I want to join the others here in showing appreciation for Daryl's efforts
in maintaining Gempak. Even now it continues to serve us well, and I still
don't see a day where Gempak will fully go away. You're absolutely one of
the sharper tools in the shed and seriously, thank you for all you do!
I've tried to pitch in on Gempak's github as time allowed but that's been
tough for me this past year. I'm still watching the issues and chime in
when/how I can, and agree with Daryl that all are welcome to contribute.
Something I've been kicking the tires on lately is to create a Gempak
expert AI agent. So far it's just been to supplement the Gempak docs &
core code to local LLMs such that they can be referenced during chats
(RAG). I want to try fine-tuning/training local models on those docs, the
code base, and even gembud & gempak support exchanges and github issues.
There is a wealth of intel there, and I've seen encouraging signs that we
can create domain experts for tools like Gempak that could help us not only
use it, but possibly modernize it as Daryl suggests. I haven't gotten far
on this side-project yet but I am encouraged.
Best,
-Mike
P.S. Speaking of netcdf, this is a _little_ off topic for Gembud but I
don't care... Is anyone working on a WASM build of netcdf? This is
another project I'm trying to devote time to, building browser-native
utilities for our data, and I'd be interested to hear if others are working
on something similar.
On Thu, Mar 5, 2026 at 11:38 AM Tyle, Kevin R <ktyle@xxxxxxxxxx> wrote:
> Amen to that, Robert!! Daryl, your ongoing work to ensure GEMPAK never
> "gp"ends is hugely appreciated!
>
> Now, given your entreaty to invite more community participation via
> GitHub, who wants to be the first to add "GEMPAK needs an Apple Silicon
> build" as an Issue? (this is not a rhetorical question, but maybe it is).
>
> Speaking of issues, I'm sure myself and the rest of the community would
> strongly take issue with your self-descripton of you being "one of the
> least-sharp tools in the shed"!
>
> Thanks again!!
>
> Best regards,
>
> Kevin
>
> _________________________________________________
> Kevin Tyle, M.S.; Manager of Departmental Computing
> NSF ACCESS Campus Champion
> Dept. of Atmospheric & Environmental Sciences
> UAlbany ETEC Bldg - Harriman Campus
> 1220 Washington Ave, Room 419
> Albany, NY 12226
> ktyle@xxxxxxxxxx | 518-442-4578 | @nywxguy | he/him/his
> _________________________________________________
>
> -----Original Message-----
> From: gembud <gembud-bounces@xxxxxxxxxxxxxxxx> On Behalf Of Mullenax,
> Robert R. (WFF-820.0)[PERATON INC]
> Sent: Thursday, March 5, 2026 11:34 AM
> To: Herzmann, Daryl E [AGRON] <akrherz@xxxxxxxxxxx>; Mullenax, Robert
> (PERATON) via gembud <gembud@xxxxxxxxxxxxxxxx>
> Subject: Re: [gembud] [EXTERNAL] Re: Problem compiling GEMPAK 7.20 on RHEL
> 9.10
>
> Hello Daryl,
>
> Thank you so much for all you do for the community!
>
> Regards,
> Robert Mullenax
>
>
> Robert Mullenax
> Meteorologist
> Columbia Scientific Balloon Facility
> Peraton
> 1510 E. FM 3224
> Palestine, TX 75803
> 903-723-8037
> robert.r.mullenax@xxxxxxxx
>
> -----Original Message-----
> From: Herzmann, Daryl E [AGRON] <akrherz@xxxxxxxxxxx>
> Sent: Thursday, March 5, 2026 10:24 AM
> To: Mullenax, Robert R. (WFF-820.0)[PERATON INC] <
> robert.r.mullenax@xxxxxxxx>; Mullenax, Robert (PERATON) via gembud <
> gembud@xxxxxxxxxxxxxxxx>
> Subject: Re: [EXTERNAL] Re: Problem compiling GEMPAK 7.20 on RHEL 9.10
>
> CAUTION: This email originated from outside of NASA. Please take care
> when clicking links or opening attachments. Use the "Report Message"
> button to report suspicious messages to the NASA SOC.
>
>
>
>
> Hi Robert,
>
> To followup on this, I am thankful you pointed out this problem and it
> provoked me to (hopefully) fix the Github repository CI so that these types
> of failures don't happen again. For the rest of the interested GEMPAK
> Users here a bit of commentary on this situation.
>
> - I have never liked the idea of GEMPAK's Makefile system silently
> proceeding after encountering errors. I can see the point that GEMPAK is
> so massive that having a few packages fail may not mean much to most
> users. The year is 2026 though and yours truly has taken steps to stop
> this madness going forward. The github main branch has many more instances
> now were compilation failures will stop the full build.
>
> - While NCEP continues to make NAWIPS releases, the code changes are
> rather small and I am not as worried anymore about merging these code
> changes back into GEMPAK, so the idea of Community GEMPAK diverging more
> from upstream NAWIPS is no longer a non-starter to me.
>
> - I was lazily not generating binary builds of GEMPAK releases, but will
> try to start doing that going forward. I have a ways to go on this yet,
> but am hoping to make a 7.20.0.1 release with binary artifacts available ,
> famous last words.
>
> - I have also started to consider replacing various extlibs content with
> user/system provided libraries. This may not be practical for ancient
> library versions like netcdf and hdf5, but I had some luck with png,
> libxslt, and libxml2. None of this has happened on github yet, but if
> others listening here are interested....
>
> I would encourage folks to collaborate on GEMPAK on github. I'm one of
> the least sharp tools in the toolbox, but enjoy tinkering with GEMPAK and
> would like to see its legacy continue!
>
> daryl
>
> ________________________________________
> From: Mullenax, Robert R. (WFF-820.0)[PERATON INC] <
> robert.r.mullenax@xxxxxxxx>
> Sent: Tuesday, March 3, 2026 3:23 PM
> To: Herzmann, Daryl E [AGRON]; Mullenax, Robert (PERATON) via gembud
> Subject: RE: [EXTERNAL] Re: Problem compiling GEMPAK 7.20 on RHEL 9.10
>
> Hi Daryl,
>
> dfmfrq.o is missing from gempak/source/diaglib/df/Makefile and pdmfrq.o
> is missing from gempak/source/diaglib/pd/Makefile
>
> I added those to the respective Makefile and recompiled and it seems to
> have all worked.
>
> Thanks for pointing me in right direction!
> Robert Mullenax
>
> Robert Mullenax
> Meteorologist
> Columbia Scientific Balloon Facility
> Peraton
> 1510 E. FM 3224
> Palestine, TX 75803
> 903-723-8037
> robert.r.mullenax@xxxxxxxx
>
> -----Original Message-----
> From: Herzmann, Daryl E [AGRON] <akrherz@xxxxxxxxxxx>
> Sent: Tuesday, March 3, 2026 2:33 PM
> To: Mullenax, Robert R. (WFF-820.0)[PERATON INC] <
> robert.r.mullenax@xxxxxxxx>; Mullenax, Robert (PERATON) via gembud <
> gembud@xxxxxxxxxxxxxxxx>
> Subject: Re: [EXTERNAL] Re: Problem compiling GEMPAK 7.20 on RHEL 9.10
>
> CAUTION: This email originated from outside of NASA. Please take care
> when clicking links or opening attachments. Use the "Report Message"
> button to report suspicious messages to the NASA SOC.
>
>
>
>
> Hi Robert,
>
> My initial thought is that dfmfrq is missing from
> gempak/source/diaglib/df/Makefile COBJS
>
> Will have to research this some more!
>
> daryl
>
> ________________________________________
> From: Mullenax, Robert R. (WFF-820.0)[PERATON INC] <
> robert.r.mullenax@xxxxxxxx>
> Sent: Tuesday, March 3, 2026 2:25 PM
> To: Herzmann, Daryl E [AGRON]; Mullenax, Robert (PERATON) via gembud
> Subject: RE: [EXTERNAL] Re: Problem compiling GEMPAK 7.20 on RHEL 9.10
>
> Hi Daryl,
>
> No this is a new install of RHEL 9 and GEMPAK has not been built before on
> this machine. After the first failed attempt I just wiped the GEMPAK
> directory and unzipped main.zip again.
>
> No, the only errors are the undefined reference below referring to
> df_mrfq. Dfmfrq.c does not show in make.out.
>
> Thanks,
> Robert
>
>
>
>
>
> Robert Mullenax
> Meteorologist
> Columbia Scientific Balloon Facility
> Peraton
> 1510 E. FM 3224
> Palestine, TX 75803
> 903-723-8037
> robert.r.mullenax@xxxxxxxx
>
> -----Original Message-----
> From: Herzmann, Daryl E [AGRON] <akrherz@xxxxxxxxxxx>
> Sent: Tuesday, March 3, 2026 2:17 PM
> To: Mullenax, Robert R. (WFF-820.0)[PERATON INC] <
> robert.r.mullenax@xxxxxxxx>; Mullenax, Robert (PERATON) via gembud <
> gembud@xxxxxxxxxxxxxxxx>
> Subject: [EXTERNAL] Re: Problem compiling GEMPAK 7.20 on RHEL 9.10
>
> CAUTION: This email originated from outside of NASA. Please take care
> when clicking links or opening attachments. Use the "Report Message"
> button to report suspicious messages to the NASA SOC.
>
>
>
>
> Hi Robert,
>
> Is the file `/home/gempak/GEMPAK/os/linux64/lib/gemlib.a` old from a
> previous build? Otherwise, do you see an earlier error around compilation
> of `dfmfrq.c`
>
> daryl
>
> ________________________________________
> From: gembud <gembud-bounces@xxxxxxxxxxxxxxxx> on behalf of Mullenax,
> Robert R. (WFF-820.0)[PERATON INC] <robert.r.mullenax@xxxxxxxx>
> Sent: Tuesday, March 3, 2026 2:09 PM
> To: Mullenax, Robert (PERATON) via gembud
> Subject: [gembud] Problem compiling GEMPAK 7.20 on RHEL 9.10
>
> Good afternoon,
>
> I am having an issue compiling GEMPAK 7.20 on RHEL 9.10. The build
> completed and I did an install, but when I open NTL I found that NMAP2 and
> NSHARP did not build, along with other programs. Here is the error I am
> seeing. This is seen several times in the make.out Anyone have any ideas?
>
> /usr/bin/ld: /home/gempak/GEMPAK/os/linux64/lib/gemlib.a(dgdriv.o): in
> function `dg_driv':
> /home/gempak/GEMPAK/gempak/source/diaglib/dg/dgdriv.c:367: undefined
> reference to `df_mfrq'
> collect2: error: ld returned 1 exit status
> make[5]: *** [Makefile:185: nmap2] Error 1
>
> Thanks very much.
> Robert Mullenax
>
> Robert Mullenax
> Meteorologist
> Columbia Scientific Balloon Facility
> Peraton
> 1510 E. FM 3224
> Palestine, TX 75803
> 903-723-8037
> robert.r.mullenax@xxxxxxxx
>
> _________________________________________________________
> NOTE: All exchanges posted to NSF Unidata maintained email lists are made
> publicly available through the web. Users who post to any of the lists we
> maintain are reminded to remove any personal information that they do not
> want to be made public.
>
> NSF Unidata gembud Mailing List
> (gembud@xxxxxxxxxxxxxxxx)
> For list information, to unsubscribe, or change your membership options,
> visit: https://mailinglists.unidata.ucar.edu/listinfo/gembud/
>
> _________________________________________________________
> NOTE: All exchanges posted to NSF Unidata maintained email lists are
> made publicly available through the web. Users who post to any of the
> lists we maintain are reminded to remove any personal information that
> they do not want to be made public.
>
> NSF Unidata gembud Mailing List
> (gembud@xxxxxxxxxxxxxxxx)
> For list information, to unsubscribe, or change your membership options,
> visit: https://mailinglists.unidata.ucar.edu/listinfo/gembud/
>
--
- Mike Zuranski