W3C home > Mailing lists > Public > public-gld-wg@w3.org > May 2013

Re: Vocabulary visualization - can you help?

From: Dave Reynolds <dave.e.reynolds@gmail.com>
Date: Thu, 02 May 2013 16:39:41 +0100
Message-ID: <518288BD.2040503@gmail.com>
To: public-gld-wg@w3.org
Hi Joćo Paulo,

On 02/05/13 16:19, Joćo Paulo Almeida wrote:
> Dear All,
>
> Thanks for you comments.
> I attach a new version of the diagram.

Great, thanks again.

> I've addressed most of the comments:
>
> - Single colon is used instead of '::' to represent namespace scoping
> - vcard is no longer the range of siteAddress (in the absence of a range,
> I've used rdf:Resource, is this OK?)

I would prefer a greyed-out vcard class. The point is that we recommend 
use of vcard but don't require it.

I realize this is not part of UML but the point of this diagram is not 
to be UML but to be a useful guide to help people understand ORG :)

I guess an alternative would be a caption under your rdfs:Resource box 
stating that use of vcard:VCard is suggested.


> - navigability is always shown
> - foaf:Group is aligned vertically with foaf:Agent
>
> With respect to the aesthetics, I am not in favour of using rounded
> corners, as this would no longer be a UML diagram. I am in favour of
> standards :-)
> Drop shadows are not possible in the tool.
>
> About clickable parts, this is just a matter of making an area map in
> html. I can do that once we agree on the diagram.
>
> Regards,
> Joćo Paulo

Cheers,
Dave

> On 1/5/13 10:00 AM, "Richard Cyganiak" <richard@cyganiak.de> wrote:
>
>> Dave, Joćo,
>>
>> Comments on the proposed diagram, in addition to what Dave said:
>>
>> 1. I understand the semantics of the distinction between gray and white
>> boxes. But why is the Organization class colored differently? Just
>> because it's the most important one? If that's the case, its size and
>> central position are sufficient to communicate that, and it doesn't need
>> to have a different color.
>>
>> 2. I agree with Dave that the diagram looks a bit bland. Can it be made a
>> bit more visually interesting? Rounded corners, drop shadows, anything?
>>
>> 3. Personally I would prefer if the connections that have inverses would
>> have arrows on both ends, rather than on no end.
>>
>> 4. I think the member connection and the hasMember connection both lack
>> an arrow.
>>
>> 5. Minor point regarding the layout: Organization, foaf:Group and
>> foaf:Person have something in common: They are all subclasses of
>> foaf:Person. But in the arrangement, they have nothing in common, and are
>> positioned quite differently.
>>
>> 6. The colors besides black and white should be picked, if possible, from
>> the color palette already used in http://www.w3.org/TR/vocab-org/ , to
>> make it fit in nicely.
>>
>> Best,
>> Richard
>>
>>
>> On 1 May 2013, at 12:35, Dave Reynolds <dave.e.reynolds@gmail.com> wrote:
>>
>>> Hi Joćo Paulo,
>>>
>>> [Sorry to be slow to respond, just too busy :(]
>>>
>>> Many thanks for this. It is definitely an improvement over the earlier
>>> version. Does this technology offer clickable images as as well?
>>>
>>> The diagram shows the range of org:siteAddress as being vcard:Vcard yet
>>> this is no longer the case, vcard is now simply a recommended option.
>>> With your tooling is it possible to grey out boxes?
>>>
>>> The use of double '::' is incorrect from an RDF point of view. With
>>> your tooling is it possible to use single ':' instead?
>>>
>>> Aesthetically it's a little uninspiring but acceptable.
>>>
>>> Do other working members having opinions on whether to adopt this (with
>>> above tweaks) in preference the current diagram?
>>>
>>> Dave
>>>
>>> On 18/04/13 15:15, Joćo Paulo Almeida wrote:
>>>> Dear All,
>>>>
>>>> Please find attached our proposal of diagram for ORG. It is a complete
>>>> diagram (only a transitive derived property is ommitted, the rest is
>>>> all
>>>> in).
>>>>
>>>> I've tried to address the issue that Dave raised with respect to the
>>>> representation of attributes.
>>>>
>>>> Of course, we could produce a simplified version (leaving some elements
>>>> out).
>>>>
>>>> We have followed a number of conventions to represent the ontology in
>>>> UML:
>>>> - Classes in white are imported from other vocabularies
>>>> - Navigability is only shown (arrows) in case the property does not
>>>> have
>>>> an inverse
>>>> - Non-disjoint subclass specialisation is shown with different arrows
>>>> to
>>>> favour correct interpretation
>>>>
>>>>
>>>> Best regards,
>>>> Joćo Paulo
>>>>
>>>>
>>>> On 12/4/13 9:02 AM, "Joćo Paulo Almeida" <jpalmeida@ieee.org> wrote:
>>>>
>>>>> Dear Phil,
>>>>>
>>>>> We can do that. That is, we'll make a proposal and bring it to the
>>>>> group.
>>>>> I hope we'll be able to address the concern Dave raised with respect
>>>>> to
>>>>> the diagram we produced earlier for ORG, and I believe we can build
>>>>> consensus on some form of graphical representation.
>>>>>
>>>>> I am sorry I was not able to join in the discussion today on
>>>>> ORG/RegOrg.
>>>>> This is because Brazil is -5 hours with respect to Dublin time.
>>>>>
>>>>> Regards,
>>>>> Joćo Paulo
>>>>>
>>>>>
>>>>> On 12/4/13 7:06 AM, "Phil Archer" <phila@w3.org> wrote:
>>>>>
>>>>>> During the face to face meeting (still ongoing), we've been
>>>>>> discussing a
>>>>>> comment concerning the diagram for the ORG ontology. This highlights
>>>>>> the
>>>>>> fact that all those of us who have created diagrams for our vocabs
>>>>>> use
>>>>>> different tools and create different-looking diagrams.
>>>>>>
>>>>>> Ideally, we'd like them all to have the same look and feel. And even
>>>>>> more ideally we'd like the diagrams to be clickable so you can jump
>>>>>> to
>>>>>> the relevant definitions etc. That's a nice to have, not a
>>>>>> requirement.
>>>>>>
>>>>>> Do you have the tooling and/or the time to help create these please?
>>>>>>
>>>>>> Phil.
>>>>>>
>>>>>>
>>>>>> --
>>>>>>
>>>>>>
>>>>>> Phil Archer
>>>>>> W3C eGovernment
>>>>>> http://www.w3.org/egov/
>>>>>>
>>>>>> http://philarcher.org
>>>>>> +44 (0)7887 767755
>>>>>> @philarcher1
>>>>
>>>
>>>
>>
>>
>
Received on Thursday, 2 May 2013 15:40:11 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:52:08 UTC