The cult of design dictatorship (2012)
alexcabal.com
alexcabal.com
The problem is not the cult of design dictatorship, the problem is bad design and bad designers.
Apple and 37signals are two examples of design oriented thinking (and I don't mean visual design), but there are many others, even from within the world of Free & Open Source Software.
After all, why else would a term like "Benevolent Dictator For Life" exist if it weren't for design dictatorships in programming language development?
The real claim that this post is making is that "You are not Steve Jobs and you are not 37 Signals". And... well that may be true, but it also may not. And if you can't have frank discussions about the utility of the things you make, and whether or not you have evidence to back up why/how you are doing the right thing, then yeah, you may be a bad designer.
That doesn't mean that being a designer or a dictator is a bad thing inherently, and arguing against central organizing authority in creative works is highly problematic, especially in the absence of any concrete alternatives to suggest.
As an aside, I wouldn't describe either Steve Jobs or the 37signals as "nice". Smart, pretty determined, resolutely sure of themselves, but nice is not the first adjective that springs to mind.
The question is, are you going to be sadomasochist follower or an individual (which ironically is precisely what a person with a healthy ego is).
Have good opinions. There is much to improvise in this world, nothing is final ... every class is open.
The Gnome 3 discussion from last year sparked this post because Gnome 3 was becoming an example of that: designers creating a product that many vocally disliked, and insisting on sticking to their original vision, torpedoes be damned. The result, many argue, is a deeply flawed product.
The thrust of the post is, "be humble and open-minded as a product leader, because chances are you're probably not the genius that these exceptional success stories were/are."
DHH is (and I use this term unironically) a visionary. And on top of that he had a keen enough political sense (and arrogance) to be able to deflect or mute criticism of Rails's early flaws.
Ruby is a slow, memory hog? Doesn't matter, developer time is worth more than machine time. Buy bigger machines.
Rails concurrency model sucks? Doesn't matter, fire up more processes!
These were real problems, which have been subsequently addressed in Rails (by and large, by other people in the community who cared about those subjects). But what was important about what DHH did was define a vision and aesthetic for what web development should look like (and please note i'm no DHH fanboy. I jumped ship during the Merb/Datamapper split), and kept on pushing on his priorities even in the face of legitimate criticisms of things he thought were less important.
So, Rails succeeded because DHH has sensible enough taste in terms of prioritizing concerns, and a strong enough user base to fix Rails's shortcomings.
I'm not as familiar with the specific battles over Gnome (partially because i'm on OSX, and also because i find the interlocutors in the Gnome discussion to be so furiously inarticulate), but the conflict really only comes down to two possibilities. Either the leaders of Gnome genuinely are shit designers and aren't meeting the needs of their community, or they're really terrible at politics.
That's a really important distinction to be mindful of. I'm not amongst their target userbase, so frankly I can't say one way or the other, but all of the complaints I've heard against Gnome's leadership are things that I as a user have never ever cared about (granted I haven't used Gnome in a while, but all the times i've used systems w/ Gnome installed by default in the past, i've been satisfied).
And really, if Gnome's leadership is so inadequate, i'm puzzled why a critical mass hasn't risen up and forked the community. That's really the ultimate vote of dissatisfaction.
I think it comes down to two other factors: community and hype.
Technologically, it wasn't anything special, and still isn't. Many of us who'd been doing web development for years at that time had either used or created similar or better frameworks in languages like Perl, Tcl, Python and even Java. These frameworks were usually kept internal to the organization that developed them, however, so they were obviously nowhere near as widely used.
DHH was of minimal importance, too. Yes, he was somewhat of a visible figure head and spokesman for the project, but that was about it. For the average Ruby on Rails user, DHH didn't have much of an impact.
In my opinion, Ruby on Rails brought together several distinct groups of young men (women are still very rare in the Ruby on Rails community) who'd typically been outcasts within the computing industry. They included:
1) Less-talented UI, web and graphic designers. These people, unable to find work in more traditional software development, print media, and other fields, ended up moving toward web development, where the bar to entry was set much lower.
2) Less-talented software developers. These people, either due to age, a lack of experience, a lack of education, or a lack of natural ability, were inherently drawn to Rails. It provided the rigid structure ("convention over configuration") that they needed in order to get anything done. It also allowed them to continue to avoid learning SQL and proper database design techniques, while creating something that partially worked (even if the result lacked severely in terms of performance and reliability).
3) Attention-seeking youth. We all know who these people are. They're the ones who repeatedly wrote loud, profanity-ridden "articles" full of anger. Or they created absurd, cryptic writings and art, and then spontaneously vanished, creating much unnecessary drama. Many of them were also self-styled "hipsters", who just went out of their way to be different merely for the sake of being different.
Ruby on Rails provided something these people could all rally around. It gave them a common cause, if you will. And they rallied around this cause quite loudly, which generated an immense amount of hype relative to what they were able to accomplish, or what their software provided. This helped draw in more and more of these outcasts, making the community larger and larger.
I think that GNOME 3, for instance, is a result of spillover from this newly-formed community into existing, established open source communities. Members of a community formed solely around a lack of merit forced their way en masse into what was once a near-total meritocracy, and as would be expected, disaster was the result.
Do you think bad designers will be more or less drawn to a culture where they don't need to defend their decisions, than one where what they say will be discussed and validated?
Bad designers seem like a constant to me, so there's no point wishing they didn't exist. Instead, design culture ought to respond such that good design can still happen in the presence of bad designers. That can't happen in a dictatorship.
There are a lot of places where having a centralized ultimate authority is very very very important, but in which individual autonomy is still preserved and let flourish. Newspapers are a good example. There's an editor who's in charge and they're in charge of the tone, direction and disposition of the sorts of things the paper is going to cover. They may even write for their own paper, but the sum of the paper is always going to be more than just what they've directly put in.
Nevertheless, it's really important that the editor sets priorities for the newsroom, and manages what everyone is aiming at.
Are there bad newspaper editors? Hell yes. Are there strategies to circumvent bad editors? Yes again. Is it sometimes easier just to find another job than deal with an editors bullshit? Yep again http://theadvocate.com/home/5922742-125/pulitzer-winners-lea...
Where software tends to differ from things like a newspaper hierarchy (or something like the police department from the Wire), is that it is okay to have discussions of efficacy and criticize a project's leadership (which tends to happen ALL THE FREAKING TIME).
And this is where it's relevant to discuss the way that designers prioritize and deflect criticism of their priorities (as i mentioned in reply to acabal: https://news.ycombinator.com/item?id=5767301 ).
Successful designers try to figure out what their usebase actually needs and in an priority order that makes sense to deliver them in. People complained bitterly that the iPhone didn't even have copy & paste which is a feature that didn't show up until iOS3. Didn't seem to hurt the iPhone's sales, though.
In that vein, i'm still entirely unsure as to whether Gnome is doing the wrong thing (particularly because i haven't seen anyone actually articulate what's wrong w/ Gnome3 in a meaningful way, and i have no idea who Gnome3's current actual userbase is, and who their intended userbase is).
What's the alternative? Well, there's a couple. Someone should articulate a reasonable case to the Gnome guys as to why their vision is wrong. If that doesn't work, fork the project and let the userbase decide on where the project should go.
If you decide you don't like centralized authority in creative endeavors, i guess you can try to invent some new method of project direction (or heck borrow Debian's i guess), but all of these endeavors have their tradeoffs. Debian moves slow, community forking is divisive, and trying to articulate why you are right and the Gnome guys are wrong is potentially futile.
-------------------------------------------------
I guess the tl;dr is this. Design and (organizational) politics are interlinked, but they're not the same thing. You can be good at design and bad at politics. You can be good at politics and bad at design. In the Gnome debate, it is unclear to me which side is good at either of these qualities.
Which is precisely what the "cult of design dictatorship" describes.
Your correction amounts to saying "the problem is not crime, it's criminals" -- it's obvious and it's already conveyed in the original phrasing.
>That doesn't mean that being a designer or a dictator is a bad thing inherently
There's two ways to see this.
The pedantic ("is it true 100% percent and for all persons?") which I find useless, and the pragmatic ("is it true for most people that try to play Jobs/37 Signals that they are not fit for the role?").
The second, I believe holds. And that's why the idea of "being a dictator designer" is an inherently bad thing. Because it gives the majority of people the wrong idea about what they should do.
I don't care if the same thing works for a tiny minority. We should not perpetuate memes that only work for a tiny minority.
If that minority has the talent and stubbornness of Jobs/DHH, they will do it by themselves anyway.
The memes ("be a dictator and force your opinion to your team/users") only serve the mediocre, those that do harm.
No, that is distinctly not what I am saying. My assertion is that centralized authority is important, and perhaps even necessary in creative endeavors.
Furthermore complaints about "bad" designers tend to actually amount to "i disagree with your design vision/priorities/choices, therefore you are bad". But disagreeing with someone's design choices does not mean that they are being a bad project leader.
Edit: having read over my original post, it's a bit muddled so i can see where you're coming from. If you read some of the other sibling posts, you'll get a clearer sense of what i mean.
To that, I agree.
I don't have a problem with centralize authority (one guy being in charge etc, like Linus or Guido or Jobs).
I have a problem when the person being in charge does it with some hazy "You'll listen to me because because I know good design" thing -- Jobs, DHH etc proved their worth in practice, they didn't merely claim they have the knowledge.
Whereas today you see people convinced they are the next Jobs/DHH that are just shallow imitators with no talent, and they try to force this at their teams / companies / communities.
There are great designers who have a fantastic intuition for aesthetics and horrible intuition for anything else.
They are a problem because they seem to be solving problems (because they make things look good) but are really just making things pretty.
The problem is interfaces that look good statically, that are flashy but are bad UI's for actually doing things.
Dictatorship of design is good shorthand for this.
That's it. We're looking at a rolling freight train of disastrous desktop designs over the last ten years. A somewhat cohesive, simple explanation would be good. This sounds like it. OK?
I also cringe when I see "You are not Steve Jobs" etc. indeed I'm not nor would I want to be, I disliked many things about the man intensely when he was alive and that hasn't changed one iota since his death.
It fascinates me how we continue to set the bar of leadership based on a man who judged by his actions was a borderline sociopath, I guess success by whatever measure truly does forgive all sins.
http://www.businessinsider.com/steve-jobs-jerk-2011-10?op=1
The only thing that amazes me about his career is that he didn't get punched in the face more often.
I would disagree with this. He was certainly an incredibly successful business man but there are plenty of others that have matched him: Richard Branson, Bill Gates, Henry Ford, Sam Walton, Jack Welch etc.
Take (for example one of my favorite) industrial designers - Raymond Loewy - This is a man who created the Shell and BP Logos, designed the Scenicruiser greyhound bus (iconic), coca cola vending machines, the GG1 (in my opinion one of the most beautiful trains of it's generation and they ran for just shy of 50 years) and the PRR S1 (which I think is the single most beautiful train I've ever seen) and just to finish it off he designed the livery for Air Force One.
That's kind of a pedantic distinction, isn't it?
Well, in business issues he IS a "once-in-a-century-genius".
He bootstrapped 4 companies. Apple (which he built from nothing), NeXT (which he build for a pittance and sold for 400 million), Pixar, and Apple v2 (which he built up from near bankruptcy to total dominance).
Even just taking a company from near bankruptcy to the highest ever profits and cash held in the history of enterprise in a decade, would be enough. But he also changed (and owned) 3-4 market segments (mp3 player, music downloads, app stores and tablets. And having the profit of all other top 5 PC makers combined for it's computer lineup I'd count as "owning" that market too.
>I also cringe when I see "You are not Steve Jobs" etc. indeed I'm not nor would I want to be
Noticed how the statement was not targeted at you, but to people who DO believe they are like Steve Jobs and that mimic him?
For that crowd, the reminder "you are not Steve Jobs" is totally appropriate.
>It fascinates me how we continue to set the bar of leadership based on a man who judged by his actions was a borderline sociopath, I guess success by whatever measure truly does forgive all sins.
And it fascinates me how you don't see that leadership is all about success.
Would you have a cuddly, lovable by all, Army General in World War II that lost the war, or a cursing, bad-tempered sonofabitch that won it?
Plus, what other "sins"? Shouting at some engineers? Closing favorable deals? Insisting he gets components in time? Losing his temper at meetings? Yes, because not 100% of us do the same things and worse at times...
It's not like he had employees beheaded, tortured children in his basement or paid prostitutes to piss on him or something. He yelled at some people, fired others. IMHO, the worst thing he did was taking advantage of Woz. And that was like 35 years ago.
(I also find the BI article you linked to extremely grasping at straws. He once left the hotel he had booked a room in because he didn't like the room"? Sure, borderline psychopath behavior. And for a while he didn't want to acknowledge his daughter and pay child support, but he later made up. Yeah, that's Jeffrey Dahmer material...).
It seems odd to copy one without the other.
Let me tell the obvious: Gnome 3 doesn't suck because it focus on design, it sucks because it has horrible design process. If it's not fulfilling user requirements, that's bad design by definition.
The problem with Gnome is management. Last time I tried improving the font selector (which I think still is utterly broken for selecting weights), nobody cared. They though the only weights people need are "bold" and "italic". Now compare to the font selector on Mac and say Apple is "design dictatorship" with no regards to user requirements...
According to the author, gnome3 and unity are flawed. Well, many people like unity now. It's not that different from the other operating systems. And sure, there'll be people who won't like it too.
But if these are examples of design dictators screwing up things, what about the interfaces that existed before unity? They had a ton of issues too. And while I may not have had issues with them, unity is certainly easier for the non-technical crowd. And the rest of us know how it needs to be tweaked to our liking.
Nothing's wrong really.
Yes, there are a small number of people who like the changes that have taken place. But their numbers are indeed quite small compared to the much larger number of people who have been driven away completely by these changes.
I don't think that you appreciate how many GNOME 2 users (including former developers and other contributors) are either still using GNOME 2, or have moved on to KDE, Xfce, or other non-GNOME desktops.
The same goes for Ubuntu. With Unity and their other recent changes, many users have instead moved to Linux Mint, Debian, and other distributions.
These are the worst users to lose, especially the contributors, because they're the ones who did do things correctly, leading to the initial success of the project. Once they start to leave, we end up with GNOME 3-style debacles where it's one bad design or decision after another.
There's probably no good evidence available about how many people like Gnome 3 or Unity. For abandonment rates, have a look at the graph on popcon.ubuntu.com. Those numbers come with caveats, like the fact that not every installation sends data, but none of the lines have any discernible downturn.
> This cult is insidious. Its two main tenant are: 1. The designer is always right. 2. If you don’t like what the designer is doing, you’re wrong, and you should go somewhere else. Doesn’t sound very friendly, does it?
1. Couple of fallacies here: you've set up a scenario in which your conclusion is supported (gosh, that doesn't sound friendly! This guy's a genius!), but it's also ignoratio elenchi: it doesn't fucking matter whether it's friendly or not, because who gives a shit whether the philosophy by which you design a product is friendly or not? It's like asking whether the philosophy is crunchy or gooey.
2. I would argue that if the designer is doing their job properly and working with and for users, then they will be usually right, and if you don't like it, you're wrong, and you should go somewhere else (because you're probably a neckbearded engineer trying to design something with zero user empathy).
> Steve Jobs made a zillion bucks cramming his design decisions down peoples’ throats.
1. In the same way as any designer, living or dead, who has shipped something to consumers, was "cramming [their] design decisions down peoples' throats."
2. In addition to the hugely biased language used, it's a gross oversimplification of design at Apple. A good example of Steve Jobs designing something is the iDVD anecdote. The iDVD team spend weeks working on a user interface that they think works. Jobs comes into the meeting, stops them halfway through, ignores their complicated workflows, and draws a simple rectangle which has a single "BURN" button on it. His great skill wasn't design, but editing and empathy.
> and now one of its founders spends his days custom-building and racing F1 cars.
Just another casual misrepresentation. DHH has not retired and is still working hard at 37signals.
> They did all this by being design dictators.
Yes. Forget the brilliant engineering, marketing, thought leadership, branding, etc. It was this cult thing you've conjured out of nowhere.
> Steve Jobs had a vision, and if you didn’t like his vision, you could go home.
Yes. But a vision is nothing to do with design. Example: Steve Jobs had the vision for MobileMe/iCloud. The vision was a cloud-based software product that allowed you to synchronise your devices and keep data across all of them. The design is terrible. Design being architecture and the implementation. Vision != design.
And what do you mean by "go home"? Isn't the same true of any product? If you don't like Android you can "go home". If you don't like Ferrari you can "go home". With "go home" you're implying that the consumer loses out. In reality because there IS ONLY ONE WAY A PRODUCT CAN POSSIBLY WORK, you're criticising them for not disrupting the space time continuum in order to offer two different products so you can not like Steve Jobs' vision and still like Steve Jobs' vision. Fuck me.
> 37 Signals made its products like it wanted to, and if you didn’t like it, you could suck it.
WHAT THE FUCK ARE YOU TALKING ABOUT? I'm giving up on the rest.
That's not editing, that's literally throwing everything away and forcing his vision on them. Precisely what I'm talking about. And "empathy" is not a word I'd use to describe Steve Jobs, who by all accounts was a terrible jerk.
> Design being architecture and the implementation. Vision != design.
Maybe a better title for this post would have been, "The cult of product leader dictatorship." When I wrote design, I didn't mean industrial design or UI design, but product design and leadership. My fault for being unclear.
> WHAT THE FUCK ARE YOU TALKING ABOUT?
On reading this post again 6 months after I wrote it, those are indeed strong words that don't really reflect 37 Signal's attitude. My words, not theirs.
Throwing stuff away is editing. Sigh.
> and forcing his vision on them. Precisely what I'm talking about.
1. His vision was "a simple experience for the user". iDVD didn't end up with a simple rectangle and a single button marked "Burn". He refocused his teams around simplicity and ease of use.
2. I note that you've avoided responding to the other parts of my response in which I inquire as to whether you believe all designers in history are "forcing" their "vision" on others.
> And "empathy" is not a word I'd use to describe Steve Jobs, who by all accounts was a terrible jerk.
Product design is all about empathy with users. Steve Jobs was blessed with that in abundance. If you attribute Apple's success to his vision and design, then you have to acknowledge that users love their hardware and software. Jobs was great at cutting through bullshit and getting himself, his team, his engineers and designers -- who are all highly technical 'power users' -- to be humble and remove themselves from the equation and build something for mass consumption.
Uh this is working really well...
Let's see, Unity, Gnome 3, Windows 8... there is a pattern where static design has prevailed over the coherent, overall functional design of products. This is especially painful in the desktop PC market because people use them to do thing, even ordinary people have a use for a keyboard OMG.
The thing about overall functional design is that here you might talk to the engineers a bit. You might windup with a slightly less slick product but one that does what the users want it to - assuming you have users who to do more.
Canonical claim that they user test the Unity interface. The test tasks appear to be geared to 'end users', although there is a nod to 'power users'[1]. This focus on testing and iteration would seem to be at variance with the idea of a 'design dictator'. Does anyone have hard evidence of user testing of the Gnome 3 desktop?
[1] http://design.canonical.com/topic/user-experience/ [list of blog posts showing range of activity]
So wide off the mark, that you hit a guy on a wheelchair 2 blocks away.
It DOES fucking matter if "the philosophy by which you design a product is friendly or not" when you're trying to build that product in a community (as is the case with Gnome 3). He specifically mentions Gnome 3 as his example, and in the dynamics of Open Source communities friendly DOES matter. A lot.
Perhaps you could spend "fainting" (sic) and more time in reading and understanding where TFA comes from.
>2. I would argue that if the designer is doing their job properly and working with and for users, then they will be usually right, and if you don't like it, you're wrong, and you should go somewhere else (because you're probably a neckbearded engineer trying to design something with zero user empathy).
And I'd argue that he describes cases where the designers are NOT doing their job properly.
It's not like most designers working in web/software startups are trained professionals, profficient in the history of design, psychological and cognitive issues, HCI principles and user testing methodologies. A lot are just glorified graphic designers, if that. Especially the ones TFA article describes, and which we've all met, are the "I know better" BS types.
Also your whole argument is based on the premise of: "if the designer is doing their job properly and working with and for users". Well, TFA describes a case where the users are not happy with the designers work -- a case where a project (namely Gnome 3) lost momentum and users.
>Just another casual misrepresentation. DHH has not retired and is still working hard at 37signals.
Just another pedantic correction of a comment made in jest that was not supposed to be taken literally. Not to mention he doesn't say he's retired at all.
>>37 Signals made its products like it wanted to, and if you didn’t like it, you could suck it. >WHAT THE FUCK ARE YOU TALKING ABOUT? I'm giving up on the rest.
He talks of the same "FUCK" (sic) that 37 Signals themselves admit (and are proud of). That they make opinionated products, and that if somebody isn't pleased with them, so be it.
Ah the whine of the open source developer who wonders why the average joe finds OSS unusable and impregnable. It doesn't matter if there's a great product at the end of it. If the product is suboptimal then we can all moan about the reasons why.
Also worth pointing out that you're misrepresenting his article. He says:
> It’s a problem that doesn’t just apply to Gnome 3—though Gnome 3 certainly suffers from it. It’s the problem of the Cult of Design Dictatorship.
So either you don't understand his article, or he's applying a generalisation drawn from Gnome3 and then deliberately applying it to the wider world. Which?
> Perhaps you could spend "fainting" (sic)
I'm obliged to point out, with a sense of delicious irony, that you are using sic incorrectly ;)
> and more time in reading and understanding where TFA comes from.
I read it and understood it. It's horseshit.
> It's not like most designers working in web/software startups are trained professionals,
What? Yes, they are.
> profficient in the history of design, psychological and cognitive issues, HCI principles and user testing methodologies. A lot are just glorified graphic designers, if that.
A lot, or most?
> Especially the ones TFA article describes, and which we've all met, are the "I know better" BS types.
I guess there's an inherent risk with drawing _massive, sweeping generalisations_ from "I've met some of these".
> Also your whole argument is based on the premise of: "if the designer is doing their job properly and working with and for users".
Actually my argument is better summarised as:
1. If a designer is doing their job properly, then the characteristics described in the original article all hold true, but the outcome is successful so nobody cares. 2. The point laboured in the article is, essentially, that design is not a democracy and you should "go home" if you don't like a product. I counter that this is a platitude: it could be true of any product ever made.
> Just another pedantic correction of a comment made in jest that was not supposed to be taken literally. Not to mention he doesn't say he's retired at all.
"spends his days". It's an unfair characterisation made in support of a tawdry argument which boils down to a bunch of plattitudes
> He talks of the same "FUCK" (sic) that 37 Signals themselves admit (and are proud of). That they make opinionated products, and that if somebody isn't pleased with them, so be it.
My point was: But so what? Every single piece of software is made with opinion behind it. If you don't like it you can choose another piece of software. That's true of everything from bottled water to delivery pizza. Who gives a shit? It's not worth pointing out.
No, the whine of the open source developer who knows why the average joe finds the OSS project he talks about unusable and impregnable: the BS decisions of "designer dictators" that alienate the users.
>So either you don't understand his article, or he's applying a generalisation drawn from Gnome3 and then deliberately applying it to the wider world. Which?
Or, you don't understand the obvious: that things that hold for Gnome 3 can also hold for a wider world -- and that some of them (like a friendly decision process) can be especially important for OSS projects.
Reading comprehension problems?
> Perhaps you could spend "fainting" (sic) >I'm obliged to point out, with a sense of delicious irony, that you are using sic incorrectly ;)
Wrong again. I only forgot the words "less time" between spend and fainting. As for "sic" I used it as it is meant to be used. It has no hard dependency on the tense or form of the word used -- ("feel faint" vs "fainting"). It just means that those were your words (even if adjusted to fit the quotation's syntax).
>>It's not like most designers working in web/software startups are trained professionals, >What? Yes, they are.
Oh, I'm afraid they are not. You can count the number of them that have a university degree in HCI in one hand.
>I guess there's an inherent risk with drawing _massive, sweeping generalisations_ from "I've met some of these"
Yes, it's called deductive reasoning. You should try it. It's not like there's an infinite number of startups either -- so it can also be exhaustive or near exhaustive sampling.
>1. If a designer is doing their job properly, then the characteristics described in the original article all hold true, but the outcome is successful so nobody cares. 2. The point laboured in the article is, essentially, that design is not a democracy and you should "go home" if you don't like a product. I counter that this is a platitude: it could be true of any product ever made.
OK, so your argument is wrong. (1) is a statistical noise (a few highly talented designer arseholes) so for all practical cases the bad end result the article laments holds true.
>My point was: But so what? Every single piece of software is made with opinion behind it.
Which is a useless platitude in itself. The important thing is the degree of opinion in each software. Others are "taking or leave it and fuck you", and others listen to their team/community/users.
>That's true of everything from bottled water to delivery pizza.
Not really. There are "designer" pizza places that only sell you a specific kind and fuck off if you don't like it. And there are places that let you design your own pizza, sandwich, coffee, whatever. And places in between.
And after the nearly sensible point, my imagination wished Jobs had decided to become an Evil Super Villain, and set up a volcano base, with a moon based super weapon, preferably a "LASER"....
"Strong opinions held loosely." That's the key, not being a virtuosic genius; no one has infallible vision.
A lot of people think they have good ideas and want to moonlight as designers. But when the data comes in evaluating their ideas, too often those people have moved on to other things. That's because they're idea guys - when their idea fails, they lose interest in the problem. Real designers stick with it, learn, iterate and find better solutions.
> The best software has a vision.
...now just make sure the vision is not too narrow and that it's a vision and not an edict (now about Linux DEs, the Gnome 3 and Unity teams OP is referring to just keep turning a narrow copy-cat vision into a bunch of edicts... while the KDE vision tends to be so freaking all-encompassing that you get lost in it and so full of corner use-cases for bugs to hide that no developers can keep up with the bug hunt... sigh, and thanks god for xfce)
And then there is this huge community of interface-conservatives, they are against change per se...
A good designer understands that design is a discussion and that there will always be pragmatic compromises to be made.
However, having the implementer make the important design decisions about a user interface is nearly always the wrong thing to do. The implementor knows too much about the implementation to see it the problem like a user would. It is this which causes UI's that look like database tables when they should resemble faces and buckets and controls that mould to engines rather than fingers.
It is not the implementors' fault. They are just to close to the metal and too far from the users.
"Terrorists in the murder of design: Why you're wrong and failure is beneficial."
It's sure to encourage a good discussion.
At the same time, no one is going to argue with yr Hypermedia hubris or API spec. =P
Or namespacings. Or syntax preferences. Or hacks.
Look, no One Person (or role) wins (or loses -- and none of this "dictatorship" sensationalism is really needful, methinks). It's all sausage factory at the end of the day. Just try to create enough black boxes before lunch so you have some dignity at dinner and can sleep after the midnight snack.
A CEO gently reminded me one day, "No one lives or dies by this." It's frustrating, not a "dictatorship." In just the same way that no one "killed the coffee"; and no your computer did not just "die." This sensationalist writing makes it difficult for us chill developers who just want to make a simple critique without having to be pigeon-holed with all the hyperbolic-complaint-machine because all of our critiques have the same content (this modal or that button) but the hyperbolic-complaint-machine suggests what's beyond frustration.