[time-nuts] Stable32 now available
Dr. David Kirkby
drkirkby at kirkbymicrowave.co.uk
Wed Jan 3 05:24:08 EST 2018
On 3 January 2018 at 04:09, Tom Van Baak <tvb at leapsecond.com> wrote:
> I asked Bill for clarification and here's some of what he shared:
> > Tom:
> > I’m glad that the word is getting out that Stable32 is now freely
> > My donation to the IEEE UFFC included all source code, and it is up to
> > to decide whether and how to make it available. That has not yet been
> > My motivation was to see that it lived at least in its present form
> > my having to remain involved.
That's really good of Bill.
I will continue to support the commercial
> > version for a year more, but not the free one.
> > The IEEE UFFC Stable32 distribution is released under a form of the “MIT
> > License” which is as free as can be. That was made possible last year by
> > Scientific Endeavors Corporation who agreed to end my obligations under a
> > royalty payment agreement for the GraphiC scientific plotting functions
> > Stable32 uses. No other such issues exist for the Stable32
> distribution, so
> > it became possible to make it free.
Nice of Scientific Endeavors Corporation.
> > The Stable32 source code is organized into two basic parts, the
> top-level Windows
> > user interface and a DLL that contains the core analysis functionality.
> It is
> > likely that the latter is the more valuable for future versions and
> other purposes.
> > For example, I have ported that Windows FrequenC.dll (which is
> distributed with
> > Stable32) to Linux as a libfrequenc.so shared object library that can be
> used by
> > GCC/G++. I have also created a wrapper function for that so it can be
> used with
> > Python. I hope to make these (and the critical FrequenC function
> > available soon.
Several technologies come to my mind, that could give a multi-platform
1) WxWidgits in C++
2) Qt in C++
3) Mathplotlib in Python.
5) Command line version for UNIX and UNIX-like systems.
> I hope this explains things a bit. The Stable32 distribution package is
> now freely
> > available for all to use. The Stable32 source code is under the control
> of the
> > IEEE UFFC AdCom. I plan to make the core Stable32 FrequenC Library
> > and functionality available for Windows, Linux and Python.
I would not be surprised if Poul-Henning Kamp would be interested in
porting to FreeBSD too.
Solaris is pretty much dying since Oracle bought Sun, so there are
arguments for not bothering with a Solaris port. But from my experience,
testing software on multiple platforms often highlights bugs that exist in
all platforms, but have not been discovered. I lost count of the number of
hours I spent porting Sagemath to Solaris, but a number of bugs were
discovered by me whilst porting to the Solaris. Those bugs existed on
Linux, but had just not manifested themselves.
> > Best regards,
> > Bill Riley
> My comments:
> 1) I use both Stable32 and TimeLab equally; and now that Stable32 is free,
> everyone can now enjoy both of them too. They overlap somewhat, but each
> has its own set of strengths and target audience. Both show evolutionary
> bloat by now but for the small subset of features one typically uses the
> learning curve is not high. Each includes a comprehensive user guide. Both
> are native Windows apps, and run under emulation on Unix.
I assume, but perhaps are mistaken, you mean Linux here. Strictly speaking,
a system is not UNIX unless certified by The Open Group. Linux is certainty
not UNIX, and in several cases the code breaks UNIX compatibility. For
example, on Linux, the df command reports sizes in kb, but for UNIX system
it is 512 byte blocks. I don't know, but I suspect FreeBSD conforms to the
UNIX standard more than Linux. Also on Linux, people tend to use tools like
drkirkby at hawk:~$ which ls
to find the path to an executable. But 'which' is not part of the UNIX
standard, so may not be available on a UNIX system. But the following
gives the same information
drkirkby at hawk:~$ command -v ls
but works on UNIX systems in addition to Linux systems.
More information about the time-nuts