Verification is the process of determining whether software products have been built correctly according to their requirements and design specifications. It includes testing, refactoring, and quality assurance.
The Verification Playground is a competition that focuses on problems that can’t be fully automated and require human expertise “in the loop.” It’s part of the TOOLympics (Competitions in Formal Methods) track of ETAPS.
We’re a community of engineers, researchers and students working together to build verified software that delivers value to customers. We’re committed to making the best possible product that meets customers’ needs, while delivering on our promise of high quality, reliability and security.
Our 먹튀검증소 is designed to make the verification process easier for everyone involved. By enabling teams to quickly add media items that are relevant to a verification task - including Tweets, YouTube videos, Facebook posts and Instagram photos - teams can start fact-checking right away.
Use Check’s browser extension to easily add media items that are relevant to a fact-checking task on your team and answer annotations about them right from the extension’s sidebar.
When you add a media item, Check automatically creates an issue in the appropriate project and assigns it to a specific team. Afterwards, you can see your progress in Check.
MR Rate: shared performance indicator for team collaboration, iteration, customer results
At Verify, we value MR Rate as a shared performance indicator for all teams. It’s a measure of how well we collaborate on a problem statement, how well it’s being iterated to deliver customer results, and how much progress we’ve made on the issue.
It’s a great way to gauge team progress, and it’s an important shared performance indicator for all team members: the UX and engineering teams are responsible for creating clear problem statements connected to user insights, and then adding interaction specifications and acceptance criteria for that statement. It helps product management, UX and engineers prioritize issues and iterate on them to meet the customer’s needs.
Challenges correlate with ease of verification
One of the main goals of Verify is to encourage participants to use their favorite tools when solving verification challenges. As a result, participants are likely to tackle challenging algorithms and data structures that they may not otherwise have seen or used before.
However, even the most popular verification tools struggle with some common challenge types. For example, proving that an algorithm is recursive requires filling in a lot of detail about it that may not be necessary for its behavior but can be distracting. In some cases, participants also had to prove that their solutions worked correctly on non-recursive data structures.
When we compare the difficulty of different challenge types, we find that challenges that entail a natural recursive algorithm are significantly more difficult than those that don’t. Perhaps this is because recursive algorithms are harder to implement and prove in tools that support formal verification.
Another interesting observation is that the more a challenge’s input data structure is scalar, the more successful teams are. This is probably because the organizers are aware of the difficulties associated with mutable input, and so they tend to stick with scalars for their challenges.
