This is an approximately 10 minute condensed version of the plenary talk from the 2024 JMP Discovery in Manchester. It's posted here as both a resource for JMP 18 content and as an example of a presentation technique that can be used in certain situations with medium to large audiences.
Here's a transcript from the presentation that can be used as a starting point for a script if needed:
I did have a little, I guess a comment on it from the beginning here, just so that we know the context of it kind of taking the spirit of Manchester to heart.
This is what I would call a stage one demo. This is a marketing demo where the goal is to get people into the bucket, get them excited, get him foaming at the mouth to come and talk to the sales team and get them qualified to where we can actually do a more focused Technical demo where we can actually know what the situation is - tailor our notes about what new, what's new in 18, to the specific pain in the customer and that's really what we need to do with this particular product is we need to to kind of speak to what else is doing next is focus in on what aspects of the software the new features are going to resonate with our specific audience.
And this is one of those very few places where we just kind of take a wide net because we want to get as many people into the funnel as possible to then move on to that.
Uh, in that spirit I shamelessly mooched from John Sall's Discovery talk. So, that's where this is a 10 minute version of the plenary talk from Manchester.
Peaks, pipes, profiles, presets and Python.
And let's go ahead and dive right into it.
The first thing that we want to talk about is the peaks.
This is a topic that's near and dear to my heart because of spectroscopy and what we've done is in functional data Explorer.
We've added a new feature in modeling that allow us to go in and add peaks manually to be able to go in and extract information really quickly.
This is all completely interactive like what we would expect to see in our normal.
People work peak finding tools.
You can see that we highlight the region, we get the start, we get the end, we get location and height information and we get area under the curve.
All those standard parameters that were used to seeing the one thing here that we need to remember is that we're gonna be doing this a little bit differently than we would with other types of methods.
And right here, automating and running through many, many different peaks all at once, being able to get all of the same information out of them simultaneously.
So we've gone from manually grabbing peaks or having to figure out what distribution those peaks come from to one click, and we got the majority of the information that we need.
So again jump 18, we add an expanded set of spectroscopic capabilities to both fit curve.
I didn't show that today, but all the the highlight features are in functional data Explorer as well.
Next, pipes types.
I mean data connectors we used to call bringing data into a system or routing it through a bit of logic.
A pipe.
When I was listening to my dad speak about learning latex or not latex, the Linux he talked about piping I things together using a little pipes so data connectors we're going to talk about pipes.
We've had those for forever, but what we've done now is actually made it really easy to standardize across an institution and also share that information as well.
So here you're gonna see me.
Go ahead and take a bunch of data connectors that I've already built and I'm going to start by building a little library that I can send across to everyone else in my group.
So I wanna share my biochemistry while I wanna sell my share, my excel and my mold map.
I can save those out as an add in that will then be automatically installed when I send it into someone else and they open it up and jumping team if I want to create a brand new data connector, I can use an existing data, connect an existing data connector or one of the prototypes.
Here you can see me.
I've got access to all these different pieces of information that I can find, tune and customize to the needs of a particular database or the configuration of a particular database.
This is a bit more for your database walks the guys who really know what they're doing in the databases, but again we've got access to quite a bit of information would come to this.
This info we can also test it to make sure that everything behaves the way we expect it to.
So again, key point here is we've gotten the new bility to be able to manage tune and more important most importantly share data connectors across teams of businesses.
I can't tell you how many customers I have that have different versions of a driver or have different versions of a situation where just having this capability built in will erase a lot of issues that we have across our across that customer base.
Next, profilers.
So profilers have been in jump for forever, but what we've done in 18 is we've gone and done some a little bit of refresh in there.
Well, first thing, that's kind of out of the gates is we built into the ability of have prediction intervals so different than confidence intervals.
They give us a different way of looking at problems and use and a different set of numbers to use in forecasting with the green numbers at the bottom and the green bars on the profile.
Let's turn those off for the moment.
For some decoration for some augmenting of the graphs, we've added the ability to overlay the interactions.
You can see we did that incredibly quickly and this again, we're actually going from jump running Python, pulling the image back.
So we're leveraging the speed and power of both of those tools to help us quickly visualize a fairly complex interact fairly complex algorithm.
So again, the five kind of marquee features of jump 18 peaks, pipes, profiles, presets and Python.
