This is petty and shallow, but I'm annoyed with programmer culture in the sense that people always phrase this sort of question in terms of "what if the person that gave us so much kicked off" rather than "what if they were unavailable or retire"? It's morbid and sort of implies that the only way that person would ever stop is if they died.
It leaves me feeling slimy. It's intended to say this person is so dedicated they would never ever abandon a software project of their own free will. In the case of free (as in beer) software projects, the author is often not even compensated. It really makes it seem like the people that are helping you are entirely subservient to the machines and the wanton desires of others. It also appeals to the serious business person that lives in the hearts of so many developers where only the harshest framing can be considered as dealing with the world in real terms.
I guess the underlying assumption is the person in charge would retire or become unavailable with some deliberation and planning. Somebody gone due to accident is entirely different thing.
I partly agree with you, but I think people are also interested in how a sudden and unexpected departure would play out. If Linus Torvalds wanted to retire and spend the rest of his life diving rather than maintaining the kernel, I expect there would be a planned transition.
You're reading too much into the statement in my opinion. "What if XYZ got run over by a bus," is a colloquialism that simply takes the farthest position in order to talk about the worst case. It just means, "what would be the fall back position if everything broke and vapor came out"?
You're right of course, but there is a reason the colloquialism takes this form as opposed to another. Colloquialisms, the intelligible ones at least, sublimate some underlying element of culture in order to communicate sensibly and easily.
This very idea of stretching logic to extremes without consideration for real world effects is very much a foundational element in programmer culture. It's born out of rationalism and certain elements of enlightenment thinking. However, this pure (im)practical reason is often applied in all circumstances irrespective of the context. After all, it's possible to reason about a person's absence without also implying the worst, and this kindness is valuable. If SV could somehow change its foundational thinking from reason + corrosive business practices to reason + kindness we'd all be better off.
When talking about backups, DR, and institutional knowledge, I always use death in the example for the simple reason that it is final and there is absolutely no way to recover the lost individual. If they retire or otherwise move on, you can go visit and ask questions, pay them, or even threaten them for what you need, but death prevents that and is also much less predictable.
The question on the lips of everyone in the Linux community is, "What if Linus Torvalds (affectionately known to the trade press as 'Linux Torvalds') gets hit by a bus?" Reams of virtual paper have been wasted on idle speculation about the results of what is a fairly simple experiment. We decided it was high time someone actually took the trouble to find out what would happen if Linus Torvalds were to get hit by a bus. Our preliminary findings are printed below; we hope that this study will eventually be published in peer-reviewed Linux publications such as Linux Journal and Slashdot. Well, Linux Journal.
Methodology
Our sample consisted of 200 Linus Torvaldses separated into a test group of 100 and a control group of 100. 97 members of the test group were hit by buses at speeds ranging from 5 MPH to 70 MPH. The other three test subjects ran into stationary buses on their own power. A different bus was used for each trial so as to ensure that the trials would be independent.
The control group was observed over a period of six months, in which none of the subjects was hit by a bus.
Extremely valid point. It survives, but loses some of the essential character or traits that made it great.
So perhaps the result with Linux might be that the project continues, but the kernel forks, and even the official Linux branded kernel doesn’t hold to certain essential concepts like don’t break userspace. Perhaps quality control goes down. Perhaps corporate interests start driving the direction of the kernel away from user interests.
I would also suggest that the change might not be a bad thing. With Linus, the kernel follows a relatively predictable path forward. Projects like Fushia might make the current kernel design obsolete. We might need to make bold design decisions moving forward that Linus may not approve of if the project is going to continue to lead the industry.
I suspect the very first thing that would happen is that someone would finally break ABI in the kernel and forever cement Linux in its place as the operating system most incompatible with itself.
19 comments
[ 3.2 ms ] story [ 30.2 ms ] threadIt leaves me feeling slimy. It's intended to say this person is so dedicated they would never ever abandon a software project of their own free will. In the case of free (as in beer) software projects, the author is often not even compensated. It really makes it seem like the people that are helping you are entirely subservient to the machines and the wanton desires of others. It also appeals to the serious business person that lives in the hearts of so many developers where only the harshest framing can be considered as dealing with the world in real terms.
Anyway, carry on. I just wanted to say my piece.
This very idea of stretching logic to extremes without consideration for real world effects is very much a foundational element in programmer culture. It's born out of rationalism and certain elements of enlightenment thinking. However, this pure (im)practical reason is often applied in all circumstances irrespective of the context. After all, it's possible to reason about a person's absence without also implying the worst, and this kindness is valuable. If SV could somehow change its foundational thinking from reason + corrosive business practices to reason + kindness we'd all be better off.
+1 ;)
The question on the lips of everyone in the Linux community is, "What if Linus Torvalds (affectionately known to the trade press as 'Linux Torvalds') gets hit by a bus?" Reams of virtual paper have been wasted on idle speculation about the results of what is a fairly simple experiment. We decided it was high time someone actually took the trouble to find out what would happen if Linus Torvalds were to get hit by a bus. Our preliminary findings are printed below; we hope that this study will eventually be published in peer-reviewed Linux publications such as Linux Journal and Slashdot. Well, Linux Journal.
Methodology
Our sample consisted of 200 Linus Torvaldses separated into a test group of 100 and a control group of 100. 97 members of the test group were hit by buses at speeds ranging from 5 MPH to 70 MPH. The other three test subjects ran into stationary buses on their own power. A different bus was used for each trial so as to ensure that the trials would be independent.
The control group was observed over a period of six months, in which none of the subjects was hit by a bus.
"""
https://www.crummy.com/writing/segfault.org/Bus.html
So perhaps the result with Linux might be that the project continues, but the kernel forks, and even the official Linux branded kernel doesn’t hold to certain essential concepts like don’t break userspace. Perhaps quality control goes down. Perhaps corporate interests start driving the direction of the kernel away from user interests.
I would also suggest that the change might not be a bad thing. With Linus, the kernel follows a relatively predictable path forward. Projects like Fushia might make the current kernel design obsolete. We might need to make bold design decisions moving forward that Linus may not approve of if the project is going to continue to lead the industry.