Category Archives: Tutorial

Describing the progress of projects – a drama in 5 acts

Some time ago, I was looking for a nice, vivid way to describe and present the progress of our projects we do (for a meeting with VIPIs – very important PIs). Usually, when people have to come up with this, they present some numbers – percentages to show the progress. I don’t like this, because of the following reasons:

  • Percentages are often given without greater context. Ask yourself: What is the measure behind it actually? Are we talking about the (relative) amount of time, work volume, …?
  • Percentages are often thrown at you without giving a reference/base (i.e. the total time, work volume, …). It is actually almost impossible to give this information – unless you already completed the project. Then, and only then you would KNOW how much time or work a project costs (has cost). This means also that percentages are only right at the two extremas: 0% (not started) and 100% (completed). The rest is pure assumption.
  • Individual tasks that have to be done for progressing are hard to compare and judge in terms of the actual contribution to the project. You just have overcome an issue – is the plus in percentage now 50%? 30%? 53%? Has every task the same weight?

You will find out for sure in the end. When you have done the project. But this is never the point when you show the progress of a project. You show the progress if you are RIGHT IN the project, don’t you?

What a drama!

Let me tell you something about the classical drama, common in France during the 17th-19th century. It consists usually of five acts:

  1. Exposition (Introduction/Prologue): Setup story, character, setting
  2. Complication/Conflict (increase until climax/rising action)
  3. Peripeteia (climax): Inversion of circumstances/feeling of hero
  4. Retardation (falling action, decrease until conclusion): Slow part from climax to final resolution
  5. Dénouement (conclusion/resolution): final status of all persons (damned/happy end)

When you read this, does it not come to your mind that every project you did, do, and will do, is actually a drama? Replace hero with researcher, story with hypothesis, character with molecules, and settings with instruments, and so on…

Hence, I decided to use this structure to show the progress of our projects. First, let me rephrase the terms a little bit:

  1. Exposition: Setup of hypothesis, review of literature, first ideas, plans, …
  2. Complication: First experiments, revealing issues (i.e. complications), first attempts to solve these issues, … This is the part of the project in which you feel that the project is controlling and working against you.
  3. Eureka: developing solutions to issues, overcome them (inversion of complications). This is the part of the project in which you gain control of the project. Step by step you increase your confidence.
  4. Protocol: Slow part, in which all experiments are systematic and „nice“ (i.e. according to protocol). Here, you control the project. You wonder from time to time why you had problems in the first place.
  5. Dénouement: Publishing (any form), planing next subproject/sequel, …

I assigned colors to these acts (or phases as I will call them from now on) and show the progress of projects like this:

2016.08.05 - project overview

As you can see, I use a gradient between the individual colors to visualize the progress. Project 5 is practically finished (just published the paper for this) – it has the colors of all phases. Project 7 is in the protocol phase. Project 2 and 6 are both in the Eureka phase. However, you might see a little difference in the shading of the gradient. The achievment of a phase is a fluent process. Here, I wanted to express that we already worked a little longer/more in the Eureka phase on Project 2 than on Project 6. Project 3 and 4 have not really been started yet (we practically know only that these projects exists and read a paper or two on them). For Project 1 we just started with the very first experiment.

How to setup the gradient? Well, since we describe here a process that has already taken place, we can simply use the spend time on every phase to balance the gradient. It does not need to be too accurate, though. If you spend a week on the hypothesis and literature, a month for the initial experiments, and now just started the third phase (let’s say for two weeks), then your gradient ratio would be 1:4:2 or 14%:57%:29%. Simple, isn’t it? And no more GUESSING.

I realized, however, that just colors do not work alone – I needed kind of a “progress bar”. The length of the progress bars you see in the above picture represent the progress according to the total phases (i.e. 2 out of 5, 3 out of 5, etc) – wherever I feel it fits better in. This can be seen for Project 1, which just finished phase 1 and already has a slight shade of phase 2. However, the length is still 1 out of 5 (20%).

Note that I also listed some minor projects in the slide, which I did not evaluated in terms of progress. The reason is that these projects are always “present”. We are always working on new mathematical models or little things to improve our daily work.

I just wanted to share this here in case anyone else finds this useful 🙂

PC-222 multimeter, RS-232 protocol

For a project we needed to read out our PC-222 multimeter. It supports RS-232, but the protocol is not documented at all and the software is proprietary. After a little research on the web, we found some information in a comment on Amazon. With this and a little trial–and–error we found out how it works.

PC-222 multimeter

PC-222 multimeter

Continue reading

Howto: Replacing axes in Origin without changing the data

Recently, my co-worker and me had to replace an axis of one of our plots in Origin [1]. We measured a signal over time and, actually, wanted to just add a little offset without changing the data. We didn’t want to go through the get-a-nice-matrix-out-of-your-data-for-3D-plotting-torture, again.

Continue reading

References

  1. " OriginLab - Origin and OriginPro - Data Analysis and Graphing Software " http://www.originlab.com/

Howto: Fit sigmoid functions in OpenOffice Calc and Excel

Last week, I posted an article about sigmoid functions and how to use them. Nevertheless, it is hard to guess the parameters for a given problem. So, people use software such as Origin [1] or QtiPlot to fit.

Personally, I use Origin/QtiPlot only for plotting and Excel/OO–Calc for evaluation/calculation, because both programs are much more comfortable and powerful. However, both lack the possibility to fit sigmoid functions, automatically.

It is possible to do such fits for nearly any function using the solver, though, but only a few people I’ve met knew how to do this or that it is even possible. But, I think the solver is a very handy feature and, therefore, I want to give here a short introduction into using it for fitting a sigmoid function to a set of data.

Continue reading

References

  1. " OriginLab - Origin and OriginPro - Data Analysis and Graphing Software " http://www.originlab.com/