Who could've thought the question of deciding whether @Ref/Absolute or @Ref/Relative should assume the default role would become such a hotly discussed issue :-)
Cheers, Bergey
________________________________________
From: Markus KARG [markus_at_headcrashing.eu]
Sent: 02 August 2011 19:23Who could've thought the question of deciding whether @Ref/Absolute or @Ref/Relative should assume the default role would become that
________________________________________
From: Markus KARG [markus_at_headcrashing.eu]
Sent: 02 August 2011 19:23
To: jsr339-experts_at_jax-rs-spec.java.net
Cc: 'Jan Algermissen'
Subject: [jsr339-experts] Re: Hypermedia - Take 2
> Jan,
>
> The way I understand it, Sergey is just supporting the idea to make the
> absolute references a default option in JAX-RS
> API, so that by default the API can support also (potential) ill-
> behaved clients. Such behavior is IMO a variation to
> Postel's Law. Do you have any particular concerns with this?
>
> Marek
You are proposing that we shall squander valueable resources out of the hypothetic risk that there might be a ill-behaving client? I would understand this if client and server would not have any knowledge of each other as it is typical in the web. But here, things are different. Client and Server actually need to share the same view of a RESTful service interface that is defined exactly by one party -- the application programmer. So if a client is not working with that server, is it in the interest of that programmer that it silently will work when squandering additional bandwith? I wouldn't it be more in the interest of the application programmer that bandwidth is saved and that one client programmer opens a case with the service provider, then learning about his fault?
To: jsr339-experts_at_jax-rs-spec.java.net
Cc: 'Jan Algermissen'
Subject: [jsr339-experts] Re: Hypermedia - Take 2
> Jan,
>
> The way I understand it, Sergey is just supporting the idea to make the
> absolute references a default option in JAX-RS
> API, so that by default the API can support also (potential) ill-
> behaved clients. Such behavior is IMO a variation to
> Postel's Law. Do you have any particular concerns with this?
>
> Marek
You are proposing that we shall squander valueable resources out of the hypothetic risk that there might be a ill-behaving client? I would understand this if client and server would not have any knowledge of each other as it is typical in the web. But here, things are different. Client and Server actually need to share the same view of a RESTful service interface that is defined exactly by one party -- the application programmer. So if a client is not working with that server, is it in the interest of that programmer that it silently will work when squandering additional bandwith? I wouldn't it be more in the interest of the application programmer that bandwidth is saved and that one client programmer opens a case with the service provider, then learning about his fault?