[foaf-dev] xmlns.com and foaf site downtime / hosting

Just to let you know that I'm aware the sites are/were down.

The root cause was http://wiki.dreamhost.com/index.php/Bad_httpd_conf 
... Dreamhost (where the http services reside) switched IP address, and 
the DNS services for these sites were elsewhere (at Gandi and Joker 
registrars fwiw). For the simple xmlns.com case I've moved this so that 
DH are now the DNS server. I didn't switch foaf-project.org over yet as 
things are a little more complex there, and I want to get the namespace 
document serving again first. A task which was complicated by DH control 
panel behaving really rather oddly, but I think I'm onto it now.

http://xmlns.com/foaf/spec/ should now be visible. The FOAF wiki, svn 
and (more or less useless anyway) homepage are not yet back.

Sorry for any inconvenience.

Meanwhile, I'm less and less enamoured of Dreamhost, although I do 
appreciate that they give hacker-friendly shell access (even without 
root). Can folk here recommend other candidates? Now that Amazon EC2 has 
some save-state-to-S3 options, is that worth considering? The FOAF sites 
are essentially static files currently, but I've been experimenting with 
Drupal as a way of getting a maintainable site whose administration 
could be more distributed. Suggestions welcomed.

cheers,

Dan

--

Re: [foaf-dev] xmlns.com and foaf site downtime / hosting

Dan Brickley wrote:
> 
> Just to let you know that I'm aware the sites are/were down.

It should all be back now, including wiki and svn; however there might 
be DNS propagation/caching issues (although timeouts should be short).

Dan

Re: [foaf-dev] xmlns.com and foaf site downtime / hosting

Hi Dan,

> Meanwhile, I'm less and less enamoured of Dreamhost, although I do
> appreciate that they give hacker-friendly shell access (even without root).
> Can folk here recommend other candidates? Now that Amazon EC2 has some
> save-state-to-S3 options, is that worth considering? The FOAF sites are
> essentially static files currently, but I've been experimenting with Drupal
> as a way of getting a maintainable site whose administration could be more
> distributed. Suggestions welcomed.

We're big fans of both EC2 and Drupal, as you'll find here:

 <http://webbackplane.com/technology/ec2>
 <http://webbackplane.com/technology/drupal>
 <http://webbackplane.com/service/hosting-and-management>

Also, some gushing praise for RightScale is in one of my blog posts, here:

 <http://webbackplane.com/mark-birbeck/2008/02/sugarcrm-rightscale-and-ec2>

Regards,

Mark

--
Mark Birbeck, webBackplane

mark.birbeck@...

http://webBackplane.com/mark-birbeck

[foaf-dev] Proposed changes to Relationship vocabulary

_______________________________________________
foaf-dev mailing list
foaf-dev@...
http://lists.foaf-project.org/mailman/listinfo/foaf-dev

Re: [foaf-dev] Proposed changes to Relationship vocabulary

On Jul 3, 2008, at 6:25 PM, Ian Davis wrote:

> This would be a breaking change, since any reasoners encountering  
> these properties may be relying on them being sub properties of  
> foaf:knows. However, it is a bug in the vocabulary to keep them as  
> they are so I'd like to make the change.
>
> What's the feeling among the FOAFers about this proposed change?

Sounds mostly good to me, but I'm a bit unsure of a few of the terms  
you mention:

> rel:knowsInPassing

Not sure how much information "in passing" is meant to convey, but  
seems like it might satisfy foaf's "indicating some level of  
reciprocated interaction".

> rel:lostContactWith

Similarly here, lostContactWith seems like it implies that at some  
point in the past, there was "reciprocated interaction" in the form of  
the "contact".

> rel:apprenticeTo
> rel:mentorOf

These are less clear, but I'm having a hard time imagining an  
apprenticeship that didn't involve some sort of interaction. FWIW, the  
OED says of an apprentice, in part: "... in which the employer is  
reciprocally bound to instruct him".

Other than that, I'd say go for it.

.greg

--

-- 
"I think what would be welcome at this point is a global outbreak of
  (de)constructive nihilism, so that people could stand up and say ...
  I'm not mad as hell anymore, and I'm not gonna take it!"    - TorgoX

_______________________________________________
foaf-dev mailing list
foaf-dev@...
http://lists.foaf-project.org/mailman/listinfo/foaf-dev

Re: [foaf-dev] Proposed changes to Relationship vocabulary

Hi Ian,

A subproperty of a symmetric property not necessarily be symmetric. It
can be, but isn't entailed to be.

Therefore, I guess there is no need to change definitions
{relationship vocab properties} rdfs:supPropertyOf foaf:knows.

2008/7/4, Ian Davis <iand@...>:
> Although foaf:knows is not declared to be symmetric it _is_
> implied by the prose and Dan has suggested making this explicit in the
> schema. This would cause problems with properties such as rel:childOf which
> are clearly not symmetric.
>
> Therefore I propose that I change the relationship vocabulary so that the
> following properties are no longer sub properties of foaf:knows:

--

-- 
 <at> prefix : <http://www.kanzaki.com/ns/sig#> . <> :from [:name
"KANZAKI Masahide"; :nick "masaka"; :email "mkanzaki@..."].

Re: [foaf-dev] xmlns.com and foaf site downtime / hosting

On Wed, 2008-07-02 at 15:33 +0100, Dan Brickley wrote:
> Drupal as a way of getting a maintainable site whose administration 
> could be more distributed. Suggestions welcomed.
> 

Dan,

Can I also suggest possibly using Chisimba as an alternative to Drupal?
Chisimba has a built in FOAF module as well as an option to use the AWS
as well (although still a little in development but that can be finished
really quickly).

Check out http://avoir.uwc.ac.za for more information as well.

--Paul

All Email originating from UWC is covered by disclaimer 
http://www.uwc.ac.za/portal/public/portal_services/disclaimer.htm 

_______________________________________________
foaf-dev mailing list
foaf-dev@...
http://lists.foaf-project.org/mailman/listinfo/foaf-dev

Re: [foaf-dev] Proposed changes to Relationship vocabulary

2008/7/4 KANZAKI Masahide <mkanzaki-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>:

Hi Ian,

A subproperty of a symmetric property not necessarily be symmetric. It
can be, but isn't entailed to be.

Of course I trust your statement on this, but would be grateful for a pointer the appropriate part of the specs where it appears (I'm overdue a total re-read of the whole spec suite, but there are rather a lot and I'd completely forgotten this part).
 

Therefore, I guess there is no need to change definitions
{relationship vocab properties} rdfs:supPropertyOf foaf:knows.

In principal, sure. But I wonder if this might lead to some confusion (especially among other people like myself who have forgotten this non-rule

Perhaps a possible way of making this clearer might be to add a term to either the foaf of relationship spec something like:

x:asymmetricRelationship rdfs:subPropertyOf foaf:knows

(with a better name!)

then make

{relationship vocab properties} rdfs:supPropertyOf  x:asymmetricRelationship

with a little explanatory prose.

Cheers,
Danny.

--
http://dannyayers.com
~
http://blogs.talis.com/nodalities/this_weeks_semantic_web/

_______________________________________________
foaf-dev mailing list
foaf-dev@...
http://lists.foaf-project.org/mailman/listinfo/foaf-dev

Re: [foaf-dev] Proposed changes to Relationship vocabulary

Hi Danny,

> Of course I trust your statement on this, but would be grateful for a
> pointer the appropriate part of the specs where it appears (I'm overdue a
> total re-read of the whole spec suite, but there are rather a lot and I'd
> completely forgotten this part).

It's not easy to show a pointer... "A subproperty of a symmetric
property = symmetric" is not necessarily the case, because there is no
such definition in OWL semantics that leads such interpretation.

OWL Semantics 5.2 says, "if P is SymmetricProperty, then P(x, y) =>
P(y, x)", and nothing more than that.

You might be interested in taking a look at new SKOS Reference draft
[1], section 8.6.3, where the editors put this note:
[[
Note that, although skos:related is a symmetric property, this
condition does not place any restrictions on sub-properties of
skos:related. I.e. sub-properties of skos:related could be symmetric,
not symmetric or antisymmetric, and still be consistent with the SKOS
data model.
]]

> {relationship vocab properties} rdfs:supPropertyOf  x:asymmetricRelationship
>
> with a little explanatory prose.

I think machine (application) won't be confused that {relationship
vocab properties} are symmetric, but humans might be. So, some
explanatory prose will be a good idea.

cheers,

[1] http://www.w3.org/TR/2008/WD-skos-reference-20080609
--

-- 
 <at> prefix : <http://www.kanzaki.com/ns/sig#> . <> :from [:name
"KANZAKI Masahide"; :nick "masaka"; :email "mkanzaki@..."].

Re: [foaf-dev] Proposed changes to Relationship vocabulary

Danny Ayers wrote:
> 2008/7/4 KANZAKI Masahide <mkanzaki@... 
> <mailto:mkanzaki@...>>:
>
>     Hi Ian,
>
>     A subproperty of a symmetric property not necessarily be symmetric. It
>     can be, but isn't entailed to be.
>
>
> Of course I trust your statement on this, but would be grateful for a 
> pointer the appropriate part of the specs where it appears (I'm 
> overdue a total re-read of the whole spec suite, but there are rather 
> a lot and I'd completely forgotten this part).
The RDF sepc doesn't contain a list of all the rules that do not exist 

But the wrong assumption might be caused by the following rule [1]:

uuu |rdfs:subPropertyOf| vvv |.|
vvv |rdfs:subPropertyOf| xxx |.|

==>

uuu |rdfs:subPropertyOf| xxx |.

if one wrongly things of owl:SymmetricProperty as a superproperty of 
foaf:knows this would imply that any subproperty of foaf:knows is 
symmetric as well, but owl:SymmetricProperty is a class and not a 
property and so foaf:knows might be an instance of owl:SymmetricProperty 
not a subproperty. There is no rule that supproperty inherit any 
property from their super-property except 
inverseOf(||rdfs:subPropertyOf) as in the rule above.

|

uuu |rdfs:subPropertyOf| vvv |.|
vvv |rdf:type| xxx |.|

does not imply:

uuu |rdf:type |xxx |||

|
>  
>
>     Therefore, I guess there is no need to change definitions
>     {relationship vocab properties} rdfs:supPropertyOf foaf:knows.
>
>
> In principal, sure. But I wonder if this might lead to some confusion 
> (especially among other people like myself who have forgotten this 
> non-rule 
>
> Perhaps a possible way of making this clearer might be to add a term 
> to either the foaf of relationship spec something like:
>
> x:asymmetricRelationship rdfs:subPropertyOf foaf:knows
>
> (with a better name!)
>
> then make
>
> {relationship vocab properties} rdfs:supPropertyOf  
> x:asymmetricRelationship
>
> with a little explanatory prose.
|what is an asymmetric property, does this mean that i cannot be the son 
of my son or just that by being his father I'm not necessarily a son of 
him as well? Intuitively fatherOf would be strongly asymmetric while 
inLoveWith just doesn't have to be symmetric (but theoretically a 
property of the same type in the other direction is possible).

Cheers,
reto

1. http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#RDFSRules|