SWGFAST Standards for Conclusions Draft for Comment

[Boyd Baumgartner]

*edit* I updated the title

Any one care to offer up thoughts on the new Standards for Conclusions Draft for comment?

Let me state the disclaimer of ' I know the discussion is best directed towards SWGFAST considering it's a draft for comment'. Specifically, I'm curious as to what people think of the flowchart.

If you haven't seen it, there's a link in The Weekly Detail, but I'll post it again:

http://www.swgfast.org/documents/method ... FT_1.0.pdf

My personal thoughts, are that it's a good first step. I have never seen the process outlined in such a fashion, and if it gets people thinking about it in a new way, that's good. I personally am a flowchart type nerd.

From a critical aspect, I think that it suffers from:
a visual orientation perspective (difficult to follow due to lack of alignment),
non conventional flowchart designations (Y/N's coming off of the rectangles (which are typically reserved for predefined processes and not decisions which are what the diamonds are reserved for)),
the flowchart type (it's a system flowchart represented as a general flow chart),
some less than best organization of thoughts (the interaction between boxes 540,550 and 600 to be exact),
the lack of information contained in the Verification process,
the lack of recognition of the 'holistic' elements (only levels 1,2,3 are represented),
odd terminology (box 100 says 'latent print enters the examination system' but then goes into the analysis phase. When in the profession, examination and comparison are used interchangably),
lack of clarity that could be interpreted as less than best practices (box 75 specifically)

[BobS]

Boyd,

What do you do for a living?

[Boyd Baumgartner]

I'm the drummer for Spinal Tap.

[David Fairhurst]

I'm going to address one of Boyd's points on the flowchart and then make general comment on the guidelines.

non conventional flowchart designations (Y/N's coming off of the rectangles (which are typically reserved for predefined processes and not decisions which are what the diamonds are reserved for

As far as I can see the only example of this box 50 which I interpreted as a mistake; there's a Y arrow but no N leading from that box. There's an N leading in from 60 but that's perfectly acceptable.


My only Comment, so far, about the guielines is on the assertion that "quantity" is the number of minutiae.
As I have previously put forward (http://clpex.com/phpBB/viewtopic.php?f=2&t=1311&start=0) the size or area of a print is a much better indicator of the quantity of detail. I hope no-one would disagree with the idea that a larger print contains more detail (more ridges) than a smaller one that contains the same number of minutiae.

I mean, using swgfast's own quality grades, let's consider a Medium-High quality latent print. It has 2 minutiae; each a ridge ending in the same orientation. There are 3 clear ridges flowing without event on the outside of each of the minutiae and 3 between them. Quality of 3LD is limited and your're inconclusive.

Now take another print. This one is also Medium-High in quality. This one also has 2 ridge endings, each 3 ridges from the edge of the print. This time there are 9 intervening ridges. I've just given you a whole load more detail, both 2LD and 3LD to work with without increasing the number of minutiae.

I know it takes a lot of time to get people to get away from numbers. We made a huge step when we realised that a numerical threshold was not valid. It seems that we've been looking around for something else to do with this number that we can get from our examination.
We thought "Well, if the number isn't everything, then it has to be something else; doesn't it? Wait a minute, now were Quantitative-Qualitative, so the number must be the quanitity. That's a good use for it. After all, I'm not going to go to all the trouble of counting the minutia and then not use it for anything. That's just a waste of time."
Why not just stop counting. It'll save you a whole lot of time.

Sorry that turned into a bit of a soapbox moment, but I'm going to post it anyway.

[briano]

What Boyd was trying to say,

He thinks it suffers from:

It’s difficult to follow due to lack of alignment,
uses non-conventional flow chart designations,
It’s a system flow chart represented as a general flow chart,
lacks organization of thoughts,
the verification process lacks information,
no recognition of the holistic elements,
odd terminology,
box 75 really needs to be looked at.

Next week we will isolate and capture a fast moving neutral atom. Dang, already done.

[Boyd Baumgartner]

Here Brian, maybe this will clear things up for you...

Flow.jpg

[briano]

There you go!

Elephant:

Mouse built to government specifications.

[Bill Schade]

Boyd, I like your thinking. It's pretty straight forward but I think you left something out

[Boyd Baumgartner]

Bill, you're always one-upping me. And is that Visio I smell? I had to use the shapes in Excel.....

[Bill Schade]

Boyd, I'm stung that you think I was "one-upping" your excellent work.

I thought I was building on the prior work of another the way "scientists" do

[Boyd Baumgartner]

Bill, you must be mistaken. As Baumgartner et al. JFI 151 (53), 2007 states," 'True Scientists' (whatever that means) just circularly self reference their previous non-validated works."

[Les Bush]

Apologies fellas for dropping in on your verbal judo. The SWGFAST documents reveal a lot about the differences in the approach taken by their experts compared with those of the IAI. The most important I think is the absence of recognition toward statistical modelling and probability. Thankfully the SWGFAST experts have decided to deal with what is validated by the huge amount of experience in proving personal identity. Hopefully the IAI will similarly drop the inclusion of conditions allowing for probability in conclusions. If in the future the probability model/s prove they are rigorous enough to be reliable in real world examinations then they deserve to be accepted. Until then SWGFAST are stating their support for what we currently rely upon and it affirms what we already know. Individualization of humans by means of fingerprint examinations is not novel but an accepted and proven reality. The SWGFAST experts use plain English expressions with an economy of words. They state that the likelihood of an impression being made by a different source is so remote that it is considered as a practical impossibility. I'm interpreting the word 'practical' to mean 'physical' in the sense the friction skin pattern of detail is not repeated and unique. They use one expression only for Inconclusive and incorporate Insufficient which is reasonable since the outcome of these examinations is the same, no Hit.
I also like that they have defined Sufficiency better than IAI who include the word 'adequate'. By stating the condition of Sufficiency to mean a product of the quality and quantity of objective and observable data they have incorporated what it is we do and record. The flow chart is an example of what goes on in the black box of the experts mind so Simon Cole will now know it isnt just a 'leap of faith' but includes analysis and measurement. Well done SWGFAST the USA is well placed to improve the science.