Skip to end of metadata
Go to start of metadata

Change in Venue & Dinner details

Dates: 6-7th June 2013
Location: Poznan, Poland


Please confirm your attendance for catering requirements using *this doodle poll.

Objectives (expected meeting outcomes)

  • Advance in productization and exploitation
  • Ensure alignment between TF outcomes and user requirements in order to maximize uptake possibilities
  • Outreach and community engagement


Skype log

[06/06/2013 9:36:22] khalid.belhajjame: Are we using this skype channel for the Poznan Plenary? Pinar wants to join
[06/06/2013 9:36:37] Piotr Holubowicz: Sure, why not
[06/06/2013 9:36:51] Sean Bechhofer: I'll also add @finn.
[06/06/2013 9:36:53] ** Stian Soiland-Reyes ha cambiado el tema del chat a "2013-06-06/2013-06-07 Wf4ever Plenary, Poznan" **
[06/06/2013 9:37:08] ** Sean Bechhofer ha agregado a finn.bacall **
[06/06/2013 9:37:23] Oscar Corcho: thanks, that's great to be updated
[06/06/2013 9:37:25] Piotr Holubowicz: For all the people that want to join remotely, please see
[06/06/2013 9:37:59] ** Raul  Palma ha agregado a cmazurek **
[06/06/2013 9:38:04] Stian Soiland-Reyes: Agenda:
[06/06/2013 9:38:45] ** Piotr Holubowicz ha agregado a Filip Wi?niewski **
[06/06/2013 9:49:15] Daniel Garijo: Kristina: the creation of ROs is difficult. New users cannot use it as it is right now.
[06/06/2013 9:54:15 | Editado 9:54:22] Daniel Garijo: ... please, tell us what is (realisticly) going to be implemented by the end of the project.
[06/06/2013 9:55:11] Daniel Garijo: Carole: it all comes to: In this paper how do I get this image/chart, etc.
[06/06/2013 9:55:41] Daniel Garijo: ... how many of these workflows can be embedded in VMs?
[06/06/2013 9:56:14] Daniel Garijo: GK: I tried an it is not as easy as it sounds.
[06/06/2013 9:56:47] Daniel Garijo: Carole: I want the button that gets me the report of the RO of the paper.
[06/06/2013 9:57:14] Daniel Garijo: ... RDF is NOT a solution. It can't be read by users.
[06/06/2013 9:57:53] Daniel Garijo: hi Pinar!
[06/06/2013 9:58:04] pinar.alper: hi all
[06/06/2013 9:58:06] khalid.belhajjame: We can see you Pinar
[06/06/2013 9:58:08] khalid.belhajjame: :)
[06/06/2013 9:59:48] Daniel Garijo: Jose: We have these needs, we have these tools. What can we do with that?
[06/06/2013 10:00:00] Daniel Garijo: Carole: We have done that already ( a lot)
[06/06/2013 10:00:19] Daniel Garijo: Instead, what we want to do is 2 or 3 things very well.
[06/06/2013 10:00:39] Daniel Garijo: Sean: myExp is a catalog of workflows.
[06/06/2013 10:01:08] Daniel Garijo: ... possibility to move to a richer catalog of workflows (related to publications).
[06/06/2013 10:01:21] Daniel Garijo: ... support for collaborative research based on workflows.
[06/06/2013 10:02:21] Daniel Garijo: ... clarifying which of these options we are going for is key
[06/06/2013 10:03:21 | Editado 10:07:07] khalid.belhajjame: Carole: myExperiment is not a fully fledged research management system, which Seek, for example, is
[06/06/2013 10:05:25] Daniel Garijo: Jose: 3 levels of exploitation:
[06/06/2013 10:05:52] ** Katy Wolstencroft se ha retirado **
[06/06/2013 10:05:54] Daniel Garijo: ... the lite, the myExperiment and then the extended part (archival, etc).
[06/06/2013 10:06:26] Carole Goble: myExperiment is a library of workflows
[06/06/2013 10:06:46] Carole Goble: its myWorkflow
[06/06/2013 10:07:01] gklyne: Trouble is, as I see it, ROs are *very* generic, very malleable, so it's hard to have meaning ful discussions without clarity of what you want to *do* with your ROs.
[06/06/2013 10:07:30] Carole Goble: that is why I am grounding to workflows backing papers and workflow exchange
[06/06/2013 10:08:25] Sean Bechhofer: Hi @pinar!
[06/06/2013 10:09:02] gklyne: @Carole - good answer - so that's what we need to focus on.
[06/06/2013 10:09:54] gklyne: FWIW, we've (Jun and I) been using github to create an RO that backs our in-use submission.  The rest is just tools we use.
[06/06/2013 10:10:47] Piotr Holubowicz: I'm not sure that the terms "core" and "full" are most precise. It suggests that full includes core, which is not necessarily true?
[06/06/2013 10:11:05] Oscar Corcho: My opinion without following the real live discussion (so feel free to throw it away if it does not fit - I do not want to compromise the flow of discussions without been there, as not having the context of the conversation makes it difficult):
1. We have proposed a general RO model which is great, as it has attracted lots of attention from people everywhere (I was constantly asked to explain ROs to people in SePublica, for instance).
2. It is time that the technology that we provide focuses on a very specific specialisation of ROs that contain X, Y, Z. We know that the model may have more, but those X,Y,Z are the ones that we want to focus on. It could be how to insert ROs in publications (as Carole mentions, since this is a niche not covered currently by the market), or it could be other things.
3. So it is mainly about exploitation of results (and making our users happy when creating the ROs anyway)
And again, sorry since I am not there and I do not want to break the discussion that you are having (probably I am talking nonsense)
[06/06/2013 10:11:12] Carole Goble: in the reproducibility workshop the RO as a framework for what the tools are doing and as a framework for interop between systems was feasible
[06/06/2013 10:11:13] Sean Bechhofer: I think Raul is proposing that the portal provides functionality relating to "option C" -- the management of the general scientific process.
[06/06/2013 10:11:38] Sean Bechhofer: THe RO model/API can support  both the concrete workflow centric and the bigger bits.
[06/06/2013 10:11:40] Carole Goble: all the systems are supporting aspects of RO but not using the actual models, and neither will they
[06/06/2013 10:14:00] gklyne: "myExperiment not aimed at supporting ROs in a living environment" - did I get that correct?  (sounds good to me)
[06/06/2013 10:14:14] Carole Goble: no -
[06/06/2013 10:15:39] Carole Goble: its myExperiment as we are working now  is aimed at supporting workflow ROs in a living environment but NOT the way a project is organised and scientists working together to manage an investigation
[06/06/2013 10:18:26] Sean Bechhofer: @carole: myExp may link data relating to workflow execution, but is not primarily about data relating to an entire investigation.
[06/06/2013 10:19:35] Alan R Williams: It also does not reflect progress of a workflow to publication for others to use
[06/06/2013 10:21:12 | Editado 10:21:21] Daniel Garijo: Kristina: what is the boundary between me having a wf in myExp and having an RO.
[06/06/2013 10:21:36 | Editado 10:22:16] Daniel Garijo: Carole: some ROs for papers have nothing to do with workflows.
[06/06/2013 10:22:04] Piotr Holubowicz: is the users' presentation in dropbox?
[06/06/2013 10:22:23] Stian Soiland-Reyes: Carole: ROs are just as applicable for instance for iPython notebooks  - which has nothing to do with workflows, but python scripts. Same requirements on provenance, dependencies, etc
[06/06/2013 10:23:47] Stian Soiland-Reyes: Daniel: BeyondPDF thoughts - RO as a meta-container
[06/06/2013 10:26:38] Daniel Garijo: Pique: the RO sketch is not linked an uniform way to the model
[06/06/2013 10:27:00] Carole Goble: For that we need notions of project management.
[06/06/2013 10:27:08] Carole Goble: meta workflows
[06/06/2013 10:28:12] gklyne: Am I missing something?:  as far as I'm aware, the only modelling of dependencies we've done is for workflows (inputs, services, etc.) - and that's been prettyy ad hoc as far as I can tell from my quality analysis work.
[06/06/2013 10:28:28] gklyne: Yet it appears it shoukd be more central to the RO model?
[06/06/2013 10:30:09] Carole Goble: Graham - my point is dependencies are pretty damn important  - my reproducibility slides point out that to run a Taverna workflow depends on availability of plug ins, services, scripts and the execution DCI. Some of this is captured in the wf bundle....
[06/06/2013 10:30:32] Carole Goble: its been on my shield picture for a long time....
[06/06/2013 10:32:09] Carole Goble: I would like a "produce RO report" to produce a report about the RO
[06/06/2013 10:32:24] gklyne: Carole, I accept it's been there as a requirement, but I'm not seeing the generic models, unless you count the checklist work. (I suppose one could view a checklist as a dependency description, but I think it's more than that.)
[06/06/2013 10:33:09] Carole Goble: this is what other systems (distribution modeller, etc) have so people can read it and put it in their notebooks. even if its hyperlinks, like the one Dani emailed round
[06/06/2013 10:34:06] Daniel Garijo: an html page that can be linked/snapshotted maybe.
[06/06/2013 10:38:10] Alan R Williams: For some domains, the dependencies can be used to provision.
[06/06/2013 10:38:19] Alan R Williams: Also, commonly used for setting up AMIs
[06/06/2013 10:42:54] Daniel Garijo:
[06/06/2013 10:43:00 | Editado 10:44:32] Daniel Garijo: ^^ that is an example (even if the links don't work)
[06/06/2013 10:43:01 | Editado 10:43:09] Carole Goble: thanks Dani
[06/06/2013 10:44:38] Carole Goble: Stop figuring out the generic design of such things and start thinking about a standard view
[06/06/2013 10:46:13] Sean Bechhofer: @carole -- my reference to templates wasn't about generic views/mechanisms, but rather having two or three standard views (for the different users that Julian refers to). Agree on something simple/concrete.
[06/06/2013 10:51:08] Carole Goble: @Sean Phew
[06/06/2013 10:51:10] Carole Goble: :)
[06/06/2013 10:53:16] Daniel Garijo: Julian: Assistance on attribution.
[06/06/2013 10:53:43] Daniel Garijo: ^^This is what matters most for scientists
[06/06/2013 10:54:34 | Editado 10:59:05] Pique: Just because we're talking about ipython notebook, have a look at wakari
[06/06/2013 10:54:47] Carole Goble: do you know
[06/06/2013 10:55:23] Carole Goble: Attribution was raised for
[06/06/2013 10:56:00] Alan R Williams: Is also a real need for BioVeL
[06/06/2013 10:57:38] Carole Goble: the most important thing is being able to cite
[06/06/2013 10:59:10] Carole Goble: citing for backing and citing for credit
[06/06/2013 10:59:39] Carole Goble: in that case we need to submit ROs to impactstory and altmetrics
[06/06/2013 11:00:34] Carole Goble: is a VM approach @Graham
[06/06/2013 11:03:16] Stian Soiland-Reyes: We should work with DataDryad, they have citable data sources, including folders of such - but not much in terms of the relationships between them or descriptions of the resources - just a recommendation for a README file (which is low entry requirement, though!)
[06/06/2013 11:06:11] Carole Goble: We have talked with them many times. Also DataVerse is an option
[06/06/2013 11:08:25] pinar.alper: Provenance Dashboard
[06/06/2013 11:10:36] Pique: Provenance browser as it is already implemented in Taverna would be the ideal functionality I will use more in a Myexp provenance tab
[06/06/2013 11:11:18] Carole Goble: in fact I asked for roundtriping to put the RO into Taverna to view past provenance
[06/06/2013 11:11:22] Alan R Williams: @Pique - disagree strongly
[06/06/2013 11:11:24] Stian Soiland-Reyes: That's doable - but step by step. Getting a summary table first. Then input outputs. Browsing of individual steps would take more effort and come later.
[06/06/2013 11:11:33] Don Cruickshank:
[06/06/2013 11:11:39] Stian Soiland-Reyes: Taverna 3 will have that roundtripping
[06/06/2013 11:12:23] Carole Goble: but ths could still be an option
[06/06/2013 11:12:41] pinar.alper: @Stian Have you seen the way KNIME displays results, worth looking at before starting an implementation
[06/06/2013 11:13:11] Pique: @allan, could you detail a little bit more..? :)
[06/06/2013 11:13:29] Carole Goble: it doesn't do comparison but it would be great as a "rerun using this  as an example I can edit"
[06/06/2013 11:13:35] Alan R Williams: I don't think the current provenance browsing in Taverna is good at all.
[06/06/2013 11:13:47] Carole Goble: @Alan - its better than nothing
[06/06/2013 11:13:51] Alan R Williams: Yes, as I first baby step it would be OK
[06/06/2013 11:14:22] Carole Goble: But I agree we need a portfolio of provenance tools
[06/06/2013 11:14:36] Carole Goble: Vistrails uses spreadsheets
[06/06/2013 11:14:40] Alan R Williams: I'd prefer to put the stuff into JSON and use one of the many different JSON browing/charting/graphing tools
[06/06/2013 11:15:07] Carole Goble: @Alan - we are on the same page of leveraging other tools
[06/06/2013 11:15:20] Alan R Williams: But, yes, need to round-trip.
[06/06/2013 11:15:26] Pique: @allan I'm glad to know you're more exigent than me
[06/06/2013 11:15:34] Carole Goble: GeoFlow plugins, for example, in Excel would be powerful for GIS
[06/06/2013 11:16:29] Stian Soiland-Reyes: but JSON is not a magic wand - you would have to write it out specifically for something those tools would make sense of.

PROV-JSON could for instance faithfully represent all the executions that occurred in a workflow with data used and generated - but that is probably not what you want to browse or chart in such a tool.
[06/06/2013 11:16:48] Stian Soiland-Reyes: now discussing folder support and folder templates
[06/06/2013 11:17:06] Alan R Williams: @Stian - really no. Can discuss later.
[06/06/2013 11:18:07] Carole Goble: re: data enables you to run your own dataverse. They have a new Big Science project
[06/06/2013 11:20:10] pinar.alper:  conventions for folder structure is useful. like maven and its archetypes
[06/06/2013 11:20:39] Carole Goble: @Stian - so we know the tools and we can do concrete specific examples rather than waffle in generic terms. General is poison to simplicity and perfect is the enemy of the good.
[06/06/2013 11:20:48] Stian Soiland-Reyes: Pinar: +1
[06/06/2013 11:21:53] Carole Goble: Alan's point is that we don't even do the simple
[06/06/2013 11:21:55] Alan R Williams: Pinar: +2 :)
[06/06/2013 11:26:34] Carole Goble: I gave a keynote at BOSC about software and science - it includes the "Jam" based adoption model and the two reasons why people don't use software: 1. It Sucks and 2. it doesn't suck but its too costly to use. The talk is in the dropbox in the Keynotes folder in Presentations
[06/06/2013 11:30:09] Carole Goble: is also an essential read
[06/06/2013 11:35:27] Carole Goble: I am astonished that dependencies are not explicitly recorded, given that this is a *preservation* project...
[06/06/2013 11:56:23] Daniel Garijo: Occupy gezi?
[06/06/2013 11:56:41] Piotr Holubowicz: --- coffee break is over ---
[06/06/2013 11:58:01] pinar.alper: we have had a week of protests here initiated by harsh police response to a peaceful demonstration. I think it has had some international coverage
[06/06/2013 11:58:18] Daniel Garijo: Ahhhhh!
[06/06/2013 11:59:01] khalid.belhajjame: @Pinar, Stian is making fun of you, he is asking if you are protesting :)
[06/06/2013 11:59:47] pinar.alper: @khalid not as much as I would have liked to
[06/06/2013 11:59:48] Carole Goble: @GRaham chat at break. Using RightField for making Minim templates, making Minim definitions
[06/06/2013 12:00:52] Carole Goble: can we have a recognition that dependencies are part of the RO model .....
[06/06/2013 12:06:40] Daniel Garijo: we have the data dependencies recorded. We have the script/web services dependencies recorded. We do not have particular tools/libraries recorded.
[06/06/2013 12:07:30] Jose Manuel Gomez Perez: Kristina re myEXp :Need to decide on how user interaction will be from now on
[06/06/2013 12:07:40 | Editado 12:07:57] gklyne: Sean talking about myExperiment futures ... implementing the RODL API.  Would it be appropriate to review the effectiveness of the API before proliferating it? (I have come to some views, not sure if we'll have opportunity to discuss)\
[06/06/2013 12:29:31] Pique: I see many overlapping functionalities btw RODL and MyExp in most of the research lifecycle process: create, search, inspect, edit, publish ROs
[06/06/2013 12:30:46] Jose Manuel Gomez Perez: +1 @Pique
[06/06/2013 12:32:33] Jose Manuel Gomez Perez: Exploitation-wise and also for our own clarity we need to clearly position the different systems in different "markets", audience, etc.
[06/06/2013 12:34:19] gklyne: @piotr, I'd be interested to chat about your experience with authn/z (OpenId, etc.).  Increasingly, I'm thinking that there should be no new user databases - OpenId helps there.  Also, Authz model that allows (safe) handoff to "proxy" non-web process (bearer token does the handoff, but "safe" is debatable.
[06/06/2013 12:34:43] Sean Bechhofer: There needs to be synchronisation between the vocabularies that are being offered for, e.g. relationship annotation.
[06/06/2013 12:35:26 | Editado 12:35:38] Daniel Garijo: I say we should avoid make users work twice. If I annotate my Ro in one system, I am not going to do it in the other. At least I would not do it.
[06/06/2013 12:35:42] Don Cruickshank: absolutely
[06/06/2013 12:36:11] Don Cruickshank: if the RO URI for a pack is "external" then myexp acts only as a cache of that RO
[06/06/2013 12:36:48] gklyne: @piotr, @don, how do you see changes being propagated between myExp and the Portal?
[06/06/2013 12:36:52] Don Cruickshank: changes are made on the external RODL and then cached back in
[06/06/2013 12:37:36] gklyne: Conflict resulution?  (Sorry, this is getting detailed, but I think therre may be bearpits here)
[06/06/2013 12:37:43] Sean Bechhofer: Kristina: Use Deliverables to help us structure the rationale for differentiation.
[06/06/2013 12:38:17] Don Cruickshank: if the RO is external then it will always fetch at least the manifest which contains most of the information needed to validate the freshness of local data
[06/06/2013 12:38:41] Don Cruickshank: the exception that springs to mind is that the manifest doesn't allow me to validate the local knowledge of the ORE structure
[06/06/2013 12:39:10] gklyne: @Don, so myExp will be responsible to maintain coherency when working with external ROs.  Sounds reasonable, I think.
[06/06/2013 12:39:15] Don Cruickshank: yes
[06/06/2013 12:39:34] gklyne: If non-trivial ;)
[06/06/2013 12:39:47] Don Cruickshank: well, i addressed it mostly with alpha already so i know the issues :)
[06/06/2013 12:40:54] Don Cruickshank: there are some resources that don't have MD5 hashes which is the main issue that i think needs addressing in that area
[06/06/2013 12:40:56] gklyne: @Don - this all goes back to that discusion we had in Southampton, oh, a lifetime ago?
[06/06/2013 12:41:15] Don Cruickshank: +1
[06/06/2013 12:42:05] Don Cruickshank: the state i want to get the manifest in is to be able to use it to validate local knowledge about the entire RO, including the ORE structure
[06/06/2013 12:42:32] Don Cruickshank: and then the hash (e.g. etag) of the manifest is something i can use to validate the freshness of the entire RO
[06/06/2013 12:43:12] gklyne: @don My nagging concern is that this will require quite tight coupling between the myExp and RODL implementations - which in many ways I see is potentially an API issue (for which I must shoulder some blame), exposing too much internal detail of the model... make sense?  Can we envisage a way to loosen the coupling?
[06/06/2013 12:43:36 | Editado 12:44:05] gklyne: (Carole just said: imagine replacing myExperiment with X, Y or Z ... that's part of what I'm nagging at)
[06/06/2013 12:44:17] Don Cruickshank: i see it as a necessary coupling that makes the RO API work
[06/06/2013 12:44:25] Don Cruickshank: hmm
[06/06/2013 12:49:38 | Editado 12:50:12] gklyne: @Don ... part of my thinking is trying to be clearer about what kind of RO-related functionality is being exposed.  If it's just access, I think the API works well and a loosely coupled environment (e.g. access for checklist evaluation works well).  For detailed management of the RO structure/model, I agree the coupling is probably unavoidable.  But I'm wondering if there's a middle ground for, say, exchanging ROs.
[06/06/2013 12:50:11] Carole Goble: when I am thinking of reporting ROs I am also thinking of summaries
[06/06/2013 12:56:19] gklyne: I'm wondering if folders (in ROs) are ever anything more than a presentation hint - can they carry real semantics?
[06/06/2013 12:56:46] Sean Bechhofer: Portal as general mechanism for managing an RO.
[06/06/2013 12:56:53 | Editado 12:57:17] Sean Bechhofer: myExp provides a specific tailored view
[06/06/2013 12:57:27 | Editado 12:57:38] khalid.belhajjame: I think that the point made by Stian makes sense and identify the spaces where myExperiment and RODL play distinct and complementary roles
[06/06/2013 12:57:33] Sean Bechhofer: swiss army knife vs. tin opener
[06/06/2013 12:57:55] Jose Manuel Gomez Perez: @Sean: good way to put it
[06/06/2013 12:58:09] Carole Goble: please look at which tells you what scientists thnk they need
[06/06/2013 12:59:06] Jose Manuel Gomez Perez: Pique: myExp and RO Portal need to be more friendly for RO creation by end users
[06/06/2013 12:59:27] Jose Manuel Gomez Perez: otherwise it is death to do it!
[06/06/2013 13:00:48] Carole Goble: and
[06/06/2013 13:01:29] Jose Manuel Gomez Perez: @Carole: yep, those are reference. I'm aware Raúl and Piotr have been looking into them
[06/06/2013 13:02:59] Jose Manuel Gomez Perez: @Khalid: can you elaborate on the complementarities?
[06/06/2013 13:03:06] Raul  Palma: @remote users you can use from 13:00 to 14:00 this number 0048617001437
[06/06/2013 13:05:06] Carole Goble: Zenodo - Sharing Research Data across Europe - Making Science More Visible

Newly launched, Zenodo<> offers a one-stop-store for research output. Created by OpenAIRE and CERN, and supported by the European Commission, this new-generation online repository offers its service from the OpenAIRE pan-European initiative, which expands the linking of research output to datasets and funding information, in European and national contexts.

*Enabling everyone to Share and Cite Data*

Zenodo welcomes multi-disciplinary research data from any individual, scientific community or research institution. Upload allowance is generous (1GB) and can be used by institutions without their own data repository. Based on the same concept as OpenAIRE, which gathers Open Access publications across a variety of funding schemes, Zenodo provides a rich interface to link objects together with funding information.
[06/06/2013 13:11:13] gklyne: Dependencies... random thought... if we have a common way to indicate a generic dependency (e.g. a single specified property), then maybe tools that know about more specific kinds of dependency (e.g. workflow analyser) also generate general dependency links.  A standard checklist could then do some dependency analysis/reporting.
[06/06/2013 13:11:24] Don Cruickshank: @raul: it says it doesn't recognise the conference code
[06/06/2013 13:13:52] Don Cruickshank: ok - it let me in that time
[06/06/2013 13:14:04] Raul  Palma: great
[06/06/2013 13:16:53] Jun: hi, I am in now. I can hear Raul talking
[06/06/2013 13:17:16 | Editado 13:17:23] Daniel Garijo: Off to lunch!
[06/06/2013 13:17:36] Daniel Garijo: We'll be back in 45 min Jun
[06/06/2013 13:17:38] Jun: okay. i will lunch too then and read the logs. thanks dani!
[06/06/2013 13:17:41] Daniel Garijo:
[06/06/2013 13:17:46] Daniel Garijo: no prob
[06/06/2013 13:17:49] Don Cruickshank: me too
[06/06/2013 13:19:29] Stian Soiland-Reyes: LUNCH
[06/06/2013 14:00:03] Jun: Are we going to have a new code after 2pm?
[06/06/2013 14:00:19] Jun: This is the one: 0048617001422?
[06/06/2013 14:08:41] Jun: Hi, I think I've joined the conference. but your end is very quiet. much quiet than 10 mins ago. Please let me know that I am in the right place
[06/06/2013 14:09:04] pinar.alper: they might not have come back from lunch
[06/06/2013 14:09:31] Jun: hi Pinar
[06/06/2013 14:09:38] pinar.alper: hi jun
[06/06/2013 14:09:38] Jun: are you dialing in too?
[06/06/2013 14:09:54] pinar.alper: I used linphone in the morning
[06/06/2013 14:10:01] pinar.alper: not connected right now though
[06/06/2013 14:10:15] Jun: okay. I managed to dial in using skype in the end. with a lot of effort
[06/06/2013 14:18:31] Jun: oh, now I heard all the noise:)
[06/06/2013 14:18:40] Raul  Palma: we are back
[06/06/2013 14:19:11] Raul  Palma: i just started the call :)
[06/06/2013 14:44:48] Carole Goble: make the story around the RO and then tell the stories around the RO
[06/06/2013 14:49:18] Carole Goble: link the stories to different uses of the Ro, different ages of the RO and different stakeholders
[06/06/2013 14:49:52] Carole Goble: what if the RO is a backing a paper or published insight?
[06/06/2013 14:50:14] Carole Goble: what if the RO is a training object for a new PhD using the RO?
[06/06/2013 14:50:27] Carole Goble: I want to add something to a previously created RO
[06/06/2013 14:50:35] Carole Goble: I want to repair an RO that is broken
[06/06/2013 14:50:57] Carole Goble: I want to make an RO for a workflow I already made
[06/06/2013 14:51:23] Carole Goble: I want to make an RO while I am making a workflow
[06/06/2013 14:55:14] Sean Bechhofer: A demo that uses everything can end up looking forced -- all the tools get shoe-horned in.
[06/06/2013 14:55:39 | Editado 14:58:43] Sean Bechhofer: Better to have smaller things that illustrate (realistically) how tools get used in scenarios/stories.
[06/06/2013 14:58:33] Sean Bechhofer: Nice user presentation could be one of those little stories.
[06/06/2013 15:09:41] Sean Bechhofer: Need validation of round tripping between backends.
[06/06/2013 15:11:41] Sean Bechhofer: Sharing of vocabularies between RODL and myExp (cf Stian/Khalid google spreadsheet identifying vocab).
[06/06/2013 15:12:02] Jose Manuel Gomez Perez: OBJECTIVE IS USERS NOT HAVING TO DO THINGS TWICE!
[06/06/2013 15:12:21] Jose Manuel Gomez Perez: Tools must be aligned
[06/06/2013 15:13:52] Jun: I hear very well all the time
[06/06/2013 15:13:58] Don Cruickshank: mostly clear to me
[06/06/2013 15:14:02] Piotr Holubowicz: (y)
[06/06/2013 15:16:48] Don Cruickshank: regarding the benefits of the RO Portal development to myexp, i think it forces us to address federation and how users might move or copy an RO from one store to another
[06/06/2013 15:17:33] Piotr Holubowicz: @Don true, though I don't know if it's an immediate need - let them first create ROs in a shared environment, and then think about how to move them around?
[06/06/2013 15:17:41] Don Cruickshank: that they're different implementations means that it should highlight areas where we could be relying on technical conventions in the implementation rather than how they are described in the spec
[06/06/2013 15:18:23] Carole Goble: GigaScience (galaxy) and Nature Scientific Data ?
[06/06/2013 15:18:33] Sean Bechhofer: Similar to the W3C imp requirements (>=2)
[06/06/2013 15:18:41] Don Cruickshank: also RFC
[06/06/2013 15:24:39] khalid.belhajjame: Carole: Dependencies are not supported by the RO model although they are a key element for preservation
[06/06/2013 15:25:46] Jun: @khalid, do you have any plan to extend support for this part in the RO model?
[06/06/2013 15:26:47 | Editado 15:27:27] khalid.belhajjame: @Jun, we already did that in the Component Profile, it should not be difficult, and I think we will be able to do that
[06/06/2013 15:27:54] Jun: Mmmm, be nice we have some discussion about how this works together with the RO model or how this can be more exposed on things like the RO guidelines and
[06/06/2013 15:28:14] pinar.alper: pom for the RO
[06/06/2013 15:28:33] Piotr Holubowicz: +1 for the pom :)
[06/06/2013 15:28:53] Jun: :)
[06/06/2013 15:29:13] gklyne: Why not a simple property (a roerms:dependsOn b)
[06/06/2013 15:29:31] khalid.belhajjame: @Graham, I am thinking of something simple as that
[06/06/2013 15:29:43] Don Cruickshank: T2 dependencies are literal strings iirc
[06/06/2013 15:29:51] gklyne: Good!  Then generate it from wfdesc, etc as possible/available
[06/06/2013 15:29:53] Raul  Palma: have you seen
[06/06/2013 15:29:54] Don Cruickshank: at least we pull them out and show them in myE as such
[06/06/2013 15:30:00] khalid.belhajjame: what we dont want to do is to try have a long list of properties for every kind of resource and environement
[06/06/2013 15:30:05] Raul  Palma: we suggested dc:requires
[06/06/2013 15:30:11] Raul  Palma: in the annotation mapping
[06/06/2013 15:30:17] Jose Manuel Gomez Perez: @graham @khalid: but we should probably provide the methods that inform or allow evaluating such dependencies, too, right?
[06/06/2013 15:30:18] Raul  Palma: for the execution envieronment
[06/06/2013 15:30:26] Jun: this dependencies sounds quite overlapping with checklist also
[06/06/2013 15:30:39] Jose Manuel Gomez Perez: @jun: that weas my point too
[06/06/2013 15:31:05] Jun: it's another level of information representation than just recording what's inside an RO
[06/06/2013 15:31:06] gklyne: @Jun - yes.  Checklists are one way of detecting implicit dependencies.  This discussion is about making them explicit and visible to other tools
[06/06/2013 15:31:17] Jun: Ack Graham
[06/06/2013 15:31:53 | Editado 15:32:19] khalid.belhajjame: @Raul, yes, we did those, but I don't think all of the properties there are supported yet, some of them were at the proposal level, and were included in the component profile but not in the RO ontologies
[06/06/2013 15:32:14 | Editado 15:32:25] gklyne: @jun With explicit common dependency pattern, our minim repository could have a standard check-dependencies entry.
[06/06/2013 15:32:43] Raul  Palma: @khalid, true, but we have already a draft to start with :)
[06/06/2013 15:32:51] khalid.belhajjame: yes :)
[06/06/2013 15:33:11] Jun: I don't think it will in the line of *minimal* thoughts we have had so far, but quite the other end
[06/06/2013 15:33:46] Jun: I think Carole wants something to exhausively and completely represents all the dependencies information related to reproducibility
[06/06/2013 15:34:01] khalid.belhajjame: @Jun, I agree, it is hard to get right an identify the right balance between minimality vs completness
[06/06/2013 15:34:04] pinar.alper: +1 jun
[06/06/2013 15:34:13] Jun: checklist model can be used for that. but our minimal approach so far is not doing exactly what Carole wants
[06/06/2013 15:34:25] Carole Goble: not necessarily exhaustive
[06/06/2013 15:34:30] Raul  Palma: +1
[06/06/2013 15:34:35] Jun: @khalid, yes, and to know when is enough is hard
[06/06/2013 15:34:37] Carole Goble: just best effort is better than  none
[06/06/2013 15:34:53] gklyne: @Jun that's not what I'm hearing (other than as an ultimate ideal), so much as having a way to express such information when it is available.\
[06/06/2013 15:35:07] Carole Goble: Listen to the users -they are putting key information in readme and configs
[06/06/2013 15:35:46] Sean Bechhofer: PResumably that informaiton in the readme and configs is not exhaustive. What I'm hearing is that we need to be able to expose what is there so that it's not implicit.
[06/06/2013 15:36:23] Carole Goble: read the readme files
[06/06/2013 15:36:33] Raul  Palma: @graham +1
[06/06/2013 15:37:12] pinar.alper:  back in 15 mins
[06/06/2013 15:37:17] Jose Manuel Gomez Perez: by the way, these "dependencies" are part of the decay causes that we identified in showcase 7
[06/06/2013 15:37:20] Jose Manuel Gomez Perez:
[06/06/2013 15:38:03] Carole Goble: TIMBUS, SHIWA, ERFlow,
[06/06/2013 15:39:13] Carole Goble: so we know they are important. For preservation more important than other features.....
[06/06/2013 15:39:31] Jose Manuel Gomez Perez: definitely
[06/06/2013 15:40:01] Jose Manuel Gomez Perez: we need to represent them explicitly
[06/06/2013 15:42:09] Rafael González-Cabero: I think that Idafen, the phd student that came to Granada, is working in VM annotation
[06/06/2013 15:42:47] Rafael González-Cabero: it will be nice to have ROs annotated in such way that we can later make a semantic matching with the VM that are able to run them
[06/06/2013 15:44:13] gklyne: @esteban - this preservation checklist - would it be a way to think about approaching your user evaluation question?
[06/06/2013 15:46:01] aleixgarrido: @graham, this could be used for validation of functionalities that it provides
[06/06/2013 15:47:21] aleixgarrido: @graham though I was thinking also in other type of validation as how the chechlist can improve end-users daily work and then try to use end-users validation
[06/06/2013 15:48:39 | Editado 15:49:10] gklyne: (@esteban - I emailed a response to that.  Talk later?)
[06/06/2013 15:49:07] aleixgarrido: @graham I am on Aleix computer since mine does not manage to conect, I will check it out tonight and mail back
[06/06/2013 15:49:21] aleixgarrido: @graham yes
[06/06/2013 15:50:02] Sean Bechhofer: Significant properties:
[06/06/2013 15:55:19] Jun: it's not my fault :)
[06/06/2013 15:55:27] Sean Bechhofer: It's the JUN9000
[06/06/2013 15:55:35] pinar.alper: :)
[06/06/2013 15:55:49] Sean Bechhofer: I'm sorry Dave, I can't do that.
[06/06/2013 16:03:08] Daniel Garijo: :D
[06/06/2013 16:03:17] Daniel Garijo: please mute guys :)
[06/06/2013 16:03:18] Pique: I cannot help of thinking of cyborgJun when she talks.. someone to mute the mic p,lease :)
[06/06/2013 16:03:23] Jun: I muted myself
[06/06/2013 16:03:34] Sean Bechhofer: robots can do that....
[06/06/2013 16:03:42] Rafael González-Cabero: :D:D:D
[06/06/2013 16:03:50] Jun: God! I can't believe that! I am becoming a cyber person
[06/06/2013 16:03:58] Daniel Garijo: Zakim-Jun?
[06/06/2013 16:04:08] Daniel Garijo: ;)
[06/06/2013 16:04:38] Jun: Don has sounded normal, hasn't he? It must be my fault then
[06/06/2013 16:04:39] Don Cruickshank: Don Cruickshank is somewhat glad that Metal Mickey hasn't been shown on TV for some time now.
[06/06/2013 16:05:38] gklyne: (Jun - are you using speakers or headset?)
[06/06/2013 16:05:57] Jose Manuel Gomez Perez: speakset :P
[06/06/2013 16:06:41] Jun: speaker. I rarely use headset
[06/06/2013 16:08:26] Jun: We should also consider this call:
[06/06/2013 16:08:37] Jun: which I can talk more next Friday, when I might sound more normal
[06/06/2013 16:08:54] Jose Manuel Gomez Perez: @Jun: looks perfectly suited, yes!
[06/06/2013 16:09:48] Jose Manuel Gomez Perez: plus I'm quite aware Amit would welcome contributions
[06/06/2013 16:12:47] pinar.alper: I need to leave for today. bye all
[06/06/2013 16:12:53] Daniel Garijo: bbye pinar
[06/06/2013 16:12:58] Jun: bye Pinar!
[06/06/2013 16:13:05] Jose Manuel Gomez Perez: bye!
[06/06/2013 16:13:26] Raul  Palma: bye!
[06/06/2013 16:14:45] Stian Soiland-Reyes: Jun, I like your star trek voice!
[06/06/2013 16:14:58] Stian Soiland-Reyes: you can just tell us anything and it sounds like it MUST be true
[06/06/2013 16:15:13] Jun: @stian, everyone is making fun of my voice. I am not going to talk any more, at least for today
[06/06/2013 16:15:45] Pique: :(
[06/06/2013 16:16:12] Rafael González-Cabero: @jun: can you say loud: EX TER MI NATE
[06/06/2013 16:20:23] Daniel Garijo: I like your voice jun! Don't be angry :P
[06/06/2013 16:22:10] Jose Manuel Gomez Perez: Rafa: EPNOI
[06/06/2013 16:27:49 | Editado 16:28:00] khalid.belhajjame: @Jun, dont take it badly, you could be a candidate for the new female doctor Who :)
[06/06/2013 16:28:27] Sean Bechhofer: A female dalek as Dr Who? Interesting twist.
[06/06/2013 16:29:30] Jose Manuel Gomez Perez: guys...
[06/06/2013 16:30:28] Piotr Holubowicz: interesting: Rafa's epnoi found pack 398. RODL's search found pack 398 and Hyperleda Luminosities
[06/06/2013 16:32:02] Daniel Garijo: so which one is "more" right?
[06/06/2013 17:00:09] Don Cruickshank: Don Cruickshank signs off.
[06/06/2013 17:19:34] Jose Manuel Gomez Perez: Rafa or Dani
[06/06/2013 17:19:41] Jose Manuel Gomez Perez: are you coming to the PMB?
[06/06/2013 17:19:50] Jose Manuel Gomez Perez: we are already in the room
[06/06/2013 17:20:06] Jun: Jose, do you need me to be in the PMB?
[06/06/2013 17:20:16] Jun: someone from OXF will be in the room?
[06/06/2013 17:20:28] Jose Manuel Gomez Perez: Kevin is already here
[06/06/2013 17:20:37] Jun: If you want, I am around
[06/06/2013 17:20:42] Jose Manuel Gomez Perez: but we can arrange it by skype
[06/06/2013 17:20:49] Jun: okay
[06/06/2013 17:24:50] Sean Bechhofer: Jun. Kevin is here
[07/06/2013 9:00:37] Piotr Holubowicz: === The videoconference has started ===
[07/06/2013 9:32:05] Raul  Palma: @Marco: please dial Skype user pionier_vc_system and enter 0048617001423 followed by # sign
[07/06/2013 9:34:19] Raul  Palma: @Marco, and all other by skype, just make sure you enter the number, one by one, slowly and ended with #
[07/06/2013 9:35:41] Marco Roos: Is it working for anyone else?
[07/06/2013 9:36:22] Raul  Palma: yesterday Don & Jun did use it
[07/06/2013 9:36:24] Jose Manuel Gomez Perez: Jun and Don, are you online?
[07/06/2013 9:36:27] Piotr Holubowicz: Raul connected just now
[07/06/2013 9:36:32] Raul  Palma: i tried it today as well just now
[07/06/2013 9:36:59] ** Julián Garrido ha agregado a Lourdes **
[07/06/2013 9:40:43] Sean Bechhofer: Piotr: Alignments for Portal/myExp
[07/06/2013 9:40:55] Sean Bechhofer: 1/ Annotations: June
[07/06/2013 9:41:02] Sean Bechhofer: 2/ Notifications: July
[07/06/2013 9:41:11] Sean Bechhofer: 3/ Folders: Portal in June, myExp in July.
[07/06/2013 9:41:25] Sean Bechhofer: feedback from portal experience to myExp.
[07/06/2013 9:41:46] Jose Manuel Gomez Perez: Communication bw. teams on Monday calls
[07/06/2013 9:41:49] Sean Bechhofer: Use Monday calls for communication
[07/06/2013 9:41:58] Jose Manuel Gomez Perez: changes announced on Wed calls
[07/06/2013 9:43:43] Sean Bechhofer: Pique: Q about overlap of attribution.
[07/06/2013 9:44:07] Sean Bechhofer: reconciling identification of people from different systems.
[07/06/2013 9:44:54] gklyne: Isn't this an issue generally about user information management?  OpenId is just authentication.
[07/06/2013 9:48:31] Stian Soiland-Reyes: About user unifications - ORCID just accepted my patch to add RDF/FOAF support:
[07/06/2013 9:48:36] Stian Soiland-Reyes: so that could be one easy win
[07/06/2013 9:49:37] Stian Soiland-Reyes: not live yet.. but I guess it would be out in a week or so
[07/06/2013 9:49:59] Jose Manuel Gomez Perez: MArco?
[07/06/2013 9:50:00] Stian Soiland-Reyes: MArco - could you mute?
[07/06/2013 9:50:09] Jose Manuel Gomez Perez: tx :)
[07/06/2013 9:50:19] Jose Manuel Gomez Perez: unless you were trying to talk...
[07/06/2013 9:50:27] Jose Manuel Gomez Perez: manifest yourself
[07/06/2013 9:51:06] Sean Bechhofer: Piotr: For June take into account properties.
[07/06/2013 9:51:14] Sean Bechhofer: Store things in a consistent fashion.
[07/06/2013 9:51:40] Sean Bechhofer: Then worry about the user management/identification piece later.
[07/06/2013 9:52:20] Sean Bechhofer: Jose: Transactions? Issues with consistency?
[07/06/2013 9:53:17] gklyne: One for the evaluation?
[07/06/2013 9:53:49] Jose Manuel Gomez Perez: address all this in the next release
[07/06/2013 9:56:46] Carole Goble: it is not really working -- massive echo and massive background noise
[07/06/2013 9:57:54] Jose Manuel Gomez Perez: MArco, do you have any question???
[07/06/2013 9:58:19] Marco Roos: A cure for the common cold, pehaps? (Or man-flue as some say)
[07/06/2013 9:58:21] Carole Goble: about to start the deliverable run through
[07/06/2013 9:58:57] Jose Manuel Gomez Perez: man-flue=very much bio-ish terminology :)
[07/06/2013 9:59:06] Carole Goble: can you check the immediate plan - esp from the user point of view. the slides are in the dropbox.
[07/06/2013 9:59:43] Marco Roos: I can hear you, but not see you, unfortunately. And I won't talk, because of the noise. But Kristina will do that for the Bio's :)
[07/06/2013 10:00:25] Carole Goble: @Jose Man-flu is pseudo flu that men get when they have a cold. Women just call it having a bit of a cold...
[07/06/2013 10:00:48] Jun: Morning guys. I can hear Raul talking
[07/06/2013 10:01:00] Jose Manuel Gomez Perez: @Carole: ack!
[07/06/2013 10:08:44] Carole Goble: reference implementation could include other components that have been used?
[07/06/2013 10:10:20] Carole Goble: vignettes not big stories
[07/06/2013 10:10:32] Carole Goble: D1.4v2
[07/06/2013 10:10:48] Carole Goble: use cases from BioVeL?
[07/06/2013 10:11:17] Alan R Williams: How much can be checked separately? e.g. checking a run can be encapsulated in an RO v showing it in myE++ v re-running it
[07/06/2013 10:12:36] Carole Goble: checked separate from what?
[07/06/2013 10:12:44] Kevin Page: the "example architecture orchestration" is in the last (final) arch doc. so doesn't need repeating here (though that one was more techy than related to user vignettes)
[07/06/2013 10:12:48] Alan R Williams: each other
[07/06/2013 10:13:24] Alan R Williams: I mean checked not in the tick box sense but "does that part work"
[07/06/2013 10:15:53] Carole Goble: can we reorganise the deliverables?
[07/06/2013 10:29:11] Piotr Holubowicz: @All if you need to print anything, i.e. boarding passes, send it to me by email
[07/06/2013 10:29:17] Piotr Holubowicz: I will print it during the coffee break
[07/06/2013 10:30:10] Sean Bechhofer: @piotr -- just sent you mine :)
[07/06/2013 10:30:19] Piotr Holubowicz: thx, got it :)
[07/06/2013 10:30:34 | Borrado 10:30:38] Daniel Garijo: Este mensaje ha sido borrado
[07/06/2013 10:34:03] Sean Bechhofer: Khalid: ALignment of, e.g. descriptions of RODL UML etc.
[07/06/2013 10:34:44] Sean Bechhofer: Carole: no overview of the models.
[07/06/2013 10:35:04] Sean Bechhofer: Needs the "shield" view.
[07/06/2013 10:35:33 | Editado 10:37:18] Sean Bechhofer: ...view from 5,000 feet
[07/06/2013 10:35:46] Sean Bechhofer: Hi @jun!
[07/06/2013 10:36:38] Sean Bechhofer: ...illustrate how alternative views could be integrated
[07/06/2013 10:37:36] Marco Roos: @Carole +1
[07/06/2013 10:37:36] Jun: I am listening and thinking how we can improve
[07/06/2013 10:37:50] Jun: we did try to think on that line, but obviously we didn't do enough
[07/06/2013 10:38:16] Jun: I thought this picture was on this sort of line:
[07/06/2013 10:39:23] Carole Goble: NOOOOOO!!!!
[07/06/2013 10:39:53] Carole Goble: its an ontology view that does not tell a developer or user how this works
[07/06/2013 10:40:43] Carole Goble: that is a partial, technical, ontology driven view.
[07/06/2013 10:41:06] Sean Bechhofer: THis one?
[07/06/2013 10:41:08 | Editado 10:41:14] Jun: do you mean something to show developer/user how to extend the RO model?
[07/06/2013 10:41:14] Sean Bechhofer: That's what Herbert used.
[07/06/2013 10:41:22] Carole Goble: I mean that we use the "onion model" to explain how this works together and people get that
[07/06/2013 10:41:40] Carole Goble: @Sean yes!
[07/06/2013 10:41:46] Jose Manuel Gomez Perez: got it!
[07/06/2013 10:41:46] Jun: We have that in the "About" page
[07/06/2013 10:41:59] Jun: maybe it's not placed in the right place or explained sufficiently?
[07/06/2013 10:42:00] Carole Goble: Its waht HERC understood, what Herbert understood
[07/06/2013 10:42:37] Sean Bechhofer: I think Carole's point is to use it to explain the ontologies and how they fit together.
[07/06/2013 10:42:48] Sean Bechhofer: And how other things might fit in too.
[07/06/2013 10:42:56] Marco Roos: I lost what is the target audience for this?
[07/06/2013 10:44:15] Carole Goble: The link between the about and the ontologies part is not clear. The idea that someone might replace part of the suite with their own model is not clear.
[07/06/2013 10:44:22] Carole Goble: the dependencies part is missing
[07/06/2013 10:45:42] Carole Goble: @Marco the target is anyone planning to  use an RO in their platform. Like publishers etc. People who deliver scholarly comms infrastructure
[07/06/2013 10:45:45] Jun: q+ Provenance part will be talked at WP4?
[07/06/2013 10:45:52] Carole Goble: its is NOT researchers
[07/06/2013 10:47:32] Carole Goble: @Jun I like the site and I am actively promoting it. But it has not been reviewed by people who are trying to understand ROs who are not us. It is also too workflow centric
[07/06/2013 10:47:33] Marco Roos: @Carole. To make it even more tangible: Rebecca or her developers?
[07/06/2013 10:48:02] Daniel Garijo: @Carole: Alex García (the one who is trying to jailbreak the pdf to get rid of it and replace it with containers) always uses a very similar picture. He uses an RO for the bibliography (using his ontology), then another ontology for defining laboratory protocols (not workflows), then wfdesc for workflows, and all of it is grouped as an RO using our model. Like bits of lego pieces.
[07/06/2013 10:48:03] Carole Goble: I would separate research object and the how to specialise for workflows, how to specialise for data etc.
[07/06/2013 10:48:29] Carole Goble: You claim ROs are general and then you drop right down into workflow
[07/06/2013 10:48:37] Carole Goble: @Marco yes
[07/06/2013 10:48:51] Carole Goble: or someone like Herbert or John Ainsworth
[07/06/2013 10:49:11] Carole Goble: @Dani yes
[07/06/2013 10:50:35] Carole Goble: And you could replace each of the components with your own ontology, and specialise each component for your own data type
[07/06/2013 10:50:59] Daniel Garijo: plug and play :)
[07/06/2013 10:51:02 | Editado 10:52:00] Jun: okay, I htink I got some ideas about how to refactor the RO web site
[07/06/2013 10:51:11] Carole Goble: yes
[07/06/2013 10:51:31] Carole Goble: @Jun the web site will be the jewel of the project. thank you for doing it.
[07/06/2013 10:51:43] Jun: we got more material when preparing our conversation with publishers, and haven't put them on the web site. We can let them give a more high level drive to the current web site
[07/06/2013 10:52:07] Jun: Thanks Carole
[07/06/2013 10:52:10] khalid.belhajjame: @Carole, your feedback is also good for the RO journal paper, we will consider it in the presentation of the ontologies
[07/06/2013 10:52:24] Carole Goble: Ratner's video about Research Objects gives you a publishers insight
[07/06/2013 10:52:32] Marco Roos: @Jun/Carole: it is most welcome for the ISMB workshop, the better we do this, the more Scott and Rebecca will advocate ROs.
[07/06/2013 10:53:08] Carole Goble: then give examples of ROs forthe publishers infrastructure. Not ours.
[07/06/2013 10:53:25] Carole Goble: show how what they are already doing is an RO
[07/06/2013 10:53:46] Carole Goble: you can do the same for other infrastructures
[07/06/2013 10:55:10] Carole Goble: Its too "here is our ontologies, take it or leave it" and they will leave it. At last weeks reproducibility infrastructures meeting its clear noone will change their infrastructure but they might take parts of the RO models and they might use it as a exchange mechanism
[07/06/2013 10:55:16] Stian Soiland-Reyes: exactly - many of the approaches now by data publishers etc. are really about building ROs - they generally just lack the box around it
[07/06/2013 10:55:26] Carole Goble: @Stian yes.
[07/06/2013 10:55:36] Stian Soiland-Reyes: we can say "Look, we have a box! Shiny."
[07/06/2013 10:55:41] Stian Soiland-Reyes: don't show what's in the box
[07/06/2013 10:56:48] Stian Soiland-Reyes: that said - I wish we could make a non-washing-machine version of the 5000 miles view picture... :)
[07/06/2013 10:56:56] Carole Goble:
[07/06/2013 10:57:34] Stian Soiland-Reyes: Bechhofer 2010
[07/06/2013 10:57:50] Jun: Can Esteban speak a bit lounder please?
[07/06/2013 10:58:06] Jun: Thanks Carole!
[07/06/2013 10:58:24] Jun: better
[07/06/2013 10:58:50] Pique: cyborgJun is on the space, far away :)
[07/06/2013 10:59:12] Jun: well, let's see how I will sound like today :(
[07/06/2013 10:59:19] Carole Goble: Ratner's slideshow is important - its claiming ROs as a to watch trend. Its also why we need to pursue Susanna ...
[07/06/2013 10:59:39] Jun: I think we should also add all these related material on the web site
[07/06/2013 10:59:48] Jun: presentations by others
[07/06/2013 11:00:10] Jun: and I would like to have a page named as "RO at 5000 feet"
[07/06/2013 11:00:51] Carole Goble: @Stian/Sean - we need a more beautiful picture in a different tool (not my knackered ppt 2003) - it will then become iconic like the OWL-S picture or the Semantic Web layer cake
[07/06/2013 11:02:02] Sean Bechhofer: @Carole: Let's see if we can sketch it out on the way back.
[07/06/2013 11:02:18] Sean Bechhofer: Might need a proper designer to make it look beautiful.
[07/06/2013 11:02:20] Sean Bechhofer: Rena?
[07/06/2013 11:02:48] Sean Bechhofer: Could do a first cut in graffle (which is nicer than ppt imo).
[07/06/2013 11:03:08] Jose Manuel Gomez Perez: @jun: good point
[07/06/2013 11:03:57] Stian Soiland-Reyes: I think is quite brilliant - a slightly simplified version could be useful (for instance only the box on the upper right)
[07/06/2013 11:05:12] Raul  Palma: <please select you choice for lunch in the list. btw pancakes (called nalesniki) is also common polish dish>
[07/06/2013 11:05:21] Sean Bechhofer: @stian -- that picture is indeed nice, but is still at an "instance" level rather than the conceptual view that I think Carole is looking for.
[07/06/2013 11:11:15] Stian Soiland-Reyes: Like ?
[07/06/2013 11:11:32] Stian Soiland-Reyes: just cropped iut
[07/06/2013 11:11:33] Stian Soiland-Reyes: it
[07/06/2013 11:13:06] Jun: Carole, we have a separate session in D4.2v2 to discuss the different aspects of provenance, provenance of run and of RO. and give a list of applications of provenance in our project, including completeness assessment, provenance presentation in myexperiment (possibly), workflow abstraction, and Khalid's idea on provenance profiling (using provbench)
[07/06/2013 11:15:03] Stian Soiland-Reyes: In\!/view.aspx?cid=37935FEEE4DF1087&resid=37935FEEE4DF1087%21320&app=PowerPoint I used a different figure which I also found on - it's suitable for audiences who are not scared of seeing "RDF" in there. For everyone else, just a few arrows (relations) are all you need.
[07/06/2013 11:18:13] Jun: @jose/@esteban: You need a large user group for this kind of evaluation
[07/06/2013 11:19:20 | Editado 11:19:39] Carole Goble: the killer app for provenance: (i) compare n  runs, what was different; (ii) how did workflow X get used in a series of runs as part of a study; (iii) what was used in a run that will need credit; (iv) can I patch over a run in order to start a workflow part way through? (v )  reporting what I did methods and materials in the paper/lab book; (vi) can I use provenance for distributed resource repair (or could I just use the workflow spec rather than the provenance?
[07/06/2013 11:20:35 | Editado 11:20:45] khalid.belhajjame: @Carole, we have explored (vi) in a previous work
[07/06/2013 11:21:01 | Editado 11:21:14] khalid.belhajjame: but it is still a valid application, I agree
[07/06/2013 11:22:27] Jun: @Carole, yes I agree. but we have 5 months left, we need to think of ways to fit "provenance" app in the exisiting checklist/stability framework.
[07/06/2013 11:22:45 | Editado 11:24:15] Jun: We thought of implementing vi/ as part of following-up of the decay analysis
[07/06/2013 11:28:19] Marco Roos: We have a break now?  I need one. Be back later...
[07/06/2013 11:28:20] Jun: Jun go fetch a coffee too
[07/06/2013 11:32:15] Carole Goble: @Stian that is a nice slide. Can you put it in the dropbox?
[07/06/2013 11:34:44] Carole Goble: @Stian/Sean     the picture on the researchobject site is lovely, but its 50,000 feet. The ontologies are 2000 feet. how the model works and plugs these together is 5000 feet. Three pictures. But they don't fit together.
[07/06/2013 11:35:22] Carole Goble: The RO examples should bridge 5K feet to 2K feet
[07/06/2013 11:49:33] Raul  Palma: --we are back---
[07/06/2013 11:52:16] Carole Goble: @Jun/Stian/Pinar/Khalid  the true killer app from the BioVeL perspective is: compare n workflow results
[07/06/2013 11:53:08] Carole Goble: to my mind the cases are presuming that workflow runs are independent and isolated. But they are not
[07/06/2013 11:53:32] gklyne: gklyne imagines a series of images showing RO from 50K ft, 20Kft, 5Kft, 2Kft, etc...
[07/06/2013 11:54:41] gklyne: Pique: Workflows as documentation (NOT as automation)
[07/06/2013 11:54:50] Carole Goble: wfs for running scripts vs wfs as replacement of scripts
[07/06/2013 11:54:58] Carole Goble: interaction steps
[07/06/2013 11:55:14] Carole Goble: the provenance recording the trail and error
[07/06/2013 11:56:45] Carole Goble: Pique's example is exactly the same as the BioVeL cases
[07/06/2013 12:04:53] gklyne: Beyond the PDF?  The "executable movie" (Pique's demo)
[07/06/2013 12:06:49] Carole Goble: its the executable paper - like recompute the graph in the paper
[07/06/2013 12:08:01] Carole Goble: @Alan, pique's astrotaverna has 25 "Snippets" which are components
[07/06/2013 12:08:14] Carole Goble: maybe snippets is a better word!
[07/06/2013 12:12:54] gklyne: ( - I assume was mentioned)
[07/06/2013 12:14:29] Sean Bechhofer: Pique: Need to credit specific parts of work from a paper.
[07/06/2013 12:14:44] Sean Bechhofer: There's not one person who really understands all the detail.
[07/06/2013 12:29:10] Kevin Page: the earlier deliverable with the identified (at the time) community groups/subgroups is D8.2
[07/06/2013 12:29:27] Kevin Page:
[07/06/2013 12:30:21] Jose Manuel Gomez Perez: WP8
[07/06/2013 12:30:32] Jose Manuel Gomez Perez: Jun, are you ready?
[07/06/2013 12:31:54] Jun: Are we going for WP8 now?
[07/06/2013 12:32:20] Jun: the slides are in the dropbox
[07/06/2013 12:32:33] khalid.belhajjame: @Jun, you sound fine
[07/06/2013 12:34:06] Carole Goble: Khalid - didn;t we have yours and Pinar's paper?
[07/06/2013 12:34:24] Daniel Garijo: Carole, this is just journals
[07/06/2013 12:34:30] Carole Goble: ah
[07/06/2013 12:34:31] Carole Goble: :)
[07/06/2013 12:36:11] Carole Goble: is the accelerating knowledge turns
[07/06/2013 12:36:23] Carole Goble: its a book chapter
[07/06/2013 12:37:51] Carole Goble: Jun - what about OTHERs refering to our work, like Ratner and van der Sompel
[07/06/2013 12:38:31] Marco Roos: Our ERflow contact is Silvia Ollabarriaga.
[07/06/2013 12:39:20] Pique: Claudio Vuerli for astro ErFlow
[07/06/2013 12:41:02] gklyne: I note overlap betwen ROSC and ResourceSync too
[07/06/2013 12:41:08] Marco Roos: NB It is difficult to find the number of attendees of ISMB. I just did a Google search. I found 1600 in Toronto, which was the American version. ISMB/ECCB should be somewhat higher.
[07/06/2013 12:41:11 | Editado 12:41:29] gklyne: (Rob Sanderson, me, maybe others)
[07/06/2013 12:44:59] Jun: Ack Carole. I did realise that the word "workflow" always needs a clarification when pitching it to users
[07/06/2013 12:46:27] Jun: Carole: keep the content evolved
[07/06/2013 12:46:46] Jun: ... and use this web site to gather different reproducibility web sites together, as a  community resource
[07/06/2013 12:46:56] Carole Goble: Peter Kacsuk,  Gabor Terstyanszky,  Tamas Kiss, Gergely Sipos
[07/06/2013 12:47:04] Carole Goble: for our ER Flow contacts
[07/06/2013 12:47:28] Carole Goble:
[07/06/2013 12:47:54] Carole Goble: the workshop I was at is,_2013
[07/06/2013 12:48:11] Marco Roos: NB From Silvia I understood that ERflow does not really develop anything anymore, mostly dissemination now.
[07/06/2013 12:48:11] Carole Goble: note:
[07/06/2013 12:48:47] Carole Goble: SciBus does stuff
[07/06/2013 12:48:54] Carole Goble: rather than ER Flow
[07/06/2013 12:49:28] Piotr Holubowicz: I find a bit messy. If you're a non-technical user looking for software, you may get lost IMHO. I will try to suggest some changes, especially once we make the release of the RO Portal in a week
[07/06/2013 12:49:51] Carole Goble: @Piotr agree.
[07/06/2013 12:49:54] Carole Goble: +1
[07/06/2013 12:50:59] Carole Goble: the basic problem is that its all about wf4ever and not about research objects.
[07/06/2013 12:51:51] Carole Goble: wf4ever web site is about wf4ever ...
[07/06/2013 12:52:02] Jun: @pitor, that will be great!
[07/06/2013 12:52:53] Carole Goble: the is about ROs and then an example is how wf4ever use it. THen separate the tools from tools in wf4ever example and tools in general for packaging and reproducibility and preservation.
[07/06/2013 12:53:31 | Editado 12:53:38] Carole Goble: as discussed at break - you are too close to the project; this is not about a project its about a new approach
[07/06/2013 12:53:45] Jun: @carole, yes, that's going to be the main principle to be applied to the refactoring process
[07/06/2013 12:53:50 | Borrado 12:53:59] Carole Goble: Este mensaje ha sido borrado
[07/06/2013 12:54:01] Carole Goble: great
[07/06/2013 12:59:44] Kevin Page: other wordy thoughts (provocative, not necessarily specific suggestions): local, specialised, comprehensive
[07/06/2013 13:01:31] Julián Garrido: @jose, is your slide for the wf4ever product in dropbox?
[07/06/2013 13:01:32] Piotr Holubowicz: lunch time? :)
[07/06/2013 13:03:10] Jun: I think I will not join after lunch. will use the time to work on refactoring and many other engagement stuff that I am terribly lagging behind
[07/06/2013 13:03:24] Jun: Jun waves bye and see you soon!
[07/06/2013 13:03:30] Piotr Holubowicz: bye!
[07/06/2013 13:03:42] Don Cruickshank: bye
[07/06/2013 13:03:44] Piotr Holubowicz: there will still be a presentation by Graham
[07/06/2013 13:03:57] Piotr Holubowicz: we'll go for lunch in 20 mins
[07/06/2013 13:04:56] Jose Manuel Gomez Perez: @julian: it's now in dropbox
[07/06/2013 13:05:09] Sean Bechhofer: graham: "Where do ROs come from Mummy?":)
[07/06/2013 13:09:11] Carole Goble: Graham
[07/06/2013 13:09:22] Carole Goble: might be interesting...
[07/06/2013 13:10:22] Marco Roos: Is there a specific slideset that I could look at?
[07/06/2013 13:10:34] Marco Roos: (Graham's presentation I mean{
[07/06/2013 13:10:36] Carole Goble:
[07/06/2013 13:16:50] Carole Goble: @Graham would you use underneath your ideas?
[07/06/2013 13:23:13] Marco Roos: I feel hungry on your behalf, and I have already had lunch ;)
[07/06/2013 13:23:18] Jose Manuel Gomez Perez: lunch
[07/06/2013 13:23:24] Alan R Williams: Pancakes!
[07/06/2013 13:23:27] Jose Manuel Gomez Perez: coming back to Graham's proposal later
[07/06/2013 13:23:31] Carole Goble: ok
[07/06/2013 13:23:39] Don Cruickshank: -> lunch
[07/06/2013 14:52:49] Jose Manuel Gomez Perez: back
[07/06/2013 14:55:05] Stian Soiland-Reyes: tl;dr: Graham is proposing a RDF-based Microsoft Access?
[07/06/2013 14:55:32] Carole Goble: @Stian I am not sure....
[07/06/2013 14:56:13] Piotr Holubowicz: @Stian never used Access really, but seems similar
[07/06/2013 14:56:52] Carole Goble: @Piotr It reminds me of are all too young to remember that
[07/06/2013 14:58:27] Sean Bechhofer: @Carole -- I'm not :)
[07/06/2013 14:58:43] Pique: @carole maybe because of the grey colors of the mockup? :)
[07/06/2013 14:59:01] Carole Goble: ah! you are right
[07/06/2013 14:59:33] Jose Manuel Gomez Perez: It reminds of something I did loooong ago (WebDB), though there we were more focused on presentation
[07/06/2013 15:01:39] Stian Soiland-Reyes: would have almost all the building blocks for this. I don't think it has that kind of dynamic schema-definition.
[07/06/2013 15:01:45] Pique: Resource content manager... also for external resources
[07/06/2013 15:03:09] Jose Manuel Gomez Perez: @pique: exactly
[07/06/2013 15:08:40] Sean Bechhofer: It's like Dragon's Den! Ah'm Oot!
[07/06/2013 15:09:22] Carole Goble: LOKL
[07/06/2013 15:09:26] Carole Goble: LOL
[07/06/2013 15:16:26] Sean Bechhofer: The RDF sell should be about what it supports/enables (for the user).
[07/06/2013 15:17:04] Sean Bechhofer: WHat I mean is that there shoildn't bne an RDF sell :)
[07/06/2013 15:18:34] Carole Goble: template library
[07/06/2013 15:19:19] Carole Goble: @Sean - agreed, RDF is powering the felxibility of the data describing, and linking into data collections
[07/06/2013 15:19:50] Jose Manuel Gomez Perez: @carole @sean: RDF as a commodity (finally)
[07/06/2013 15:37:15] Marco Roos: Are we done?
[07/06/2013 15:38:46] Julián Garrido: yes, we are done
[07/06/2013 15:40:20] Marco Roos: Ok, bye to everyone!!  Enjoy your remaining time in Poland. Sorry that I could not be there in person.
[07/06/2013 15:41:03] Don Cruickshank: bye!
[07/06/2013 15:45:52] Sean Bechhofer: Bye all. Thanks for the remote participation.
[07/06/2013 15:46:05] Sean Bechhofer: And good luck in space Jun9000!
[07/06/2013 17:36:06] gklyne: @Carole, I'm looking at OODT.  I've just spent 15 minutes digging around the project site and googling, and I'm still not sure what it does.  But phrases like "Grid: components to describe or profile entities such as resources with metadata and "retrieve products (resources) or metadata in standard format."  In addition to CORBA and RMI, OODT also includes WebGrid  which uses REST for providing profile and product services." give the impression that it's very heavyweight.  It's possible thatr it could be used as a substrate for my ideas, but I'm not seeing any indication of what value it would provide, and can't be sure that it won't get in the way.  My gut feel from what I'm seeing is that it's aimed at a very much more large-scale use than I want to tackle.
[07/06/2013 17:43:56] gklyne: @Carole I found a presentation: - It really looks to me as if it's intended to support large scale data, which is the opposite kind of target to what I'm considering.  Lots of technology for a substrate, but I'm seeing anything I want offered that plain old HTTP doesn't also offer, except maybe a user authentication/authorization module.  I could imagine Annalist being used with OODT-hosted data, in the sense of it being part of a wider linked data ecosystem, but not any advantage as a substrate.
[07/06/2013 17:49:21] gklyne: @stian, @Carole - Filemaker is probably a reasonable coparison, except that was desktop only.  Access too, but simpler, with with less emphasis on presentation.  I never used filemaker myself, but I knew lots on non-technical people did, so I actually think that's an encouraging comparison ;)
[07/06/2013 17:52:00 | Editado 17:52:53] gklyne: Everyone - thanks for listening, and the comments.  They do help.  I clearly need to improve the presentation in many ways if I'm to interest end-users directly.  A more compelling example would help too.  I maybe need to think about what is the minimum I could actually implement (and be prepared to throw away) to build a more compelling use case/demonstration.


The meeting will take place at the OMEGA building offices of PSNC: ul. D?browskiego 79a. Room 1111 (11th floor). 

See additional information about PSNC. Click here to see another street level view of the venue building.


For accommodation we would like to recommend you one of the following options:

  • Guest rooms in the Institute of Bioorganic Chemistry
    ul. Noskowskiego 10, 61-704 Poznan
    Reservation: phone/fax: (+48 61) 8 516 841, e-mail:

All hotels mentioned above are situated in a walking distance from the PSNC. (More hotels in Poznan)


The best options to arrive to Poznan are by plane, directly to Poznan Lawica airport, or by train, from Berlin or Warsaw.

If you arrive to Poznan by airplane, the best (and simplest) way to get to the city centre is to take the taxi. Please use the taxi parking at the airport, or call one of the following numbers: +4861 8222 222 or +48 61 8219 219 -- they should understand English ;). The cost of the taxi from the airport to the city center should be around 10EUR (~40 polish zloty). If they want more, they perhaps want to rip you off ;) in such case you should ask for receipt - the taxi drivers are obliged to have taximeter and cash register in the car. Some taxis accept credit card (those having a sticker of credit cards on the passenger windows). If you need polish zloty the currency exchange office is located in the airport hall. You also can find ATM in the airport hall.

Alternatively you may use bus singed as "L" (the bus stops in front of the airport building) and  get off at the Rondo Kaponiera stop. From this stop you may go by walk to the NH Hotel (it's around 5-7 minutes, 600 meters).  Mercure and Sheraton hotels are located very close to this roundabout (less than 100 meters).

If you arrive to Poznan by the train (from Berlin or Warsaw for instance),  again you may take a taxi in front of the main station (be careful as such taxis may be more expensive than they should), or call +4861 8222 222 or +48 61 8219 219. Taxi should cost up to 4-5 EUR. While ordering you should say that you need the taxi from the main entrance of the railway station 


Remember to also fill in the doodle poll.

  • Jose Enrique Ruiz
  • Julian Garrido
  • Lourdes: will follow by skype (health issues :( )
  • Stian: Arrive late Wed, leave early Sat. Staying at Mercure.
  • Sean: Arrive late Wed, leave early Sat. Staying at Mercure. No meat please. Vodka and vegetarian dumplings welcome though (smile)
  • Khalid: Arrive late Wed, leave early Sat. Staying at Mercure. No pork.
  • Kevin (arrive Wed eve, depart 1805 flight Fri eve)
  • Jun: will follow remotely
  • Don: will join remotely
  • Kristina: Arrive late Wed, leave 1805 flight Fri eve, staying at Mercure. Gluten free diet.
  • Graham (arr Thu AM, leave Fri PM, by train, accomm. Mercure)
  • Oscar: apologies
  • Jose, Esteban, Aleix, arriving Wednesday night, staying at NH, and leaving on Saturday afternoon.

Join remotely: Lourdes, Jun, Don, Oscar?, Marco?

Remote Access

In order to enter the videoconference, you can use either:

If you use your H.323 terminal, call the GDS number: (see Nr. below) during the videoconference time.  GDS number and PIN code (if used) both should be followed by a # sign. In case of any problem with calling GDS numbers you may connect directly to MCU: ( and choose your conference from the list (ensure to enable far end control option in your terminal). In case of private videoconference use GDS number as conference number.

If you use linphone, see instructions here

If you use Skype, see instructions here

If you use telephone,  please dial +48-61-850-1826 and enter (see Nr. below) followed by # sign. PIN code (if used) should also be followed by a # sign.

GDS Number:


The social dinner will take place on Thursday night at Villa Magnolia, located close to Pozna? International Fair ground (opposite the main railway station - Pozna? G?ówny). It is about 15 min walk from Mercure hotel in the direction to the Fair (1.2km) (see map below). 

Address: ul. G?ogowska 40, Pozna?, Poland.  60-736 Pozna?. 

  • No labels