Difference: DiVergence (1 vs. 7)

Revision 72012-06-26 - root

 
META TOPICPARENT name="IvoaTCG"

How to stop divergence of Parameter-TAP and ADQL-TAP?

Comments from ChristopheArviset

Changed:
<
<
I would like to propose that after the discussions about this topic at Trieste interop and the conclusion from Keith at the closing plenary http://www.ivoa.net/internal/IVOA/InterOpMay2008DAL/TAPconclusion.pdf
>
>
I would like to propose that after the discussions about this topic at Trieste interop and the conclusion from Keith at the closing plenary http://wiki.ivoa.net/internal/IVOA/InterOpMay2008DAL/TAPconclusion.pdf
 we don't need to put that topic in the 2008 Roadmap

Would people agree ?



There are two versions of TAP under development, one based on parameters, one based on the ADQL language, and the danger is for these to diverge into two overlapping, but incompatible, mechanisms. The diagram on the DAL blog shows an eventual convergence, but I would like to know how it will work.

I would state this more strongly: neither the Param or QL documents are so far along that they cannot be joined now. The Param document actually includes the ADQL queries. There are minor differences in the interface specifications, but these could be resolved now quite easily, I would think. If we wait until later, when groups have done implementations, there will be a lot of resistance to making changes. (Bob)

cosponsor BobHanisch

Changed:
<
<
I thought that it had been decided a year ago at Beijing to focus on the QL version of TAP - see page 5 of the Plenary summary http://www.ivoa.net/internal/IVOA/InterOpMay2007Plenary/dal-plenary2.pdf
>
>
I thought that it had been decided a year ago at Beijing to focus on the QL version of TAP - see page 5 of the Plenary summary http://wiki.ivoa.net/internal/IVOA/InterOpMay2007Plenary/dal-plenary2.pdf
 for documentary proof - Quoting

"Remove explicit data model support (POS, SIZE, etc.)標ill use ADQL plus UTYPE plus REGION instead (etc.) "

My "solution" is in http://www.ivoa.net/forum/dal/0805/0866.htm - in summary Parameter-TAP is not TAP, but Simple Cone Search V2.

So I third this motion... PaulHarrison

I agree with you all. It seems to me that people do NOT take into account priorities in development. The primary purpose of the IVOA would be to realize interoperability among the VO projects in the world, and we need "non-paper" standards as early as possible. ... MasatoshiOhishi


Revision 62008-06-25 - ChristopheArviset

 
META TOPICPARENT name="IvoaTCG"

How to stop divergence of Parameter-TAP and ADQL-TAP?

Added:
>
>
Comments from ChristopheArviset
I would like to propose that after the discussions about this topic at Trieste interop and the conclusion from Keith at the closing plenary http://www.ivoa.net/internal/IVOA/InterOpMay2008DAL/TAPconclusion.pdf we don't need to put that topic in the 2008 Roadmap

Would people agree ?



 There are two versions of TAP under development, one based on parameters, one based on the ADQL language, and the danger is for these to diverge into two overlapping, but incompatible, mechanisms. The diagram on the DAL blog shows an eventual convergence, but I would like to know how it will work.

I would state this more strongly: neither the Param or QL documents are so far along that they cannot be joined now. The Param document actually includes the ADQL queries. There are minor differences in the interface specifications, but these could be resolved now quite easily, I would think. If we wait until later, when groups have done implementations, there will be a lot of resistance to making changes. (Bob)

cosponsor BobHanisch

I thought that it had been decided a year ago at Beijing to focus on the QL version of TAP - see page 5 of the Plenary summary http://www.ivoa.net/internal/IVOA/InterOpMay2007Plenary/dal-plenary2.pdf for documentary proof - Quoting

"Remove explicit data model support (POS, SIZE, etc.)標ill use ADQL plus UTYPE plus REGION instead (etc.) "

My "solution" is in http://www.ivoa.net/forum/dal/0805/0866.htm - in summary Parameter-TAP is not TAP, but Simple Cone Search V2.

So I third this motion... PaulHarrison

I agree with you all. It seems to me that people do NOT take into account priorities in development. The primary purpose of the IVOA would be to realize interoperability among the VO projects in the world, and we need "non-paper" standards as early as possible. ... MasatoshiOhishi


<--  
-->

Revision 52008-05-16 - PaulHarrison

 
META TOPICPARENT name="IvoaTCG"

How to stop divergence of Parameter-TAP and ADQL-TAP?

There are two versions of TAP under development, one based on parameters, one based on the ADQL language, and the danger is for these to diverge into two overlapping, but incompatible, mechanisms. The diagram on the DAL blog shows an eventual convergence, but I would like to know how it will work.

I would state this more strongly: neither the Param or QL documents are so far along that they cannot be joined now. The Param document actually includes the ADQL queries. There are minor differences in the interface specifications, but these could be resolved now quite easily, I would think. If we wait until later, when groups have done implementations, there will be a lot of resistance to making changes. (Bob)

cosponsor BobHanisch

I thought that it had been decided a year ago at Beijing to focus on the QL version of TAP - see page 5 of the Plenary summary http://www.ivoa.net/internal/IVOA/InterOpMay2007Plenary/dal-plenary2.pdf for documentary proof - Quoting

"Remove explicit data model support (POS, SIZE, etc.)標ill use ADQL plus UTYPE plus REGION instead (etc.) "

Changed:
<
<
>
>
My "solution" is in http://www.ivoa.net/forum/dal/0805/0866.htm - in summary Parameter-TAP is not TAP, but Simple Cone Search V2.
  So I third this motion... PaulHarrison

I agree with you all. It seems to me that people do NOT take into account priorities in development. The primary purpose of the IVOA would be to realize interoperability among the VO projects in the world, and we need "non-paper" standards as early as possible. ... MasatoshiOhishi


<--  
-->

Revision 42008-05-16 - MasatoshiOhishi

 
META TOPICPARENT name="IvoaTCG"

How to stop divergence of Parameter-TAP and ADQL-TAP?

There are two versions of TAP under development, one based on parameters, one based on the ADQL language, and the danger is for these to diverge into two overlapping, but incompatible, mechanisms. The diagram on the DAL blog shows an eventual convergence, but I would like to know how it will work.

I would state this more strongly: neither the Param or QL documents are so far along that they cannot be joined now. The Param document actually includes the ADQL queries. There are minor differences in the interface specifications, but these could be resolved now quite easily, I would think. If we wait until later, when groups have done implementations, there will be a lot of resistance to making changes. (Bob)

cosponsor BobHanisch

I thought that it had been decided a year ago at Beijing to focus on the QL version of TAP - see page 5 of the Plenary summary http://www.ivoa.net/internal/IVOA/InterOpMay2007Plenary/dal-plenary2.pdf for documentary proof - Quoting

"Remove explicit data model support (POS, SIZE, etc.)標ill use ADQL plus UTYPE plus REGION instead (etc.) "

So I third this motion... PaulHarrison

Added:
>
>
I agree with you all. It seems to me that people do NOT take into account priorities in development. The primary purpose of the IVOA would be to realize interoperability among the VO projects in the world, and we need "non-paper" standards as early as possible. ... MasatoshiOhishi
 


<--  
-->

Revision 32008-05-15 - PaulHarrison

 
META TOPICPARENT name="IvoaTCG"

How to stop divergence of Parameter-TAP and ADQL-TAP?

There are two versions of TAP under development, one based on parameters, one based on the ADQL language, and the danger is for these to diverge into two overlapping, but incompatible, mechanisms. The diagram on the DAL blog shows an eventual convergence, but I would like to know how it will work.

I would state this more strongly: neither the Param or QL documents are so far along that they cannot be joined now. The Param document actually includes the ADQL queries. There are minor differences in the interface specifications, but these could be resolved now quite easily, I would think. If we wait until later, when groups have done implementations, there will be a lot of resistance to making changes. (Bob)

cosponsor BobHanisch

Added:
>
>
I thought that it had been decided a year ago at Beijing to focus on the QL version of TAP - see page 5 of the Plenary summary http://www.ivoa.net/internal/IVOA/InterOpMay2007Plenary/dal-plenary2.pdf for documentary proof - Quoting
 
Added:
>
>
"Remove explicit data model support (POS, SIZE, etc.)標ill use ADQL plus UTYPE plus REGION instead (etc.) "

So I third this motion... PaulHarrison

 


<--  
-->

Revision 22008-05-15 - BobHanisch

 
META TOPICPARENT name="IvoaTCG"

How to stop divergence of Parameter-TAP and ADQL-TAP?

There are two versions of TAP under development, one based on parameters, one based on the ADQL language, and the danger is for these to diverge into two overlapping, but incompatible, mechanisms. The diagram on the DAL blog shows an eventual convergence, but I would like to know how it will work.
Added:
>
>
I would state this more strongly: neither the Param or QL documents are so far along that they cannot be joined now. The Param document actually includes the ADQL queries. There are minor differences in the interface specifications, but these could be resolved now quite easily, I would think. If we wait until later, when groups have done implementations, there will be a lot of resistance to making changes. (Bob)
 
Added:
>
>
cosponsor BobHanisch
 


<--  
-->

Revision 12008-05-14 - RoyWilliams

 
META TOPICPARENT name="IvoaTCG"

How to stop divergence of Parameter-TAP and ADQL-TAP?

There are two versions of TAP under development, one based on parameters, one based on the ADQL language, and the danger is for these to diverge into two overlapping, but incompatible, mechanisms. The diagram on the DAL blog shows an eventual convergence, but I would like to know how it will work.


<--  
-->
 
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback