[foaf-protocols] WebID - mandated syntax or market solution? was WebID Incubator Charter draft

Melvin Carvalho melvincarvalho at gmail.com
Sat Dec 18 16:37:16 CET 2010


2010/12/18 Jiří Procházka <ojirio at gmail.com>:
> On 12/18/2010 02:25 PM, Henry Story wrote:
>>
>> On 18 Dec 2010, at 13:27, Jiří Procházka wrote:
>>
>>>>
>>>> I don't quite see how a badge is going to be useful for robots though. Who is going
>>>> to "see" the badge here. Presumably robots will get a request for a client side certificate
>>>> and send one over.
>>>
>>> When user installs the software he should expect it to work with all
>>> WebIDs in basic functionality, regardless how many additional syntaxes
>>> it doesn't understand, sacrificing potential performance gains and
>>> additional features.
>>
>> Yes, part of the Working Group will be to develop test suites.
>> I think we even have a few allready.
>>
>>>
>>>>> You don't want your automated
>>>>> backup tool to fail, except when it looses connectivity. Maybe browsers
>>>>> will have no problem being updated with latest WebID syntax
>>>>> developments, but tools running on server or routers certainly will,
>>>>> also they often need to be as small as possible. These use cases are the
>>>>> edge on other similar technologies, so please consider them.
>>>>
>>>> You have developed a good practical reduction ad absurdum. If too many
>>>> syntaxes develop with no way of knowing their semantics, then clearly it
>>>> won't work. Therefore one should not develop syntaxes with no clear semantics.
>>>
>>> I haven't said that. That is wrong because there is a different option
>>> which I advocate - have 1 simple syntax with simple semantics defined in
>>> a spec, accounting for basic functionality, while welcoming conneg of
>>> different syntaxes.
>>
>> Yes, that is well understood that you are defending the one syntax
>> option. My belief is that in the syntax language wars that is untenable.
>> Just with XML there are many options: liberty, SAML, Portable Contacts,...
>> then there is XHMLT/HTML debate, then the JSON, N3, and other formats and so
>> on.
>>
>> Whatever one we choose each of the other one's will cry holy war, and find
>> a reason to do it their way anyway. And there is no way we can stop them.
>> I don't think badges are that powerful. What is powerful are good working
>> implementations with wide usage. That is where you gain traction.
>
> Well I suppose best would be to make a separate (but linked) spec for a
> thing called for example "WebID-I" as "WebID - Interoperable" specifying
> such minimal required syntax to be compliant with it, for the sole
> purpose of the badges. Hopefully enough people will be reasonable and
> seeing its purpose and its separateness from the core WebID spec and
> will save the holy war for some other event.

The Web is all about interoperabiility.  That's why we have Web
standards, it's not limted to WebID.

"Badges" seem like a neat idea, to advertise some of the things you
can do, but that's not the only way to interoperate.

Perhaps the concept needs fleshing out a bit more.  Could be a good
topic of discussion when the XG kicks off.

>
> Best,
> Jiri
>
>> Ok, so I hope that shows that Peter Williams as wrong when he says that I am
>> absolutist. Far from it. Don't forget that on every e-mail I write the symbol
>> of the babel fish is hidden.  :-)
>>
>>       Henry
>>
>>
>> Social Web Architect
>> http://bblfish.net/
>>
>
>
> _______________________________________________
> foaf-protocols mailing list
> foaf-protocols at lists.foaf-project.org
> http://lists.foaf-project.org/mailman/listinfo/foaf-protocols
>


More information about the foaf-protocols mailing list