6 ms·

This post is unfortunate. This post is so extremely unfortunate because first, it is wrong, and second, it highlights the wrong problem (and the post's author

This post is unfortunate. This post is so extremely unfortunate because first, it is wrong, and second, it highlights the wrong problem (and the post's author even admits it in the post). 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.

ThomPete 13y ago
Bad design and bad designers is never the problem. If you have bad designers you have bad management. 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.
prollyignored 13y ago
The post is really about rock-star-developer-ninja ego, which is a non-problem really. Ego / Persona / Flair are all part and parcel of Hollywood / Music / Fashion / Chefs and now Software. 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.
acabal 13y ago
I don't disagree that dictators can make great software. My point, perhaps poorly communicated, was that using these exceptional success edge-cases as an excuse to uncompromisingly drive your own design vision will probably do more harm than good--precisely because they were successful as edge cases. 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."
- knowtheory 13y ago
  Right, but you're targeting the wrong problem still. 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.
hythloday 13y ago
"The problem is not the cult of design dictatorship, the problem is bad design and bad designers." 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.
- knowtheory 13y ago
  So, actually I let acabal hoist himself by his own petard a bit. Dictatorship is the wrong choice of words for what I think he's describing and serves to obfuscate the manner in which decisions are actually made in creative projects. 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-leave-times-picayune-for 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 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.
coldtea 13y ago
>The problem is not the cult of design dictatorship, the problem is bad design and bad designers 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.
- knowtheory 13y ago
  > 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. 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.
  - coldtea 13y ago
    >No, that is distinctly not what I am saying. My assertion is that centralized authority is important, and perhaps even necessary in creative endeavors. 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.
joe_the_user 13y ago
What? 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?

ThomPete 13y ago

Bad design and bad designers is never the problem. If you have bad designers you have bad management. 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.

prollyignored 13y ago

The post is really about rock-star-developer-ninja ego, which is a non-problem really. Ego / Persona / Flair are all part and parcel of Hollywood / Music / Fashion / Chefs and now Software. 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.

acabal 13y ago

I don't disagree that dictators can make great software. My point, perhaps poorly communicated, was that using these exceptional success edge-cases as an excuse to uncompromisingly drive your own design vision will probably do more harm than good--precisely because they were successful as edge cases. 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."

knowtheory 13y ago
Right, but you're targeting the wrong problem still. 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.

hythloday 13y ago

"The problem is not the cult of design dictatorship, the problem is bad design and bad designers." 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.

knowtheory 13y ago
So, actually I let acabal hoist himself by his own petard a bit. Dictatorship is the wrong choice of words for what I think he's describing and serves to obfuscate the manner in which decisions are actually made in creative projects. 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-leave-times-picayune-for 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 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.

coldtea 13y ago

>The problem is not the cult of design dictatorship, the problem is bad design and bad designers 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.

knowtheory 13y ago
> 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. 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.
- coldtea 13y ago
  >No, that is distinctly not what I am saying. My assertion is that centralized authority is important, and perhaps even necessary in creative endeavors. 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.

joe_the_user 13y ago

What? 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?