Tag Archives: Progress Updates

Dogfooding DatPlot

I’ve been fortunate enough to be dogfooding DatPlot on some engineering work I’m currently doing for a client. Dogfooding is a term used when you use your own software on a real-world project to see how it works. So far, I’m very happy with DatPlot. The things I like about DatPlot and what have made me far more efficient than in the past doing similar work:

  • Data Source Swap: It took me 45 minutes to generate a dozen strip chart plots by simply swapping source data once I had the original chart set up, moving the three event lines to the proper event locations and rewording the title information. That’s under 4 minutes per plot on average! Using MS Excel to achieve similar plots previously took me 3-4 times as long, maybe even longer (ok, not quite similar, as generating strip chart plots in Excel is a pain in the rear verging on impossible to say the least).
  • Event Lines: In this case they have come in very handy as I can place them, move them where I want them and simply read off the curve intersection points to extract the desired information. Previously I had to manually read off the data points – often using pencil and ruler on a paper printout (also a pain if the scale is tight) – or dig in the raw numeric data and try to find the exact row I’m looking for where the event occured.
  • Zooming/Panning: It’s very refreshing to be able to dynamically zoom and pan on your chart to get the look you want. I did find myself setting axis scales manually once I knew exactly what scale I wanted to display. It simply provides more exact control this way.
  • Chart Annotation: I’m glad I decided to add a chart annotation feature, as I’ve been able to use it to convey additional information on the chart. This way everything is available electronically on one chart.
I’ve also discovered several shortcoming, some of which I have already addressed, some of which I plan on addressing soon:
  • DPM Saving Mechanism: DatPlot does not save the full data source file (typically a CSV file or similar) as part of the DPM (DatPlot Master) file, it only saves the numeric data data for the parameters plotted in the DPM and the path to the data source file. Why? To keep DPM file sizes small and prevent data duplication. When opening a DPM file, DatPlot expects the data source file to be in the saved location, otherwise it cannot load the full data set, and no new data can be added to the plot. Well, this is not a good solution when DPM files are to be shared with other people using different file folder structures.
    I’m changing the setup where the DPM still does not save the full data source file, but now allows the user to browse and select the data source file when DatPlot cannot find the original data source file, either because it was moved or renamed. Alternatively, you can choose not to update the file location and DatPlot will still show you the saved plot, but you will not be able to add additional parameters to the plot, because the source data is not available.
  • Line Addition: I needed to add a simple straight, hand drawn line on the charts to show a slope. I added this feature to DatPlot.
  • Curve Fitting: I want to get a curve fitting system implemented sooner rather than later as I think it will be helpful and needed.
It’s been a great experience using my own software for a real-world project. I’m still hoping to make version 1.0 of DatPlot by the end of August, all depending on how much time I have to finalize things. Keep checking this blog for updates!

August 2011 Update

DatPlot Changes & License Decision

DatPlot has been in the hands of four beta testers for a couple of weeks now. Through their feedback, I’ve been able to add a few useful features to DatPlot in addition to fixing a few bugs found. Specifically the following changes were made to DatPlot (thank you to all beta testers!):

  • Added ability to add/move/edit/remove text annotations on graph pane surface
  • Added ability to manually set X, Y and Y2 (right Y axis) axis minimum and maximum range values
  • Added ability to use logarithmic scale for X, Y, and Y2 axis
  • Added ability to plot data with partially missing (blank) parameter values
  • Added help documentation
  • Numerous bug fixes….
I’ve also made a decision on how I want to offer DatPlot. I decided to go with a simple freeware license. Basically you will be able to download for free and use DatPlot for your own (legal) needs, however you wish. For the full legal juiciness of the license, you will have to wait until you can download and install DatPlot, hopefully by the end of August 2011. More soon!

Strip Chart Plot Demonstration

The following 7.5 minute video demonstrates how simple and quick it is to take raw engineering data and create a strip chart plot using DatPlot. At the end of the video, the Data Source Swap feature is also demonstrated to give the viewer an idea how multiple charts using similar source data can be rapidly generated once the initial chart is setup.

For best viewing quality, select 720p HD and Full Screen setting after starting the video playback.

 

July 2011 Update

As of mid July 2011, the initial version of DatPlot is feature complete and beta testing has begun. Not only am I in the process of testing DatPlot and squashing bugs, but I also am in the process of having two engineering colleagues use DatPlot in their day-to-day plotting needs. I want to know if DatPlot is a useful and effective tool for them to create their strip chart and scatter/line plots. Feedback so far has been positive.

Going forward, I plan on following the “Release early, release often” philosophy. The idea is to release the initial version with the features most requested by users, but make that version available as soon as possible and as stable as possible. After the initial version is released, I hope to release updated versions on a more or less regular basis with new features added, as requested by users. Some feature requests may never make the cut for one reason or another. But the point is that I envision DatPlot’s future to be guided by the users’ needs based on what they require in their daily work plotting numeric data.

I would love to hear your thoughts on any of this, feel free to contact me at info@datplot.com