Water Resources Division
ITAC Home Page
|
|
|
| |
Water Resources Division |
From: |
Gregory E Granato |
To: |
Stephen F Blanchard/WRD/USGS/DOI@USGS |
Cc: |
GS-W ITAC, Robert M Hirsch/DO/USGS/DOI@USGS, Gregory E Granato/WRD/USGS/DOI@USGS, jim@zope.com |
Date: |
10/16/01 12:58 PM |
Subject: |
Re: Computer graphics in WRD (and the other disciplines?)--Thank you |
Mr. Blanchard:
Thank you for your reply. I hope the ITAC committee will consider Mr Fulton's suggestion, his experience, his background and his knowledge of our needs. (I have no stake in this issue except the desire for excellent and efficient public-domain graphing software). I know money is tight at all levels, I hope the ITAC has a way to weigh the small (the cost of one WRD district project) development costs against the productivity costs (GS 7,9,11,12,13,14, and 15's spending hours doing graphs). I can not speak for Mr. Fulton, but maybe he would be willing to contribute to the January ITAC meeting.
Outsourcing this task may be warranted since many computer folks in the USGS seem to be over tasked and understaffed. Also, the Technology Enterprise Office has methods and contacts for cooperative research and development agreements (CRADAs) with the private sector, which can be a source of funds and or expertise.
Thanks again for your consideration of this issue,
Greg
**********************************
Gregory E. Granato
Hydrologist
U.S. Geological Survey, MA-RI District
10 Bearfoot Road
Northborough, MA 01532
email: ggranato@usgs.gov
http://ma.water.usgs.gov/ggranato
phone: 508 490 5055
fax: 508 490 5068
******************************************
From: |
Stephen F Blanchard |
To: |
Gregory E Granato/WRD/USGS/DOI@USGS |
Cc: |
Robert M Hirsch/DO/USGS/DOI@USGS, GS-W ITAC |
Date: |
10/16/2001 12:10 PM |
Subject: |
Computer graphics in WRD (and the other disciplines?)--Thank you |
Greg:
Bob Hirsch asked me to respond to your email regarding graphic in WRD.
You comments are correct regarding the loss of G2 as our DGs go away. Your statements regarding the fact that S-Plus and TkG2 offer less capabilities than G2 are also correct. Unfortunately, there is little that we can do in the immediate future to alleviate this problem. There are no commercial off the shelf graphics packages, that we are aware of, that will do what we need in WRD. That is why G2 and TkG2 were developed in the past. The Information Policy Advisory Committee (formerly CPAC) previously addressed this issue and decided that WRD should not be in the business of developing graphics software packages. That is partly why S-Plus was selected to provide us with some graphics capabilities with the hopes that S-Plus capabilities could be enhanced. The groups here in HQ that are most directly responsible for the graphics software, DIS and NWIS, both think that it would cost hundreds of thousands of dollars to develop the graphics software in house and that it probably would have to be done using Java. We are not prepared to go down that road yet but we do acknowledge that the issues you raise are valid. What I will do is to put this topic on the ITAC meeting agenda for our next meeting (in Jan) so that we can re-look at the issues you raised. Thanks for sharing your thoughts with us. If you would like to discuss further, pls give me a call. Thanks
*************************************************************************
Steve Blanchard
Acting Chief, Office of Information
U.S. Geological Survey Phone: 703-648-5629
440 National Center Fax: 703-648-5644
Reston, VA 20192
**************************************************************************
From: |
Gregory E Granato |
To: |
Robert M Hirsch/DO/USGS/DOI@USGS |
Cc: |
Gregory E Granato/WRD/USGS/DOI@USGS |
Date: |
10/09/01 09:06 AM |
Subject: |
Computer graphics in WRD (and the other disciplines?)--Thank you |
Dear Mr. Hirsch:
In the districts, and at HQ our Data General servers are dying. The program G2 will die with them. G2 is currently the only program that will produce graphs with the proper format (for example,proper fonts, proper layouts, with the ticks on the inside on all four sides, etc.). G2 will also produce graphs in postscript formats compatible with publications graphics formats. When the unix servers die thousands of hours of training and work will be lost because existing templates and batch programs for graphing hydrologic data will be lost. Neither of the proposed replacements tkG2 or S+ are an improvement (actually both are a step backward in terms of productivity and the ability to create layout-quality graphics in proper USGS formats). There is also a risk that data may be distorted or changed each time an illustrator has to redraft a figure to bring it into compliance with publication standards.
The cost for providing a top-notch graphics package should be weighed in terms of the fact that scientific graphics are our life blood for communicating with the public and with decisionmakers. Also, the cost of a robust graphics package would be small in comparison to the thousands of hours USGS scientists at the GS-7,9,11,12,13,14,and 15 spend graphing each year and the thousands of hours that GS-7,9, and 11 illustrators spend processing these graphics for publication.
In the districts, we are being constantly squeezed to do more with less resources (and growing overheads). Therefore, good tools are necessary to increase productivity to meet these challenges. HQ as the unifying source of information and hydrologic tools should step forward to work with a private company to modify an existing graphics package to our specifications (as the default) or to create a package for us. As one option, I have searched for Mr. Fulton (the original USGS author of G2) to see if he would consider upgrading the program for new operating systems.
I am sending this to you because I have contacted our local computer folks and several computer groups at headquarters. I have also approached my own chain of command. Each person and group seems to be busy at their own task, and because these folks do not face the difficulties of graphing data for interpretation and publication there is no motivation to mount a response.
I hope that this information is helpful. I know that you have many (more) important issues to consider, but a small pebble kicked from the top of a mountain may move more earth than a large boulder pushed from the bottom.
_______________________________________________________
| Greg,
The following is Mr. Fulton's response to my query about the possibility of a G2 upgrade. If this is a viable alternative, an improved version of G2 in an open source graphics package published as a USGS OFR and distributed freely over the web with the USGS logo in the interface may be a positive addition for outreach to the scientific and education communities. It could put us on the desktop of thousands of high school and college computers. "I really don't think that this is an opportunity for Zope Corporation, however, I'd be willing to try to help you out. Here is an option you might consider. Have G2 rewritten/ported to Python. This is necessary because, unfortunately, GNU Smalltalk is no longer supported or widely used. Python, OTOH, has extremely wide usage and support. It is the second most popular very-high-level language, after Perl, and I predict it will surpass Perl in a few years. Python has extensive libraries for interfacing with Graphics libraries, databases, and so on. Python would certainly provide much greater performance and stability over the GNU Smalltalk implementation you have now. I would definitely have moved G2 to Python had I remained at USGS. Moving to python would require:
A *wild* guess at the level of effort is about $100,000. This just gives an order-of-magnitude estimate effort. I would imagine i'd accomplish this by assembling a team of Python developers outside of Zope Corporation. This is something that I would do as a favor to USGS. It would not be a significant financial opportunity for me, and Zope Corporation would not be involved. Also note that I would only entertain this if the code was Open Source. This project would be managed as an Open Source project. I could see reducing the cost in three ways:
We could control risk by making this the goal of a first phase. If the first phase is successful, then later phases could introduce more portable graphics or winding libraries. - Get some of the work done by USGS personnel, under my supervision. - Make sure USGS bears the risk. That is, rather than me committing to a price, we come up with some sort of suitable hourly contracting/ consulting arrangement and we use mutual project management to control risk.If something like the above works for USGS, then we can talk in more detail. I'll reiterate that I'd be doing this as a favor. I wouldn't be making a lot of money. If USGS wants my help, they're going to have to make it easy for me. I'm not going to spend a lot of time selling this and I'm not going to have a lot of tolerance for contracting/procurement hassles. This later means that if we do decide to do something, I'd probably need to sub to one of your existing contractors. Take care, Jim P.S. I'm always interested in USGS news. I'd love to hear what's going on with reorganizations, DIS, NWIS, etc. --Jim Fulton mailto:jim@zope.com Python Powered! CTO (888) 344-4332 http://www.python.org Zope Corporation http://www.zope.com http://www.zope.org |
Address questions or comments to Charles Merk, ITAC Secretary
cfmerk on (703) 648-5680