Frequently asked questions

What is Tare Program?

Tare Program is evaluation-as-a-service for long-baseline GNSS clock comparison data. You send agreed comparison inputs; we apply a proprietary off-line transformation and return post-processed outputs with before/after statistics. The method is proprietary and patent-pending—we do not disclose it. You receive results only: no algorithm, parameters, or code.

Who can participate in the pilot?

Participation is limited to qualified organizations: universities and research labs, government agencies, and vetted private corporations with a clear need to evaluate GNSS timing or clock comparison. Applications are reviewed individually. Not open signup.

What data can be submitted?

Only data your organization is authorized to share; non-classified and export-compliant. Accepted formats: IGS clock files (.clk, .clk.gz), CGGTTS time-transfer files (.cggtts), and structured CSV as specified at onboarding. Max 500 MB per file; compression accepted. Raw receiver logs, proprietary binary formats, and streaming feeds are not accepted. Full details are in the pilot terms.

What are the pilot scope limits?

Per participant: up to three datasets (each up to 90 days of data), up to three processing runs per dataset, results typically within five business days. One primary technical contact per organization. Deliverables: post-processed series, RMS comparison report, summary metrics—no code or parameters. Additional datasets or runs may be considered by mutual agreement.

How long is the pilot and when does it run?

Each engagement is 90 days. The program is invite-only: we review applications and extend invitations when we have capacity (up to ten concurrent participants). Start and end dates are set in your pilot agreement after acceptance—not a single fixed public window. Extensions may be possible by mutual agreement after the pilot.

What do participants receive?

Post-processed datasets and before/after statistical comparisons (e.g. RMS on agreed metrics). That is the evaluation deliverable. The proprietary transformation is never disclosed. Results are for internal evaluation under the pilot agreement—we do not guarantee the same movement on your panel, and we do not warrant operational timing performance. Pilot participants may receive priority access to licensing discussions if the technology proceeds to commercial deployment.

How is data handled?

Submitted data is processed in a controlled, offline environment. Transfer is encrypted. The transformation runs on systems that do not expose the method. Your data is not used for any purpose other than the pilot and is not shared with third parties.

What does the pilot cost?

Pilot fees are set by segment. For universities, research labs, government agencies, and non-profit research institutions: $45,000 USD for the 90-day period. For commercial organizations: $195,000 USD for the 90-day period. Payment terms are set in the pilot agreement.

Who owns the data and the method?

Your submitted data remains yours. The post-processed results we return are for your evaluation; use is limited to evaluation under the pilot agreement. The proprietary transformation method is Banlys intellectual property.

Are published results representative of what I will see in a pilot?

Representative of the evaluation protocol and the GNSS clock-comparison domain—not of a guaranteed effect size on your panel. Published runs are fixed, public IGS-class exhibits (pinned windows, null tests in the evidence room). Your pilot uses the same train/holdout frame on your agreed inputs under confidentiality. Public statistics show what we observed on those panels; your pilot answers whether similar movement appears on yours.

What do the demo results show?

In published IGS-class holdout runs: typical RMS reduction ~45%, strict-window examples up to ~81%, post-process residuals commonly 17–22 ns. Your panel may differ.

Our public demonstration uses International GNSS Service (IGS) data. Same data, same metric—only the transformation applied. No method is disclosed on the demo page. Results on your data may vary.

How do I apply?

Use the pilot application form. We will review your organization, dataset description, evaluation goals, and budget approval. Qualified applicants receive a pilot agreement, scheduling information, and dataset submission instructions. Applications are typically reviewed within 5–10 business days.

What happens after I apply?

We review your application within 5–10 business days. If accepted, we send a pilot agreement and payment instructions. After we receive payment and the signed agreement, we send upload access, the metadata template, and file-naming convention.