Re: Hash vs Hashless URIs

On 12/11/12 9:14 AM, Melvin Carvalho wrote:
>
>
> On 11 December 2012 14:48, ☮ elf Pavlik ☮ 
> <perpetual-tripper@wwelves.org <mailto:perpetual-tripper@wwelves.org>> 
> wrote:
>
>     Excerpts from Melvin Carvalho's message of 2012-11-21 15:51:49 +0000:
>     > On 21 November 2012 15:43, ☮ elf Pavlik ☮
>     <perpetual-tripper@wwelves.org
>     <mailto:perpetual-tripper@wwelves.org>>wrote:
>     >
>     > > Excerpts from Kingsley Idehen's message of 2012-11-21 14:18:13
>     +0000:
>     > > > Please call vote on this matter. I have better things to do
>     with my
>     > > > time. If a majority want a WebID to be defined a hash based
>     HTTP URI,
>     > > > let's get that in the open right now. We all have decisions
>     to make
>     > > > about how we spend our time and priorities across our respective
>     > > > development efforts.
>     > >
>     > > $ curl -I http://sebastian.tramp.name
>     > > HTTP/1.1 303 See Other
>     > > Date: Wed, 21 Nov 2012 14:33:16 GMT
>     > > Server: Apache/2.2.14 (Ubuntu)
>     > > X-Powered-By: PHP/5.3.2-1ubuntu4.18
>     > > X-Pingback: http://pingback.aksw.org/
>     > > Location: http://bis.informatik.uni-leipzig.de/SebastianTramp
>     > > Vary: Accept-Encoding
>     > > Content-Type: text/html
>     > >
>     > > personally i would also like to use WebID without #i or #me or
>     #self or
>     > > #elf or #wtf ;)
>     > >
>     >
>     > elf, can you explain why you would want this ... do you have any
>     specific
>     > reason, or is it simply a 'vanity' url?
>     first of all apologies for this super delayed reply!
>
>     to start in very general way, i like when we design technology
>     that serves people and make sure that we don't expect people to
>     stretch their comfort ONLY to fit some technological solution (at
>     the same time i don't mind encouraging some stretching but
>     motivated with more extensional reasons!) i also realize that we
>     design with certain group of people in mind as well - *developers*
>     - but if someone needs to stretch from technical reasons ... ;)
>
>     when i think of a concept of ID, i see most people identifying
>     themselves nowadays with either their email address (including
>     unfortunate @twitter thing) or their homepage / online profile. i
>     see various challenges people from webfinger/simple web discovery
>     face with using email like identifiers, and i agree that we can
>     simplify certain things by sticking to HTTP. when it comes to
>     identifying onself with certain HTTP URL, may people may have
>     their homepage, blog, various online profiles. sometimes people
>     also print them on their personal cards. i've never seen someone
>     using a URL with #something it for personal homepage or online
>     profile. i understand distinction between referencing an online
>     profile and entity which this profile describes, still requiring
>     people to use #whatever feel to me like putting burden on those
>     who *just want to use this damn thing*, and very likely brining
>     confusion to broad audience.
>
>     besides making pools and discussions within this group, mostly
>     with people having strong and at least medium (like myself)
>     understanding of underlying technologies. i would like to look for
>     feedback from people like *my auntie* or at least *arts student*
>     who can find their way in interacting online but see this whole
>     thing as a magical black box.
>
>     from my perspective sticking to identifiers like
>     http://funkyjohny.me or http://funkyclub.org/johny can make it
>     simplest for people to really identify oneself with such URL,
>     which to my understanding mean that even using 303 may bring
>     confusion if we type http://funkyjohny.me and end up somewhere
>     let's say on http://funkyjohny.me/profile
>
>     looking at present tendencies in online interaction, besides
>     crafting technologies with solid technical foundations we MUST
>     also make sure that we make them SEXY!
>
>     once more i see it very impractical to design beautiful
>     *Interaction Experience* just within circle of geeks and such
>     *front end* related issues i would like to confront with people
>     who consider internet as magical thing :)
>
>     last but not least i would like to say that i find great
>     appreciation to all the work people do in this and other related
>     working groups! and i would find it a big shame if lack of
>     SEXYness become and obstacle to a broad adoption of this work...
>
>
> FWIW I have tried explaining # URIs to people of all ages, male and 
> female.  I start by saying that we have the concept of web pages.  
> Then I tell them that it is possible to split web pages up into 
> sections using the # character.  Each section can be about a different 
> topic.  Then each topic can have properties and attributes in it's own 
> right.
>
> I have *never* once failed to explain this to a non technical person 
> in less than 5 minutes.  In fact most people come back to me and say, 
> that was very easy.
>
> It's only the "cool kids" that find this concept very hard to grasp.  
> That is because it requires unlearning what they have learnt.  It's 
> sad to say that some web developers will never understand this.  It's 
> not about lack of knowledge, it's about preconcieved notions.  This is 
> not unique to the web, it's true in many fields.
>
>
>
Melvin,

The issue isn't hash or hashless HTTP URIs. The issue is simply about 
URIs (Uniform Resource Identifiers) where the "R" is broken because it 
means "Resource" and the problem with "Resource" is that its really 
supposed to be a synonym of "Entity" .  Resources or Entities exist in 
different realms. For example a Web document is a Web realm entity, each 
of us is an Entity of a different realm from the Web. Thus, the 
denotation mechanisms have nuanced implications that continue to be lost 
in these permathreads.

As I said above, the issue is that you have Entities in different realms 
associated with realm oriented denotation mechanisms. One camp wants the 
denotation mechanism to be http: scheme URIs which are unintuitive (but 
cost effective), when the realm is the World Wide Web. Another camp 
wants the denotation mechanism to be some other scheme which is more 
intuitive than http (but less cost-effective) when the realm is the 
World Wide Web, but not so when the realm is the Internet.

Remember, I explicitly said to TimBL: "I'll concede http: scheme 
specificity for WebID definition since he sees http: colloquially 
meaning Web" . Personally, I prefer URIs to be the focal point of modern 
name resolution and profile discovery  oriented protocols. 
Unfortunately, I am now much less interested in arguing about the 
virtues and wisdom inherent in such an interoperable approach, with 
regards to WebID.

Others: to really understand why myself, Ted, Toby, Stephane, and many 
others who care about interop it might be worth digesting these 
presentation excerpts:

1. 
http://www.youtube.com/watch?v=PU7nKBNR1Vs&feature=player_detailpage#t=1276s 
-- Architecture vs Engineering
2. 
http://www.youtube.com/watch?v=PU7nKBNR1Vs&feature=player_detailpage#t=1840s 
-- more of the same .


-- 

Regards,

Kingsley Idehen 
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen