Announcement

Collapse
No announcement yet.

Kates Cuts

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Sorry, the southern one is probably more along Houndsditch, or a bit west, really. The interesting thing, of course, since this includes Stride, is notice, if JtR is heading home after Stride, and Eddowes was seen leaving the police station towards Houndsditch (at the other end), then you can imagine how the two of them could have met.

    - Jeff

    Comment


    • Originally posted by JeffHamm View Post

      The victim locations are the red squares, and there's a white line linking them as well (in chronological order). Trying to get the street names visible, with the jeopardy map overlay, is tricky as the map has to be large enough to see the street names, but zoomed out enough to show all the locations and have enough surrounding area to do the calculations properly. But you can get the idea by eye-balling key locations, such as in the C5 alone, the southern peak is directly below the GSG and north-east of Mitre Square, the northern one is north-west, basically, running north and south along Goulston Str, which runs from NW-SE.

      Here's the map without an overlay. It's a bit messy as it has locations of various suspects (some, like Chapman's, reflect locations they weren't in at the time of the murders, but I didn't realize that at the time I placed them). There was a threat a while back where I listed various profile rankings for suspects (which zones they were in, etc). I was developing and improving some of the routines during the course of that threat, so the profiles change a bit over time, which can be interesting to see. Anyway, you might be able to use this, in conjunction with the overlay maps, to get an idea of what streets/area specifically things indicate. Might be easier if you print out copies, and go old school with a ruler are protractor though.

      - Jeff
      By including non-Canonical victim locations you provided a map that in and of itself proves multiple killers working within a small geographic area at around the same point in time. Something so often debated by people who want to have some mythical madman as the sole killer in that area at that time.

      Michael Richards

      Comment


      • Originally posted by JeffHamm View Post

        The victim locations are the red squares, and there's a white line linking them as well (in chronological order). Trying to get the street names visible, with the jeopardy map overlay, is tricky as the map has to be large enough to see the street names, but zoomed out enough to show all the locations and have enough surrounding area to do the calculations properly. But you can get the idea by eye-balling key locations, such as in the C5 alone, the southern peak is directly below the GSG and north-east of Mitre Square, the northern one is north-west, basically, running north and south along Goulston Str, which runs from NW-SE.

        Here's the map without an overlay. It's a bit messy as it has locations of various suspects (some, like Chapman's, reflect locations they weren't in at the time of the murders, but I didn't realize that at the time I placed them). There was a threat a while back where I listed various profile rankings for suspects (which zones they were in, etc). I was developing and improving some of the routines during the course of that threat, so the profiles change a bit over time, which can be interesting to see. Anyway, you might be able to use this, in conjunction with the overlay maps, to get an idea of what streets/area specifically things indicate. Might be easier if you print out copies, and go old school with a ruler are protractor though.

        - Jeff
        thanks jeff
        this is cool. and theres my favored suspect right in the thick of it.
        "Is all that we see or seem
        but a dream within a dream?"

        -Edgar Allan Poe


        "...the man and the peaked cap he is said to have worn
        quite tallies with the descriptions I got of him."

        -Frederick G. Abberline

        Comment


        • Originally posted by Michael W Richards View Post

          By including non-Canonical victim locations you provided a map that in and of itself proves multiple killers working within a small geographic area at around the same point in time.
          It's not quite the same point in time, though. Tabram apart, the other non canonical murders were separated by several months from the C5, all five of which happened within the space of two and a bit months.
          Kind regards, Sam Flynn

          "Suche Nullen" (Nietzsche, Götzendämmerung, 1888)

          Comment


          • Originally posted by Sam Flynn View Post

            It's not quite the same point in time, though. Tabram apart, the other non canonical murders were separated by several months from the C5, all five of which happened within the space of two and a bit months.
            I suppose I should have specified that I meant the late 1880's, the small geographical area is of course indisputable. People seem to want to add murders to the Rippers list from early spring 1888 with Emma to as late as 1889 with Alice, so that's the gist of my comment. Some even want to have him making Torsos early in the 80's, …. go figure.
            Michael Richards

            Comment


            • Originally posted by Michael W Richards View Post

              By including non-Canonical victim locations you provided a map that in and of itself proves multiple killers working within a small geographic area at around the same point in time. Something so often debated by people who want to have some mythical madman as the sole killer in that area at that time.
              The maps presented earlier here are only based on the C5, with one of them excluding Stride. The non-cannoicals are indicated on the underlying map but not entered into the analyses I've presented in this thread (so the non-Canonicals are not included in these analyses). In other threads, I have explored alternative linkage theories, and including them doesn't really change the focus a great deal, but then. I've updated the analysis routines a bit since I did these, but it's been fine tuning a few parameters. However, adding in and taking out offenses isn't a way to do linkage, so just because these don't change much isn't support that any or all of these should be included. If you enter any random set of locations you can still do maths on them, and it will produce a jeopardy map, it just won't mean anything. Linkage, the deciding which offenses are committed by the same offender, is done before the profile, not as part of the spatial analysis of those locations.

              Attached Files

              Comment


              • Originally posted by Abby Normal View Post

                thanks jeff
                this is cool. and theres my favored suspect right in the thick of it.
                There's a few proposed suspects that rank fairly highly. Barnett, Hutchinson, and PC Sagar's suspect (he mentions monitoring a butcher I think it was in Algate, and here on casebook it was narrowed down to the indicated location). In the C5 analysis, he's right in zone 1 (the southern one). About 50% of the time, if the offender is a marauder, they live inside the orange region, and about 33% of the time inside the yellow/pink region (zone 1). So, provided JtR was a local, if your suspect is in there, then this sort of analysis would suggest they are worth further investigation. Just remember, this isn't evidence they were JtR, lots of people live in that area. It doesn't solve the case, and it's not evidence, it is just a suggestion where evidence is probable to be found.

                Think of it this way, when the police go into to dust for fingerprints, they don't dust randomly. There are locations and things that are more commonly places where you will find prints, so they start there. Random dusting would mean dusting carpets, couches, etc, which are not going to produce a finger print because there's such a low (0%) chance of finding one. They'll look at smooth surfaces, glasses, mirrors, windows, walls, etc, where prints have a higher probability of being found. But that helps to prioritize where to search, and so prints are more likely to be found if you follow that priority list. Specific information at a crime scene will influence that priority list as well (just like specific information about a crime series will produce different maps). But it still requires real evidence to solve the case, and high probability zones, like zone 1 let's say, only contain the offender 33% of the time (although it only makes up 2.5% of the total search space, so much better than chance), and 50% of the offenders are found after only searching around 10% of the total search space. These are much more efficient search strategies, but they are not "proof" for any individual, though if you have a list of individuals (like we do for JtR), it can be a good way to rank them in terms of priority.

                Just want to make sure it's clear what these provide, as they are often over-hyped in media and entertainment. While that might generate interest in them, in the long run it's counter-productive because when they don't live up to the hype, people view them as useless. They aren't useless, but they aren't magic either.

                - Jeff

                Comment


                • Originally posted by Michael W Richards View Post

                  I suppose I should have specified that I meant the late 1880's, the small geographical area is of course indisputable. People seem to want to add murders to the Rippers list from early spring 1888 with Emma to as late as 1889 with Alice, so that's the gist of my comment. Some even want to have him making Torsos early in the 80's, …. go figure.
                  I'm not sure "want" is quite the right descriptor of the majority of people's reasons for including/excluding various victims. There are those who have drawn a conclusion (not always a firm conclusion mind you), that various non-C5 victims may be part of the series. I've tried to put together a general map that includes all of the Whitechaple murders, in part as a general resource. When I get time, I'm going to do a better one using the online maps as they are so nicely detailed, but this is good for now. In this thread, apart from the image of the 6 different analyses above, we're just focusing on analysing the C5, and one analysis based upon leaving out Stride, as she is the most contentious. Mind you, opinions range from "include all of these and all torso's from the past 20 years" to "only Nichols and Chapman by the same person". I've tried to do various combinations so that there's a version most people might have one to consider, if they choose to.

                  - Jeff

                  Comment


                  • Originally posted by Abby Normal View Post

                    thanks jeff
                    this is cool. and theres my favored suspect right in the thick of it.
                    I'm just starting exam marking, so will be a while before I get the chance, but I've worked out some things that I think have the potential to improve the predictive performance of Dr. Watson. Once I get the chance to implement the changes to the code, I can do the analysis to see if things do get better, get worse, or just move sideways. Given the complexity of the problem, it's hard to know how a change with regards to a predictor will interact with all the others. Sometimes it's redundant with what is already included, sometimes it interferes, and occasionally it results in an improvement. If it does result in an improved performance, I'll update things here and we can re-look at this. Most likely, it won't change things dramatically. Hopefully it won't just be another idea that gets shown to be inadequate.

                    - Jeff

                    Comment


                    • Originally posted by JeffHamm View Post

                      I'm not sure "want" is quite the right descriptor of the majority of people's reasons for including/excluding various victims. There are those who have drawn a conclusion (not always a firm conclusion mind you), that various non-C5 victims may be part of the series. I've tried to put together a general map that includes all of the Whitechaple murders, in part as a general resource. When I get time, I'm going to do a better one using the online maps as they are so nicely detailed, but this is good for now. In this thread, apart from the image of the 6 different analyses above, we're just focusing on analysing the C5, and one analysis based upon leaving out Stride, as she is the most contentious. Mind you, opinions range from "include all of these and all torso's from the past 20 years" to "only Nichols and Chapman by the same person". I've tried to do various combinations so that there's a version most people might have one to consider, if they choose to.

                      - Jeff
                      When I said want its because there is no evidence at all to connect this elusive madman to any non-Canonical murders, and only opinion that connects one to another within the Canonical Group. When you cant prove just 5 by 1 man after 130 plus years of trying, its kind of a bad joke to then add more to the same list anyway.
                      Michael Richards

                      Comment


                      • Originally posted by Michael W Richards View Post

                        When I said want its because there is no evidence at all to connect this elusive madman to any non-Canonical murders, and only opinion that connects one to another within the Canonical Group. When you cant prove just 5 by 1 man after 130 plus years of trying, its kind of a bad joke to then add more to the same list anyway.
                        The case is unsolved, whenever a series of cases are unsolved it is not proven that any two are actually connected, that is always an assumption that one makes based upon the evidence at hand. With modern investigative techniques, there are things that greatly improve making that inference of course (if you found matching DNA from the perpetrator between a number of different offenses you can be pretty sure it was the same person, except of course, if you have a set of identical twins both committing offenses). Ballistic analysis can link offenses to the same weapon, but not necessarily the same offender directly, though that becomes an increased probability.

                        We don't have anything like that with JtR, we only have things like the behaviour that was involved (possibly including strangluation to some degree, throat cutting, and mutilations, at least in 4 of the C5) and we have the type of weapon used (knife), crime scene locations (generally outdoors, but generally locations a prostitute might take a client), and details of the victims (all poor, all living a hand to mouth existence and apparently in need of money, all alcoholics or at least heavy drinkers, all having a high probability of engaging in causual prostitution on occasions), and all (except Emma Smith), apparently killed by a single person. The consideration of Tabram for example, being earlier is because her murders shows some similarities (similar victimology, similar type of location, excess violence, some indications of possible strangulation - protruding tongue etc) but some differences as well (no throat cutting, no multilations, two weapons which may suggest two rather than one attacker, etc), and later victims, like McKenzie, do show attacks on the abdoment, etc but they are less "determined"? might be the best word. People discuss and consider the evidence we have, it's not possible to be conclusive with what we have, but there are good, valid, and reasonable arguments to be made both for and against various victims.

                        You have formed your opinion, but it is only that, an opinion which I'm sure you feel is well supported, and I would agree with you that your opinion is defensible. I do not agree with you when you suggest it is a "bad joke" and employ other pejorative sophistry as if that is a valid defense of your opinion. It's not, and it is incorrect to suggest that the interpretations of others, who might consider the possibility other victims may indeed have been by the same person, are indefensible. There is a true state of things (either at least some of the offenses were committed by one person, whom we refer to as JtR, or none of them were, which makes JtR a myth), we just don't know what it is.

                        The fact the case remains unsolved might not be because the same person didn't commit all the C5, but because they also committed some of these other crimes, and by not including them we're overlooking something. In contrast, it might be because we're including too many. Or, we might have the right set. Any of those are possible, and defendable, and worth considering and keeping an open mind to.

                        But the case is unsolved, in my view, primarily because the evidence we have is insufficient to solve it, no matter which of those three descriptions is true. More, and new, evidence is required, and that may come from a better understanding of the individual crimes, and the events surrounding them, or it may come from a better understanding of the victims (if MJK is ever "identified" and her background revealed, maybe more information on "other Joe" for example, that may lead somewhere - or it may not, right now the information is unknown so we have no idea if it is or is not informative), or it may be new information that arises from a "suspect focused" line of research that results in clear links between the "suspect" and the crimes somehow. Maybe the "suspect file" will be recovered, and a fresh look at that information will reveal something. Who knows, certainly not me as I'm no fortune teller.

                        The one thing I do know, though, is that anybody who states categorically that they know exactly which victims are the only victims of JtR, is overstating their case.

                        - Jeff
                        Last edited by JeffHamm; 11-13-2019, 11:22 PM.

                        Comment


                        • Originally posted by Michael W Richards View Post

                          By including non-Canonical victim locations you provided a map that in and of itself proves multiple killers working within a small geographic area at around the same point in time. Something so often debated by people who want to have some mythical madman as the sole killer in that area at that time.
                          Nobody thinks that all women killed in the Whitechapel area at that time were victims of the same killer, so I'm not sure who you're having these debates with.

                          Comment


                          • Originally posted by Fiver View Post

                            Nobody thinks that all women killed in the Whitechapel area at that time were victims of the same killer, so I'm not sure who you're having these debates with.
                            You havent been reading very closely fiver, there are a bunch of posters here who add victims to the Canonical Group, which by itself cannot be a proven "series". Check some Torso related threads.
                            Michael Richards

                            Comment


                            • Originally posted by JeffHamm View Post

                              There's a few proposed suspects that rank fairly highly. Barnett, Hutchinson, and PC Sagar's suspect (he mentions monitoring a butcher I think it was in Algate, and here on casebook it was narrowed down to the indicated location). In the C5 analysis, he's right in zone 1 (the southern one). About 50% of the time, if the offender is a marauder, they live inside the orange region, and about 33% of the time inside the yellow/pink region (zone 1). So, provided JtR was a local, if your suspect is in there, then this sort of analysis would suggest they are worth further investigation. Just remember, this isn't evidence they were JtR, lots of people live in that area. It doesn't solve the case, and it's not evidence, it is just a suggestion where evidence is probable to be found.

                              Think of it this way, when the police go into to dust for fingerprints, they don't dust randomly. There are locations and things that are more commonly places where you will find prints, so they start there. Random dusting would mean dusting carpets, couches, etc, which are not going to produce a finger print because there's such a low (0%) chance of finding one. They'll look at smooth surfaces, glasses, mirrors, windows, walls, etc, where prints have a higher probability of being found. But that helps to prioritize where to search, and so prints are more likely to be found if you follow that priority list. Specific information at a crime scene will influence that priority list as well (just like specific information about a crime series will produce different maps). But it still requires real evidence to solve the case, and high probability zones, like zone 1 let's say, only contain the offender 33% of the time (although it only makes up 2.5% of the total search space, so much better than chance), and 50% of the offenders are found after only searching around 10% of the total search space. These are much more efficient search strategies, but they are not "proof" for any individual, though if you have a list of individuals (like we do for JtR), it can be a good way to rank them in terms of priority.

                              Just want to make sure it's clear what these provide, as they are often over-hyped in media and entertainment. While that might generate interest in them, in the long run it's counter-productive because when they don't live up to the hype, people view them as useless. They aren't useless, but they aren't magic either.

                              - Jeff
                              thanks jeff
                              totally agree.
                              interesting note: what are the only two direct sources of evidence that explicitely implicate a jew? the GSG and George Hutchinson. and there he is right in the primary zone. and right smack dab in the middle of the kill zone too. hmmmm
                              Last edited by Abby Normal; 11-18-2019, 02:27 PM.
                              "Is all that we see or seem
                              but a dream within a dream?"

                              -Edgar Allan Poe


                              "...the man and the peaked cap he is said to have worn
                              quite tallies with the descriptions I got of him."

                              -Frederick G. Abberline

                              Comment


                              • Originally posted by Michael W Richards View Post

                                You havent been reading very closely fiver, there are a bunch of posters here who add victims to the Canonical Group, which by itself cannot be a proven "series". Check some Torso related threads.
                                I am well aware that there are people who think the Torso Killer was also the Ripper. I am well aware that there are people who add other murder victims to the C5. So far, none of them have attributed every period murder in or near Whitechapel to the same killer. The idea that there were "multiple killers working within a small geographic area at around the same point in time" is the default assumption, held by virtually everyone who has posted here.

                                So again, who are you having these debates with? Who is claiming that every period murder in or near Whitechapel was done by the same killer?

                                Comment

                                Working...
                                X