2010 Users Committee Survey Results

Results Listed by Question

1. Please select your country and state/province of residence.
2. Please select your profession(s).
3. Select your individual area(s) of study/specialization.
4. What Unidata Data Visualization and Analysis packages do you use?
5. What Unidata Data Access tools do you use?
6. What Unidata Data Management tools do you use?
7. How often do you use Unidata packages/tools?

Use of Unidata Data Visualization and Analysis Packages (GEMPAK, IDV, McIDAS)

8. Unidata provides analysis and visualization software that enables creativity among educators.
9. Unidata provides analysis and visualization software that enables academic learning.
10. Unidata provides analysis and visualization software that enables scientific discovery.
11. If you have any suggestions or comments regarding Unidata Data Visualization and Analysis products, please enter them below (e.g. enhancement or future services).

Use of Unidata Data Access Tools (LDM/IDD, TDS, RAMADDA)

12. Data access from Unidata's Internet Data Distribution is provided in a timely fashion.
13. Data access from Unidata's Internet Data Distribution is reliable.
14. What data are missing from Unidata's offering?
15. I use these data for:
16. Are you familiar with RAMADDA?
17. If you have any suggestions or comments regarding Unidata Data Access tools, please enter them below (e.g. enhancement or future services).

Use of Unidata Data Management Tools (netCDF, CDM, libCF, UDUNITS)

18. The Unidata data management tools allow for management, manipulation, and sharing of the scientific data.
19. The Unidata data management tools provide a functional way to distribute and access scientific data.
20. Unidata will play a significant role within the community distributing and providing access to data in the future.
21. If you have any suggestions or comments regarding Unidata Data Management tools, please enter them below (e.g. enhancement or future services).

Educators and Students

22. How have you used Unidata's Data Visualization and Analysis Packages?
23. Do you feel the Unidata's Data Visualization and Analysis Packages have helped you advance in your endeavors as a student?

Unidata Community Services

24. Are you aware of the support and training workshops available from Unidata?
25. Rate your level of satisfaction with Unidata applications for each of the following categories:
26. What are the most critical data services, software, and cyberinfrastructure needed for improving Earth System science education?
27. Are there unmet community needs where Unidata might be able to help or provide a service?
28. What non-Unidata tools do you use for your work?
29. What hardware and operating systems do you have in your department?
30. What hardware and operating systems do YOU use?
31. What is your overall level of satisfaction with the Unidata program?

---

Question 1: Please select your country and state/province of residence.

Country of residence graph:

Country of residence breakdown:

Country of residence	Absolute frequency	Relative frequency	Adjusted relative frequency
United States	150	62.5%	63.56%
Argentina	2	0.83%	0.85%
Australia	2	0.83%	0.85%
Austria	1	0.42%	0.42%
Bosnia and Herzegovina	1	0.42%	0.42%
Brazil	5	2.08%	2.12%
Bulgaria	1	0.42%	0.42%
Canada	5	2.08%	2.12%
Cape Verde	1	0.42%	0.42%
China	5	2.08%	2.12%
Colombia	1	0.42%	0.42%
France	3	1.25%	1.27%
Germany	4	1.67%	1.69%
India	4	1.67%	1.69%
Iran, Islamic Republic of	4	1.67%	1.69%
Israel	1	0.42%	0.42%
Italy	12	5%	5.08%
Mexico	1	0.42%	0.42%
Mozambique	1	0.42%	0.42%
Netherlands	2	0.83%	0.85%
New Zealand	1	0.42%	0.42%
Norway	1	0.42%	0.42%
Panama	1	0.42%	0.42%
Portugal	2	0.83%	0.85%
Romania	1	0.42%	0.42%
Russian federation	1	0.42%	0.42%
Singapore	1	0.42%	0.42%
Spain	5	2.08%	2.12%
Sweden	1	0.42%	0.42%
Switzerland	2	0.83%	0.85%
Taiwan	1	0.42%	0.42%
Thailand	1	0.42%	0.42%
Tonga	1	0.42%	0.42%
Turkey	1	0.42%	0.42%
United Kingdom	9	3.75%	3.81%
United States Minor Outlying Islands	1	0.42%	0.42%
Not answered:	4	1.67%	-
Sum:	240	100%	100%
Total answered: 236

US state of residence graph:

US state of residence breakdown:

US state of residence	Absolute frequency	Relative frequency	Adjusted relative frequency
Alabama	4	1.67%	2.61%
Alaska	2	0.83%	1.31%
Arizona	2	0.83%	1.31%
Arkansas	1	0.42%	0.65%
California	11	4.58%	7.19%
Colorado	21	8.75%	13.73%
Florida	4	1.67%	2.61%
Georgia	2	0.83%	1.31%
Guam	1	0.42%	0.65%
Hawaii	1	0.42%	0.65%
Illinois	3	1.25%	1.96%
Indiana	4	1.67%	2.61%
Iowa	2	0.83%	1.31%
Kansas	3	1.25%	1.96%
Maryland	6	2.5%	3.92%
Massachusetts	5	2.08%	3.27%
Minnesota	1	0.42%	0.65%
Mississippi	2	0.83%	1.31%
Missouri	2	0.83%	1.31%
Montana	1	0.42%	0.65%
Nebraska	4	1.67%	2.61%
New Hampshire	1	0.42%	0.65%
New Mexico	1	0.42%	0.65%
New York	13	5.42%	8.5%
North Carolina	5	2.08%	3.27%
Ohio	1	0.42%	0.65%
Oklahoma	4	1.67%	2.61%
Oregon	1	0.42%	0.65%
Pennsylvania	3	1.25%	1.96%
Rhode Island	2	0.83%	1.31%
South Carolina	1	0.42%	0.65%
South Dakota	2	0.83%	1.31%
Texas	10	4.17%	6.54%
Utah	2	0.83%	1.31%
Vermont	2	0.83%	1.31%
Virginia	1	0.42%	0.65%
Washington	6	2.5%	3.92%
Wisconsin	7	2.92%	4.58%
Wyoming	2	0.83%	1.31%
State Not Listed	7	2.92%	4.58%
Not answered:	87	36.25%	-
Sum:	240	100%	100%
Total answered: 153

Non-US state/province of residence replies:

Non-US state/province of residence
Rio de Janeiro
Berkshire
Galicia
WI
Lazio
Europe
Milan
Aveiro
Tasmania
Maharashtra
Pv
Wien
Quebec
Madrid
Rio de Janeiro
Cundinamarca
L'Aquila
Oxfordshire
Brandenburg
La Spezia
Berkshire
Beijing
NRW
Quebec
Si Chuan Province
Espirito Santo
Sao Paulo
Ile de France
Maharashtra
Saint-Petersburg
Noord-Holland
Oslo
England
Tel-Aviv district
Jalisco
Glestan
Jujuy
Khorasan razavi
Pacific
Rio Grande do Sul
Trentino
England
Maputo
South-Khorasan
Ontario
Beijing
Zurich
Ontario

---

Question 2: Please select your profession(s). (Please check all that apply.)

List of professions graph:

List of professions breakdown:

Professions	Absolute frequency	Relative frequency	Adjusted relative frequency
Faculty	52	16.05%	21.67%
Researcher	117	36.11%	48.75%
Systems Administration	37	11.42%	15.42%
Professional	62	19.14%	25.83%
Postdoctoral Position	4	1.23%	1.67%
Graduate Student	25	7.72%	10.42%
Undergraduate	1	0.31%	0.42%
Other (please specifiy):	26	8.02%	10.83%
Sum:	324	100%	100%
Total answered: 240

Other profession replies:

Other professions listed
Forecaster
Web and data developer
Private research use
Contractor to NOAA
Staff Meteorologist
Data manager
Software Engineering
Meteorologist
Software developer
Scientific Programming
Programmer
Unidata Staff
Environmental Health Scientist/Federal Government
Gov. Contractor
Public Safety
Private co
Private
Software developer
M.Sc student
Government
Data manager
Forecaster
data integration planner
Weather Buff
Retired

---

Question 3: Select your individual area(s) of study/specialization. (Please check all that apply.)

Areas of study/specialization graph:

Areas of study/specialization breakdown:

Area(s) of study/specialization	Absolute frequency	Relative frequency	Adjusted relative frequency
Meteorology/Atmospheric Science	178	41.69%	74.17%
Oceanography	37	8.67%	15.42%
Geography	9	2.11%	3.75%
Hydrology	2	0.47%	0.83%
Earth System Science	22	5.15%	9.17%
Environmental Science	23	5.39%	9.58%
Geology/Geophysics	7	1.64%	2.92%
Climate	44	10.3%	18.33%
Computer Applications	92	21.55%	38.33%
Other (please specfiy):	13	3.04%	5.42%
Sum:	427	100%	100%
Total answered: 240

Other areas of study/specialization replies:

Other areas of study/specialization listed
Energy Policy
Space Systems
Language
Statistics
Architecture
Real-time operations
Data science
Computer Science, Chemistry, Biology
Unidata Staff
Satellite remote sensing
Ecology
Fluid dynamics
Snow Science

---

Question 4: What Unidata Data Visualization and Analysis packages do you use?

Packages used graph:

Packages used breakdown:

Unidata Data Visualization and Analysis packages used	Absolute frequency	Relative frequency	Adjusted relative frequency
IDV	122	46.39%	68.16%
GEMPAK	96	36.5%	53.63%
McIDAS	45	17.11%	25.14%
Sum:	263	100%	100%
Total answered: 179

---

Question 5: What Unidata Data Access tools do you use?

Packages used graph:

Packages used breakdown:

Unidata Data Access tools used	Absolute frequency	Relative frequency	Adjusted relative frequency
LDM/IDD	99	49.01%	66.44%
Thredds Data Server (TDS)	72	35.64%	48.32%
RAMADDA	31	15.35%	20.81%
Sum:	202	100%	100%
Total answered: 149

---

Question 6: What Unidata Data Management tools do you use?

Packages used graph:

Packages used breakdown:

Unidata Data Management tools used	Absolute frequency	Relative frequency	Adjusted relative frequency
netCDF	196	50.65%	96.55%
Common Data Model (CDM)	36	9.3%	17.73%
UDUNITS	82	21.19%	40.39%
libCF	14	3.62%	6.9%
OPeNDAP	59	15.25%	29.06%
Sum:	387	100%	100%
Total answered: 203

---

Question 7: How often do you use Unidata packages/tools?

Frequency of use graph:

Frequency of use breakdown:

Unidata packages/tools frequency of use	Absolute frequency	Relative frequency	Adjusted relative frequency
5+ hours per week	115	47.92%	48.12%
Between 3 and 5 hours per week	34	14.17%	14.23%
Between 1 and 3 hours per week	59	24.58%	24.69%
Less than one hour per week	30	12.5%	12.55%
Never	1	0.42%	0.42%
Sum:	239	100%	100%
Total answered: 239

---

Question 8: Unidata provides analysis and visualization software that enables creativity among educators.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	69	28.75%	39.88%
Agree	85	35.42%	49.13%
Indifferent	18	7.5%	10.4%
Disagree	1	0.42%	0.58%
Sum:	173	100%	100%
Total answered: 173

---

Question 9: Unidata provides analysis and visualization software that enables academic learning.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	81	33.75%	47.09%
Agree	76	31.67%	44.19%
Indifferent	15	6.25%	8.72%
Sum:	172	100%	100%
Total answered: 172

---

Question 10: Unidata provides analysis and visualization software that enables scientific discovery.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	77	32.08%	44%
Agree	80	33.33%	45.71%
Indifferent	15	6.25%	8.57%
Disagree	3	1.25%	1.71%
Sum:	175	100%	100%
Total answered: 175

---

Question 11: If you have any suggestions or comments regarding Unidata Data Visualization and Analysis products, please enter them below (e.g. enhancement or future services).

Comments regarding GEMPAK & AWIPS 2:

• Retain GEMPAK/ Ensemble Functionality as you migrate into AWIPSII.
• GEMPAK is a great tool, but has become outdated and its online documentation is oftentimes incomplete.
• GEMPAK is the most robust and mature meteorological analysis package in existence. Please continue to maintain it, regardless of its future within the NWS.
• Please continue to maintain GEMPAK, regardless of its future within the NWS.
• I'm looking forward to AWIPS II (going from GEMPAK to AWIPS II).
• Continue support of GEMPAK even if development ceases. GEMPAK is far too valuable to the meteorological community for it to be abandoned completely.
• Looking forward to using GEMPAK's replacement.
• I am somewhat concerned about the status of the GEMPAK sunset and transition to AWIPS2.
• Provide access and support for AWIPS-II while maintaining access for GEMPAK.
• Keep GEMPAK fully functional.
• GEMPAK 6.2.0 is calling...
• Please continue to support the GEMPAK programs until a solution that performs as well (i.e. quickly) is available.

Comments regarding the IDV:

• IDV rocks. And the ease of creating a modified version is very useful and important.
• IDV functionality is still limited and lacks features of GEMPAK critical for class room use.
• The primary goal should be to have software that can create publication quality figures with ease. I cannot comment on academic learning and educator creativity since these are not myareas, but we use Unidata tools as building blocks for a rich web-based tools for scientific discovery.
• 3-D stuff is nice, but we need more current weather displays. WXP did a nice job of displaying surface and upper air plots and contours; I want more of that.
• Although I am using IDV more for some quick views of my own research model output, it seems Unidata is lacking in a software tool like grads that is pretty easy to use and allows researchers to come up with publication-quality graphics. I don't think IDV is there yet, but it would be great if it could be brought to that stage quickly.
• IDV faster?
• IDV software package is often unstable and difficult to use.
• Support for fine spatial and temporal scale observation visualization / analysis within the IDV (I need to revisit IDV to see if this has improved recently, so this may be a non-issue).
• More development of IDV needed so that it equals the capabilities of GEMPAK since GEMPAK is going away.
• As the CF Radial convention gels it would be nice if IDV supported it. The CFRadial convention is for stationary and mobile, scanning and fixed radar and lidar data.
• IDV has a steep learning curve and it is not easy to use for forecasting like GARP is.
• The IDV is still too slow to use regularly, and I still don't think you can get publication quality graphics from it.
• IDV: Improvement of startup times. Better memory management.
• Should be more user friendly.
• Why indifferent above, when I use Unidata tools 40+ hours/ week?
1. If the purpose of Unidata is to provide tools, services, and infrastructure leadership to advance etc. etc.. Why would an inordinate amount of time, money, and resources be allocated to create an application outside the scope of existing operational processes? You already know the answer, but let us take this a step further. The IDV, utilizing far inferior architecture, has been forced onto the Unidata community, for reasons beyond the scope of this mission.
2. To prove the above, is the IDV a product developed to such a level that it will be adopted by the NWS as an operational application?
3. If the IDV will not be adopted operationally, should the resources that have been allocated for its development, been more appropriately applied elsewhere?
4. Should our future generations of students learn from applications that are not used operationally, or those adapted from those real-world applications, so they are more prepared to enter the workforce?
5. If those resources where allocated with more forethought, so that simultaneous integration and development could occur with the AWIPSII release, is it likely that a superior NAWIPSII application would be released?
6. Will the time, money, and resources that have been used in the AWIPSII release be equal to or greater than those allocated for the IDV, since it will more closely resemble an operational product; thus better preparing future generations for operational and research applications?
7. Will this survey be read with the true intent that it was submitted, or simply discarded as it does not agree with the status-quo?
• Some of the tools that used to be in the Ferret gui would be great to add to IDV. The Ferret gui is no longer supported and no longer works at my institution. I've tried to use IDV to replace the uses I previously made of Ferret to preview parts of a netCDF file. It can do some things, though I found the graphics and menus less intuitive with IDV. (I also use IDV for classroom instruction, mainly overlays and forecast products tasks that I used to use McIDAS for.)
• The IDV tool seems less universally functional than GEMPAK. I can't run IDV on my desktop(~ 3 years old - maybe a memory problem - freezes up), or on my laptop (~2 year old running VISTA) but can use it on my tablet which has the slowest processor but the most memory and also runs VISTA.

Comments regarding McIDAS:

• We use much for the web. I think McIDAS is relatively easy to use for this purpose. I would hate to see it lose support.
• I have been working mainly with the McIDAS-V package. It has a number of powerful tools but the number of inconsistencies and lack of complete tools can outweigh the gain of the new tools.

General comments we received:

• Integrate products with any GIS.
• Unidata needs to be more proactive in maintaining data compatibility. Too often, it takes a crisis in users seeing a change in data, then having to get a quick fix. It appears that the UPC is not fully engaged.
• I am greatly dismayed at the departure of the entire talent pool behind IDV. How can Unidata let two such talented and motivated individuals slip away?
• Without Unidata, our education and research would be much poorer in quality.
• May benefit from a more commercial bias (or industry input).
• I think all members of the Unidata community would benefit from a central repository of user-created GEMPAK scripts / IDV bundles / "how to" articles. A wiki'ish kind of setting. I have probably spent weeks searching through the mailing lists and support requests trying to find examples, bug fixes, etc. An organized location "hub" of information would reduce the amount of time users search for answers, and provide a new platform of sharing scripts /bundles / tricks / bug fixes.
• Need better documentation and on-line training for use of the tools.
• An ability to script the generation of meteorological products from packages like GEMPAK and McIDAS is vital! Interactive GUIs are nice but I do a lot of heavy lifting by running batch scripts to produce output for web pages and other tools.
• There needs to be better tools for educators to learn and understand the power of these tools. The Meted/COMET program should be used for this purpose. These powerful tools need to be publicized to the people who can get the most out of them.
• Provide more technique and documents support.
• Overcome the IDV crisis asap.
• I principally use ArcGIS for geographic data displays. Any tools to connect ArcGIS to Unidata data would be helpful.
• Better interfaces to user provided code.
• I really don't use the tools for analysis and visualization. I use the tools for data access extensively.
• General: An IDV-quality tool for time series data (i.e., non-gridded datasets), would be a superb addition to the Unidata software stack.

---

Question 12: Data access from Unidata's Internet Data Distribution is provided in a timely fashion.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	49	20.42%	52.69%
Agree	39	16.25%	41.94%
Indifferent	5	2.08%	5.38%
Sum:	93	100%	100%
Total answered: 93

---

Question 13: Data access from Unidata's Internet Data Distribution is reliable.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	48	20%	51.06%
Agree	41	17.08%	43.62%
Indifferent	4	1.67%	4.26%
Strongly Disagree	1	0.42%	1.06%
Sum:	94	100%	100%
Total answered: 94

---

Question 14: What data are missing from Unidata's offering?

Response given:

• Brazilian radar and lighting network.
• Up to date decoding tables (station ID, grib) necessary for use with data.
• Raw radar data files from Canada/Mexico and elsewhere around the world.
• Not aware of any.
• We receive everything we really want right now.
• There are always more datasets that could be offered, but rather than offering everything via"push", Unidata and Unidata software should be developed in a way to enable pull access for many of these datasets.
• Models. More models.
• EUMETSAT radar for Europe; regional models for Europe.
• Do not use data from IDD.
• Satellite image at full resolution using different formats other then McIDAS.
• While it is not Unidata's fault, we had always struggled with getting reliable service from the IDD due to latency issues on the internet routes to our campus. We purchased a NOAAport sat. receiver system to circumvent these issues.
• Radar data in netCDF format.
• Nothing.
• I represent a private interest that does not have access to IDD; we only use the LDM, not the IDD.
• Air quality data that can be overlaid.
• Climate data for reference datasets and anomaly calculations.
• Near real-time, high resolution ECMWF GRIB data full ECMWF, Meteosat et. al.
• None.
• Hurricane ATCF track forecasts.

---

Question 15: I use these data for (please check all that apply):

Response graph:

Response breakdown:

Data usage	Absolute frequency	Relative frequency	Adjusted relative frequency
Collaborations	40	10.78%	44.44%
Computer Lab	28	7.55%	31.11%
Courses	41	11.05%	45.56%
Education/Teaching	48	12.94%	53.33%
Field Projects	28	7.55%	31.11%
Outreach	17	4.58%	18.89%
Presentations	39	10.51%	43.33%
Publications	35	9.43%	38.89%
Research	65	17.52%	72.22%
Thesis/Dissertation	14	3.77%	15.56%
Other	16	4.31%	17.78%
Sum:	371	100%	100%
Total answered: 90

Other data usage replies:

Other data usage listed
Production
Storm Ready
Personal forecasts
Provide forecasts to on campus users, and for our on-campus alerting system
Forecast
Work
Sponsored projects funded by government agencies
Operational Forecasting
Commercial
Forecast contests
AWIPS Testing
Forecasting
Forecast Operations
Forecasting

---

Question 16: Are you familiar with RAMADDA?

Response graph:

Response breakdown:

Familiarity	Absolute frequency	Relative frequency	Adjusted relative frequency
Yes	57	23.75%	41.01%
No	82	34.17%	58.99%
Sum:	139	100%	100%
Total answered: 139

---

Question 17: If you have any suggestions or comments regarding Unidata Data Access tools, please enter them below (e.g. enhancement or future services).

Comments regarding the IDV:

• IDV still does not read BUFR - this is a serious drawback for the (European) academic community.

Comments regarding the THREDDS Data Server:

• Remote scans in TDS. Support for the full common data model. Relative paths so that proxies and ports are not shown. Improved HTML interface.
• Some way to automate RAMADDA repositories with TDS on a regular basis would be good.
• NetCDF Subset Service for Point datasets in TDS.
• WFS and SOS services in TDS.
• ISO 19115 output service in TDS.

Comments regarding RAMADDA:

• Value of RAMADDA for my applications remains limited.
• We are just beginning to use RAMADDA and we think it might be very useful. It worries me that we would begin to depend on this tool when we know that the technical brains behind the product has left Unidata.
• Can RAMADDA servers talk to each other yet? Again, I need to play catch-up a bit, so this may be addressed already.
• Some way to automate RAMADDA repositories with TDS on a regular basis would be good.
• I have heard of RAMADDA but have not used it or looked into it enough to know what it really is/does.
• RAMADDA: A promising tool, but I'm not sure it's quite ready for production use yet.
• Unidata needs to keep RAMADDA alive!

Comments regarding the LDM/IDD:

• It would be nice to have a reference web page with links to sites that archive the IDD and ADDE offerings. Occasionally there is a need to retrieve products that may have been missed due to system downtime or unexpected problems with our LDM server or network. I always have to search the mail archives looking for mention of available short and long-term archives.
• Why is IDD only available to universities and not everyone?

General comments we received:

• Better documentation.
• The IDV/ADDE method of requesting data is an excellent idea that gets around the latency problems we've had with the IDD/LDM but it does present issues with larger datasets when we have a lab with many students simultaneously requesting data (read: painfully slow). rhaps the clients could cache data and serve each other's duplicate requests via a peer to peer system within the same subnet?
• Better CDM support. Particularly ad hoc unit-conversions (to/from) and metadata-conversion tools (to/from) to promote integration and uptake.
• Needs better documentation about its features and how to use them.
• Support Swath feature type in CDM. Expand THREDDS' Netcdf subsetting service to other CDM types.
• OpenDAP !! Good stuff.
• The maps don't play particularly well with browsers on a Linux platform.
• A comment on long term strategy and priorities: It is my belief that the future-looking place for Unidata to concentrate its energies is on tools that support data interoperability -- netCDF, CF, OPeNDAP, TDS, THREDDS, RAMADDA, etc. While recognizing that there is a significant group of "customers" that rely on Unidata applications such as IDV and McIDAS, this is in a sense a "legacy" customer base -- a legacy from a day when there weren't alternative applications that could provide the functionality these users needed. Alternatives do exist today, however -- imperfect, perhaps, but clearly positioned to become more perfect if demand and support existed. By focusing on the infrastructure that supports applications, rather than the applications, themselves, Unidata can strengthen its pivotal position within the software developers community and thereby leverage energies from the community. This kind of "scalability" thinking will be needed for Unidata to meet the ever-rising demands that will be seen when (for example) netCDF and CF gain OGC and WMO certification.

---

Question 18: The Unidata data management tools allow for management, manipulation, and sharing of the scientific data.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	107	44.58%	56.32%
Agree	76	31.67%	40%
Indifferent	4	1.67%	2.11%
Disagree	3	1.25%	1.58%
Sum:	190	100%	100%
Total answered: 190

---

Question 19: The Unidata data management tools provide a functional way to distribute and access scientific data.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	107	44.58%	56.61%
Agree	73	30.42%	38.62%
Indifferent	8	3.33%	4.23%
Disagree	1	0.42%	0.53%
Sum:	189	100%	100%
Total answered: 189

---

Question 20: Unidata will play a significant role within the community distributing and providing access to data in the future.

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Strongly Agree	111	46.25%	58.42%
Agree	63	26.25%	33.16%
Indifferent	15	6.25%	7.89%
Disagree	1	0.42%	0.53%
Sum:	190	100%	100%
Total answered: 190

---

Question 21: If you have any suggestions or comments regarding Unidata Data Management tools, please enter them below (e.g. enhancement or future services).

Comments regarding netCDF:

• With the direct link of netCDF to ArcGIS we will be able to use netCDF as a standard storage format much more in the future.
• Perhaps Unidata can work with tool providers like NCO to provide better tools.
• These tools are absolutely critical to our daily work.
• Again, without these tools, our data access would be much more difficult.
• This isn't Unidata's issue per se, but we need to converge on a method of handling irregularly gridded (e.g., FEM) data within netCDF and CDM.
• Some capabilities available only in the NetCDF-Java should be back-ported to the C/C++ and FXX API.
• MS-Win ports for current netCDF releases (3.6.3 and later) would be greatly appreciated. Right now these require significant effort to compile.

Comments regarding UDUNITS:

• UDUNITS needs continuing Fortran support!!!!!!!!!
• HTML pages for UDUNITS database for easy perusal online; currently only XML provided.

General comments we received:

• From the questionnaire, it appears that Unidata is claiming that OPeNDAP is an example of Unidata software. Most of the credit for OPeNDAP software development should actually be given to the opendap.org organization.
• Version compatibility should be considered.
• Consult industry and promote bridging between academic/research and real-world applications.
• Traditionally netCDF has a strong position in earth system modeling on global/climate scale, but it becomes increasingly important at engineering scales. This will greatly increase the adoption of these Unidata products. This holds also for the OPeNDAP and Thredds tools that I currently don't use myself as well. Higher level tools like IDV may also grow in popularity although there is more competition in the field of 2D/3D visualization.
• Not ready for primetime now, but if there is continued development in scientific lead improvements, the potential is great.
• I think it is convenient.
• I can see the large effort of Unidata scientific team to develop new procedures, tools, and other products. Please continue with this way of work, all the world could benefit of it.

---

Question 22: How have you used Unidata's Data Visualization and Analysis Packages? (Please check all that apply.)

Response graph:

Response breakdown:

Unidata Data Visualization and Analysis Package use	Absolute frequency	Relative frequency	Adjusted relative frequency
In teaching (classroom exercises, group activities, etc.)	46	30.07%	63.01%
For research (undergraduate or graduate)	65	42.48%	89.04%
As a forecasting tool	39	25.49%	53.42%
Other (please specify):	3	1.96%	4.11%
Sum:	153	100%	100%
Total answered: 73

Other package use replies:

Other package use listed
Field Studies
Fun!

---

Question 23: Do you feel the Unidata's Data Visualization and Analysis Packages have helped you advance in your endeavors as a student?

Response graph:

Response breakdown:

Level of agreement	Absolute frequency	Relative frequency	Adjusted relative frequency
Yes	54	22.5%	90%
No	6	2.5%	10%
Sum:	60	100%	100%
Total answered: 60

---

Question 24: Are you aware of the support and training workshops available from Unidata?

Response graph:

Response breakdown:

Awareness of Unidata workshops	Absolute frequency	Relative frequency	Adjusted relative frequency
Yes	187	77.92%	89.9%
No	21	8.75%	10.1%
Sum:	208	100%	100%
Total answered: 208

---

Question 25: Rate your level of satisfaction with Unidata applications for each of the following categories:

Response graph:

Response breakdown:

		Highly Satisfied	Satisfied	Indifferent	Dissatisfied	Highly Dissatisfied	Sum
Ease of Use	Absolute frequency Relative frequency Adjusted Relative frequency	38 17.92% 2.6%	138 65.09% 9.45%	25 11.79% 1.71%	11 5.19% 0.75%	0 0% 0%	212 100% 14.51%
Educational Value	Absolute frequency Relative frequency Adjusted Relative frequency	65 30.95% 4.45%	94 44.76% 6.43%	48 22.86% 3.29%	3 1.43% 0.21%	0 0% 0%	210 100% 14.37%
Research Value	Absolute frequency Relative frequency Adjusted Relative frequency	94 44.76% 6.43%	102 48.57% 6.98%	9 4.29% 0.62%	5 2.38% 0.34%	0 0% 0%	210 100% 14.37%
Technical Support	Absolute frequency Relative frequency Adjusted Relative frequency	60 28.99% 4.11%	95 45.89% 6.5%	43 20.77% 2.94%	8 3.86% 0.55%	1 0.48% 0.07%	207 100% 14.17%
Timeliness of Updates	Absolute frequency Relative frequency Adjusted Relative frequency	51 24.4% 3.49%	100 47.85% 6.84%	49 23.44% 3.35%	9 4.31% 0.62%	0 0% 0%	209 100% 14.31%
Documentation	Absolute frequency Relative frequency Adjusted Relative frequency	43 20.98% 2.94%	100 48.78% 6.84%	33 16.1% 2.26%	28 13.66% 1.92%	1 0.49% 0.07%	205 100% 14.03%
Community Support (mailing lists, etc.)	Absolute frequency Relative frequency Adjusted Relative frequency	68 32.69% 4.65%	104 50% 7.12%	29 13.94% 1.98%	7 3.37% 0.48%	0 0% 0%	208 100% 14.24%
Sum	Absolute frequency Relative frequency Adjusted Relative frequency	419 100% 28.68%	733 100% 50.17%	236 100% 16.15%	71 100% 4.86%	2 100% 0.14%	1461 100% 100%

---

Question 26: What are the most critical data services, software, and cyberinfrastructure needed for improving Earth System science education?

Comments regarding GEMPAK & AWIPS 2:

• An easy to use, "script-able" visualization utility that is stable and generates publication quality plots. IDV is not very scriptable and its plot quality, in comparison to utilities like NCL or GrADS, is poor. GEMPAK is the only software that fills this gap at the moment, but is (as mentioned) outdated. A new visualization package is needed.
• Overall, a SIMPLE, easy to use way of discovering and retrieving available data products. The current LDM/IDD method presents a barrier unless you have someone available with experience with these systems. Applications with a more user friendly interface to display or manipulate data is something that would be helpful. The current GEMPAK graphical programs have a nice simple interface for new users but this has a very uncertain future. While I have a lot of respect and appreciation for the efforts that Unidata has made with the IDV, the user interface is awkward and slow causing our students to shy away from it for most needs.
• GEMPAK Applications and services for undergraduates with no post-graduate aspirations that would prepare them for non-academic work.

Comments regarding the IDV:

• An easy to use, "script-able" visualization utility that is stable and generates publication quality plots. IDV is not very scriptable and its plot quality, in comparison to utilities like NCL or GrADS, is poor. GEMPAK is the only software that fills this gap at the moment, but is (as mentioned) outdated. A new visualization package is needed.
• Overall, a SIMPLE, easy to use way of discovering and retrieving available data products. The current LDM/IDD method presents a barrier unless you have someone available with experience with these systems. Applications with a more user friendly interface to display or manipulate data is something that would be helpful. The current GEMPAK graphical programs have a nice simple interface for new users but this has a very uncertain future. While I have a lot of respect and appreciation for the efforts that Unidata has made with the IDV, the user interface is awkward and slow causing our students to shy away from it for most needs.
• Visualization of all meteorological data is key for Earth Science education, particularly the IDV. The IDV helps students tie together everything they've learned in Dynamics / PMet /Synoptic Meteorology in one easy to use tool.
• Ease and speed of use in the infrastructure. IDV could be a really great tool for this, but I've found it's is just too slow for real-time classroom analysis of large datasets on average computers. Subsetting input data does help quite a bit, but I'd like to see it even more efficient and responsive in the future.
• Visualization packages like the IDV that are much simpler to use.
• IDV connection between Unidata tools and GIS tools.
• Support for access to and visualization of long term climate datasets. Current tools are focused on weather, not climate. Aggregation of decades/centuries of data (not just gridded) is lacking.

Comments regarding netCDF:

• Data discovery tools, common data formats *cough cough netCDF cough*.
• The NetCDF format, now in the new HDF vest, is fundamental for data archive, sharing and distribution. The availability of a WMO conventions on NetCDF could boost it.

Comments regarding McIDAS:

• McIDAS

Comments regarding the THREDDS Data Server:

• TDS: Could it be better integrated with a standalone instance of an OPeNDAP server? I.e., could it use the OPeNDAP server, instead of its internal OPeNDAP classes?
• Subsetting for massive datasets (grids but also station/timeseries data).
• Support for access to and visualization of long term climate datasets. Current tools are focused on weather, not climate. Aggregation of decades/centuries of data (not just gridded) is lacking.

Comments regarding RAMADDA:

• Subsetting for massive datasets (grids but also station/timeseries data).
• Support for access to and visualization of long term climate datasets. Current tools are focused on weather, not climate. Aggregation of decades/centuries of data (not just gridded) is lacking.

Comments regarding the LDM/IDD:

• Free distribution of data.
• Requirements GLOBAL integrated unified datasets freely available.
• Enhanced IDD bandwidth, which will imply more top-tier servers. Other methods of data distribution need to be examined, which should include compatibility with Open GeospatialConsortium systems including the evolving NetCDF standard within OGC (that's easy!) and SWE.
• Easy to use and widely supported tools for data analysis and visualization, data distribution methods focused on on-demand delivery and facilitated data discovery, with growing focus on collaboration.
• Internet (socket-based) transfer mechanisms for data distribution (like IDD, but open rather than closed/restricted to certain communities).
• Free, live, real-world data.
• Overall, a SIMPLE, easy to use way of discovering and retrieving available data products. The current LDM/IDD method presents a barrier unless you have someone available with experience with these systems. Applications with a more user friendly interface to display or manipulate data is something that would be helpful. The current GEMPAK graphical programs have a nice simple interface for new users but this has a very uncertain future. While I have a lot of respect and appreciation for the efforts that Unidata has made with the IDV, the user interface is awkward and slow causing our students to shy away from it for most needs.
• Support for access to and visualization of long term climate datasets. Current tools are focused on weather, not climate. Aggregation of decades/centuries of data (not just gridded) is lacking.

General comments we received:

• Timely service oriented architecture that provides geo-processed information on-demand.
• Finding and extracting data in a useful manner. Data extraction tools seemed to be geared towards a non-users idea of what people want.
• Facilitate use of data/displays with common desktop tools for publication, presentation.
• Manipulation without having to export to other formats. Need to transition from traditional synoptic level analysis/display to boundary layer and mesoscale uses for air pollution, energy applications, and operations management.
• Format that supports parallel IO.
• Improved visualization software.
• Integration with idl, matlab.
• The most critical need is to maintain software that can produce high quality maps/images for use in the classroom, internet, or publication in available data formats. Additionally, maintaining a database of classic weather events with their data would be an amazing service.
• As mentioned before, we need software that can not only display the information, but also make it easy to port imagery to the Web.
• Bandwidth and data storage remain a problem. It is so easy to fill up disks.
• Timely access to data and analysis tools without time-consuming training.
• The ability to access, integrate, and manipulate atmospheric and related datasets as **seamlessly** and **easily** as possible should be a priority of Unidata. There is much Unidata software and tools can do, but it can take considerable effort to do it quickly. This represents a barrier for education.
• An efficient way of sharing data (easily accessible archive), a set of clear case studies and easy to use web based plotting routines.
• Improve use of Geographic information System file support.
• Support to install and discover new software to help. Funds to have large disk storage with backup facilities. Help to organize classrooms.
• An effective and efficient search and discovery method in distributed data among different hosts, in different country.
• Improving parallel I/O and parallel post-processing netCDF data manipulation.
• Overall, a SIMPLE, easy to use way of discovering and retrieving available data products. The current LDM/IDD method presents a barrier unless you have someone available with experience with these systems. Applications with a more user friendly interface to display or manipulate data is something that would be helpful. The current GEMPAK graphical programs have a nice simple interface for new users but this has a very uncertain future. While I have a lot of respect and appreciation for the efforts that Unidata has made with the IDV, the user interface is awkward and slow causing our students to shy away from it for most needs.
• Better integration and support for conversions between discipline-based standards (eg GRIB, BUFR) and application-areas (eg. marine forecasting, model comparison, geology).
• Robust cross-platform support.
• Common data formats, accessible distribution servers.
• I'm not involved with education.
• Graphics software able to display complex phenomena so that students can visualize aspects of atmospheric science.
• Adoption of existing Unidata standards.
• Websites to run various simulation codes remotely.
• Ability to flexibly couple simulation codes (like ESMF, CCA, OpenMI, ...) based on common standards.
• Examples of use of datasets to encourage more widespread use. Perhaps pointers to Unidata training opportunities at scientific meetings and conferences - or opportunities for training for others at University programs to foster more widespread use?
• Increase the ease and intuitive use of the data interface.
• A more multi-disciplinary ESS education uses GIS more than pure meteorology. There are probably more that can be done with GIS.
• Better instruction and tutorials! Visual guides!

---

Question 27: Are there unmet community needs where Unidata might be able to help or provide a service?

Comments regarding GEMPAK & AWIPS 2:

• Displaying data/analyses/computations from multiple sources on one map (like what CIMSS- MSN does).
• Continuing strong GEMPAK support is key. I'm looking forward to the new version and your continued help desk support with it. Also, it would be great if you occasionally offered intermediate level classes with your visualization tools. After using these tools for several years there are often occasions where I'd like to use them in more advanced situations and there is little documentation or support to help with that. Some examples would be trying to process MODIS imagery with McIDAS, plotting a VIS/SWIR composite using multiple satellites, plotting ASCAT winds using GEMPAK, etc.
• Active development and expansion of GEMPAK and/or a successor program, specifically enhancing support to take advantage of increasing capabilities associated with newer hardware and software tools.

Comments regarding the IDV:

• Integration with Google Earth. Whirlwind is okay but slower and more unstable.
• Displaying data/analyses/computations from multiple sources on one map (like what CIMSS- MSN does).
• 3D visualization with live flight tracking.
• An IDV-quality time series analysis tool.

Comments regarding netCDF:

• Increased effort in common data formats.
• Releasing a final standard for netCDF for unstructured grids. A standard is needed, similar to the CF convention, for handling non-regular grids in netCDF files. Without this, there is no way for visualization programs to easily read the next generation model output, e.g. MPAS.
• Again, some advanced capabilities on NetCDF are available only using Java. A backport to the other API's could help us a lot.
• Provide a current maven directory for netCDF-java.
• Managing or updating non-standard (less-recognized standard) netCDF.

Comments regarding the THREDDS Data Server:

• Live feed access via open source GIS standards.

Comments regarding RAMADDA:

• Live feed access via open source GIS standards.
• Provide a repository of GEMPAK/NAWIPS restore scripts that can be exchanged among users and or uploaded by users.

Comments regarding the LDM/IDD:

• Live feed access via open source GIS standards.
• Free access to ECMWF data would be nice so students can see what a proper forecast looks like.
• Getting weather data available worldwide.
• Easy access to archived data, both station and model data.
• Access to large climate data holdings.
• Access to archived high resolution satellite imagery at high temporal (1 h) and spatial (e.g., US regions) resolution.

General comments we received:

• Some of the models produce enormous volumes of data. It is critical that visualization software is designed with this capability built in.
• Accurate, up to date documentation.
• Easy to use.
• Visualizing Earth System science data that are relevant to personal life style.
• GIS support and data access.
• Synergy of data types.
• More cross training and information sharing.
• Data discovery, dissemination, & analysis, along with being able to easily subset/reduce data before transit. All of which Unidata's tools do pretty well.
• Unidata serves the world... so a multilingual approach on the Unidata's website is needed to satisfy those coming from non-English background.
• OpenDap for developing countries to access and process data remotely independent of internet intermittency in their own countries.
• Access to a broad range of data ... which requires an new framework tools (APIs, development libraries, and servers) that will raise the level of interoperabilty.
• Any data visualization package with a shallow and quick learning curve initially that can also provide deeply customizable graphics for users at a higher level of production.
• Fog, Severe weather forecasting using parameters from products of different NWP model runs.
• Ease of use.
• Help get station data as NCDC can't seem to do it well.
• Unify community use of other packages NCL/NCAR Graphics/RIP/GRADS to fulfill gaps in Unidata applications.
• Creating a firm standard for meteorological data, so that the ability for anyone to read the data is not in question.
• Might be nice to team up with the MET people at DTC to create a tool to turn WRF-netCDF output into the type of netCDF that MODE within MET needs.
• Might be nice if a web-based training video/tool was created that could be used independently by first or second year students to get comfortable with key Unidata visualization tools (GEMPAK, nsharp, nmap2,mcidas, IDV) prior to when they reach upper-level courses where these tools are used.
• The production of case studies in the past was extremely helpful. I wish more of these were regularly produced.
• Unified approach to WRF post processing?
• Stronger support for GIS data, specially in IDV.
• An improved data plotting and visualization package.
• Nopoint/station based datasets need better access services, this is critical for us to make data more widely available (for example to government agencies for climate policy).
• Improve all tools with respect to thread-safety and start developing threaded tools.
• Validation of data out of the US.
• More high-level application-support libraries to promote integration and display. Including plot-libraries (eg. wind barbs, profiles, contours, vectors), conversion support libraries(to/from CDM, UDUNITS, CF-conventions, etc).
• Even though the GEMPAK / IDV / LDM community is rather small, it could benefit from a central hub of information. I feel that valuable time is lost scouring through the Mailing List /support forum archives. If this information was more organized, and there was a place where users could provide some of their own examples, the community would benefit greatly!
• I think you're already working on this, but an easier/more efficient way to subset large datasets would be useful. It sounds like opendap can do much of this, but a more simplified or generalized interface could be useful. I think its out there, but I don't know enough about it to use it.
• Would like to see further merging of OGC and Unidata domains, tighter integration of TDS with metadata standards (ISO 19115, FGDC, DIF, etc.)
• I think Unidata provides a great environment of learning.
• Assist in setting new data standards (e.g. coordinate systems, staggered grid storage, ...).
• Interact in coupling models.
• Climate and air pollution; climate model inter-comparisons?
• Yes in scripting and quantitative analysis. In McIDAS-V, there are tools for obtaining min, max, average and other data stats, but they are not readily available in the interactive mode. In addition, there are limited color tables. Large data sets also are slow to load and can make the software unstable. This is a problem because data sets will only get bigger.
• Unidata might be able to provide a service to teach tools and their use directly to students via the web (webinar-style).
• Better document the definition of various mapping projections with respect to CF-1.X.
• Unidata should begin to support more vigorously data types being produced outside of the NWP and climate modeling communities, i.e. more observational/measured data.
• More details about data format and some data converters tool. While the formats to store data are very robust, there are many formats. Some of these don't play well with each other. IDV provides a good platform for actively analyzing the data, but when you need this done automatically, eg generating maps and charts, GEMPAK has its problems. Once you have the data in the correct format, GEMPAK works excellently, but it's getting it to the correct format that is the issue. When GEMPAK is combined with LDM, it's seamless and trouble free but when you do not have access to this feed things become problematic. The other issue is that it ends up doubling the data. What is converted into GEMPAK format cannot be read by other programs so there is the need to keep the original around. Not to mention the additional processing needed to do the conversion to GEMPAK format on the fly. A better system for integrating data without duplicating it would ease the load on ingestion systems.
• Automatic update of software.
• We have an old and powerful weather radar (S-band) that can be used to display echoes of NW Argentina. Could UNICAR develop an adaptive tool to distribute our weather radar data?. And, on the other hand, we have a marine radar that can be adpated to measure rainfall with very low investment cost. Could UNICAR help us to develop this weather tool?
• Looking ahead, it'd be great to have services for digital object identifiers and dataset versioning.
• The demand for rapid development of Unidata's libraries and servers exceeds Unidata's ability to deliver. This is, of course, a symptom of SUCCESS. Unidata is to be congratulated. But the supply/demand gap is nonetheless an important challenge to meet, because it seems poised to widen in the future.

---

Question 28: What non-Unidata tools do you use for your work?

Response graph:

Responses breakdown:

Non-Unidata Tools Used	Absolute frequency	Relative frequency
MATLAB	32	9.4%
NCAR Command Language (NCL)	24	7.0%
GrADS	21	6.1%
IDL	19	5.6%
Python	11	3.2%
Google Earth API	10	2.9%
Weather Research and Forecasting (WRF) Model	8	2.3%
Bufkit	8	2.3%
In-House Developed Software (Unspecified)	6	1.8%
R	6	1.8%
Ferret	6	1.8%
McIDAS-X/V	6	1.8%
Gibson Ridge GRLevel2/3/AE/Earth	6	1.8%
ArcGIS	5	1.5%
Java	4	1.2%
Fortran	4	1.2%
Matplotlib	4	1.2%
Digital Atmosphere	4	1.2%
RAOB	4	1.2%
netCDF Operator (NCO)	4	1.2%
Compilers	3	0.9%
Microsoft Office	3	0.9%
GIS	3	0.9%
NCAR Graphics	3	0.9%
Panoply	3	0.9%
Ocean Data View	3	0.9%
VAPOR	3	0.9%
CDO	3	0.9%
Web sites & Web services (Unspecified)	3	0.9%
Word Processing	2	0.6%
Spreadsheets	2	0.6%
Apache Tomcat	2	0.6%
Global Mapper	2	0.6%
NetBeans	2	0.6%
HDF	2	0.6%
GRIB	2	0.6%
ncview	2	0.6%
suds/solo	2	0.6%
GNU Plot	2	0.6%
VERDI	2	0.6%
ERDDAP	2	0.6%
Live Access Server	2	0.6%
GMT (Generic Mapping Tool)	2	0.6%
pydap	2	0.6%
GRIB API	2	0.6%
wgrib/wgrib2	2	0.6%
AAPP/IAPP	2	0.6%
scipy	2	0.6%
numpy	2	0.6%
Postscript	1	0.3%
Ant	1	0.3%
Struts	1	0.3%
Publishing Software	1	0.3%
OpenOffice	1	0.3%
SQL	1	0.3%
SQL Alchemy	1	0.3%
PostreSQL	1	0.3%
RDBMS (Unspecified)	1	0.3%
Microsoft Access	1	0.3%
MySQL	1	0.3%
Apache HTTP Server	1	0.3%
Aptana AJAX Web Development IDE	1	0.3%
Google Web Toolkit	1	0.3%
Google Maps API	1	0.3%
HTML	1	0.3%
KML	1	0.3%
OpenLayers	1	0.3%
AniS	1	0.3%
ArcInfo	1	0.3%
ArcView	1	0.3%
SAGA GIS	1	0.3%
Grass	1	0.3%
Manifold	1	0.3%
ENVI	1	0.3%
Nokia QT Creator	1	0.3%
JDeveloper	1	0.3%
Eclipse	1	0.3%
Microsoft Visual c/c++	1	0.3%
Perl	1	0.3%
Ruby	1	0.3%
SHARP	1	0.3%
Firefox	1	0.3%
Community Earth System Model (CESM)	1	0.3%
NOGAPS Model	1	0.3%
MM5 Community Model	1	0.3%
Naval Oceanographic Office (NOO) Data	1	0.3%
Japan Meteorological Agency (JMA) Data	1	0.3%
HDF EOS	1	0.3%
HDF5	1	0.3%
GRIB2	1	0.3%
ncdump	1	0.3%
NetCDF	1	0.3%
Cygwin	1	0.3%
Ubuntu	1	0.3%
Configurable Interactive Data Display (CIDD)	1	0.3%
Tecplot	1	0.3%
WXP	1	0.3%
WINGRIDDS	1	0.3%
Metview	1	0.3%
Image Magik	1	0.3%
Golden Software Tools (Surfer Grapher)	1	0.3%
GEMPAK	1	0.3%
Hurrevac	1	0.3%
WeatherLink	1	0.3%
Paraview	1	0.3%
PCCGRIDDS	1	0.3%
CAMEO/Aloha	1	0.3
AWIPS	1	0.3%
NOMADS	1	0.3%
LDM	1	0.3%
Pangaea	1	0.3%
GeoNetwork	1	0.3%
GI-cat	1	0.3%
GI-go	1	0.3%
Virtual Ocean	1	0.3%
MapServer	1	0.3%
ncWMS	1	0.3%
pycdf	1	0.3%
PyNGL	1	0.3%
pupynere	1	0.3%
idlhdf5	1	0.3%
Magics	1	0.3%
ECMWF GRIB libraries	1	0.3%
HDFView	1	0.3%
GDAL/OGR	1	0.3%
STATISTICA	1	0.3%
MathCad	1	0.3%
SAGE	1	0.3%
OpenMI	1	0.3%
ADDS	1	0.3%
Environment Canada Tools	1	0.3%
OGC Services	1	0.3%
Sum:	342	100%

Response Type graph:

Responses breakdown:

Non-Unidata Tools Used By Type	Absolute frequency	Relative frequency
IDE Software	5	1.5 %
Database and SQL	6	1.8 %
Web Services & Web Sites	6	1.8 %
Data Formats	8	2.3 %
Math & Statistic Software	8	2.3 %
Word Processing & Spreadsheet Software	9	2.6 %
Web Development & Hosting	9	2.6 %
Data Access/Catalog Services	12	3.5 %
Data	13	3.8 %
Misc	14	4.1 %
Data Manipulation & Decoder Software	26	7.6 %
Scripts & Languages	29	8.5 %
Data Visualization & Analysis Software	197	57.6 %
Sum:	342	100%

Responses given:

Non-Unidata tools used
Text input
McIDAS-X, McIDAS-V
Thick client GIS which reads netCDF
mcidas
ArcGIS, Access (unfortunately)
Ferret, NCL, IDL
matlab sage(python) wgrib wgrib2
NCL, NCO, CDO, grads, IDL
WRF, Bufkit, Hurrevac, CAMEO/Aloha, spreadsheet, GIS, IDL, MATLAB
Matlab, NCL
matlab and idl
hdf5, IDL
IDL
NCL, PyNGL, python scientific packages
python
VERDI, matplotlib
OpenLayers; Google Web Toolkit; Apache: Tomcat, Ant, Struts; Java; Eclipse; Ubuntu; Firefox
WXP, GRLevel2/3/AE/Earth
grads, Gibson Ridge, bufkit
WRF and MM5. GrADS.
Matlab
NCL, IDL, GrADS
mySQL
NCL, FORTRAN
WRF and supporting packages. GRADS. python.
matlab
Grads. GoogleEarth. Matlab.
NCL, VAPOR. I've worked with SWE-derived tools, as well, for moving data that originated from IDD. However, this could be looked on as a testbed for using SWE services to distribute data in a different manner.
VAPOR and NCL scripts
MATLAB, Google Earth (GIS integration / eye-candy), GIS software (Manifold)
GrADS, NCL, MATLAB, R
Many software packages such as GR-levelX. This radar tool is superior to other radar visualizations.
personal analysis routines in C, python (specifically numpy, scipy, matplotlib, pupynere); various GNU applications, open source applications, the GNU toolchain, etc.
matlab R for analysis
FORTRAN
Postscript
html
Bufkit
Matlab, Grads
R and matlab
Environment Canada software for viewing model data
apache web server
Ocean Data View, ArcView, Google Earth, etc.
MatLab
Matlab
ArcGIS, IDL, Ruby
SSEC McIDAS
GrADS
grip-api (ECMWF), cdo and related libraries (MPI Meteorology)
NCL Graphics, GrADS, Panoply
suds/solo, NCL, Google Earth, WRF
Too many to list.
wgrib, wgrib2, WRF
Java, Matlab
ECMWF GRIB libraries, GrADS, NCL, NCO. In my view all WMO binary standards (GRIB, BUFR, etc) should be abandoned with the writing of a WMO convention on NetCDF (extending the CF?).
Matlab, JDeveloper
Vapor5D (like IDV but better support for model grids - eg. Arakawa-C).
Digital Atmosphere (for analysis plots etc).
ESRI/ArcGIS, RESC/ncWMS (standalone), GrADS.
ERDDAP layer ontop of THREDDS (to address deficiencies/gaps in TDS capabilities and integrate dissparate datasets better).
BUFKIT
pycdf, pydap and matplotlib
ncl, nomads
ESRI
GrADS
GI-cat, GI-go, geonetwork, VirtualOcean
ERDDAP
Matlab, python,
Matlab, Java
Primarily NCL, especially since it works so well with WRF output and netCDF data.
* OGC services
* Google Maps API, Google Earth, KML
* GDAL/OGR
* GMT
* Global Mapper
Google Earth, ArcGIS, Bufkit, RAOB, Excel
MATLAB, STATISTICA, document publishing tools
Programming/data analaysis & display packages like IDL, MatLah, MathCad
IDL, ENVI, SQL, Google Earth, Map Server
locally written software
RAOB, GrLevel3, GRLevelAE, soloii, CIDD
Ferret, matlab
General Mapping Tool, NCAR Graphics, IDL, Grass, ArchInfo.
python, locally written Fortran
Publishing software, word processing, statistics software, math software, etc.
AWIPS
MATLAB, Paraview, OpenMI, ...
Unix environment open source tools as well as Mac graphing software
IDL, RAOB
Java development environment (NetBeans),
ncview, ncdump, McIDAS-X, ImageMagik
Matlab, IDL, software written at EOL, and ncl.
R/MATLAB programming
GRLevel2 Analyst.
IDL,NCL
microsoft visual c/c++, cygwin, gnu c++ and nokia qt creator, mathworks matlab, aptana ajax web development ide, etc.
Several websites (e.g., Weathertap).
Live Access Server
NCL, GrADS
netbeans, tomcat
McIDAS-V
I use NCL, CESM,WRF and ferret
Pyton, Gnu-Plot, Grads,IDL
Proprietary meteorological display and data distribution tools developed by our company.
BUFKIT
GNU compilers (C, Fortran 77/90, C++), Perl, Python
IDL, MatLab, Web, NOAA Web services
IDL, Python, HDF
IDL
Gibson Ridge, Digital Atmosphere.
GrADS, NCL, surfer, grapher, etc.
HDFView
Global Mapper, GIS, SSEC McIDAS, NOAA Climate & Weather, and many video/animation software packages
R, Google Earth, SAGA GIS
aapp (atovs and avhrr processing package) from Eumetsat
ferret, nco
Google Earth
in-house (legacy) software for view/manipulate/plot NetCDF files,
Matlab,
Ferret.
ODV (occasionally)
Digital Atmosphere, WRF, ATOVS software (like AAPP+IAPP, from UK-MetOffice and Wisc.Univ.respectively), RAOB, Weatherlink (Davis Instruments).
Python, pydap, PostgreSQL, SQL Alchemy.
ncview
OppenOffice, Microsoft Office
BUFKIT, Fortran, spread sheets
NOGAPS Models, Naval Oceanographic Office Information and Data, USGS, ADDS,
grads, matlab
Various RDBMSs, Gnuplot, NASA Panoply, Pangaea, NCO suite, Google Earth.
Tecplot, IDL, Matlab
R, cdo, ncl, parallel netcdf
EPA tools such as VERDI
Matlab, NCL
grib, grib2, python, ncl, grads, matplolib, numpy. scipy
office productivity tools, image processing tools, ocean applications (ODV, Ferret, ...) Matlab, LAS,
Bufkit
NCL
Office software, various gridded datasets
WINGRIDDS
HDF, HDF-EOS, Panoply, GrADS
Matlab
compilers, grads, Metview, Gribapi, Magics
GEMPAK, LDM
NCAR graphics and NCL
Reanalyses from ECMWF, JMA
Grads
Aniscode, PCGRIDDS, Digital Atmosphere, SHARP

---

Question 29: What hardware and operating systems do you have in your department?

Response graph:

Responses breakdown:

Hardware/OS	Absolute frequency	Relative frequency
Windows	80	29.9%
Linux	113	42.2%
Mac/Apple	39	14.6%
Sun/Solaris	14	5.2%
AIX	5	1.9%
HPUX	6	2.2%
BSD	1	0.4%
HPC	1	0.4%
OS/2	1	0.4%
Unix (Unspecified)	8	3.0%
Sum:	268	100%

Responses given:

Hardware and operating systems used within department
Linux, iOS, Windows, OS/2
Linux based
Dell, RH linux, Mac OSX
MSWindows, Linux, HPC IBMs
linux
Windows, Ubuntu, Linux.
Linux, Mac
Dell desktops -- Linux (RHE 4/5)and Windows
IBM supercomputers -- AIX
linux/macs/pcs redhat linux. latest max. don't know about PC.s
Linux, Windows
Linux clusters
RHEL and Windows XP Pro
linux, pc, mac
Linux, OS X, Windows
linux and winXP.
Linux
Linux on Dell
PC hardware, windows/linux
Unix, Linux, Mac, Windows
Commodity desktops running Fedora Linux
CentOS 5.5, Windows 7, Quad core intel processors and Core i7 3 GHZ processors, minimum 500 GB hard drives
mostly linux, 64-bit Dell PCs. These machines are dual-boot so that windows can also be used. We have a 72 node linux cluster for more serious computing
2 Linux clusters, several Linux workstations for modeling, data management, LDM, Gempak, and many Windows PCs for student use (display data, run Gempak through X-windows) and for sharing of images. Several flavors of Linux, and Windows XP.
PC, Mac, Linux/Unix
Linux, Mac, PC
Dell, LINUX
RHEL, Windows, MacIS
Multicore workstations. SuSE and Debian Linux, Windows. Mac
Linux cluster
MacOS, Linux (RedHat), Windows 7.
Over 90 TB of spinning media, about 35 systems for data acquisition, analysis, archival, and data-serving, and several supercomputers. Note that this is not in the ATMO dept here.
Llinux box (X64, Fedora based, Quadcore) and MacBookPro (Snow Leopard) for my personal use, a JS21 IBM blade server as the front hardware, but most of the people use W7 or WXP
Linux, Windows
Operational servers - ranging from older Pentium to Xeon quad-core for numerical modelling, running Debian and CentOS - , workstations - Dell thin-clients and others, Windows and Linux (Debian, mostly for researchers / grad students)
AIX, Redhat Linux, SUN OS, Windows XP, Windows 7, and access to various supercomputing clusters
PC's, Macs, Linux Servers running Debian.
Various OSs (Linux, OSX, BSD, Windows [virtualization]), simple to advanced desktop machines, various sized webservers, small computational clusters.
mostly linux and unix some windows
Windows XP
Mac laptop OS 10.6
pc's, windows xp
Several unix workstation and servers and few Windows PCs
We have IBM P series and X series machines , HP Server. operating system AIX, HPUX ,Linux and Windows.
PCs+Windows+Linux
sun4u sparc SUNW,Sun-Fire-V440+SunOS+Linux
Nec
5 Linux servers, 2 for data processing, one with 1 Tb hard-disk bay.
Linux SLES 11, IBM and Hp
Windows XP,
Linux, windows and osx
PC with Windows XP
Both Linux and Windows
Sun workstations, Windows
pc with linux
HP
Dell
Linux
Windows
Linux using AMD/Intel
AIX, Linux, MacOSX
Servers and workstations with Debian, Ubuntu and Suse Linux
Our workstations are PCs running Ubuntu Linux, Windows 7 via dual boot. Several Linux based servers.
Linux, Solaris, MacOS-X, AIX, HP-UX, Windows
Linux PC Clusters
Mac OS, Windows (XP and 7), Linux (Ubuntu and RHEL), HP-UX
Windows XP-Pro, Linux on PC-based workstations and servers.
75% Linux (desktop & clusters), some 20% prefer Microsoft on desktop, sparse Macs, some legacy Unices (mostly Sun's and Alphas) for the nostalgic.
Linux, Windows
HP ProLiant-series, RocketCalc cluster appliances, Viglen HPC clusters, Dell storage servers, Lacie and Buffalo NAS appliances.
Linux, Windows XP/Vista/7, Windows Server 2003/2008.
Linux and Windows PCs
mac (osx), linux (redhat, ubuntu) , pc (win xp, vista, win7)
WinXP, Win7, MacOS
64bit Redhat (Centos) PC network on one HP server
Linux / Windows / DELL
OS used: Windows XP, Windows 7, Ubuntu, RedHat, Fedora, CentOS.
Hardware: mostly Dell servers and desktops.
PC & linux
Windows, Linux, OS X
linux, windows
Windows, Linux
Mac OSX, Windows
Linux/Windows PCs, Macs, Solaris machines. Lots of incoming data processed through multiple machines.
* Linux (CentOS)
* Windows
PCs, clusters
Linux, Windows, Mac
various flavors of Linux, Microsoft Windows 7
Mainly Windows, some Linux
Many Linux servers, mac and pc desktops
linux, Windows
Various windows os, Debian and Redhat Linux flavors, Apple OS versions and a wide variety of hardware ranging from laptops through server clusters.
x86 servers, Linux CentOS.
Linux, Mac, and Windows - Intel based hardware.
Sun, Dell intel-based hardware
Solaris, Linux, Windows XP
Most are Linux > CentOS 5.4 as well as MS Windows on older machines.
Intel-based computers running either Win XP Pro or Redhat Enterprise Linux
Windows XP, Windows 7, Various Linux distributions
SuSE Linux i586
Centos Linux i386 and x86_64
Solaris Sparc
Mac, Unix (Gentoo), PC (Windows XP & 7), Solaris Unix
intel linux sun sparc
Linux, Mac, Windows
Servers are all Linux (CentOS 5 I think). Desktops are mixture of Linux, Windows, Mac and Solaris.
mac, PC, Linux, Unix
Redhat Linux, Windows, OSX on 32 and 64 bit x86
workstations and servers.
Linux
Windows XP, 7 and Linux (Red Hat)
windows, unix
Linux 64 bit system computers with OS Mandriva
Mac OS(Snow Leopard), Windows(XP,Windows 7).
Solaris Sparc, Solaris x86
mcgill university geograsphy department's hardware, would be too long to describe in here, (in the lab I work) unix machine from sun, intel based sun workstation, multiple ms-win os desktops, unaware of the depatment-level big picture.
~30 PCs, dual-booted Windows and Linux
Linux virtualized in an IBM Blade
Red Hat Linux Enterprise; Dell servers.
Mac OS-X, Linux
windwos7(64), Centos5
UNIX
MS Windows 7 and Linux, both 64-bit
I have a 64 bit Linux PC running on the Fedora 12 platform; I also have access to NCAR bluefire and a local Linux cluster.
Mac, Windows, Linux
Numerous Intel x86 and x86-64 servers
Linux (CentOS, Debian)
Windows XP, Vista, 7
Windows-based PCs and server; a few MAC
AIX, Solaris, Windows
LINUX, Windows
Single and multi-core x86 systems as well as multi-core x64. I use a variety of dedicated servers as well as end-user pc systems.
windows/linux;
PC and PC cluster, IBM super computer
i386 CentOS/Windows7/MacOS
pc, mac, linux/cluster
SUNOS, UNIX, MacOS
Multiple PC, laptop, networking, etc.
Windows, OS X, Linux, Unix
PC/linux(centos) and SUN/Solaris
SGI-Altix ICE, SUSE-enterprise Linux
Linux workstations
MacosX personal comupters
Centos/AMD/Intel
Linux servers/ RedHat / Apache
Microsoft Win7(XP,2K)
Apple OS-X
Pentium processors.
MS XP & 2003/2008 Server, and VMWare with Ubuntu and Debian virtual machines.
Macs, OSX, HP blades, RHEL5, Windows.
linux, windows
PC, Windows
Dell Precision Workstation 7500 plus 32 PC's with
GEMPAK5.11.4 for students access.
Windows Linux -- Not sure about hardware
pc's - linux
cluster - pc's - linux
Linux, Mac OS X, UNIX, Windows.
Mostly MS-Win platforms, some Linux
Series of small PC's and multi-core computers. Some systems are running MS Windows and the rest is using various distributions of Linux
Solaris
linux cluster
Linux, Sun Solaris, OSX, Windows XP & Windows 7
A few legacy Sparc machines; rest is Intel with a few AMD; mostly 64-bit multi-core
Linux
linux & Windows
Custom built quad-core processor, 8Gb of memory, 500gb hard-drive. Fedora 13 64-bit. Bufkit installed on Windows 7 partition.
Linux, Linux cluster
Numerous desktops and laptops, mainly Unix-based with a few Sun/Solaris based systems for good measure.
HP Windows XP
Linux, Mac OS X, Windows
LINUX Redhat
Sun, IBM, Fujitsu
Solaris, Linux, Windows
PC Client/Server environment running both Linux and Windows
Solaris and Linux
linux workstations
Redhat Linux, Win XP
Linux (Various Distributions)
Windows XP

---

Question 30: What hardware and operating systems do YOU use?

Response graph:

Responses breakdown:

Hardware/OS	Absolute frequency	Relative frequency
Windows	76	29.9%
Linux	128	50.0%
Mac	30	11.7%
Sun/Solaris	13	5.1%
AIX	2	0.8%
HPUX	1	2.3%
Unix (Unspecified)	6	2.3%
Sum:	256	100%

Responses given:

Hardware and operating systems used personally
Dell desktop with Windows 7 64-bit
Ditto
Dell, RH linux, Mac OSX
See 17
pc hpc
Windows, Ubuntu
Linux, Mac
Dell desktops and servers -- Linux RHE 4/5
MAC and linux.
Linux
MAC OS
linux and pc
Linux, OS X, Windows
linux
Linux/Windows desktop quadcore
Linux on Dell
Mac hardware with OS X
Mac, Linux, Unix, Windows,
Commodity desktops running Fedora Linux
Ubuntu
Same as above
linux OS (windows on laptop for interaction with others at university). Dell optiplex 745 desktop, Dell Studio XPS laptop. I also use the 72 node linux cluster.
Windows XP, Vista, Linux on the same collection of hardware mentioned above -- mostly x-86 workstations/PCs, and the 2 clusters.
Sun/Linux
Linux
Dell, LINUX
RHEL, Windows
Multicore workstations. SuSE and Debian Linux, Windows
Linux
MacOS, Linux (RedHat)
Linux, Ata-over-Ethernet and iSCSI RAID, Intel and AMD-based PC's and servers
MacBookPro 15'' and Linux box (Fedora 10 X64 running on a Quadcore)
Linux, Windows, PC
1) Core i3, 6GB RAM, 1TB disk, Windows 7 64bits + CentOS 5.4
2) HP tx2500us (AMD Turion X2, 3GB RAM, 250GB disk), Windows 7 32bits, Windows Vista Pro 32bits, Fedora 13 (virtualized)
Supercomputing clusters, AIX, Redhat Linux, Windows XP
Same as above.
Intel Core2 Duo CPU T9300@2.50GHz, 8GB RAM.
lesserwhirls@microsat-xps:~$ uname -a
Linux microsat-xps 2.6.32-24-generic #42-Ubuntu SMP Fri Aug 20 14:21:58 UTC 2010 x86_64 GNU/Linux
laptop with linux
Windows 95/98/XP
DOS
Mac G5 OS 10.5
Unix
same
Both MacOsX and Windows XP
Unix workstations and a personal Vaio Windows laptop
Xeon workstation, Linux
HPUX,AIX,Linux and windows. Mainly use Linux
PCs+Windows+Linux
sun4u sparc SUNW,Sun-Fire-V440+SunOS+Linux
Linux Fedora Core systems / Ubuntu.
Dual-core PC servers (outdated, bought in 2005).
OpenSuse, Win Xp
AMD Athlon, 64 bit
linux and osx
Same
Both Linux and Windows
Ditto
pc with linux
HP
Dell
Linux
Windows
Linux and Windows using AMD/Intel
AIX, Linux, MacOSX
Ubuntu
PC running Linux.
Solaris, MacOS-X, Linux
Linux Debian PC
Mac OS X (MacBook Pro)
Multi-processor Dell workstation running Scientific Linux
Just Linux
Intel-based
Rocketcalc Saturn mini-clusters, Viglen HPC rackmount clusters, HP DL360 and DL380 servers running linux.
Linux, Windows XP/7, Windows Server 2008.
Linux, Windows 7, Windows Vista, Windows XP
mac (osx), linux (redhat, ubuntu) , pc (win xp, vista, win7)
WinXP, Win7, MacOS
64bit Redhat (Centos) PC network on one HP server
Linux / Windows / DELL
linux
All of the above. I work on multiple operating systems / servers.
PC and Linux
Windows, Linux, OS X
linux
Window, Linux
Mac OSX
Primarily Linux/UNIX for data processing and plotting, and Windows for visualization and analysis (easier for writing papers and PowerPoint presentations). Dual-booting makes this easy, though.
* Linux (CentOS)
* Windows Vista, Dell
Linux, Windows
AMD Phenom II X4 with Ubuntu
Mainly Windows, some Linux
Red Hat Linux servers, mac OSX laptop and desktop
linux, Windows
WinXP, Debian Linux.
x86 servers, Linux CentOS.
Linux and Mac.
Intel-based solaris
Same
Intel-based computers running either Win XP Pro or Redhat Enterprise Linux
Mostly Windows XP, but for time to time all OSs mentioned above
SuSE Linux i586
Centos Linux i386 and x86_64
Solaris Sparc
Mac + Gentoo; less Solaris and Windows XP
Mac and Linux
Dell Latitude D430 Laptop, Windows XP
mac, PC, Linux, Unix
Linux and Windows
Linux
Windows XP, 7 and Linux.
windows
Mandriva OS linux computer
MacOSX, Ubuntu Linux, Windows XP, Solaris
linux redhat fc12 and fc13, redhat linux centos distrib, ms-windows xp, 2003 server and ms-windows 7
PCs and Mac laptop; Windows, Linux, Mac
Linux, Windows
Red Hat Linux Enterprise; Dell servers.
Mac OS-X, Linux
Centos5, Windows7
UNIX
HP notebooks
I use Fedora 12 on the Linux PC
Linux,Ubuntu,i686 GNU/Linux,x86_64 x86_64 x86_64 GNU/Linux
Linux (Debian)
Windows XP Pro
MAC and Windows OS PC
See above
LINUX, Windows
All x86 and x64 based hardware. I primarily use linux with windows on one machine for proprietary software. As far as linux flavors, I prefer Gentoo but am required to use CentOS and Ubuntu for some ingestion software.
windows/linux, PC and PC cluster
linux, windows
i386 CentOS/Windows7
linux/cluster
MacOS, UNIX, SUNOS
Windows & OS X
PC/linux(centos) and SUN/Solaris
SGI-Altix ICE, Cray XT6, IBM Power 6
Linux variants
CentOS/Intel
Linux/RedHat
MS Win7
the same of the Item#17
Macs, OSX, HP blades, RHEL5, Windows.
linux
PC, Windows
Linux Fedora 13 on another Dell Precision workstation and a Dell desktop.
Windows, Linux
pc's - linux
Linux. UNIX.
All: MS-Win, Linux, MacOSX
Dell Precision T7500, running RHEL and a smaller dual core PC with Fedora 13 Linux.
Linux
linux cluster and Windows desktop/laptop
Linux (mostly CentOS, a little Ubuntu); OS X, Windows Xp, Vista, 7.
Intel/ AMD
Linux, 2 x Intel(R) Pentium(R) 4 CPU 3.00GHz, RAM 1025824 kB, HD 122.9 G, I mostly work on network servers tough.
various
As above
Linux
For laptops, Mac. For desktops, Fedora. Personally own two of each setup, with one aging Windows machine in case of dire necessity.
HP Windows XP
Mac OS X
LINUX Redhat
Sun, IBM, Fujitsu
Solaris, Linux, Windows
High-end PC running Windows XP and Cygwin
Solaris and Linux
Win PC, linux
Linux and Win XP
Intel Core2 Duo E8600
Both Linux and Windows XP

---

Question 31: What is your overall level of satisfaction with the Unidata program?

Response graph:

Response breakdown:

Level of satisfaction	Absolute frequency	Relative frequency	Adjusted relative frequency
Highly Satisfied	91	37.92%	45.05%
Satisfied	102	42.5%	50.5%
Indifferent	7	2.92%	3.47%
Dissatisfied	2	0.83%	0.99%
Sum:	202	100%	100%
Total answered: 202

Comments regarding GEMPAK & AWIPS 2:

• Response to GEMPAK problems are pretty lacking.
• Generally very happy, although minor issues sometimes come up (e.g., right now nmap2 is having problems accessing archived data on our 64-bit machines, but not on 32-bit ones, garp doesn't state IR cloud top temp. on 64-bit machines but does on 32-bit.) Sometimes help when issues have been emailed never comes or takes a long time. Just to reiterate, without Unidata, our data access and display capabilities would be much fewer and poorer quality. I have been using Unidata for more than 2 decades, and wouldnot be able to do my jobs of research and education without it.
• Unidata does a great job providing training updates and support to all of the packages I have used. The only reason I don't give you a highly satisfied rating is that the GEMPAK support has fallen off noticeably. The timeliness of replies to help-gempak@unidata.ucar.edu has not been what it was and current staff seems much less willing to help with problems than previously was. No offense to current staff but it would be nice to seem them be more interactive with folks in the GEMPAK community - especially with the new release. All of the other Unidata support folks have bent over backward to help and you really felt they would help you resolve the problem no matter what. I don't get that same feeling in the GEMPAK support arena these days.
• Support for GEMPAK has been lacking.
• The quality of Unidata's GEMPAK support has not been what it used to be.
• I'm very pleased with the Unidata program as it pertains to individual weather buffs. How it relates in the academic field is something I have no experience to comment on. I've always been able to find problems I may have in running or updating GEMPAK/Nawips and support staff has been of tremendous service to the community.

Comments regarding the IDV:

• IDV is a very high quality package provided freely.
• Good, however I find that IDV is a little unstable.
• I like IDV.
• IDV works very well in our daily, hourly publishing to my web site.
• As mentioned earlier on IDV appears to be in deep trouble after losing the 2 key developers.

Comments regarding netCDF:

• The documentation for netCDF and the conventions are mostly why I go to Unidata. Would like to see more support for Python.
• UDUNITS has issues with Fortran. The way the netCDF libraries were supplied with the latest netCDF4 made it harder for users to compile as old makefiles could not be used.
• Unidata data tools, mostly related to TDS and netCDF (with CF) are central to our dataservices.
• NetCDF is clean data format, having data and meta data together in single file.
• NetCDF works nicely, without bugs, noticed by me, at least. It is the best data format for exchange of information I have used.
• I just cannot work without NetCDF. Our storage mechanisms are based on NetCDF.
• As I work mostly with netCDF, it has been relatively easy to retry and manipulate data. I do data conversion and analysis with other software such as NCL and GrADS.
• The netCDF4 format provides an efficient, machine independent data format that is ideal for storing and sharing output from our very large computer simulations. I have installed netCDF4 on a large number of architectures and always found adequate support for the build process. The chunking and compression options now available are invaluable in reducing storage requirements whilst still allowing easy access to subvolumes of data for analysis. Netcdf4 is my main interest in Unidata; I have little to do with your other products or services.
• A lot is possible for working with and distributing data once it is in NetCDF and TDS.
• We deal in more and more high-volume datasets (local and remote) - so OPeNDAP support is absolutely required, and THREDDS (TDS) is preferred as a baseline concentrator for coredata.
• The core support for NetCDF, TDS and other base level libraries that Unidata provides is critical for our operations and the desire we have for data interoperability.
• Having used netCDF for ~15 years I value netCDF as having potential for long-term data archival and exchange. Support beyond the Linux/Unix-derivative environment could help in further establishing a widespread use of the data format.
• I mostly use netCDF to get my stuff going. So far I have been able to configure and use it on the multiple platforms available at my department, just by using Unidata's FAQ.

Comments regarding UDUNITS:

• UDUNITS has issues with Fortran. The way the netCDF libraries were supplied with the latest netCDF4 made it harder for users to compile as old makefiles could not be used.

Comments regarding the THREDDS Data Server:

• Unidata data tools, mostly related to TDS and netCDF (with CF) are central to our dataservices.
• A lot is possible for working with and distributing data once it is in NetCDF and TDS.
• We deal in more and more high-volume datasets (local and remote) - so OPeNDAP support is absolutely required, and THREDDS (TDS) is preferred as a baseline concentrator for coredata.
• The core support for NetCDF, TDS and other base level libraries that Unidata provides is critical for our operations and the desire we have for data interoperability.

General comments we received:

• In the last few years, more posts to Unidata lists go unanswered. Even though community support is a focal point, it would be useful for Unidata management and governing committee members to acknowledge some of the posts where appropriate. A few instance s of posts by governing committees and UPC management were extremely valuable in describing positions on AWIPS II etc, but these are few. Social networking access is not allowed into corporate networks, so email lists, digests and archives/blogs are still appreciated. There is a disconnect among packages which is evident when new data evokes fixes/bugs that trickle in with each package independently and not communicated among groups.
• I'm very happy Unidata exists, keep up the good work, thanks.
• Lots of smart dudes.
• I don't know what the meteorological community would do without the necessary bodies to maintain data formats, data distribution, etc. It is hard to put a value to something that is so well done and needed.
• We depend on Unidata heavily. I'm somewhat concerned about the seemingly sudden departure of some of the best talent there.
• Customer support at Unidata is what it should be everywhere. That's one area that Unidata shines. They also listen to the community. I wish, however, there were software that does better at Web publishing easier and also gives more examples at scripting. The latter I would like to see more emphasis on. "Show me, and I'll understand" should be a mantra of Unidata.
• Excellent support from the team. Congratulations!
• Overall, the work from the techies is excellent and supports the users well. I only miss a faster and more powerful tool (IDV) to work with for research purposes.
• Exceptional and dedicated community, strong commitment to researchers and students inc. outside the US, excellent tools to visualize and manage data with strong focus on facilitated usage, reliable support and maintenance of tools offered.
• The issue surrounding plotting and visualization software is a big one that still remains to be addressed by Unidata.
• Unidata is an excellent service to all educators. Their assistance is invaluable.
• The Unidata program understands that the community is its lifeblood; Unidata takes this very seriously and it shows.
• Tools provided are useful, simple to implement and have a good level of support.
• Am connected at home to the Internet with a half-speed dial-up line and don't have enough time at school (where I have fast wireless) to finish your programs.
• We have been receiving data through the IDD for over 15 years. The service has been very reliable. We have also used extensively the GEMPAK plotting package with great success. We are in the process to transition to the IDV, but we are finding it quite difficult.
• Continuity in the program, community is very strong.
• Unidata program gives me access to very powerful tools and a lot of data from my country that I don't have access in other way.
• Unidata continues to make my job easier.
• The tools we use are very good and offered in the right way to the community - that's making especially netCDF to a real success-story!
• Overall Unidata does a great job - very helpful to a small institution like ours.
• Use of functions could be better explained and the relation of each function with the other ones, in order to save time while programming.
• More programming tools/libraries for high-level functionality would be useful (eg. station-plots, generic wind-profile, etc).
• The Unidata program has provided me with the tools to succeed in both undergraduate, graduate, and my current job in the private industry. The freedom of information, dedicated support, small but active community, and continual development have provided and will continue to provide students and professionals with the tools they need to advance their understanding of Earth Sciences!
• I've really only used NetCDF; I've found the sign-up process mildly intrusive (I hate having to remember a username and password for a site I rarely visit), and have never found the installation for more advanced data sharing systems to be trivial enough to outweigh the annoyance of linux installation details. While I consider myself a decent sysadmin for very small research clusters, I find the ease of use of data service programs is buried behind a lot of jargon that makes it difficult to identify the precise system I need to implement for my needs.
• I'm a big fan of Unidata, and have enjoyed working with them to this point. I hope that relationship will continue for many years to come!
• The tools work well and openly distributed. Feedback is taken and used.
• I find that Unidata is representative of the goals presently in place; they have been receptive to "outsider" suggestions and queries without complaint.
• They provide me with the base from which I can build analysis tools and visualization of meteorological data. It also allows collaborations with NWS.
• LDM, netCDF, GEMPAK and IDV are the packages we use most often. They have all become integral components of our operations to varying degrees.
• Great independent high-quality components/tools.
• I believe Unidata needs more resources to serve its expanding community and to ensure continued good engineering of software products. The potential is there, but further development is needed. Documentation is needed. Better scripting tools are needed.
• I remember what it was like before Unidata and the internet. It is much easier to teach and
• to do field research now. The ability and ease that one can access realtime data in the classroom and remote field sites is amazing. Much of that can be attributed directly to Unidata.
• I LOVE Unidata. I wish I was able to use the tools more often, but I really find myself struggling when it comes to working with it. I think there needs to be better documentation within the Unidata family... but not just long text documents with a few figures here and there. I'm thinking something like NWS's Warning Decision Training Branch's DLOC: http://wdtb.noaa.gov/courses/dloc/index.html
• I'm not working in the Earth System science sector, so to me, the tools offered by Unidata are often too focused on Earth System science. Many of the visualization tools would be great, if they were just a bit more capable of defining custom views from the data (e.g. based on netCDF). This would involve indexed reads (row compressed data), as well as general heat map visualizations etc.. Maybe I just overlooked those possibilities, but they would sure add a lot of benefit to the viewer applications offered by Unidata, also for other disciplines.
• Sadly, we were not trained on site at Unidata.UCAR.EDU. Given that, it remains hard to comply to CF-1.0+ and COARDS specifications for various cartographic projections.
• Unidata provides an excellent suite of tools. It could do more marketing to the climate community and help them along.
• I am impressed with the improvements to IDV in recent years. I am grateful for the large archive of mailing list messages for troubleshooting LDM and NetCDF (other developers use the lists for other packages). Also grateful that the tools and data are provided free of charge... they are a boon to our company and to my use privately at home.
• Unidata has little or no competition within the atmospheric sciences community in the area of data and software services, and while for some organizations this could lead to a retentive approach to providing services, Unidata on the other hand continues to be leaders in innovative approaches to data services, software development, and leadership for your community. Thank you.
• I think it helps me a lot in my research work.
• We've been users of Unidata products since the beginning and have always found their services to be excellent. It would be nearly impossible to function without them as we rely on the data and software they provide for education, outreach, and research.
• Minimal use, but would like to expand to have better tools for manipulating/viewing our NetCDF data collection. They probably exist, I don't know much about them.
• Is a very important source of weather technologies that help us to study of meteorology status and trends.
• The meteorological community is fortunate to have the Unidata Program and dedicated professionals to serve our data access and weather analysis needs.
• Unidata's software stack provides an integrated environment of well designed, high quality, powerful data management & analysis tools. Frankly, we could never allocate resource to develop similar tools in-house. You provide atruly great service to the Earth science community. Thank you!
• Ease of use of netCDF compared to grib1/2, but no compression.
• Unidata is to be commended for the pioneering work it does, and for the high productivity it achieves with modest staffing levels. The challenge for Unidata will be to cope with the demands that come from its successes.
• It is a very versatile product. It should be more user friendly.
• Unidata tools are important to our dataflow and data management infrastructure, in particular LDM is used extensively. GEMPAK is used for data visualization, analysis and some automated/scripted product development but less so than LDM.

---