Realtime Linux: academia v. reality [LWN.net]

Realtime Linux: academia v. reality

Posted Jul 26, 2010 23:03 UTC (Mon) by fuhchee (guest, #40059)
Parent article: Realtime Linux: academia v. reality

"... Both are pretty good examples of the pragmatic approach of the Linux kernel development community and its key figures. Linus especially has always silently followed the famous words of the former German chancellor Konrad Adenauer: "Why should I care about my chatter from yesterday? Nothing prevents me from becoming wiser." ..."

This appears to necessitate the use of a special LKML dictionary when reading categorical superlatives. For example, it would map "never, ever, this is stupid" means "I currently don't see why, sell it to me better.". Perhaps an unLKML decoder script should ship in kernel/tools/, together with a coefficient-of-exaggeration database for main participants.

to post comments

Realtime Linux: academia v. reality

Posted Jul 26, 2010 23:12 UTC (Mon) by SEJeff (guest, #51588) [Link] (19 responses)

don't feed the trolls people.

Realtime Linux: academia v. reality

Posted Jul 27, 2010 3:53 UTC (Tue) by butlerm (guest, #13312) [Link]

Exaggeration aside, I think the GP has a point. This has happened more than once.

Realtime Linux: academia v. reality

Posted Jul 27, 2010 12:26 UTC (Tue) by nye (guest, #51576) [Link]

Sense of humour malfunction?

Realtime Linux: academia v. reality

Posted Jul 27, 2010 13:00 UTC (Tue) by fuhchee (guest, #40059) [Link] (16 responses)

In plain English, it is a problem if leaders regularly make categorical statements about the future, and then flip-flop. It means that any particular statement / prediction has a credibility cloud over it. This makes planning more difficult, never mind the unnecessary social friction.

Realtime Linux: academia v. reality

Posted Jul 27, 2010 15:28 UTC (Tue) by martinfick (subscriber, #4455) [Link]

It is a worse problem when leaders refuse to change their opinion simply for the sake of being consistent with the past when assumably they were less knowledgeable.

And, of course, one has to consider the amount of times (%centage wise) this happens before passing judgment. I would suspect that this is rather low for Linus, and that the whole point (which seems to have been missed) is: that he doesn't sway easily, and that the few subjects where he has being swayed drastically show both his original good judgment on most things, along with the ability to be convinced of opinions better than his first (usually correct) opinions.

Realtime Linux: academia v. reality

Posted Jul 27, 2010 17:49 UTC (Tue) by pflugstad (subscriber, #224) [Link] (8 responses)

In plain English...

Actually, this statement identifies another problem - many non-native-English speaking readers won't understand Linus' comments, or take them at face value.

Or in other cultures, it's literally the worst possible thing to change your mind as publicly as Linus' occasionally does, so they assume he really means it.

So I certainly don't see the OP as a troll, but asking a legitimate question.

Realtime Linux: academia v. reality

Posted Jul 28, 2010 10:02 UTC (Wed) by nye (guest, #51576) [Link] (6 responses)

>Or in other cultures, it's literally the worst possible thing to change your mind as publicly as Linus' occasionally does

The unwillingness for a public figure to acknowledge a past mistake and fix it is the cause of a great variety of societal ills. I don't believe anyone should ever pander to a culture that encourages this kind of poisonous behaviour.

Realtime Linux: academia v. reality

Posted Jul 28, 2010 22:39 UTC (Wed) by njs (subscriber, #40338) [Link] (5 responses)

> The unwillingness for a public figure to acknowledge a past mistake and fix it is the cause of a great variety of societal ills.

Okay, I'm with you so far...

> I don't believe anyone should ever pander to a culture that encourages this kind of poisonous behaviour.

...but you lose me here. We're talking about people who have certain expectations, and those expectations are, in fact, valid for their culture. These people may or may not like this aspect of their culture, but it doesn't matter -- so long as they have those expectations, then they are not worthy to participate in kernel development? Explaining to them that things work differently here is somehow "pandering" to their culture?

Realtime Linux: academia v. reality

Posted Jul 29, 2010 1:15 UTC (Thu) by dlang (guest, #313) [Link] (1 responses)

the problem is that if changing your opinion is the worst possible thing you can do you are in one of two categories

1. you make the perfect decision the first time, every time

2. you do things wrong, even when you know better.

Since there is no developer who ever qualifies for #1, avoiding changing decisions at all costs would lead to having a bad system, and knowing that it was bad.

so yes, the kernel development _is_ better off by being willing to change decisions, even if that excludes some cultures from participating.

Realtime Linux: academia v. reality

Posted Jul 29, 2010 3:41 UTC (Thu) by njs (subscriber, #40338) [Link]

I've seen two suggestions made in this thread:

1) It's somewhat problematic if leaders regularly make very strong, emphatic statements, specifically saying that this is not an ordinary decision but rather one that will never be changed under any circumstances, and then change their minds.

2) If they're going to do that anyway, then maybe that should be explained to newcomers, since their default understandings will otherwise be wildly miscalibrated.

You seem to be arguing about something else, not entirely sure what, and I don't have much to say about it.

Realtime Linux: academia v. reality

Posted Jul 29, 2010 12:25 UTC (Thu) by nye (guest, #51576) [Link] (2 responses)

>We're talking about people who have certain expectations, and those expectations are, in fact, valid for their culture. These people may or may not like this aspect of their culture, but it doesn't matter -- so long as they have those expectations, then they are not worthy to participate in kernel development? Explaining to them that things work differently here is somehow "pandering" to their culture?

It seems I had misinterpreted the previous post in haste. An explanation of the differences between cultures to the honestly ignorant (I use this word in a purely descriptive way with no unstated implications intended) would be a worthwhile exercise in education (especially when, as I believe, the one culture is clearly superior to the other in a particular way).

So, I retract that statement.

Realtime Linux: academia v. reality

Posted Jul 29, 2010 12:27 UTC (Thu) by nye (guest, #51576) [Link] (1 responses)

>So, I retract that statement

Which, now that I think of it, is sort of ironic in the circumstances.

Realtime Linux: academia v. reality

Posted Jul 29, 2010 13:05 UTC (Thu) by fuhchee (guest, #40059) [Link]

two words: epic win

Realtime Linux: academia v. reality

Posted Aug 2, 2010 12:52 UTC (Mon) by miku (guest, #35152) [Link]

Reality doesn't care about 'other' cultures. That is why Linux kernel is so good: it tries to map reality instead of bow to authority/ego.

It is best possible thing for a person to change their opinion when it contradicts with reality. Linus, thankfully, is among the sane people who practice this =)

Realtime Linux: academia v. reality

Posted Jul 27, 2010 23:37 UTC (Tue) by aliguori (guest, #30636) [Link]

Categorical statements are never accurate in the long term. You can either avoid them entirely which makes life dull, embrace them with unwaivering dedication to avoid the appearance of "flip-flopping", or simply admit when you're wrong and keep going.

The last characteristic is one that's extremely appreciated in a maintainer. It's much more fun to contribute to a project if you think that with a sufficiently compelling argument, you can convince the leadership to agree with you.

Categorical statements

Posted Jul 27, 2010 23:48 UTC (Tue) by jejb (subscriber, #6654) [Link] (4 responses)

> In plain English, it is a problem if leaders regularly make categorical
> statements about the future, and then flip-flop. It means that any
> particular statement / prediction has a credibility cloud over it. This
> makes planning more difficult, never mind the unnecessary social friction.

Actually, I'm afraid history really doesn't support this view.

Stevenson was told by all medical authorities that people would suffer seizures if they travelled at more than 30mph, so the Rocket was a stupid idea.

Max Planck despised Ludwig Boltzmann's statistical mechanics because of the challenge it gave to classical thermodynamics. He went as far as to attack Boltzmann both verbally and in print for the heresy. Planck was ultimately forced to use statistical mechanics to solve the ultraviolet catastrophe and lay the basis for quantum mechanics ...

Einstein famously and vehemently denied the conclusions of the EPR paradox with his "spooky action at a distance" comment. He recanted very reluctantly when the Bell inequalities proved it.

I could go on, but you get the idea.

Great discoveries are made by challenging the accepted and laid down "facts". The corollary to this is that if no-one lays down the "facts" to be challenged, the human instinct for contrariness doesn't get aroused as much as it should and some of our brilliance sinks into the mire of mediocratic reasonableness.

Being wrong is a recoverable error. Never daring to be wrong is an opportunity missed and a life never lived

Categorical statements

Posted Jul 28, 2010 2:15 UTC (Wed) by fuhchee (guest, #40059) [Link] (3 responses)

I don't think your analogy works the way you intended. My comment is equivalent to cautioning those who told Stevenson about seizures, or Max Planck for his accusations of "heresy", or Einstein about his vehemence. One should be more humble.

"The corollary to this is that if no-one lays down the "facts" to be challenged, the human instinct for contrariness doesn't get aroused as much"

Now this "corollary" needs somewhat more evidence to convince that naysayers are a necessary (or necessarily positive) factor in innovation.

Categorical statements

Posted Jul 28, 2010 15:29 UTC (Wed) by jejb (subscriber, #6654) [Link] (2 responses)

> I don't think your analogy works the way you intended. My comment is equivalent to cautioning those who told Stevenson about seizures, or Max Planck for his accusations of "heresy", or Einstein about his vehemence. One should be more humble.

I'm failing to see your point. I gave Einstein and Planck as examples of people who made categorical negative but wrong statements and later admitted they were wrong (without, incidentally, incurring a "credibility cloud"). You seem to now be saying that it's OK for the likes of Einstein and Planck to do this, but everyone else should be humble?

> Now this "corollary" needs somewhat more evidence to convince that naysayers are a necessary (or necessarily positive) factor in innovation.

A corollary is a logical deduction from a proposition. If there's enough evidence to support the proposition then, ipso facto, there's enough to support the corollary.

If you think the proposition needs more evidence, there's enough in google to supply virtually any amount of it going back to the beginnings of recorded history.

Categorical statements

Posted Jul 28, 2010 15:46 UTC (Wed) by fuhchee (guest, #40059) [Link]

> You seem to now be saying that it's OK for the likes of Einstein and
> Planck to do this, but everyone else should be humble?

You misread. "cautioning" is not saying "it's OK".

> A corollary is a logical deduction from a proposition.

Well, thanks for the lesson, but it doesn't quite work here. This "corollary" is your main point, and it is supported by exactly three historical anecdotes. By the way, in none of those stories has there been any indication that the "should-have-been-humbler" people performed a useful service in naysaying. IOW, there has been no argument that without those "laid-down-wrong-facts", the discoveries would not have been made.

If you can't make that argument stick, perhaps your original argument/corollary is not quite as logically sound as you believe it is.

Categorical statements

Posted Jul 28, 2010 15:50 UTC (Wed) by egk (guest, #50799) [Link]

Although this does not have much bearing on the general discussion, it should be said that the story about Einstein seems very doubtful. He died in 1955 and Bell's paper only appeared in 1964. Moreover, Bell's work is theoretical: it could not "prove" anything about the real world, except the fact that something was testable, in principle. Einstein, if he had been alive, could very well have said that he expected experiments to go one way instead of another. And the actual experiments came quite a bit later.

Realtime Linux: academia v. reality

Posted Jul 26, 2010 23:29 UTC (Mon) by dlang (guest, #313) [Link]

it's less 'sell it to me better' than it is 'fix it so it won't hurt anyone else (including those maintaining the rest of the kernel)'

what was eventually merged was not the same thing that was discussed earlier.

Linus is very pragmatic about accepting oddball things if they really don't hurt anything else. Unfortunately many of the things presented this way really do hurt the kernel, but mostly less in performance than in maintainability.

Realtime Linux: academia v. reality

Posted Jul 27, 2010 10:59 UTC (Tue) by nix (subscriber, #2304) [Link] (5 responses)

systemtap will never ever ever go in until one day it suddenly does.

Realtime Linux: academia v. reality

Posted Jul 27, 2010 14:35 UTC (Tue) by mjthayer (guest, #39183) [Link] (4 responses)

> systemtap will never ever ever go in until one day it suddenly does.

SystemTap or uprobes? For SystemTap itself, it looks to me as though Frank and friends have worked around requiring anything they can avoid in the kernel itself. And done it pretty well given the stones laid in the path of anyone doing kernel modules outside the official kernel tree.

Realtime Linux: academia v. reality

Posted Jul 28, 2010 13:01 UTC (Wed) by nix (subscriber, #2304) [Link] (3 responses)

Yeah, but even now you have people like Christoph Hellwig turning up on e.g. the glibc list when probe point adding patches were proposed, saying 'no, uprobes will never exist in this form'. So uprobes may exist but some kernel hackers are trying as hard as possible to make sure they stay useless and out of *every* package's upstream.

Realtime Linux: academia v. reality

Posted Jul 28, 2010 16:08 UTC (Wed) by mjthayer (guest, #39183) [Link] (2 responses)

> Yeah, but even now you have people like Christoph Hellwig turning up on e.g. the glibc list when probe point adding patches were proposed, saying 'no, uprobes will never exist in this form'.

I thought I read about work to make SystemTap work with procfs. Any idea what is happening there (Frank?) Although something seems to be happening currently with uprobes, and Christoph Hellwig seemed to be actively (positively) involved in the discussion (see http://lkml.org/lkml/2010/7/27/121).

Realtime Linux: academia v. reality

Posted Jul 28, 2010 16:18 UTC (Wed) by fuhchee (guest, #40059) [Link] (1 responses)

> I thought I read about work to make SystemTap work with procfs.
> Any idea what is happening there (Frank?)

I'm not quite sure which effort you may be referring to.

> ... Christoph Hellwig seemed to be actively (positively) involved

Yes. It is unfortunate though that instructions of the form "Don't SYMBOL_EXPORT this facility [since you-know-who might use it]" are still the order of the day.

Realtime Linux: academia v. reality

Posted Jul 28, 2010 16:27 UTC (Wed) by mjthayer (guest, #39183) [Link]

>> I thought I read about work to make SystemTap work with procfs.
>> Any idea what is happening there (Frank?)
> I'm not quite sure which effort you may be referring to.
I can't seem to find any reference to it now. If you are not aware of it then I strongly assume I misunderstood something at some point.