JSON-P vs. JSR-353?

[James Carman]

I just happened to notice the article on JavaWorld about MicroProfile:


http://www.javaworld.com/article/3152057/enterprise-java/eclipse-embraces-java-microservices-initiative.html



It says "The first release of MicroProfile was based on JAX-RS (Java API for RESTful Services) 2.0, CDI (Contexts and Dependency Injection) 1.2 and JSON-P (JSON with Padding) 1.0."  My understanding was that MicroProfile includes/requires JSR-353 (JSONP?) not JSON-P.  

[Antonio Goncalves]

The noun "JSon-P" (P for Processing) was a terrible mistake. It makes sense in the Java EE world (because it tastes like "JAX-P") but outside it, indeed it sounds like JSON with Padding

Antonio

--
You received this message because you are subscribed to the Google Groups "MicroProfile" group.
To unsubscribe from this group and stop receiving emails from it, send an email to microprofile+unsubscribe@googlegroups.com.
To post to this group, send email to microp...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/microprofile/159333ae-9012-43b0-a802-662e3d2bb74c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Antonio Goncalves
Software architect, Java Champion and Pluralsight author

Web site | Twitter | LinkedIn | Pluralsight | Paris JUG | Devoxx France

[Rahman USTA]

I think so, Antonio. It is not also a nicer choice for SEO. When we search JSON-P then we get Json-Padding related content mostly.

To view this discussion on the web visit https://groups.google.com/d/msgid/microprofile/CA%2BZZq98-37n86vHSXCzuHgGzO4tWV7aB2jYT_9pqC7HW6018Aw%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

--

Rahman USTA

Istanbul JUG

https://github.com/rahmanusta

[James Carman]

Perhaps we should just refer to it by JSR-353 from now on or something, since the acronym is extremely unfortunate.

On Wednesday, January 4, 2017 at 10:36:27 AM UTC-5, Antonio Goncalves wrote:

The noun "JSon-P" (P for Processing) was a terrible mistake. It makes sense in the Java EE world (because it tastes like "JAX-P") but outside it, indeed it sounds like JSON with Padding

Antonio

On Wed, Jan 4, 2017 at 4:30 PM, James Carman <jwca...@gmail.com> wrote:

I just happened to notice the article on JavaWorld about MicroProfile:


http://www.javaworld.com/article/3152057/enterprise-java/eclipse-embraces-java-microservices-initiative.html



It says "The first release of MicroProfile was based on JAX-RS (Java API for RESTful Services) 2.0, CDI (Contexts and Dependency Injection) 1.2 and JSON-P (JSON with Padding) 1.0."  My understanding was that MicroProfile includes/requires JSR-353 (JSONP?) not JSON-P.  

--
You received this message because you are subscribed to the Google Groups "MicroProfile" group.

To unsubscribe from this group and stop receiving emails from it, send an email to microprofile...@googlegroups.com.

To post to this group, send email to microp...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/microprofile/159333ae-9012-43b0-a802-662e3d2bb74c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Werner Keil]

I'm afraid that makes no sense, because JSR numbers come and go. JSR-353 is final, the version I referred to the new JSON Processing JSR 374, not 353 in the article.

JSON-P and JSON-B both also in official GitHub projects or with domains like json-b.net now were inspired by JAX-P vs. JAX-B.

One cannot avoid ambiguities.

If people look for "Microprofile" they get some matches pointing to this effort, but among the top hits is 

https://github.com/zeux/microprofile

microprofile - embeddable CPU/GPU profiler

And the term "Micro" in the Java ecosystem has long been a synonym for "Mobile" or "Embedded". JSR-361 is called "Java^TM ME Embedded Profile". Or "Java Micro Edition Embedded Profile". Aka "Microprofile" is just as easy to confuse Mobile and Embedded with efforts like this that are more on the EE side.

When terms like "Embedded" or "Micro" were brought up in the Java EE JSRs, Antonio and others will know quite well I was among those warning of misunderstandings with other technologies like Java ME.

It seems more marketing-affine people picked the name because "Microservices" are everybody's darling. Yet there are other projects especially on GitHub with the same name, so not a better situation than JSON-P or JSON-B.

Werner

[Ondrej Mihályi]

I agree that the name JSON-P has not been the best choice, but it's already there in the name of the spec: https://json-processing-spec.java.net/

If we want to avoid confusion, I think the only option is to refer to it as JSON-Processing. The search for it in google brings the JSR project page as the first hit, which is optimal and much better compared to searching for "JSON-P".
The JSR numbers change with versions, as Werner pointed out.

--Ondrej

Dňa streda, 4. januára 2017 20:34:47 UTC+1 Werner Keil napísal(-a):

[Werner Keil]

At least the Twitter account I reserved for the EG/Spec Lead of JSR 374 is called JSONProcessing. Twitter has no "-", 

Nobody knows what will happen with java.net. And Dmitry moved ahead with JSON-B by creating a GitHub organization by that name. I simply replicated it for JSON-Processing and also used the short form json-p. I doubt, he or Oracle will change that, but it is fine to refer to "JSON-Processing" whenever possible.

It's a bit like saying JavaEE instead of JEE or J2EE ;-)

Werner

[Werner Keil]

Hi,

JSON Processing 1.1 (JSR 374) just passed its Public Review Ballot yesterday:
https://jcp.org/en/jsr/results?id=5917

Interestingly, Tomitribe although eligable did not vote.
Was there a special reason for that? Given this is the update to a key ingredient of MicroProfile which is likely to join a future MicroProfile release as soon as it's Final, curious if Tomitribe (David/Amelia) were just too busy, travelling somewhere or had a technial reason not to vote?

Also having missed the JSF ballot that closed on the same day, it was probably just a scheduling conflict or some issue with the EC account, please try to sort it out as another vital ingredient, CDI 2 just started its PR today;-)

Cheers,
Werner

[Mark Struberg]

Funny that. We are currently still applying changes to JSONP. Just check the mailing lists.

There are still a few open tasks which are on the TODO list and agreed to be added before 1.1-PFD. 

Probably that's the reason why Tomitribe didn't vote YET.

[Werner Keil]

If they had problems they should discuss it with the Spec Lead. He and other EG members of both JSON-B and JSON-P are very responsive.

All EC members are supposed to vote (Tomitribe is relatively new to the EC so maybe still learning)
If a member finds a problem they are always welcome to put a comment there. E.g. I abstained from a couple of votes if the Spec Lead did not come back to me/us in time, but made clear in the comment why.

As Tomitribe also missed the JSF 2.3 ballot and there Ed said last night he already pushed the PFD to MavenCentral (thus it seems no open issues or change requests there) there is little reason to believe it has to do with JSON-P 1.1 in particular.

Btw. if anybody comes to JavaLand this year (Tomitribe has at least Andy talking about MicroProfile) please come to the Community ground, at least for JSON-P I'm happy to offer live hacking or "Hackergarten" (whether it's called that or different no matter) beside my JSON-P talk in the community track. So we can look at any issues that still exist and have not been solved before then.

[Werner Keil]

There is even one Ed Burns filed a while ago, but they all seem to be on a "wishlist" for post 1.0, I would consider that a backlog:
https://java.net/jira/browse/JSON_PROCESSING_SPEC/fixforversion/16272/?selectedTab=com.atlassian.jira.jira-projects-plugin:version-summary-panel

Ed is in the Java EE 8 panel at JavaLand, so I'll ask him what he thinks, but some of those if still relevant could become "post 1.1" IMO.

[Werner Keil]

There are only 3 in that group, all others were resolved.