What ICs are your peers using these days? What languages and operating systems do they favor? Over the next year, we'll post the results of the Embedded Systems Programming survey online and in the magazine. Here's a little background behind the study.
You can't improve what you can't measure. That's the basis behind scientific examination and engineering. We need to measure, to quantify, to identify the world around ourselves in order to better understand it. When we can quantify data it succumbs to mathematics and analysis. We can sort, sift, slice, and dice the numbers to tease out new data. From data comes information; from information comes knowledge; from knowledge comes wisdom.
Every year Embedded Systems Programming conducts a survey of working embedded developers from around the globe. We pick people randomly from the ranks of our subscribers, EE Times readers, and attendees of the Embedded Systems Conference. Overall we poll thousands and thousands of working engineers, programmers, and managers who know the realities of embedded systems development because they're living it. You may have received an invitation yourself. If you participated in our survey, thanks.
If you didn't, prepare for some surprising data. Our survey was lengthy and it asked lots of detailed questions (and sometimes the same question in subtly different ways) about hardware, software, tools, suppliers, schedules, budgets, and more. We asked about the future and the past. We asked about current projects and upcoming ones. We asked about operating systems, open sources, managers, and microprocessors. And thousands of developers responded, providing Embedded Systems Programming with the most detailed and accurate picture of embedded development anywhere.
Beginning with this issue, we're turning that data around and presenting it to you, along with our analysis of what the numbers all mean. Converting data to knowledge.
The results were surprising in some cases—which is good. If you aren't surprised by the results, what's the point in collecting the data? We'll present the results in installments, with each month's issue focusing on a different topic. This month, we look at embedded microprocessors and how developers chose the one they're using now. In other months we'll examine the same question for operating systems, review vendor preferences, rank programming languages, discuss schedules and budgets, and much more.
There's a lot to learn from all this and it'll take us most of a year to get through it. If you have specific questions you'd like to see answered, send me an e-mail at with your request. That'll be a form of data collection in itself: what survey results do readers want to see?
In the meantime, enjoy the issue.
Here are a few questions:
1. Overview of chip architectures in current use, or is the 8051 still the 800lb gorilla?
2. High-level debuggers, or do people still count on “wince and shoot,” that is if it don't work, look at the source code AGAIN, real careful like and make small changes?
3. High-level debuggers, again, or do people use printf, or LEDs?
4. RTOS, roll-your-own infinite while(1) loop, or do you purchase an RTOS?
5. RTOS again, at what price point do off-the-shelf RTOS become a reasonable idea?
6. Interrupt service routines. What percentage stay away from ISRs like the plague, vs. what percentage embrace ISRs?
7. High level languages, what percentage still use: assembler only, assembler/C, C only, C/C++, C++ only?
Ken WadaSr. Embedded Systems ConsultantAurium Technologies Inc.