Changes between Version 4 and Version 5 of Taskforces/FCS/VidConf20150624


Ignore:
Timestamp:
06/24/15 13:25:38 (9 years ago)
Author:
Leif-Jöran
Comment:

Some restructuring and proofing, but more is needed

Legend:

Unmodified
Added
Removed
Modified

  • Taskforces/FCS/VidConf20150624

    v4 v5  
    1414
    1515== Outcome: Minutes, Decisions and Actions (currently draft) ==
    16  Main focus Dataview return format
     16Main focus of today's Dataview return format
    1717
    18 Not much examples except for Oliver's.
     18Not much examples except for Oliver's. We need more.
    1919
    20 Advanced dataview
    21 Adhoc or based on something?
    22 Oliver mental model one dataview will convey all layers etc
     20Matej, are these exampel formats for Advanced dataview adhoc or based on something?
    2321
    24 So no msd dataview
    25 explict elements for some layers
    26 container format.
     22Oliver: mental model of that one dataview will convey all layers etc. E g no msd dataview or explict elements for some layers. It is a container format.
     23
    2724Pavel, looks nice. Reason behind it?
     25
    2826Any tree? Complex tree view? Multiple layers?
    2927
    30 Press stop 2-3 secs after stop talking.
    31 Hierarchical structure.
    32 Proper scope of work.
     28Press stop broadcasting 2-3 secs after you stop talking.
     29
     30== Questions on Hierarchical structure ==
     31
     32We need to define a proper scope of work.
    3333Transport hierarchy
    3434Matej, really another complexity level. Start with query annotation layers, and only when we master that move forward.
    3535
    36 Pavel, second suggestion multi annotations. Fine with forgetting trees for now. structured/hierarchical attributes.
    37 I think it can be nice in terms of ease of understanding that some thing belong together.
     36Pavel, second suggestion multi annotations. Fine with forgetting trees for now. Structured/hierarchical attributes.
     37I think it can be nice in terms of ease of understanding that some things belong together.
    3838
    39 Matej, what do you mean? Example.
     39Matej, what do you mean? Provide example.
    4040
    4141case
    4242
    43 Oliver (IDS): question of course is, if structured, how would it be structured ;)
     43Oliver: question of course is, if structured, how would it be structured ;)
    4444a client like the aggregator needs to make sense of that
    4545
    46 harmonize if they are structured.
    47 No good solution yet.
     46We need to harmonize if they are structured. No good solution yet.
    4847
    4948Dieter: Pavel: is there a standard way for that in the universal dependencies?
    5049
    51 Controlled vocab
    52 Like for POS the UD-17 we decided on.
    53 Jörg not every endpoint will provide UD-17 we are not going to retag all corpora.
     50Matej: If we go this route we need to come up with even more controlled vocabularies, like for POS the UD-17 we decided on.
    5451
    55 Pavel, agree hierarchical attributes is too complex for now.
     52Jörg: not every endpoint will provide UD-17. We are not going to retag all corpora.
    5653
    57 Jörg bidrectional translation.
     54Pavel: agree hierarchical attributes is too complex for now.
    5855
    59 Dieter, limited list
    60 choice
    61 first proposal, syllables.
    62 finer granularity. In worst case every single character.
     56Jörg: bidrectional translation is needed.
    6357
    64 Candidate for speech
    65 or primarily for textual resources?
     58Dieter: limited list layers
     59before making a choice
    6660
    67 time lined signal
    68 "textual" annotations
     61== Question about first proposal, syllables ==
     62Oliver: Finer granularity. In worst case every single characters.
    6963
    70 Dieter concrete example. Silence is best covered by example 1.
     64Dieter: Candidate also for speech or primarily for textual resources?
     65
     66Basically time lined signal "textual" annotations
     67
     68Dieter: provides concrete example. Silence seems to be best covered by proposal 1.
    7169Transcribed speech corpora.
    7270
    73 Matej, generic way to describe atomic units. So time stamp could be fitted with example 2.
     71Matej: Not really, proposal 2 is better but needs other unit like time stamps. Generic way to describe atomic units.
    7472
    75 Oliver, yes, offsets can well be timestamps
     73Oliver: yes, offsets can well be timestamps
    7674
    77 silence background noice
     75silence, background noice are interpretations/computed annotations
    7876
    79 Combination of dataviews
    80 allow to reference other layer items
    81 non-superior layers
    82 order is not defined, but
     77What about combination of dataviews
     78allow to reference other layer items. This was disussed at the workshop as one alternative LJO wants to provide proposal.
     79
     80Oliver: non-superior layers, order is not defined in the current proposals
     81
    8382Dieter: I have the impression we are re-inventing formats like EAF: https://corpus1.mpi.nl/media-archive/demo/Ams_Demo/versioning_demo/Annotations/118_fishing2-fire-2011.eaf
    8483
    85 Comparison matrix.
    86 
    87 Keep things simple.
    88 
    89 Keep it fairly straight forward.
    90 
    91 Keep non-textual formats in mind.
    92 
    93 Doing concrete proposals. More are needed.
    94 
    95 transform to standard format.
    96 
    97 What about parallel corpora?
    98 another annotation layer trans
    99 
    100 More discussion is needed. Concrete proposals.
    101 
    102 Hanna, concrete examples? Propose new layers?
    103 Mappings are welcome.
    104 
    105 Use second view in examples
    106 
    107 Need attribute for unit.
    108 
    109 Matej, lexical resources.
    110 
    111 Toying
     84LJO action: Add EAF to comparison matrix wehen we have more examples.
    11285
    11386
    114 LJO: add page for advanced datawiew to trac.
    115 Next meeting, 2-3 weeks
    116 Might be flash or adobe connect
     87Keep things simple. We need to keep it fairly straight forward and still keep non-textual formats in mind.
     88
     89We need to work on getting more concrete proposals available. More examples are needed.
     90
     91The transport format and its data could be transformed to standard format if needed.
     92
     93== Question about parallel corpora ==
     94Adds another annotation layer, potentially translation also
     95
     96Consensus for actions:
     97* More discussion is needed.
     98* Concrete proposals.
     99* concrete examples
     100
     101Hanna: propose new layers? Mappings are welcome.
     102
     103Decision: Use second proposal for new examples or another proposal with examples
     104
     105Diter: Need attribute for unit.
     106
     107
     108== Question about lexical resources ==
     109Matej raises the question about how to handle lexical resources.
     110
     111
     112== Actions summary ==
     113* ALL: Toying with the current proposals and example
     114* ALL: Provide proposals
     115* ALL: Provide examples
     116* LJO: add page for advanced datawiew to trac where these can be put for reference, still important to use mailing list for discussion
     117* Next meeting, 2-3 weeks, might be flash again or adobe connect
    117118
    118119