14:57:53 <RRSAgent> RRSAgent has joined #rch
14:57:53 <RRSAgent> logging to https://www.w3.org/2022/10/12-rch-irc
14:57:55 <Zakim> RRSAgent, make logs Public
14:57:56 <Zakim> please title this meeting ("meeting: ..."), phila
14:58:20 <phila> zakim, please title this meeting "RCH Bi-weekly meeting"
14:58:20 <Zakim> I don't understand you, phila
14:58:47 <phila> phila has changed the topic to: RDF Canonicalization and Hash Working Group - bi-weekly meeting
14:58:49 <gkellogg> present+
14:58:53 <phila> agenda: https://www.w3.org/events/meetings/51e8f278-b556-4090-b538-7928b3c628b6/20221012T110000
14:58:57 <phila> chair: phila
14:59:03 <AndyS> AndyS has joined #rch
14:59:23 <yamdan> yamdan has joined #rch
14:59:52 <Kazue> Kazue has joined #rch
15:00:18 <phila> present+
15:00:25 <yamdan> present+
15:00:35 <Kazue> present+
15:01:03 <AndyS> present+
15:01:44 <dlongley> present+
15:02:03 <markus_sabadello> markus_sabadello has joined #rch
15:02:04 <manu> zakim, pick a victim
15:02:04 <Zakim> Not knowing who is chairing or who scribed recently, I propose yamdan
15:02:26 <manu> present+
15:02:28 <markus_sabadello> present+
15:02:28 <AndyS> Life of Brian
15:02:37 <AndyS> scribenick AndyS
15:02:41 <phila> scribeNick: AndyS
15:02:58 <AndyS> scribe+
15:03:07 <Tobias_> Tobias_ has joined #rch
15:03:09 <manu> s/scribenick AndyS//
15:03:15 <TallTed> present+
15:03:20 <manu> s/Life of Brian//
15:03:40 <manu> rrsagent, draft minutes
15:03:40 <RRSAgent> I have made the request to generate https://www.w3.org/2022/10/12-rch-minutes.html manu
15:03:41 <AndyS> phila: Any new attendees?
15:04:09 <manu> rrsagent, make logs public
15:04:11 <manu> rrsagent, draft minutes
15:04:11 <RRSAgent> I have made the request to generate https://www.w3.org/2022/10/12-rch-minutes.html manu
15:04:32 <TallTed> TallTed has changed the topic to: RDF Canonicalization and Hash (RCH) Working Group — bi-weekly meeting — 2022-10-12 Agenda: https://www.w3.org/events/meetings/51e8f278-b556-4090-b538-7928b3c628b6/20221012T110000
15:04:35 <AndyS> Today: comparing the docs, deciding editors,
15:04:47 <AndyS> s/docs/algorithms/
15:04:54 <phila> https://github.com/w3c/rch-rdc/issues/6
15:04:55 <pchampin> present+
15:05:09 <AndyS> phila: How do we compare the two algorithms
15:05:21 <AndyS> ... described as being similar
15:05:25 <manu> q+ to suggest one approach to compare algorithms.
15:05:49 <AndyS> ... as a group we need to decide how to proceed
15:06:10 <AndyS> ... various ideas - need to be open and fair.
15:06:58 <phila> ack manu
15:06:58 <Zakim> manu, you wanted to suggest one approach to compare algorithms.
15:07:05 <AndyS> ... if the graph is simple then it is a simple algorithm, as bnode structures increase it gets harder.
15:07:34 <gkellogg> q?
15:07:52 <AndyS> manu: Aiden presented his algorithm. Could compare algorithm A and B work as the stages are similar.
15:08:29 <AndyS> ... run both in parallel
15:08:39 <gkellogg> q+
15:08:42 <AndyS> ... guilded tour of the process
15:08:46 <phila> ack gkellogg
15:09:22 <AndyS> gkellogg: if we able to pick some examples to run through would be useful
15:10:09 <dlongley> q+ to ask about criteria for making choices if we knew what the differences were to understand what differences to look for
15:10:11 <AndyS> ... complex category of graphs - set of promlematic use cases - performance testing
15:10:12 <manu> q+
15:10:17 <phila> ack dlongley
15:10:17 <Zakim> dlongley, you wanted to ask about criteria for making choices if we knew what the differences were to understand what differences to look for
15:10:17 <AndyS> q+
15:10:49 <AndyS> dlongley: if we figure out the differences - need to keep on mind criteria.
15:11:03 <phila> q+
15:11:11 <AndyS> ... how formal do we need to get in understanding the differences
15:11:17 <phila> ack manu
15:11:24 <AndyS> ... worried about the amount of work
15:11:38 <AndyS> manu: add an input:
15:11:56 <AndyS> ... formal analysis
15:12:09 <manu> Technical Report on the Universal RDF Dataset Normalization Algorithm: https://lists.w3.org/Archives/Public/public-credentials/2021Apr/att-0032/Mirabolic_Graph_Iso_Report_2020_10_19.pdf
15:12:20 <AndyS> ... we might consider bringing in that group
15:12:35 <aalobaid> aalobaid has joined #rch
15:12:50 <AndyS> ... introduction to the formal analysis
15:13:11 <phila> ack AndyS
15:13:17 <phila> scribe+ phila
15:13:20 <gkellogg> scribe+ phila
15:13:38 <dlongley> q+
15:13:45 <manu> q+ to comment on the "all graphs" thing
15:13:46 <phila> AndyS: You mentioned - might not be all graphs that were covered. I think we ought to target all graphs as you don't know what you'l encounter in the real word
15:13:56 <phila> ack manu
15:13:56 <Zakim> manu, you wanted to comment on the "all graphs" thing
15:14:07 <phila> ack dlongley
15:14:12 <phila> q+ manu
15:14:17 <phila> q- later
15:14:24 <AndyS> dlongley: solve for all graphs (within resource limits)
15:14:27 <manu> q+ to comment on the "all graphs" thing -- concerns around "as big as the web"
15:14:45 <AndyS> ... by default solve all "normal graphs"
15:15:04 <AndyS> ... special flag for all graphs
15:15:07 <phila> ack manu
15:15:07 <Zakim> manu, you wanted to comment on the "all graphs" thing -- concerns around "as big as the web"
15:15:28 <phila> AndyS: I'd push back a little on that as it means deciding what is and is not normal
15:16:07 <dlongley> "don't (try to) canonicalize the Web"
15:16:16 <AndyS> manu: at a higher level ... potential formal objections on charter ... e.g. very very large graphs
15:16:41 <AndyS> ... working on documents that are bounded
15:17:24 <AndyS> ... general algorithm ... state caveats e.g. not unbounded graphs
15:17:49 <AndyS> ... "poison graphs" as an attack vector.
15:18:25 <AndyS> ... we can eat up a lot of time on this.
15:18:54 <AndyS> ... scoping of graph needed
15:19:03 <phila> ack me
15:19:27 <AndyS> phila: we have to limit the scope
15:20:05 <AndyS> ... we could create an algorithm for all but not the requirement
15:20:30 <phila> -> https://www.w3.org/2022/07/rch-wg-charter/explainer.html Explainer doc
15:20:36 <AndyS> ... UCR says what we are trying to solve
15:20:45 <AndyS> ... (editors needed)
15:20:53 <AndyS> q+
15:21:01 <manu> +1 to explainer document to set the boundary of what we're trying to do.
15:21:50 <gkellogg> SHACL doesn't do datasets, only graphs.
15:21:52 <AndyS> phila: is there a condition we can do as a preprocessing graph
15:22:07 <phila> ack AndyS
15:22:35 <phila> AndyS: If you take a FOAF graph built up from bnodes - can become complex in a small file
15:22:40 <dlongley> q+ to say i think you'd have to formally prove a preprocessing step would protect you if there will be no false safe constraints in the processing algorithm
15:22:59 <phila> ... I'd rather an approach that recognizes that sometimes you can't execute, rather than defining upfront what you can't compute
15:23:13 <phila> ack next
15:23:14 <Zakim> dlongley, you wanted to say i think you'd have to formally prove a preprocessing step would protect you if there will be no false safe constraints in the processing algorithm
15:23:38 <manu> q+ to speak about "multiple phased solutions" not THE algorithm.
15:23:53 <AndyS> dlongley: a preprocessing step would need proving
15:23:57 <phila> q?
15:24:07 <phila> ack manu
15:24:07 <Zakim> manu, you wanted to speak about "multiple phased solutions" not THE algorithm.
15:24:57 <AndyS> manu: we are not generating one algorithm. There exists today some impls in the field.
15:25:40 <AndyS> ... we might look at whether it is good enough
15:26:12 <AndyS> ... then consider next version
15:27:03 <AndyS> ... not all or nothing
15:27:07 <dlongley> q+ to say we also know that RDF-star is coming -- and we'll need another algorithm for that
15:27:18 <phila> AndyS: What are the limitations? Assumption?
15:27:23 <phila> ack dlongley
15:27:23 <Zakim> dlongley, you wanted to say we also know that RDF-star is coming -- and we'll need another algorithm for that
15:28:02 <AndyS> dlongly: current limitations/assumption URDA2015 - any bound dataset
15:28:37 <AndyS> ... bail out at cost points.
15:29:22 <phila> AndyS: I'm happy with bailing out. But you can go further and say it doesn't handle all graphs. I'm happy with all graphs, with a bail out if it takes too much computing
15:29:32 <dlongley> q+
15:29:36 <phila> AndyS: Defining a shape before hand is not something we should do
15:29:39 <phila> ack dlongley
15:29:42 <manu> +1 to what AndyS is saying -- sounds like we're agreeing :)
15:30:21 <AndyS> phila: Others?
15:30:36 <Kazue> q+
15:30:40 <phila> ack Kazue
15:31:28 <AndyS> Kazue: thinking external criteria hard to decide
15:31:36 <dlongley> q+
15:31:38 <AndyS> phila: and it is political
15:31:46 <yamdan> q+
15:31:53 <phila> ack yamdan
15:31:59 <dlehn1> present+
15:32:17 <gkellogg> q+ to suggest identifying specific categories of graphs in our hypothetical dataset that are known to create computational problems.
15:32:43 <AndyS> yamdan: also important to be clear about processing.
15:32:57 <AndyS> ... A difference of the two algorithms is scope - dataset vs graph.
15:33:11 <phila> ack next
15:33:14 <manu> +1 to yamadan's points.
15:34:19 <AndyS> dlongley: criteria important. Formally defining the differences is itself difficult.
15:34:29 <phila> ack next
15:34:30 <Zakim> gkellogg, you wanted to suggest identifying specific categories of graphs in our hypothetical dataset that are known to create computational problems.
15:34:46 <AndyS> phila; please think of two criteria
15:34:54 <AndyS> s/;/:/
15:35:32 <AndyS> gkellogg: want a collection of cases beyond test cases e.g. known expensive.
15:35:32 <dlongley> 1. ease of implementation, 2. existing incubation / use in the marketplace, 3. time / resource complexity in solving common datasets, 4. time / resource complexity in solving complex (or poison?) datasets
15:36:16 <AndyS> dlongley : not an order list
15:36:18 <manu> 5. Existence of formal proofs for the algorithms
15:36:28 <AndyS> s/order/ordered/
15:36:34 <manu> 6. Demonstration of review of formal proofs for the algorithms
15:36:41 <AndyS> phila: easy of implementation - yes.
15:36:54 <AndyS> ... incubation - yes
15:37:02 <AndyS> ... resource complexity - yes
15:37:09 <AndyS> ... formal proofs - yes
15:37:57 <phila> AndyS: Ease of implementation and complexity of algorithm can be in opposition
15:38:04 <dlongley> yes, there is a tension between ease of implementation and time complexity (sometimes)
15:38:32 <manu> +1 to create an issue to track this.
15:38:37 <dlongley> 7. reusing existing primitives that are available on various platforms
15:38:39 <dlongley> q+
15:38:46 <Kazue> coverage of target RDF?
15:38:56 <phila> ack dlongley
15:39:19 <AndyS> dlongley: reuse primitives e.g. hashing algorithms.
15:39:49 <manu> q+ to note "hashed data" as the output... for BBS.
15:39:51 <AndyS> ... existing RDF serialization.
15:40:38 <AndyS> Kazue: cover real life examples
15:41:17 <AndyS> phila: need to note that only unual graph trigger the failsafes.
15:41:22 <phila> q?
15:41:25 <phila> ack manu
15:41:25 <Zakim> manu, you wanted to note "hashed data" as the output... for BBS.
15:41:28 <AndyS> s/unual/usual/
15:41:50 <AndyS> q+ to give criteria
15:42:26 <AndyS> manu: BBS signature do a statement by statement signature
15:42:52 <dlongley> 8. allow signatures on individual statements and components of statements
15:43:03 <AndyS> ... criteria: has to support selective disclosure. Hashing alternatves.
15:43:14 <phila> ack next
15:43:15 <Zakim> AndyS, you wanted to give criteria
15:43:35 <yamdan> +1 to BBS-friendly hash
15:43:39 <phila> AndyS: Dataset, not graph, no shape excluded, cover RDF-star
15:43:41 <gkellogg> +1 to AndyS
15:43:48 <phila> AndyS: Translates as do stuff with the longest life
15:43:51 <phila> q?
15:43:55 <manu> I was with AndyS all the way up to "cover RDF-star" :)
15:44:09 <gkellogg> Also, Generalized RDF (bnode predicates, literal subjects)
15:44:41 <AndyS> +1 to gkellogg.
15:46:23 <dlongley> q+
15:46:36 <phila> ack dlongley
15:47:15 <AndyS> dlongley: RDF-star. Do existing use cases.
15:47:55 <AndyS> phila: rdf-star is a nice to have but should not fail because of rdf-star
15:48:25 <AndyS> URDNA2015 FPWD
15:48:59 <AndyS> phila: URDNA2015 as FPWD.
15:49:13 <AndyS> ... likes explanatory examples.
15:49:49 <phila> Topic: Editors
15:49:53 <AndyS> phila: need to do a test suite and an explainer.
15:49:54 <gkellogg> q+ to volunteer to edit one or both of the documents and help with the test suites.
15:49:59 <phila> ack gkellogg
15:49:59 <Zakim> gkellogg, you wanted to volunteer to edit one or both of the documents and help with the test suites.
15:50:25 <AndyS> gkellogg: have been active in CG
15:50:41 <AndyS> ... hat in the ring
15:50:53 <AndyS> For the C14N spec: ...
15:50:55 <dlongley> q+
15:51:02 <manu> Thank you, Gregg for Editor-ing the canonicalization spec! :)
15:51:04 <phila> ack dlongley
15:51:04 <AndyS> ... For the C14N spec ...
15:51:28 <AndyS> dlongley: can contribute as backup editor
15:52:23 <AndyS> phila: any one like to be an editor or contribute  in some way.
15:53:00 <AndyS> ... hash doc
15:53:21 <AndyS> ... "RDH"
15:53:22 <manu> q+ to note they might be the same doc?
15:53:24 <Tobias_> q+
15:53:26 <phila> ack manu
15:53:26 <Zakim> manu, you wanted to note they might be the same doc?
15:53:47 <AndyS> manu: might be the same doc.
15:54:01 <gkellogg> q+ to discuss testing implications.
15:54:11 <AndyS> ... hashing is C14N input, hash it. -- one  page?
15:54:11 <phila> ack Tobias_
15:54:39 <yamdan> q+
15:54:40 <manu> Woo! Thanks Tobias for volunteering to be an Editor!
15:54:43 <AndyS> Tobias_: happy to help edit esp hashing
15:54:44 <dlongley> +1
15:54:48 <manu> (for the second part)
15:54:48 <pchampin> +1
15:54:50 <phila> ack gkellogg
15:54:50 <Zakim> gkellogg, you wanted to discuss testing implications.
15:54:56 <phila> q?
15:55:08 <phila> ack yamdan
15:55:15 <AndyS> gkellogg: tesring may be easier as 2 docs
15:55:26 <pchampin> a contrario, the C14N itself may be a complex document. That could justify keeping the hashing part out.
15:55:29 <AndyS> yamdan: interested in hashing part
15:55:51 <dlongley> +1 to Phil
15:56:09 <AndyS> phila: end meeting
15:56:23 <phila> RRSAgent, draft minutes
15:56:23 <RRSAgent> I have made the request to generate https://www.w3.org/2022/10/12-rch-minutes.html phila
15:56:31 <phila> RRSAgent, make logs public
15:56:37 <AndyS> zakim, end meeting
15:56:37 <Zakim> As of this point the attendees have been gkellogg, phila, yamdan, Kazue, AndyS, dlongley, manu, markus_sabadello, TallTed, pchampin, dlehn
15:56:39 <Zakim> RRSAgent, please draft minutes
15:56:39 <RRSAgent> I have made the request to generate https://www.w3.org/2022/10/12-rch-minutes.html Zakim
15:56:42 <Zakim> I am happy to have been of service, AndyS; please remember to excuse RRSAgent.  Goodbye
15:56:46 <Zakim> Zakim has left #rch
15:57:52 <manu> Note: Ahmad Alobaid volunteered to be a first-time Editor in this group.
15:58:03 <manu> rrsagent, draft minutes
15:58:03 <RRSAgent> I have made the request to generate https://www.w3.org/2022/10/12-rch-minutes.html manu
16:21:23 <pchampin> meeting: RCH WG
16:21:29 <pchampin> > rrsagent, draft minutes
16:21:39 <pchampin> rrsagent, draft minutes
16:21:39 <RRSAgent> I have made the request to generate https://www.w3.org/2022/10/12-rch-minutes.html pchampin
17:51:26 <gkellogg> gkellogg has joined #rch
17:55:36 <gkellogg> gkellogg has joined #rch
18:39:26 <gkellogg> gkellogg has joined #rch
18:58:24 <gkellogg> gkellogg has joined #rch
19:15:06 <gkellogg> gkellogg has joined #rch
19:16:06 <gkellogg> gkellogg has joined #rch
20:05:42 <gkellogg> gkellogg has joined #rch
20:38:53 <gkellogg> gkellogg has joined #rch
20:56:13 <gkellogg> gkellogg has joined #rch
21:13:05 <gkellogg> gkellogg has joined #rch
21:48:32 <gkellogg> gkellogg has joined #rch
22:03:48 <gkellogg> gkellogg has joined #rch
23:34:43 <gkellogg> gkellogg has joined #rch