Presents a bird's eye view of the operations and basic data-flows envisaged for OPP-6. The main actors and actions are identified as well as the main building blocks of the system:
(edit this diagram in draw.io)
-
Q: Implementing security restrictions/checks will not be a target for the PoC. However, should we support some very basic user authentication? Regarding the calls in (P5) and (P6) we could have some basic, token-based authentication in place especially since (P14) will be deployed on public Internet.
A: Yes. Basic security is required. Discuss with UKMO to re-use part of infrastructure. Security will focus on users like Mouktar and not end-users. -
Q: Do we want commercial search engines to also crawl AWC?
A: Not at this point. -
Q: Do we envisage a central, single AWC or multiple ones? e.g. AWC France, AWC UK, etc. A: For the pilot, only one.
-
Q: Should the local WebApp (P11) provide a functionality to create CSVs online?
A: No. -
Q: Do we care for the format/columns of the CSVs? Is there a standard to be used?
A: No. -
Q: We will need some sample CSVs, who is going to prepare/provide them?
A: Dom.
- Q: Do we intend to provide some tool here or is it up to the user?
A: No.
- Q: Do we intend to provide some tool here or is it up to the user?
A: No.
-
Q: Which search engine will we pick for the PoC?
A: Google. -
Q: We need some info/research here regarding the format/content of the sitemap.
A: Check if someone has extensive experience in the subject. -
Q: We need some info/research here regarding the registration process, time to crawl, frequency of updates etc.
A: Check if someone has extensive experience in the subject.
- Q: Can we use the exact same sitemap as in (P3)? This is largely related to the specs of (P14).
A: TBD.
-
Q: Do we envisage anything more complicated than an HTTP/GET call?
A: Need notification for new data. Optionally, for new metadata. -
Q: Alternatively, do we need this call or we rely on the Crawler identifying new data (and proceeding to sending notifications)?
A: See (6.1)
- Q: For the PoC we need to define the Web Server to be used/supported.
A: Go for quick :)
- Q: Is the secondary dissemination channel something we consider for the PoC? If it is, we need to define what that should be.
A: No secondary dissemination channel for PoC.
- Q: How is the sitemap going to be generated? If not manually, we should investigate tools/methods.
A: Manually.
- Q: How are the pages going to be generated? If not manually, we should investigate tools/methods.
A: Use HTML templates privided for the PoC.
- Q: What should be the main functionality of the local WebApp?
A: Simple CSV visualization using existing tools.
- Q: Once we choose the search engine to use, we need to investigate how to query it according to our needs. Can you provide a few representative queries we would like to run? A: Dom will define a few use/test cases for that.
- Q: Decide on the crawling mechanism as well as the underlying storage mechanism (this is directly related to the queries to be supported by (P14)).
A: TBD after P14.
- Q: Provide a few indicative queries.
A: Jeremy will define a few use/test cases for that.
-
Q: Is the central approach for real-time notifications the one we want for the PoC?
A: Yes, central. -
Q: Real-time notification are about the existance of data, or they disseminate the data itself as well?
A: No.