I can’t cook, but I know a bad meal when I eat one. It’s also in the chef’s and restaurant’s best interest to listen if I bother to complain.

But if it was in the Ubuntu restaurant, you say I need a Michelin star or be Gordon Ramsay before I can complain to the chef?

No offence but I have entirely lost faith in the open source movement when it comes to usability. There have been grand claims about how usability is being put at the forefront for the better part of a decade yet the only thing I have ever witnessed happen is the changing of theme and the addition of gimmicks with no effort to fix the inefficient layout and usability problems with Gnome, not that you can actually really do anything anyway as the brunt of the issues lie upstream.

I’ve said it before ad nauseum but your problem (and Ubuntu’s) is a total failure to listen to your users. Is there any effort to identify why people choose not to use Ubuntu? Is there any effort where people can report problems they have (that isn’t a bug tracker) so you can collate the top N issues? Has there been any honest efforts to solicit opinion or engage the community?

Essentially ‘deeds not words’ and to be perfectly frank the FOSS community has used up all its words and is still blaming the near flat growth curve on everything from hardware manufacturers, OEM’s, Microsoft’s evil hand and inertia yet never stops to consider that the reason people aren’t using it may have something to do with needs and requirements.

Us: “Hey, we here don’t use Ubuntu as it doesn’t fit our needs, let us tell you why!”
You: “Shut up, you haven’t proved yourself worthy to give feedback.”

Fantastic!

Joe Monco: I love you too.

Kerberos: No, you’re not credible yet, because you don’t have a body of work to point to (or if you do, you’re keeping it quiet). And no, it’s not correct that “You want UX information from non experts as they are your target market, but the only way to give UX feedback is to be a technical expert willing to find and breach all the barriers.” Useful UX feedback comes from user tests and ethnographic research, which is what we do.

I’ve already had my period of criticizing software developers “publicly and tactlessly”. It lasted from 1999 to 2001 or so. It was great fun, and occasionally I relapse, but for the most part I’ve moved on to more effective tactics. Best of luck with yours.

I can sit here and declare that the problems are still there, that I’ve been waiting a decade for them to be fixed and that I don’t care what they say about project X that is going to claim to fix them all – I am simply not buying it. Nothing bad will happen to me except some hate mail by fanboys. I have my opinion and I am sticking to it and I like to think I have evidence on my side.

MPT on the other hand is now paid by Canonical, he works on what he’s told to work on and I have no doubt is making a positive impact, but as the saying goes it’s just rearranging deckchairs on the Titanic. He is in no position to say to Mark (who appears to be a fairly strong character) “You’re doing it all wrong” as no doubt this would probably jeopardise his job.

The changes which I feel are necessary* involve a significant investment in terms of time, effort, manpower and a reassessment of priorities. I think Mark is failing in what is required to fix Bug #1 and I can say this as publicly and tactlessly as I like without worry, an option that MPT no longer has.

* Do I feel I am right enough to risk my job over them? Do I stand behind them to that degree? Who knows? I certainly wouldn’t say that I would.

It’s not that there are problems separating M,V and C. (There are — they’re called human beings.) It is, after all, an academic paper (although Terrence is bloody good at getting academic stuff into the mainstream, as per ANTLR).

The interesting thing is how disparate the MVC achievements of various “industry standard” frameworks are in meeting their supposed goal. This shouldn’t be a surprise; you only have to look at ASP or vanilla PHP to realise that a lot of websites fail miserably at separation of concerns. And why not? They work. For a while. In a way.

It would be a particularly useful paper for architects and managers choosing a templating language, because it does at least offer a set of criteria to judge between them. Unsurprisingly enough, I’ve never met an architect or a manager who’s even heard of the bloody thing.

It’s not that there are problems separating M,V and C. (There are — they’re called human beings.) It is, after all, an academic paper (although Terrence is bloody good at getting academic stuff into the mainstream, as per ANTLR).

The interesting thing is how disparate the MVC achievements of various “industry standard” frameworks are in meeting their supposed goal. This shouldn’t be a surprise; you only have to look at ASP or vanilla PHP to realise that a lot of websites fail miserably at separation of concerns. And why not? They work. For a while. In a way.

It would be a particularly useful paper for architects and managers choosing a templating language, because it does at least offer a set of criteria to judge between them. Unsurprisingly enough, I’ve never met an architect or a manager who’s even heard of the bloody thing.

It’s true. It’s also fairly surprising and confusing that he would then actively promote a honeypot to draw away user feedback so it can all be more efficiently ignored:

“That’s why Brainstorm acts mainly as a honeypot drawing noise away from the bug tracker. And that’s why it’s gradually getting more difficult to report a bug about Ubuntu. We *need* to erect those barriers, so that we have time left in the day to improve the software.”

The problem with the current ‘usability experts’ is they think it’s all about user testing, but that is putting the cart way before the horse. You need a plan, you need to implement that plan then you need to be happy with your implementation, *then* you start user testing. User testing half-baked rubbish which is unplanned and cobbled together rubbish is pointless as generally user testing will only tell you (very roughly) what’s wrong, not how to fix it. If your ideas suck, or your implementation sucks, then no amount of user testing will help.

The bug report mechanism is supposed to be there to provide details in regards to unexpected behaviors observed in the software. It is simply not a place to hold the “next big thing” for designers or to serve as an IT version version of Hyde Park for anyone with two fists and keyboard. #386196, #530751 and #475773 are simply not bug reports – they are high-level ideas that designers are supposed to figure out on their own. In fact, it is supposed to be their job to come up with all the high-level ideas – not the blasting users’. Outside the insane asylum, the users only come into the picture when they are approached with specific questions in regards to what they expect from the software. Damn volunteers. Damn Brainstorm. If you are finding yourself deluged with random thoughts blurted out by the users, then, most certainly, you have already failed to conduct proper requirements elicitations.

Then again, I wonder if anyone inside Canonical actually understands how Ubuntu is supposed to work in the first place.

If any of you haven’t read MPT’s Why FOSS Projects Have Poor Usability or whatever, read it, because it affirms every point Kerberos’s points quite well.

That’s a shame. Did he get the memo about whom he was actually working for, or was he just too busy being a hypocrite?

If any of you haven’t read MPT’s Why FOSS Projects Have Poor Usability or whatever, read it, because it affirms every point Kerberos’s points quite well. Of course, the fact that even if you know what to avoid and how to go about doing things the “right” way results in utter failure, well I think that’s an interesting lesson and likely indicative of inefficient leadership or a really out of control project.

I don’t envy any company that is trying to put on the open face, only to realize all the tards are mucking shit up. But then, that’s why I would never try to build a commercial company by trying to turn fosstards into consumers.

In my view, if you want to leverage “open” and make it profitable, you need to either go after the business customer who is willing to pay for support and who likes or wants some of the benefits of the stack (see Red Hat, SugarCRM) or you’ve got to build off of the open but not pander to the overall community and instead target paying customers from the word go. Like Apple or Google. No, Apple isn’t open in the traditional sense but WebKit is a pretty well run project. WebKit doesn’t fuck around. Plus, Apple and Google will hire the community members that do good work. Hence how the two guys who started Camino (the native OS X build of Gecko that existed before Safari and way before Firefox finally went Cocoa) now work for Apple and Google respectively (ironically now both on WebKit, Pinkerton headed up the Chrome for Mac team).

But if you start off catering to the unwashed masses of foss users who will never pay you a dime anyway, you totally set yourself up for a shitty situation in trying to make something look professional.