Answers for Silva.1793831248346.xlsx
XLSX spreadsheet 15 KB Posted
- Attached to
- SILVAH software coding
- Solicitation number
- 12760426Q0195
- Issued by
-
Department of Agriculture Forest Service
About this file
This file is a Questions and Answers document addressing market research inquiries regarding repair and modernization of SILVAH, a forest stand growth simulator software application maintained by the USDA Forest Service.
The immediate scope focuses on repairing SILVAH version 8, a Windows desktop application comprising 81 files written in Visual Studio 2019 C++, which currently exhibits known defects including incorrect thinning prescription pathways and failure to incorporate middle plot data into stand tables. The contractor must fix the broken FSVEG interface, address the incorporation of sapling layer data from the fixed-area middle plot, and ensure the application runs on Windows 7 and above. The Government will provide source code, build environment details, and test data only to the awarded contractor after contract award, not during the proposal phase. The contractor should possess their own compiler and toolchain; remote performance is permitted with video collaboration available with Forest Service scientific staff for consultation on silvicultural rules and validation of results. The work is anticipated to require less than one calendar year and is expected to be awarded as a fixed-price contract with a time limit. Full source code, build scripts, and tests must be delivered as contract deliverables with Government ownership rights. Future SILVAH version 9 development—which may modernize the data format, migrate to a different platform or language, and potentially restore version 7 capabilities such as FSVeg, NED-3, and TCruise import functions—is explicitly stated as beyond the current solicitation scope and dependent upon a steering committee's future planning.
View the file
Other files
| File | Type | Posted |
|---|---|---|
| RFI_Silvah_12760426Q0195.pdf |
On GovTribe
Work with this file
- Download the original file
- Contacts named in this file
- Similar government files
- Ask GovTribe AI about this file
Text version
Sheet1
| Questions | Government Answers |
| How large is the Silva 8 codebase (files/lines of code) | SILVAH 8 is 81 files plus a few linked outside the folder it is in, some files have nearly 2000 lines of code, while others are small with 50 lines of code. |
| What compiler, tool chain, and GUI framework is it built on, and what Window versions must it run on | SILVAH 8 runs on windows 7 and above. Was made in Visual Studio 2019 I think, not sure what tool chain and GUI framework. |
| Do v7 and v8 share a codebase, or is v7 a separate implementation (and in what language)? | They have separate codes. Windows 8 was a whole new approach and was the first version not to be backward compatible. Windows 8 may use some pieces from windows7 but was edited to fit the new approach |
| Is there a known defect list for the v8 "repair" work? | There are some known bugs in SILVAH 8, one is simple where the wrong path is taken on a flow chart. The other is the 16 ft radius midle plot should be merged into the overall stand table for sapling layer trees. The plot is used for predicting stump sprouts but not incorporated into the overall overstory stand table. |
| Will the Government furnish full source, build environment, and test data? | We have the source code and will provide it, can provide test files but expect the contractor to have the software to write and compile. |
| Is the original developer or anyone who knows the code still available to consult? | The original programmer does not want to be involved as he has retired. The scientists that helped build are available, though one is retired and has limited availability. |
| Must v9 keep backward compatibility with legacy .sil/.def files and NED-2 import, or can the data format be modernized? | We would like to keep backward comaptibility in SILVAH 9 when we get to that point. We need to fix 8. SILVAH 8 will be difficult to import to/from because the data are formated differently and there is difference in the sample plot size. SILVAH 8 defaults are stored within the data file. |
| Are the user's guide and help system in scope for v9? | Users guide will be written by the Scientists in consultation with the programers if needed. |
| Are there any restrictions on using AI-assisted coding tools on the codebase? | Not really as long as they are locally stored and not in a cloud where the code can be compromised or shared without our knowledge. |
| Can the work be performed fully remotely, with collaboration with the Forest Service scientific staff by video call? | Yes. |
| Will the SILVAH 7 and 8 source code, or details of the current build environment (compiler, IDE, libraries), be available to respondents before a solicitation? | Code will not be provided prior to solicitation, but will be once awarded. It was compiled and written in Visual Studio 2019 or around that version. |
| Is the source code under version control, and what is its state (complete history, or archived releases only)? Will the code be provided to respondents during market research, or only to the awardee? | Source code is supplied to the awarrdee only. It is not under a version control system, though changes are documented within the software in comments. |
| . What is wrong with version 8? Is there a list of known defects, crashes, or incorrect results, and approximately how many are open? 5. Is the intended repair limited to fixing defects, or does it include porting to a newer compiler or operating system? | It cannot reach Thinning prescriptions and does not incorporate the "middle" plot into stand data. The intent was for the sapling layer to come from the fixed area middle plot. An interface with the FSVEG system is broken because the GOV restricted command line SQL calls. |
| Does SILVAH have an automated test suite, regression data sets, or validated reference outputs? If so, approximately what portion of the code or functions do they cover? 7. What documentation exists (design or architecture documents, algorithm descriptions and scientific references, code comments, user manual)? Is it current for version 8? | There is not an automated test suite. We compare data run in new versions to known good runs from previous versions. The data iften are hand calculated for validation where appropriate. SILVAH 8 has refences in comments as to where the equations and nmbers are derived. SILVAH 7 does too. |
| Is there a draft feature list or requirements document for version 9? | We are not yet ready to make SILVAH 9. That is a next step. We will be guided by a steering committee on what features we want. |
| How much time do Forest Service scientists expect to be available to explain the silvicultural rules and growth models, and who will validate that results are correct? | Scientists will be available as needed, and will work with the programmers to validate. |
| What target platforms and delivery format are expected (for example Windows desktop application, installer, web)? | Windows Desktop app for SILVAH 8, when we get to SILVAH 9 it may be a different platform. This scope is limited to fixing SILVAH 8. |
| What period of performance and level of effort does the Government anticipate (for example, approximate staffing over a base year)? | We cannot determine that, we expect the work to take less than a calendar year |
| Is remote performance permitted? Are there any personnel security requirements (for example a background investigation) or Government-furnished environments that vendors should plan for? | Remote work is fine. If we fix the interface with FSVEG, a user needs to authenticate into a Forest Service application, and I am not sure what is required for this. |
| Is the intended contract type fixed-price, time-and-materials, or other, and is a total small-business set-aside still anticipated for the solicitation? | Fixed Price seems most appropriate for this and with a time limit set. |
| The RFI states the goal "to produce a new SILVAH version 9 coning some features from SILVAH 7 and 8." Is version 9 expected to remain a Windows desktop application written in C++, or would the Government consider another platform or language for version 9? | SILVAH 9 is not really a part of this and should not have been included, because it depends on a steering committee and planning group to determine what it will look like and what feature it needs to have. |
| SILVAH "contains a wildlife attributes report, forest stand growth simulator" and supports testing alternative cuts. Will the Government provide reference stand data with the expected SILVAH 7 and 8 outputs, so version 9 can be tested against them for acceptance? | Yes, we can provide them, but SILVAH 9 is beyond the expected work in this solicitation. |
| Does the Government expect delivery of full source code, build scripts, and tests as contract deliverables, with Government rights to them? | Yes. |
| Which version 7 features must return in version 9, and which version 8 features must remain? If that list is not written yet, who will write it, and when? | That is dependent on a steering committee and is not yet written and SILVAH 9 is a next step beyond the scope of the work being solicited. |
| Must version 9 stay in C++, and must it keep reading the stand files and inventory databases that current SILVAH users already have? | SILVAH 9 will need to read the old files, but does not have to be in C++. It is a next step though and not part of this solicitation. |
| Is SILVAH a desktop program users install, or does version 9 need to be hosted? Is an Authority to Operate or any other security package part of the work? | It will be a program that is installed as of right now, but we are not ready to design SILVAH 9 at this point. It is a next step. |
| Is there a user manual, a published description of the prescription rules, or a current maintainer who can explain the code? | Yes to both of those things. |
| Testing: are there reference datasets with known correct outputs (for example prescriptions or growth projections) that can be used to confirm the repaired and new versions give the right results? | Yes, we can provide files to use for testing. |
| Users and support: approximately how many people use SILVAH, and is the contractor expected to support end users or only the Forest Service staff | The number of users is not known, though many people use it across the northeast and into parts of the mid-west. Support call will be to the USFS staff |
| Will the Government furnish the SILVAH 7 and 8 source code, project and build files, compiler and toolchain details, and a list of known version 8 defects? At what stage: with the solicitation, or after award | After the award, the source code will be provided. The compiler is an expected piece the programmer should have in their own toolbox. |
| Will source code be provided to interested parties before proposals are due, or only to the awardee? | No, only to the awardee |
| Does the Forest Service hold test cases or reference outputs showing correct results from the stand growth simulator and the prescription logic? Demonstrating that version 9 reproduces version 8’s numbers is the central technical risk in this work | Files run in SILVAH 7 can be used as the baseline to text a new version once we get to that point. SILVAH 9 is the next step. |
| The Resources help page indicates that model lookup tables are maintained in Access or Excel and exported to C++ source files. Is that table generation pipeline expected to carry forward into version 9? | Yes, but SILVAH 9 is a next step. |
| Does “repair” refer to defects in SILVAH 8, to restoring capability that SILVAH 7 has and SILVAH 8 does not, or both? | SILVAH 8 was a new code, not an upgrade to SILVAH 7 as such. It is a new set of logic and different code. There are some issues in 8 that were not found prior to the programmer's retirement. |
| SILVAH 8 does not appear to support FSVeg, NED-3 or TCruise import, or Access and ASCII export, all of which SILVAH 7 provides. Is restoring those data exchange capabilities in scope for version 9? | It is hoped that these features will return in SILVAH 9 once we reach that point. Our firest step is repairing SILVAH 8. |
| SILVAH 7 and SILVAH 8 stand files use different section structures and field names, so they are not interchangeable. Will version 9 use the version 7 format, the version 8 format, or a new one? | We want SILVAH 9 to read both formats and will need to determin what the new file format will be. We are not readdy for SILVAH 9 yet. |
| The published SILVAH 8 release notes end at 8.0.3.5 in December 2021, but the example stand files shipped in that installer are marked version 8.2.6. Is there a later build, and would version 9 start from it? | There is not a later build. Thise files are likely 8.0.2.6. SILVAH 8's last build was december 2021. We are not ready for SILVAH 9 yet. It should be a new fresh code when we are ready. |
| Are there reference datasets or known-good outputs (e.g., prescriptions or growth projections from earlier versions) that could serve as regression tests? | Yes we have several known good SILVAH files for testing. |
| . Are Section 508 accessibility or agency security authorization (ATO) requirements anticipated for SILVAH 9? | We had not planned on this, and are not ready for SILVAH 9 yet. Focus should be on repairing SILVAH 8 |
| Has USDA established an estimated budget range or funding allocation for the SILVAH 8 repairs, SILVAH 9 development, testing, and ongoing support? | No. |
File details come from the government source that posted it. Updated .