Page 5 of 5 FirstFirst 1 2 3 4 5
Results 101 to 111 of 111
  1. #101
    Join Date
    Oct 2005
    Location
    Sandy
    Posts
    4,683
    Quote Originally Posted by covert View Post
    It's back.

    Special snacks for anybody that can identify at least three of the technical details I should have explained differently.


    1) Your probably not happy with spatial distribution, isolated vs. specific. et all.
    You can have specific, widespread problems (that may be easy to avoid, stay off the sunny slopes)
    Non-specific, isolated problems sound scarier (the wind destroyed most of the surface hoar, also the wind shifted from the south to the north as the storm rolled in, so good luck finding it)

    But then you say likelihood is a function spatial distribution, but you never give us the function.
    Last edited by sfotex; 11-19-2017 at 11:42 AM.
    Life is a lot like climbing: there isn't anything much more comforting than a good #2.

  2. #102
    Join Date
    Oct 2005
    Location
    Sandy
    Posts
    4,683
    2 ) estimation vs. probability vs. qualitative vs. quantitative . A model estimate is an estimate of the probability that an event will occur, based on a probability model.

    I think your trying to say if you don't put a unit of measure next to your estimate, it doesn't become a measurement ,. i.e. you deliberately introduce ambiguity into your estimate to drive home the fact it's an estimate vs. a hard number made with a limited data set by using your own made up units to drive home the fact it's an estimate vs. a hard number made with a limited data set.

    Bossman: Sfotex, how long is it going to take to do the XYZ project?
    Me: Considerable time
    Bossman: Considerable time? what does that mean.
    Me: Well, I estimated the ABC project to take moderate time, and considerable is greater then moderate. If I give you estimates in days your going to hang your hat on it and make plans off of it..

    Then you kind of interchange Estimative Probability and qualitative but never bring out quantitative. When you go build your McMansion in Ophir at the bottom of slide path some home insurance company is going to looking at quantitative numbers when they come up with your insurance rate.
    Last edited by sfotex; 11-19-2017 at 12:21 PM.
    Life is a lot like climbing: there isn't anything much more comforting than a good #2.

  3. #103
    Join Date
    Dec 2005
    Location
    Closed Area
    Posts
    1,120
    Yo thanks for playing,

    I think describing a problem as specific and widespread or non-specific and isolated is not consistent with the model. The sunny slopes example that you use would be consistent with a distribution of "specific" where "The problem exists in terrain features with common characteristics." I think your surface hoar example fits precisely with the definition of isolated. I see what you're saying, but would describe those problems differently.

    I believe I say Likelihood is a function of Spatial Distribution and Sensitivity to Triggers, maybe that wording could have been more clear, but I do reference the matrix for deriving Likelihood from Distribution and Sensitivity. It gets to be kind of a word salad in the segment. Hard to get around that.

    On the other hand, yup, not happy with the discussion of estimative probability; brevity would have been better there. I could have focused more clearly on how those words convey estimates and uncertainty, but I left out quantitative measures on purpose.

    During the Story Time I reference an ECT test where I rate the fracture character. In the new SWAG, rating fracture character was specifically removed from the ECT process...because the ECT itself is designed to assess fracture.

    I think there are one or two more bits I would have changed, but I forget what they are.

    Thanks for the feedback. It's really helpful to hear how folk are pickin' up what I'm layin' down. It also gives me the opportunity to add clarity in subsequent episodes.

  4. #104
    Join Date
    Oct 2005
    Location
    Sandy
    Posts
    4,683
    Quote Originally Posted by covert View Post
    I think describing a problem as specific and widespread or non-specific and isolated is not consistent with the model. The sunny slopes example that you use would be consistent with a distribution of "specific" where "The problem exists in terrain features with common characteristics." I think your surface hoar example fits precisely with the definition of isolated. I see what you're saying, but would describe those problems differently.
    But specific doesn't imply distribution, it just says we know the problem exists, i.e. we know there is deep slab instability. Distribution could tell me where (north facing slopes) , and isolated widespread would describe the probability it is there in a given location or area.


    (you asked for 3, so I was digging hard...)
    3) Evidence and uncertainty, you say that uncertainty may undermine evidence, and there is a evidence uncertainty balance. Uncertainty would describe the gaps in our evidence. Lots of evidence may reduce uncertainty, but uncertainty may also linger. So maybe evidence -uncertainty gap would be more accurate? So I may dig a pit on a slope and gather some evidence and my uncertainty will, based on my random sampling of the slope and what I'm looking for, may stay the same. Or it may go down (yup, there's the buried surface hoar I'm worried about)

    Looked up Sherman Kent's paper, thanks for brining that up, it's interesting stuff. Nothing like curling up with Words of Estimative Probability on a Saturday night.
    Last edited by sfotex; 11-19-2017 at 07:41 PM.
    Life is a lot like climbing: there isn't anything much more comforting than a good #2.

  5. #105
    Join Date
    Dec 2005
    Location
    Closed Area
    Posts
    1,120
    Quote Originally Posted by sfotex View Post
    But specific doesn't imply distribution, it just says we know the problem exists, i.e. we know there is deep slab instability. Distribution could tell me where (north facing slopes) , and isolated widespread would describe the probability it is there in a given location or area.
    "Specific" is a Spatial Distribution rating in between Isolated and Widespread. It has defined parameters for spatial density and available evidence. "Isolated widespread" is a combination of two distribution ratings at opposite ends of the spectrum.

    Name:  Spatial Distribution.png
Views: 169
Size:  50.2 KB

    Typically a distribution rating is coupled with additional location information that references aspect and elevation or vegetation band. For instance, Deep Slab problems are possible in specific areas above 11,000 cubits on N-NE aspects.

    Quote Originally Posted by sfotex View Post
    (you asked for 3, so I was digging hard...)
    3) Evidence and uncertainty, you say that uncertainty may undermine evidence, and there is a evidence uncertainty balance. Uncertainty would describe the gaps in our evidence. Lots of evidence may reduce uncertainty, but uncertainty may also linger. So maybe evidence -uncertainty gap would be more accurate?
    Some folk like to speak of knowledge gaps, but I prefer balance. Uncertainty may reference gaps in evidence, but it can also refer to the quality of evidence, the unknowable, and shit we could in theory know but don't even realize we need to know that shit.
    Last edited by covert; 11-19-2017 at 11:07 PM.

  6. #106
    Join Date
    Oct 2005
    Location
    Sandy
    Posts
    4,683
    Quote Originally Posted by covert View Post
    "Specific" is a Spatial Distribution rating in between Isolated and Widespread. It has defined parameters for spatial density and available evidence. "Isolated widespread" is a combination of two distribution ratings at opposite ends of the spectrum.

    .....


    Typically a distribution rating is coupled with additional location information that references aspect and elevation or vegetation band. For instance, Deep Slab problems are possible in specific areas above 11,000 cubits on N-NE aspects.
    Interesting, your linky is broken but I found this with Google:

    The Conceptual Model of Avalanche Hazard
    (Statham et al, 2010) defines three levels of spatial
    distribution: isolated, specific and widespread.
    Matched with avalanche size, this information is a
    general overview of where, and how big the
    avalanches might be.

    So the UAC uses these terms (or has in the past)
    The Spatial Distribution (isolated, localized, widespread) to include our danger rose aspect/elevation diagram.
    https://utahavalanchecenter.org/blog...danger-ratings


    I guess thats why I was really really confused.....

    Well, since I fucked that one up, I'll have to submit another answer. (Monty Python style): You weren't specific enough on which cubit measurement unit you were using, a cubit can be 17.5 inches or 18 inches. Well, it can actually be a range between 17-20 inches. Context matters.

    This is a lot of work for a special snack.
    Life is a lot like climbing: there isn't anything much more comforting than a good #2.

  7. #107
    Join Date
    Dec 2005
    Location
    Closed Area
    Posts
    1,120
    It's getting much better down here, but you'll have to go to Canada for true consistency.

  8. #108
    Join Date
    Sep 2016
    Posts
    35
    Just listened to S1E10. Two questions:
    1. After you dropped in first and found unexpected windslab, where would you have told Zeppo to ski to avoid your route?
    2. Would it have been faster or easier to have your skis pointing the opposite direction to avoid the kick turn when you bailed from your post-up spot?
    Thanks.

  9. #109
    Join Date
    Dec 2005
    Location
    Closed Area
    Posts
    1,120
    I'm guessing at the story you are referring to, but I think I know the one.

    1. I would have told him to avoid the start zone entirely and skirt the path through the trees skiers right.

    2. Yes, but I flip lightning kick turns in my sleep. Pisses the wife off. On a more serious note. That would have been a concern for me if I was standing on a slab, where available reaction time would be marginal to non-existent. If you feel threatened by the snow under your feet, it's a good idea to point your skis in a safer direction...and maybe even go there.

  10. #110
    Join Date
    Sep 2016
    Posts
    35
    Quote Originally Posted by covert View Post
    I'm guessing at the story you are referring to, but I think I know the one.

    1. I would have told him to avoid the start zone entirely and skirt the path through the trees skiers right.

    2. Yes, but I flip lightning kick turns in my sleep. Pisses the wife off. On a more serious note. That would have been a concern for me if I was standing on a slab, where available reaction time would be marginal to non-existent. If you feel threatened by the snow under your feet, it's a good idea to point your skis in a safer direction...and maybe even go there.
    Thanks for the response. Any way to make a one-time donation?

  11. #111
    Join Date
    Dec 2005
    Location
    Closed Area
    Posts
    1,120
    Yup, go to www.avalanchepodcast.com and click on the Donate via Paypal button. Cheers...dk

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •