JMP Wish List

I would like an option to display the confidence interval for the median in the form of a notched box plot. Could not find out how to do it, did not receive an answer in disucssion, so I am assuming it is not implemented.   As a workaround, diamonds or bands would also be fine, as long as they give the confidence interval for the *median*, not the mean.   Fourboxplots.svg.jpg (Picture source: https://en.wikipedia.org/wiki/Box_plot#/media/File:Fourboxplots.svg) ... View more

A great function to add would be the ability to take a post-hoc analysis (Example Tukey's letter report showing treatment 1, 2,3,4 with corresponding letters A, AB, BC, A) and import those letters directly into a bar graph above the columns.  Post Hoc on Graphs.png   ... View more

Improve the actual neural network module and add a new module for deep neural networks.   Felice ... View more

We are able to specify the data format including specific figures in the Column Settings, and that setting is obeyed in most reports and plots.  However, it isn't obeyed in the Caption Box for Graph Builder.     While it is possible to force control by adding text manually, my idea is that either the significant figures for a certain column are obeyed, or that you can manually adjust the number format for the Caption Box. ... View more

I think it would be neat if we could throw and catch error types like in python/js/ruby since we have classes now.  So if I wanted to only catch specific error types I could do that.    And being able to have a builtin error type that we could have the classes inherit would be great.  Then having the exception_msg be that class with easier to parse information (including traceback with lines) rather than the {"Name Unresolved: bbb"(1, 2, "bbb", bbb /*###*/)}. That way we could do something like  Names default to here(1); try( a = bbb; , show(exception_msg["message"]) // show the relevant message show(exception_msg["caller"]) // what function the exception happened in show(exception_msg["traceback"][2]["message"]); show(exception_msg["traceback"][2]["line"]); ); Similar to this https://community.jmp.com/t5/JMP-Wish-List/Improved-JSL-Stack-Trace-in-Log-on-Error/idi-p/78104   ... View more

I have been a steadfast user of JMP for more than 20 years (in my field everybody has gone to R which to me is a throwback of 30 years but that's not my point today...). Nevertheless, there are a few things that still irritate me somewhat. Here is a list of them, in no particular order:   - Fix the problem that make it impossible to use the Matlab connect functions when JMP is running on a recent Mac.   - INtroduce a GAM modelling platform in the predictive modelling   - Introduce an Ordinary differential Equation (ODE and PDE) solver    - Invest into translating popular R packages into JMP add-ins. This will bring back users.   Just my two cents.... Best, Yves P. ... View more

When using multiple file import, I would like a way to specify how many rows i want to import from the file. ... View more

It would be very beneficial to have the ability to select p' and u' SPC charts in the control chart utility. Especially in this day and age of big data which often times equates to BIG denominators. Our company has a vast number of use cases for "primed" charts and very few options out there in the software realm.  ... View more

It'd be great if you didn't have to go back (from say Graph Builder) to the data table to find and change the Value Ordering properties, by instead simply double clicking on the X-Axis settings and editing the Vlaue Ordering from there.. ... View more

I would like to see a section for an MSA Type I Study with cg and cgk reporting. It's a very simple calculation which can be done manually, but it would be nice to have a way of reporting this with JMP.   Possibly, Expansion of the Measurement Systems Analysis section to include the 5 components of measurement error includes that idea under Accuracy / Bias. ... View more

Is it possible to add an option to automatically obtain AICc (and BIC) in "Nonlinear" to the next JMP versions? This is a big weakness of recent version. ... View more

I would like to be able to label selected points in Graph Builder with new text without having to go through the data table. @Jeff_Perkinson suggested to use right-click -> Rows -> Name Selection in Column... to get most of what I'm looking for, which is great.   Building onto that suggestion, I think the following would be a great enhancement for "Name Selection in Column": Proving tick options to label selected rows and the newly created column, as shown in the bottom of the next graph.   NewLabel Selection.JPG Result: NewLabel Selection Result.JPG   JMP Example Data: Tablet Measurements ... View more

When a column with images is set as a label, hovering over a datapoint in a graph shows the resp. image. It would be nice if it was possible to do this for multiple images (in different columns). (Setting several columns as label doesn't work, only the image of the first column is shown.) ... View more

What inspired this wish list request? We build a production JSL add-in on top of the native Data Filter() platform. Our filters commonly have 10–15+ columns (used for cluster/exclusion editing). In JMP 18, we could reflow the filter list from one long column into multiple columns by calling <<N Col(n) on the filter's internal Lineup Box(2) , which made the panel usable. In JMP 19, this call silently has no effect anymore — the filter list stays stuck in a single very long column no matter how many filters are added, even though other messages like <<spacing() on the same object still work. JMP Support (case 00344860) confirmed there is currently no supported way to control this layout in JMP 19. Example: a Data Filter with 15 columns for cluster/exclusion editing becomes a single scrollable column instead of a compact 3-column layout, making it very cumbersome to use.     What is the improvement you would like to see?  Add a supported and documented option (via JSL and/or the UI) to arrange the Data Filter's filter list in a specified number of columns, similar to what was achievable in JMP 18 via <<N Col(n) on the internal Lineup Box. Ideally exposed as an official message such as <<N Col(n) directly on the Data Filter object itself, so it doesn't depend on internal, undocumented report structure that can change between versions. Why is this idea important?  This is a must-have for us: we rely on the native Data Filter platform specifically for its built-in counts, histograms, and select/show/include modes. Without a multi-column layout option, large filters become impractical to use, and the only alternative would be rebuilding a custom filter UI from scratch — a significant development cost and a loss of native functionality. A documented column-count option would benefit any user or add-in developer working with Data Filters that have many columns. ... View more

The JMP graph builder is lacking in customization compared to other popular graphing software (graphpad prism for one). One major limitation is the inability to customize the markers to have solid fill and outlines. I searched for a way to do this but was unable to find an easy solution (other than to just have a solid white fill with a hollow shape). Full marker customization to this extent should be relatively straightforward and would add a lot of benefit to the visualization.       ... View more

Offer 24/7 technical support to better support customers running JMP Live in continuous production environments. Around-the-clock support would ensure timely response to critical issues, minimize downtime, and provide greater reliability for organizations that depend on the platform at all hours. ... View more

What inspired this wish list request?  One of the biggest bottlenecks in using JMP is creating structured formatted output to use in technical reports and presentations.  The current best practices include either a lot of scripting, journaling, and/or copy/paste actions.   It would be nice to be able to define a report structure with elements from JMP outputs included at specific locations, along with formatted text.     What is the improvement you would like to see?  Markdown is a well established approach for simple general construction of formatted output.  I would like to see a "JMP Markdown" capability, where I can create a markdown document that includes formatted text, JMP output (specific graphs from a report, specific tables from a report, and specific values from a report) in defined locations in the document.  The output would be a markdown document which could then be converted to various formats particularly PDF, Word DOCX, and Powerpoint PPTX Potential Workflow: Create blank markdown In parallel/sequence: Create JMP analysis report Select output from analysis report (with selection tool) and insert placeholder from the output in the markdown document. Add text with formatting tags  Save markdown file Run markdown conversion to desired format A GUI to facilitate creating the markdown formatting would be nice, also.  WYSIWIG interface in the markdown document would also be nice.   I could see this also be something that could be published to JMP Live (by converting the markdown to HTML). Why is this idea important?  This would greatly improve the ability to create formatted output from JMP that meets the specific needs of the user without extensive copy/paste and editing.  It would also enable standardized reporting and automation of analysis all the way from data ingestion to final reporting.     ... View more

What inspired this wish list request?  There are often times when it would be nice to have units appear as we might write them by hand, e.g. g/cm 3 instead of g/cm^3, or even with some text as a subscript: sample BKG  , but JMP does not make this easy. In fact, they seem to have taken away the copy/paste feature where you could write the text in Word and then copy/paste it into JMP and the formatting would be brought over into JMP.   What is the improvement you would like to see?  I'd like to see JMP add this text editing capability into JMP (and into JSL in particular for customization).   Why is this idea important?  It's important because it helps to improve messaging to the "consumer". People like to see additional information as subscript or superscripts. No one likes to read smapleBKG or g/cm3, they want the improved visual cues that comes with being able to move ANY text to superscript or subscript.   ... View more

What inspired this wish list request?  When analyzing data using this platform with a By variable, "the make combined data table" option does not include all the analyses if they are different types. For example, if some use Bayesian variance component estimates and some do not.   What is the improvement you would like to see?  Include all types of estimates into one data table. Consider adding a column to tell what type of estimate is used.   Why is this idea important?  It will truly deliver what it promises and will Make a Combined Data Table.  :)   ... View more