[FRIAM] Newborn Heart Rate

Tue Oct 5 12:58:44 EDT 2021

Debugging parallel systems is (still) hard, and language design and compiler support is even more important than it is with serial systems.   Type systems help here too by ensuring that there is only reference that can mutate at a time.

From: Friam <friam-bounces at redfish.com> On Behalf Of Frank Wimberly
Sent: Tuesday, October 5, 2021 9:41 AM
To: The Friday Morning Applied Complexity Coffee Group <friam at redfish.com>
Subject: Re: [FRIAM] Newborn Heart Rate

>race conditions, deadlocks, etc.

My favorite when I took an OS course 50+ years ago was "deadly embrace" which I'm sure is an obsolete problem.  It occurred when two processes were waiting for each other.
---
Frank C. Wimberly
140 Calle Ojo Feliz,
Santa Fe, NM 87505

505 670-9918
Santa Fe, NM

On Tue, Oct 5, 2021, 9:57 AM uǝlƃ ☤>$ <gepropella at gmail.com<mailto:gepropella at gmail.com>> wrote:
Yeah, I'm perfectly aligned with the freak among freaks sentiment, though I'd argue we *do* live in that world, we just deny it with our false beliefs. "The problem with communication is the illusion that it exists."

But the more important part of the argument surrounds whether consistency, itself, is a matter of degree or kind. The analog world is full of graded [in]consistency. You see it a lot with artifacts resulting from welding, baking, brewing, etc. ... I even saw it often with the level 3 drafting at lockheed. Any inconsistencies resulting from our designs, the effete knowledge engineers, were *easily* overcome by the gritty on-the-ground engineers ... like smoothing out burrs or gluing together pieces that don't quite fit.

In the special case of refined, crisply expressed propositions of digital computation, inconsistency finding becomes a (perhaps the) powerful tool. Debugging a serial program relies on it fundamentally. But it's softened a bit in parallel algorithms. Inconsistency is broken up into multiple, yet still crisp, types (race conditions, deadlocks, etc.). As approach "the real world" and move away from digital computation, it seems, to my ignorant eye, that [in]consistency softens more and more. Whether that softening takes the form of a countable set of types or something denser, I don't know. But it definitely takes on a different form.

Discussions like Frank and EricS are having about the stability of a limit point (never mind the ontological status of that point) get at this nicely. If you change the frame entirely (e.g. move to position-momentum) and the "inconsistency" of the singularities *moves* (or disappears entirely), then a focus on consistency is not as powerful of a tool. The focus becomes one of which frame expresses the target domain "less inconsistently" ... aka with fewer exceptions to the rule.

Yes, I know I've completely abused the word and its normal meaning.

On 10/4/21 12:03 PM, Marcus Daniels wrote:
> I agree with some of that.   I mentioned the dependently typed programming language as one technology to know when I am being inconsistent.   It doesn't mean I stop everything to resolve the inconsistency, but I might point the headlights in some other direction to avoid the inconsistency (breadth first search instead of depth first).   Inconsistency finding is a tool, and preferably a semi-automated one.
>
> I'd rather have the option of being a depth first searcher and not worry about shelter and food and health care.   I'm not talented enough to be among the small number of people that can survive (adequately) doing that sort of thing.   I think I wouldn't even like it in general, even if I were.   I don't like being the person that says something is irrelevant because everything is irrelevant.   I'd like to be a freak among billions of freaks that all admire the accomplishments of other freaks.   This is not the world we live in, though.
>
> -----Original Message-----
> From: Friam <friam-bounces at redfish.com<mailto:friam-bounces at redfish.com>> On Behalf Of u?l? ?>$
> Sent: Monday, October 4, 2021 10:16 AM
> To: friam at redfish.com<mailto:friam at redfish.com>
> Subject: Re: [FRIAM] Newborn Heart Rate
>
> OK. But academia is in serious trouble, not least exhibited by the rise of populism and anti-intellectual distrust of those who might be attracted to depth-first search.
>
> Another story: At the last salon, an entomologist asked me "Why do you know so much philosophy?" My guess is he was actually trying to politely criticize my incessant concept-dropping, referring to oblique discussions that only occur amongst such depth-first people. The answer is I don't know any philosophy. I'm the worst kind of tourist, trampling endangered species while snapping selfies on my iPhone.
>
> But the deeper answer is that we don't need the academy anymore. What we need are social safety nets that facilitate the diverse exploration of the information field splayed out before us. If an unemployed snowboarder wants to do the work to propose a new theory of everything, so be it. I'm willing to sacrifice some of my income to help that happen, even if, or perhaps because it may eventually be found contradictory to some other ToE somewhere. But a consistency hobgoblin would nip that nonsense in the bud at the first hint of contradiction ... like a blankface academic advisor in some Physics department at some elitist institution.
>
> A focus on consistency is nothing more than subculture gatekeeping <https://www.urbandictionary.com/define.php?term=Gatekeeping>.
>
> On 10/4/21 10:01 AM, Marcus Daniels wrote:
>> In some depth first search one might find a sub-problem that was uncrackable.   If it is one of 100 problems to solve, it is dumb to get hung-up on it, especially if it is of no practical significance.    But it is a problem that will attract a certain kind of (autistic) academic attention as well.
>
>

--
"Better to be slapped with the truth than kissed with a lie."
☤>$ uǝlƃ

.-- .- -. - / .- -.-. - .. --- -. ..--.. / -.-. --- -. .--- ..- --. .- - .
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn UTC-6  bit.ly/virtualfriam<http://bit.ly/virtualfriam>
un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:
 5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
 1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20211005/fafe1fe3/attachment.html>