Re: More Thoughts on Links and Operation Subclasses

On 2/4/14 9:24 AM, Markus Lanthaler wrote:
> On Monday, February 03, 2014 11:13 PM, Kingsley Idehen wrote:
>> On 2/3/14 4:15 PM, Markus Lanthaler wrote:
>>> On Monday, February 03, 2014 10:05 PM, Kingsley Idehen wrote:
>>>> On 2/3/14 3:19 PM, Markus Lanthaler wrote:
>>>>> I think the biggest problem is how this stuff is documented and the
> fact
>>>>> that people won't read it:-)  It's not a contract. It's a hint.
> Clients have
>>>>> to interpret the response in any case. Some will be more tolerant,
> some will
>>>>> break when they don't get back what they expected. That's similar to
> how
>>>>> people often run into troubles when parts of a website get redesigned
> and
>>>>> "nothing works anymore" because it looks different or they have to
> take
>>>>> different paths.
>>>> This is where TURTLE is your friend. The narrative is more readable in
>>>> TURTLE that it will ever be in JSON-LD.
>>> Sorry, but I don't see how this is relevant in this context.
>> It is utterly relevant. Do you want this stuff to be readable or not?
> What I was trying to understand is how the serialization format is relevant
> to the fact that return type of a hydra:Operation has to be considered as a
> hint rather than a contract. Sorry, I still can see it.

That's the problem.

Hold the discussion using TURTLE examples, then others will actually 
hone in (much quicker) into the matters at hand.

What is a *hint* ?

>
>
>>>> I encourage you to discuss in TURTLE so that more folks get involved in
>>>> these discussions.
>>> Yep.. Ruben will convert the examples to Turtle and we will probably
> include
>>> both versions in the spec eventually.
>> Good!
>>
>> We don't need a TURTLE vs JSON-LD vs any other notation distraction re.
>> Hydra. Let's focus on the entity types and relation types that
>> constitute the vocabulary in question.
> Amen :-) I thought that was what the discussion was about up to your mail...
> but perhaps I also just missed the point you tried to get across.

I think you did. My point is that the discussions will be much more 
productive if TURTLE examples are used as opposed to JSON-LD only 
examples. That's all.

JSON-LD is not about readability, its value comes into play when a 
particular profile of developer is implementing code.

>   As you may
> have noticed, I always reply in the same syntax as was used in the original
> message.

Yes, but you can also take the content expressed in one notation and the 
pass it through a transformer/distiller to produce TURTLE (when the 
example was JSON-LD). Doing this alone provides *hints* about the 
statements and relation semantics they are supposed to convey i.e., if 
the meaning is lost in translation then something is fundamentally wrong.

>   So even if the spec is still JSON-LD only, I really don't care much
> about syntaxes.

A JSON-LD only spec is an unreadable spec that will ultimately fall foul 
of  the same old problem of trying to reduce Semantics to Syntax. This 
pattern will fail, eternally!

> For some people JSON-LD works better, for others Turtle.

"Looks" isn't the issue. The issue is being able to discern and 
comprehend entity relation semantics. Developing a vocabulary 
collaborative (where participant profiles vary) isn't a syntax beauty 
contest. Don't undermine this effort with this mindset.

>
>
>> Examples should be printed in
>> TURTLE for broad and productive participation (assuming that is an
>> actual goal).
> I know you would prefer if the examples in the spec would be written in
> Turtle.

No, I prefer examples to be comprehensible to a broad profile of 
discussion participants. I am sure you know OpenLink Software has 
implemented every existing RDF notation that's been conjured up over the 
last 13+ years. Thus, I don't care about any of them on a superficial 
level, I believe in "horses for course" as opposed to "one size fits all" .


> Ruben feels similarly and volunteered a while ago to convert them.

Yes, because he understands the issues and the history.

> All of us are quite busy so if there's something you would like to get
> changed quickly, the easiest way is to file a pull request. Don't get me
> wrong but, as you stated yourself above, these discussions are quite a
> distraction. I want them to get out of the way but also my day has just 24
> hours.

I didn't say the discussions are a distraction, I said: discussion 
participation would be more productive i.e., it would be less costly 
(time-wise) if the points being made are clearly understood when 
accompanying examples are expressed using a readable notation that 
doesn't obscure the critical role of relation semantics in said 
discussions.

>
>
> --
> Markus Lanthaler
> @markuslanthaler


-- 

Regards,

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