Re: Are POIs always tangible?

Hi,

I'm with Thomas here that if we want to specify elaborate about the
nature of a POI, such as which brand it belongs to, we should look at
using linked data.

A non-physical POI is useful for use cases such as "the place where
Lee Harvey Oswald was shot". However, Andy posted a second
qualification for a conceptual POI, namely something that does not
have a geographical location. I don't see a use case for that yet, but
please surprise me :)

Jens

On Thu, Apr 14, 2011 at 3:01 PM, Thomas Wrobel <darkflame@gmail.com> wrote:
> POIs should certainly not have to be tangable - even "null" POIs would
> have their uses as other POIs could be positioned relative to them,
> making it easy to stuff to be moved/updated together.
>
> Likewise many non-physical unity's or even concepts could be a POI
> provided they have some sort of meaningful real world location(s).
>
> I'm not keen, however, on the idea of going to far into contact
> details/buisness/chain stuff...the idea aof a "parent" of a branch of
> a store being the business franchise is completely different to the
> idea of POI having a physical locational relationship with another.
> (ie,  poster POI might be positioned relatively to the bus it is on).
>
> While the first idea of "parent" is indeed usefull from a search
> perspective, it should be dealt with by existing semantic search and
> linked data solutions - as long as the POI stores metadata about it
> being a "starbucks", then its semantic relationship should be pulled
> from databases elsewhere without needing the POI standard to define
> business details at all.
>
>
> ~~~~~~
> Reviews of anything, by anyone;
> www.rateoholic.co.uk
> Please try out my new site and give feedback :)
>
>
>
> On 14 April 2011 14:41, Seiler, Karl <karl.seiler@navteq.com> wrote:
>> All,
>>
>>
>>
>> As many of the most popular POIs are “members” of chains / brands, and since
>> many a search for POIs involves “Starbuck” near me, a pure name search (with
>> all its fuzzy logic variability) is not as powerful as a chain search.
>>
>>
>>
>> Therefore, in my view, a POI should be able to be a member of a group,
>> chain, association via linking. Also, it is reasonable that the linked-to
>> entity (parent POI) carry POI attributes like contact info.
>>
>>
>>
>> _______________________________
>>
>> Karl Seiler
>>
>> Director Location Technology & Services
>>
>> NAVTEQ - Chicago
>>
>> (T)  +312-894-7231
>>
>> (M) +312-375-5932
>>
>> www.navteq.com
>>
>>
>>
>> From: public-poiwg-request@w3.org [mailto:public-poiwg-request@w3.org] On
>> Behalf Of Andy Braun
>> Sent: Thursday, April 14, 2011 7:30 AM
>> To: Public POI @ W3C
>> Subject: Are POIs always tangible?
>>
>>
>>
>> We started a discussion on the call this week[1] that has been floating
>> around for some time.
>>
>>
>>
>> Should the POI spec account for items that are not physical in nature and
>> therefore are not specifically tied to some geospatial context. Examples
>> would be things like Starbucks (the corporation) or a government body.
>>
>>
>>
>> For me the question comes down to one of linking. Does tying individual
>> Starbucks together have value? (and as Alex pointed out is a conceptual POI
>> even necessary for that).  The immediate concern is that we end up trying to
>> write a spec that covers everything. This could lead us down a slippery
>> slope.
>>
>>
>>
>> It would great to seem some other thoughts in this area
>>
>>
>>
>> Andy
>>
>>
>>
>> [1] http://www.w3.org/2011/04/13-poiwg-minutes.html#item05
>>
>> ________________________________
>> The information contained in this communication may be CONFIDENTIAL and is
>> intended only for the use of the recipient(s) named above. If you are not
>> the intended recipient, you are hereby notified that any dissemination,
>> distribution, or copying of this communication, or any of its contents, is
>> strictly prohibited. If you have received this communication in error,
>> please notify the sender and delete/destroy the original message and any
>> copy of it from your computer or paper files.
>
>

Received on Thursday, 14 April 2011 13:48:39 UTC