Re: [Discuss-gnuradio] Why Isn't GNU Radio Used More?

[Vijay Pillai]

I completely concur with what you wrote below and what Scott Johnson wrote some time ago.

USRP is an incredibly powerful platform and substantially low cost - I am somewhat befuddled by how it has not attained greater prevalence but at least some of the reasons are plainly obvious

- incomplete (or in some cases non-existent) documentation
- as a hardware engineer, my time is better spent in getting quickly to using the thing than to learn the nuances of Python
- it seems the vast majority of users today are software programmers, who may not be that averse to spending copious amounts of time on C++..

I also suspect that the folks at Ettus are somewhat stretched thin by all the support work. Having said all that I will add that GRC is very good (easy to learn) and can do most things that I need. It does have stability issues, as I frequently get some error dialog or the other.

Best regards,
-Vijay
--- On Mon, 5/9/11, Michael Dickens <address@hidden> wrote:

From: Michael Dickens <address@hidden>
Subject: Re: [Discuss-gnuradio] Why Isn't GNU Radio Used More?
To: "GNURadio Discussion List" <address@hidden>
Date: Monday, May 9, 2011, 11:51 AM

On May 9, 2011, at 11:12 AM, Marcus D. Leech wrote:
> I think there's a significant community out there that learned DSP techniques inside the envelope of Matlab/Simulink, and that's what they're comfortable with.

True; that's how I did (MATLAB; Simulink wasn't around yet).  I'd take that a step further: I'd guess that 90+% of -potential- MATLAB / Simulink / Octave / PyLab / LabVIEW / GNU Radio / GR Companion users just want the system to work as provided, without having to implement anything further beyond basic scripts -- meaning, for GNU Radio, they don't want to have to delve into writing or deciphering C++ blocks, and SWIG and XML glue necessary to use them in Python and GRC.  I don't know if GRC provides this what those 90+% need just yet in terms of blocks; good "help" files / system are a necessity, as pointed out by Scott.

> To change this, Gnu Radio has to appear in more places in academia, so that graduating engineers have already been exposed to it, and find it "natural".

I think that GNU Radio is pretty well represented in many areas & those are growing pretty quickly right now.  I don't think GNU Radio will grow in academia until it is more accessible and meets the needs to those 90+%.

I think to succeed in the way MATLAB did, GNU Radio needs to provide functionality and documentation for those 90+%, without requiring them to learn "anything GNU Radio" more than the GRC GUI.  For many potential users, writing C++ / Python / SWIG / XML is too daunting to even consider -- whether because of fear of the unknown, because they are not well documented, or just because there are too many potential areas for "difficult to debug" mistakes.

IMHO what GNU Radio needs is a stable API, internal code documentation, help files for "how to use GNU Radio", and then a well-written "how to use GNU Radio" book that includes examples of using GRC.  I think this combination would bring GR/C to the masses -- those 90+%. - MLD


_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio