Ending the cults of personality in free software
aseigo.blogspot.ch
aseigo.blogspot.ch
As the IT industry matures I can see a tendency to make everything Serious, Business Responsible, Grownup, etc. I don't like this development of things. Open Source started as a fun place. People were coding to Have Fun, in the broad hacker meaning of fun. See that recent (and reoccuring) flame against Crockford's "not for Evil" licence. Why are we denying people right to do that? It was a joke, he did it for fun. Hackers like jokes. Fun is a part of FLOSS.
I believe that Open Source community needs not to become adult. It needs place for fun, humour, role models, rockstars, religious preachers; place to agree or disagree, to judge people on their merits and just plainly do whatever one wants. It's how creativity is born.
uh... no.
to steal a joke: it used to be a syntax error to write FORTRAN while not wearing a blue tie.
IT used to be serious people with post grad degrees working with multimillion dollar machines. it keeps moving towards not away from cargo shorts, flip flops and "rockstar code ninjas"
I guess it depends on your point of view, but I personally love the fact that: a) my field is becoming more grown up, which means more recognition of its importance to the world, and b) the field is growing up, meaning the people who nowadays hold huge amounts of power over many other people, are taking this power more seriously.
I also don't think it's a problem to still have fun in our field - but there should be professional limits. When some engineers get together to build an engine that goes into a car, I'm all for them having fun while doing it, and making jokes, etc. But I also want them to be professional and for the end product to do its job. Usually the two don't conflict. In cases like the "nor for Evil" license, some people think they did conflict. Perfectly valid for Crockford to do it, perfectly valid for other people to say "this kind of humor means we can't use your product, much as we'd like to".
I suppose over time, we will have the same split in applications and services. All of 'em will use the project that Torvalds manages, which is why I imagine he lets off steam in irrelevant directions.
Look at the reaction people had when Gnome 3 was introduced, and Unity in Ubuntu. People didn't wait for Linus to complain against those changes driven by UI designers gone insane.
Lots, and lots of linux users had the SAME reaction Linus had, BEFORE he even publicly talked about it. I went to Gnome 2 from the KDE4 transition since I liked KDE3 but hated KDE4, and went to Xfce after the Gnome 3 and Unity debacle. And I didn't wait for Linus to talk about his thoughts on the linux desktop to switch everytime something bugged me.
Aaron Seigo is asking us to not care about Linus opinion ? But we don't give a crap about yours, man, you're the reason why KDE4 went to hell, the man behind Plasma. After so many years acting like you know better what your users want than they do you're getting all hurt when people listen to someone like Linus instead of you ? the thing about Linus is, he's just a user like the rest of us rather than a desktop developer, but with his fame he can make his opinion reach your ears and it seems you don't quite like what regular users think of your "innovations".
Not insane, just designing for a model of a user that does not include many here. I think the 'shock of the new' plays a part. I've been using GS 3.6 on the Gnome Ubuntu Remix (12.10) on the desktop for a bit and it isn't slowing me down much at all, but I'm basically an end user...
The short version of such a Web page would contain a Venn diagram with a large number of intersecting circles representing the needs of different users. Under the diagram would be a simple question: "Are you designing for the intersection or the union set?".
What I actually think would be really neat is a UI authoring kit aimed at interested people and not requiring huge coding experience. Such a kit would gently encourage users/UI authors to think about the flow of activities and would mirror the limitations of the widget libraries. Something visual working a bit like puredata. That would get some interesting alternatives out there, it would let people who want something different design it for themselves, and it would give a wider range of people an insight into UI design principles.
Actually, that would be a great idea. The industry in general doesn't seem to have cottoned on to that idea.
Trying to discredit Linus opinions is a new low. If you don't agree with someone you don't need to discredit him, because something he is saying could be right.
What I'm not that much sure about knowing what they are doing is the GNOME guys, it seems nobody could criticize their actions, and they keep doing strange things only pleasing their selves. Total onanism.
It is obvious to me that they can't listen to the outside world, they support each other in their actions like the Kodak execs before bankruptcy still believing that "film was still the future".
I think the main point is that: People should think for themselves, and just because someone famous acts a certain way doesn't mean we should follow blindly
I don't care about the general public. I'm not part of it (regarding Desktop Environments) nor do I develop DEs for them. I'm a developer who actually likes some of the same tools as Linus does, including those he developed (Git), so yes, he's position is potentially useful to me.
This idea that software isn't useful unless my mother can use it is a really annoying meme. It's like farmers claiming that reviews on combine harvesters aren't useful because they can't help uncle Tom harvest the tomatoes in his garden.
I.e. I think a model whereby the credibility of a person is specified by a single measure has value. The model could be improved by measuring credibility on multiple axes, but a single measure model is still useful.
Linus would clearly have a positive credibility by pretty much any measure you would reasonably want to use. Perhaps the author's complaint is really that this measure is elevated way beyond where it should be due to his celebrity status.
But how do you measure credibility of relative nobodies? Without good information on credibility, it is pretty rational for the casual observer to overweight the general opinions of people like Linus in forming their own opinions.
We are working very hard on this problem and more at backrecord.com. We don't have a go-live-notification list set up, but if you'd like to know when we make the website live, feel free to send me an email at the address in my profile.
I see this cult-like behaviour quite a lot.
Particular examples, when it comes to open source and free software and gnu/unix, vim emacs etc...
It feels like most people just follow the cult's teachings and hold the same opinions that they saw.
I'm sure you guys all think open source software is fantastic and can give me 10 reasons and examples in support of your opinion.
But how much of it is truly your own conclusion and your findings vs just repeating what you were told before and what everyone else seems to be saying?
Have you ever stopped for a second and asked yourself about it? Have you tried to challenge the popular opinions of the cult?
What would you be doing, if every other hacker was bashing open source today and praising proprietary software?
There's a factor in here that I call 'Bandwagon Threshold Theory'. People react negatively to new and disrupting ideas until a critical mass of people start admiring it. By then they jump on the bandwagon and start preaching and telling everyone about how cool this thing is.
If you look around you can see it everywhere.
Older ideas and technologies have already established their bandwagon and therefore there's strong resistance and hesitance in saying anything negatively about them.
Don't believe me? just say something negative about open source and see how religiously people will defend it.
I guarantee you, the majority of those defenders will simply be doing so not as a result of deep understanding and knowledge but because of the teachings of the cult.
I have previously written a short rant here as well: http://news.ycombinator.com/item?id=4601593
PS: I just used the open source examples etc to make a point. My comment is general about the cult-like behaviour in the industry not about something specific. In summary, I think we should all stop worrying about standing out and start challenging the things we take for granted as a result of being surrounded and constantly fed a set of accepted and popular beliefs and opinions.
You can not make an informed opinion on _everything_. There is too much information out there to fact check all bits of it. Sooner or later you will have to take somebody elses opinion on faith, i.e. to trust that he has fact checked something so you dont have to.
Unfortunately, ideas like "is open-source good" are usually very difficult to turn into pure facts, and are therefore opinions. And these are usually very different depending on what different people think the world is or should be - there is no "correct" fact which I can trust someone else to verify.
The review may be in a language you speak (and may be in a language they speak,) and they may have understood my book, but I am a clever author and they were the cheapest reviewer I could find. They are not themselves an author, nor can they read, but they have talk to authors before and have looked at the pictures in my book (I had not realized they were there!) and asked me questions I did not understand.
Oh, and I will not correct, evaluate or pay you for your sequel but readers of my book can choose to also read your book. I can cancel or up the fee for my book whenever I feel like it, so future readers may not exist.
I would say that it is a "correct fact" that it is a disadvantage to undertake the project I suggest even though you may on occasion succeed. Similarly, I would suggest you avoid building on your own dime on a closed source framework. But hey, I can't prove that golfing in lightning storms is a bad plan, so go nuts..
Yes, it's important not to jump on a bandwagon, but that goes both ways: it seems to be increasingly fashionable to attack "celebrities" in our field through thinly veiled ad hominems. Calling Linus a "grumpy-head" or describing Stallman as "waddling" and having a "beard matted into a bizarre pizza crust flapping off the end of his chin" [1] are just recent examples. It's ruse and it does not truly contribute to the point the author is trying to make anyway.
As @math and @forgottenpaswrd mentioned in their comments, Linus is a smart guy who, furthermore, has a non-negligible level of expertise and credibility when it comes to Linux-related topics in general. I'll pay attention to his opinions and try to extract facts from them, in order to form my own. I understand that the author is frustrated with people who don't do that, but dissing Linus is not the way to rectify the situation.
In the end, I don't care what Linus uses on his computer as much as why he uses (or refuses to use) it and how that might applied in general.
[1]: http://brooklynhacker.com/post/35445721578/that-moment-when-...
The reality is that Linus is good at kernels and DRCS for the same reason and is not mere "knowledge", and for the same reason you should pay attention when he talks about software in general.
OP says that whatever makes him good at kernels doesn't help to make him good at version control systems (X->K but X-/->V), whereas you say that there is a third factor that makes him good at both (X->K and X->V): I say that it is his long experience with managing a zillion patches and branches for the kernel that made him good at version control (K->V).
Otherwise I could counter-argument that people that use a lot Desktop features in their day to day work are in line to create the best desktop ever, which is not true.
For Linus handling a lot of patches was just the motivation and what also gave him the required background (that is not enough) and the opportunity to be an user itself of his ideas on RCS.
Another argument is that Git does not offer incredible advantages only to big projects, it is winning actually because its wast advantages can "scale down" very easily to small projects.
The bandwagon effect in OSS is a truth, but this is not a cult behavior, this is skills recognition. Just look at how languages and tools have changed in the last ten years, these changes are driven by users not by BDFLs.
When I compare software to, let's say politics or the media, I'm not ashamed of the open source community.
Then, when you're challenging open-source it's a whole different thing, you're criticizing a philosophy, you can't just bash it, you have to refute it logically.
I knew if I don't do that others will turn the comment thread into a open source vs proprietary war and the main point will be dismissed.
So I thought I'd highlight that I'm just using that as an example that everyone has seen and can relate to so that people will keep the conversation around the bigger picture.
'But Feynman only has expertise in Physics'. Well. A more accurate picture might be: Feynman is a hell of a smart guy, he chose to dedicate most of his time to Physics, but that doesn't make him any less smarter when he talks about other things.
Mercurial is a good example of a powerful DVCS (it can do everything git can) and yet a sane user interface that is easier to teach. Almost nobody "non-techie" that I know about learns git's command line. It's a horrible mess! Instead they rely on GUIs (Git Tower, gitk) or WUIs (Github). But almost anyone can learn hg's command line, since they go through great pains to make sure they pick user-facing names and words and concepts that map to familiar ideas.
We all have our strengths and weaknesses. It takes strength of character to acknowledge where you're weak, and it takes strength of will to follow only the good examples of those we admire.
That is fine. Those people are not Git's target market.
A) use a VCS (explicitly not as like a transparent feature in some software package)
B) use one from a command line
because I'm pretty sure managing documents with hg from the command line qualifies you as "techie" at the very least "tech-ish"
And I particularly don't like KDE Plasma, even if Linus does. But I do like Linus Torvalds. He's a straight talking guy who has consistently walked his talk, even if some don't like what he has to say. AFAIK there is no cult of personality as far as he's concerned. Historically cults of personality are nurtured by the personality involved. Linus has done nothing of that sort.
That's not to say that Seigo doesn't have some fair points, but as far as I'm concerned, everything coming from him on this subject just doesn't hold any value. If anything, it just emphasis just how out of touch the KDE devs are with their users.
EDIT: oh and even worse, discredit others usually writing meaningless trolling tweets without any actual argument.
I completely disagree with this line of argument that, you have to have worked as something before giving a valid opinion, I wish it were true in some case of tech journalists, but, in my opinion if you've been a user of a technology, then you can give an opinion.
On the flipside, positive case for cult of personality may be, a simple use of two words and middle finger can affect the decisions of major corporation like nVidia.
"His opinion on desktop software is as meaningful as his opinions on rocket ships, film production, oil recovery techniques, sociology, religious history, automobile engineering or any of the other topics he has no meaningful expertise in."
His opinion on desktop software is valuable to me. Why? Because he's obviously a very smart person. And I listen to smart people a lot more closely than I listen to just anyone. I would trust a physicist giving his opinions on the state of medical care more than I would a guy who believed the earth was flat.
On the other hand, even and especially in the conservative, corporate cultures I've experienced, there has been a small minority of people who are an order of magnitude smarter and more capable than their peers. And... this tends to generalize. They have their assignments. But sit down with them for a more general discussion, and they often have a better idea than most of what is going on and of its value or cost, in areas well outside their formal assignments. Further, when they -- one way or another -- end up responsible for addressing something "farther afield", they often do so better, quicker, and more insightfully than the dedicated, "expert" staff assigned to it.
That's why I pay attention to people like Linus. In my experience, when people like him address a topic, even casually, they've often been paying better attention than many others, and/or their ability and experience enables them to quickly cut through the crap and see to the heart of the matter.
Even when they are wrong, their thinking often brings up some very interesting and relevant points.
The culture in these places, for better or worse, are initially driven by some small number of charismatic leaders who can create either a good culture or a bad culture. Entropy demands that well-maintained culture must tend towards chaos as an organization grows. At this point, inertia helps prolong the original culture initially, but then provides a huge barrier to reclaiming a favorable lost culture.
Finally, the organization's reliance on oral history (we're horrible at really documenting these things) fails as older members of the tribe leave and are replaced with idealistic whippersnappers. The cycle is complete when these young'ins become resigned to their position as a "cog-in-the-machine" and stop trying to effect change.
I'm not pessimistic ... I've just been around this cycle a few times.
The Free Software movement is a political movement and requires leadership. The OP's critique of Linus (regardless of its merit) is entirely irrelevant to a discussion of Free Software Movement and the clear necessity for uncompromising and visionary leadership.
I've met a number of personality cultists in tech who use the example of Steve Jobs as a justification to be a jerk. I find this to be inversely correlated with their actual insight and intelligence. (I am an iOS developer and owner of several Apple devices.)
Yeah right.
Also, it's not like Linus does not provide reasoning for his opinions, and deriding him for not being a desktop developer is unfair.