of these guys myself. At the company I work for, this attitude is prevalent to a degree in most of the developers. It takes someone outside their heads (and usually, pressure from someone who makes the decisions) to put a friendly face on the application, and, dare I think it, reduce or refactor functionality to present a better interface to the user.
It's not that developers aren't to blame, but rather, it's how you'd expect developers to be. What FOSS needs is a free, open-source equivalent of the QA/Validation/UI Design department.
> The reason is, the motivation for open source is
> not because the person gets paid but the person
> gets prestige. The developers are designing for
> each other and they are so feature rich--geeks
> love features--and you get more prestige by adding
> features. For the average person fewer features is
> better and easier to understand.
This has been a constant battle on POPFile. People are forever asking me for this option, or that option, which are useful to a user community consisting of themselves and the two other people in the world who want the same thing. I've been argued with strenously for not adding various features and in general to innovating in the UI really slowly, but the lesson is clear: the average user should be guided by the software to the right behaviour. POPFile does have 100s of special options and they are available in a cfg file that a geek can get at.
The other problem with open source and GUIs are all the people who want things in very specific places. e.g. I got constant "Put button X at the top, no, put button X at the bottom, no put it at the top and bottom" type conversations. Finally, we've boiled the UI down to the things that most people like and anyone else can hack the HTML templates and make the UI just as they want it.
Overall, we've settled on:
1. Lots of flexibility exposed at the geek level
2. The every day functionality exposed in the UI.
There's still a lot to do to make POPFile's UI really friendly, but the biggest lesson has been to resist the power users when it comes to adding UI widgets.
John.
I think the state of OSS GUIs is better than he claims. A lot of work with regard to usability has gone into the major (!) projects like Gnome or KDE. That still does leave us with quite a few crappy OSS GUIs, but it doesn't really make sense to try to come up with some average value in this case.
A study on this could be interesting.
From TFA:
/. still dispute that view.
========
The second problem is that open source when they turn to the general tools they tend to be in the line of "let's implement what we already know" so they will take Microsoft Office and they will clone it. Since we've been criticising Microsoft for years for cloning Apple it is only fair to criticise open source for cloning Microsoft. The point being that you don't move ahead but you have to do something new.
========
Very nicely said - he's not the first or the last to say this, but I am puzzled how many in the OSS community and on
Of course, that is not only obvious but potentially dangerous from the legal perspective.
If/when OSS software gets close to endangering some big commercial software, I think this cloning thing will be the first the ISV will present to the court.
BTW, the Pope said something against cloning yesterday - was he complaining about OSS?
Just last night my wife and I where talking (read "arguing") about this.
To give a little background, we're going to be expanding our suso.org web hosting business to start offering Linux Support and prebuilt Linux systems in Bloomington. We'll be opening our office early next month.
She is worried that most people's expectations for functionality will be too high and will just end up throwing their machines away. However, we understand that Linux isn't for everybody yet and that we will have to turn somepeople away for the time being because they will have too high of expectations out of it. For instance, say that a family comes by and wants to by a new family PC so that the kids can play games and the parents can do their taxes, word processing and look at all those multimedia attachments that their family sends them. We'll we'll have to be up front with them and let them know that Linux probably isn't for them because those kids aren't going to be able to play off the shelf games yet (without fooling around with wine a lot or vmware) and they would get frustrated when some of those cute attachments don't work. Or TurboTax doesn't work.
That all said, I feel that OSS has come far enough that it is ready for the first batch of non-technical adopters. You know, the ones that like to tinker around with the latest technology.
I prefer a geek switch, or rather, a series of geek switches that turn the software from being a usable piece of software into a controllable piece of software. You must understand some people will NEVER want to switch into geek mode. I know my parents would never sacrifice usability for features. By the same token some geeks may know the author of the software well and be able to guess how to control the software and may never use an easier mode. Many games have this feature (RTS) so I'd really like to see it in programs.
The one mistake many programmers make it requiring people to accept things they do not want. Perhaps I want control X but not control Y. Many programs would force you to take both or neither. This must always be circumvent-able without resorting to geekhood.
I label thee "stereotypical FOSS person"
Previewing sounds in Konqueror is a feature, not usability
Having all those features enabled by default in OO.o, cluttering up the place, is not usable.
What's wrong with his Website? Looks nice to me - simple layout, clear differentiation between dynamic and static data, search box well-placed, headline stands out, and no seizures from looking at it.
kde-look... man, that's confusing. Too many choices, some instances of dark text on a dark background (thank heavens for my new monitor), and it's designed around a minimum resolution. Yuck. Not the worst I've seen, but far from good.
No, just keep writing software like you always have, with astoundingly complicated UIs, impossibly obscure configuration options (when it's not just a damn text file buried under /etc or ~/) and completely non-standard behavior that throws people who've used other GUIs off to hell.
Yeah, the software is "cool" because it has that neat bayesian algorithm that was harrrd to implement in Malbolge and it's "free", so that must make it better. Anyone who complains can either a) Go to hell b) Write their own version; c) Submit a patch; d) Ask for their money back; or b) STFU.
Keep copying Apple and Microsoft and everyone else instead of coming up with your own UI designs (badly, too), while snickering at said companies on Slashdot and IRC.
That's fine. Just don't yell at me when I question your claims that your app is "ready for the desktop" and is "better" than what the "evil proprietary" companies can come up with.
Ok, read my followup post. The guys a monkey.
Asthetics is an important part of usability, and can be worked out techinically with no need for 'graphics design'
If something looks tactile then the user is going to try an use it.
If something looks 'gareish' then the user is going to avoid looking at it.
If something is large, then it's going to draw the users attention.
If something flashes (like a TV) then the users brain is going to think that it is a possible danger (a panther just about to pounce) and the user will look at it.
Calming sympathetic colours or voilent clashing colours set off moods in the brain, this will then make the user tend toward one form of operation.
If you make good green and red bad then people tend to pick more extream values than if there was no colouring. (try it).
etc.... Colour is just as much a part of usability as making sure the letters read left to right for english readers.
thank God the internet isn't a human right.
It just occured to me why I dislike the "spatial" nature of the new Gnome.
#1) It is just like DOS. You can only be 'active' in one directory at a time. Deep dir's were hell, but shallow ones were kinda quick and easy to copy/move files around and open them.
#2) The entire "browser" style has lasted only because Suzie Soccermom has never learned that by using a "Tree-view + detailed list" is easier, (damn Windows default settings).
#3) IIRC Xtree, Norton Commander and Dosshell were all designed to quickly & easily allow of folder/file manipulation at the deep level. This is a huge improvment over the existing DOS CLI.
Is this really just a case of: "What is old is new again"?
Like a previous poster stated. Spatial systems would work very good for large numbers of files, if the OS did all the sorting for you. (Didn't MS try this with "My Documents, My Pictures, My P0rn..." and we all hated them dearly for it?)
Here's a novel idea, let's make the choices EASILY switchable. Include 999999999 different choices and let the end-user decide.
Damnit that's the mess that we have with any linux distro installer.
Ok, how about the Model-T Ford method: "You can have any colour you want as long as it's black."
Apple beat us to that one too.
ok I give up. Maybe I just go "roll my own". And no I am not talking about linny.
=) I love Mondays
"The price good men pay for indifference to public affairs is to be ruled by evil men." ~Plato (427-347 BC)
Some of the comments here seem blithe about the prospect of a "geek only" OS, but the beauty of the linux design and the OSS philosophy is that the entire system can be made completely user-friendly and never be 'dumbed down', no one has to lose. The problem comes in creating a mechanism for bringing naive users ( or non-technical users, or whatever you want to call these targeted users) into the interface design process. In short, how do we get open-source users as well as open source programmers? And how do they talk to each other. I am certain that there are ways of achieving this, but they will probably require business/programmer partnerships, rather than inspired risk-taking grandmothers, which is a shame since the latter probably have better ideas.
I have nothing to hide. So, why are you spying on me?
Over the past couple of years, lots of people have written about the subject of poor usablility in open source software. Lot's of people, some of them quite well known and well respected in the FOSS community keep saying it, but nobody seems to be listening.
The fact that people keep writing about poor usability would seem to indicate that nothing is changing. It would appear that FOSS pgrammers aren't taking it seriously and not making much effort to change things.
Why not?
Your whole livelihood revolves around solving other people's problems.
Which is exactly what he gave an architecture to do and you didn't.
Usability is one problem, features or functionality are another. The solutions to these two separate problems should not be coupled, so that each can be changed easily without affecting the other and so that both converge as rapidly as possible to what works best for the customer. This is little more than a restatement of tried and true "separation of concerns" object oriented programming mindset.
The fact that he names the different software roles as "programmer" and "designer" does not mean that one person cannot play both roles if necessary. On the other hand, NOT using this architecture does mean that if two people are available who each have specialized skill (one in usability and the other in business functionality), then you cannot take advantage of it. The coupling forces someone with both skills to do all the work thereby forcing quality to the least common denominator level.
The parent post should be modded Insightful, not funny.
The fact is, many, many OSS projects spend lots of efford into usability. But sadly, Slashdotters keep nailing the same thing from 2 years ago over and over!
Let's take a look:
- The GNOME usability project. Anybody can discuss usability issues, or request a UI review of their app.
- Ximian, Sun and RedHat have professional usability experts. Sun did professional usability researches. But ironically and sadly, Slashdotters keep bashing them for creating bad UIs and do not recognize them as usability experts... even though they are professionals, and it's the Slashdotters who have been whining about the need for professional usability experts in the first place.
In other words: Slashdotters kept coming with the same old crap over and over again, no matter what happens. It goes like this:
1. Slashdotters: "OSS interfaces suck! Programmers must be banned from UI design!"
2. RedHat/WhateverCompany hires usability expert and revamp UIs.
3. One of the following things could now happen:
A) Slashdotters: "OSS interfaces suck! Programmers must be banned from UI design!" (completely ignoring that usability experts revamped the interfaces of many programs).
-or-
B) Slashdotters: "OMG those UIs are designed by an idiot! Waaah!" (completely ignoring the fact that they are designed by a usability expert).
Bah, I'll get modded down for stating this since it goes against the common "all OSS apps suck at usability" Slashdot mentality, but heck, this is the truth. It's sad that Slashdotters don't see this and keep ignoring the huge amount of time that has been put into usability.
In my experience, even with a good templating system you're unable to let the designer do the interface 100%. Of course designers can make fairly sophisticated sketches, but I'm always astonished by the amount of logic that's needed to create a user interface.
In my experience web application interfaces in particular are dynamic, not static; they often change and adapt dynamically to the context, user credentials, etc.
So I think a user interface will always be the product of both designers and programmers, and as such there is no sharp separation between logic and design.
"But linux is already there for me (mandrake 10)"
In a default install of Mandrake 10, KMail now displays the images in spam (I prefer just the source-code so I can delete it quicker), it puts a big red header at the top of each HTML message filling the preview pane with something unrelated to the message, it's got a crap default font, it has icons next to each menu item making it more difficult to read, it might have slightly-transparent menus making it difficult to read, the default icons are so bland you can't see what's what (they're all bluish-white circles), I'm not sure it's possible to insist on always-non-HTML outgoing mail, XChat is now ugly, MandrakeUpdate doesn't work, rpmdrake does work, but displays error messages even when it succeeds, and it displays dialog-boxes when you start it "I'm about to run rpmdrake, okay?" - "yes, dammit, that's why I typed rpmdrake, duh!". 3-hour install time, advertising during the installer and an end-user license agreement. Didn't configure X properly, and X crashed with "no screens found" first time I ran it. Mouse-wheel doesn't work, extra buttons on the mouse don't seem to be setup to do anything, WindowMaker has a crap default configuration with useless mandrake menus and an ugly config. Text-mode terminals have a distracting white star in the corner of each screen, no matter what you're doing. Default X terminal seems to be Konsole, the slowest program ever, takes about 20 seconds to load on my machine for a console. Xterm doesn't seem to have been installed by default, neither was kppp (the internet dialler, would be nice), neither was kedit. No I didn't use "individual package selections", I've wasted too many hours in there before. New KDE theme is as horrible as the fonts, and good luck finding how to change it from if you're not in KDE at the time (it's 'kcontrol', I should note to myself, and 'gnome-control-center'). There doesn't seem to be any way of turning off the "send this picture" menu in mozilla which is so annoying to hit by accident, nor the "close other tabs" menu, cause of many a day's lost work. Oh, and KMail must have sprouted a new menu or something, because I always seem to have trouble finding anything in what used to be a nice clear layout.
Did I mention that I use KMail so many hours per day that making it even slightly harder to use is a big kick in the teeth from the people working on default configuration at Mandrake? Or that I use rxvt frequently enough that disabling the WindowMaker icon to open a terminal after you've used it once is frustrating enough to make you want to install gentoo, even if it takes you a month of downloading?
And that's just the stuff which has changed since the "quite nice but crashes with a USB key" Mandrake 9 (apart from the EULA which was always offensive). Seriously, I use Mandrake all the time, it's the best OS I've found so far, but after installing Mandrake 10, I've spent nearly every waking hour looking at pictures of Apple iMacs, and counting the cash required to buy one.
Does this comment even get 10 minutes before disappearing beneath the waves of moderation?
Well, his points are well known, and you can just buy a book where they're spelled out. So you don't _have_ to pay 10,000$ to get that good advice. You don't even need to make an 8-ball, you could just as well make a small bulleted list and put it next to your monitor.
It just requires some minimal clue and an open mind.
The problem is that a lot of web sites are just some PHB's ego trip site. The more clueless the PHB, the more he's convinced that the site:
- _has_ to be in his favourite colours, the users be damned if they don't like it. (Light orange or oragnge-ish yellow, and cyan on neon blue are actual colour schemes that the client's PHB demanded.)
- and some underadable font, while they're at it. (Surely everyone will have their screen set in 800x600, like that PHB does.)
- _has_ to have some convoluted navigation that noone understands. (Another actual example: a client actually wanted the site to have what looked like a heap of papercuts. You literally had to find the link in that heap.)
- _has_ to have the info divided to reflect the feudal fiefdoms inside the company, instead of by relevance to the user. (E.g., no way we'll let you get the hardware and the software information in the same place. Or even have links between them. It's different divisions!)
- _has_ to cut down costs by the most damaging means possible. (E.g., true story: _everyone_ I've worked for starts with ideas like "oh, we'll leave the search engine for next year. We'll start with the important parts first.")
- _has_ to had 1 MB of pictures, animations and roll-overs per page. Preferably Flash. Or flashing.
Etc, etc, etc.
Sure, they could get some clue without a $10,000 consultant. Sometimes their own programmers and designers tried giving them that clue already. But they won't.
Because it's an ego trip. The mere thought that His Royal PHB Highness could have been wrong is a blasphemy. The site must first crash and burn before he even considers getting any advice.
And even at $10,000, I don't envy him. As I've said, those sites are someone's ego trip. That someone already got that advice from internal sources, and chose to ignore it. Quite often why they bring the consultant is as a last hope to prove that their retarded design was right after all.
And then you have to explain to them that, no, their design does suck after all. (In more diplomatic terms, of course.)
And listen to them spewing bullshit in defense of their retarded idea, basically boiling down to "I know better than you, because I'm the boss here." (Another true story, quoted loosely from memory: "1 MB of graphics is nothing these days. Everyone has broadband. We need _more_ graphics, not less.")
It's not a fun job.
A polar bear is a cartesian bear after a coordinate transform.