ADQL-2.1 internal draft

[Dave Morris]

Hi Markus,

I think I agree with pretty much all of this :-)

On 2015-06-12 10:04, Markus Demleitner wrote:
> 
> As in -- "there's no way to teach them bitwise operations, whether via
> operators or functions"?  Note that if you'd have to write a few
> lines of C for a DB extension I'd still count ADQL as implementable
> on that platform.  I have a hard time believing there's a reasonable
> platform that wouldn't let you do that.
> 
> ...
> 
> I'm not fully sold on this, but I suspect you could talk me into it.
> Problems I can see include:
> 
> * how much work could be offloaded to DB extensions to still count as
>   "implementable"?  Would we require PoC implementations for them?
> * do we state minimal (maximal?) versions?
> * is there an "efficiently" somewhere in there (e.g., plain CTEs can be
>   simulated through copying subtrees, but unless you have a good
>   query planner, performance may go down the drain)?
> * how do we take things off that list?  Put things on it?  Given that
>   lifetime(standard]>>lifetime(RDBMS version), this can be a bit
>   tricky if we went about things legalistically.
> 
> Still, I think the idea is good, and it would help rationalise
> development.
> 

All valid questions - and I don't have the answers.

All of this is up for discussion at the meeting next week.

I expect we would end up having one or two experts for each platform who 
would do the verification, but it should be up to the person proposing a 
feature to organise the tests and solve any issues.

My own preferences would be to set the bar fairly low for an optional 
feature, so a proof of concept implementation would be fine.

For a mandatory feature, I think the bar should be quite high.

For a large data provider with a dedicated developer team, compiling a C 
program and installing it as an extension is an everyday task.

For a individual astronomer who may not have much IT experience, it 
might be too much to expect and could put them off.

We also need to bear in mind that TAP and ADQL are required to implement 
a registry.

Ideally, it should be easy for an independent researcher with a table in 
a spread sheet to publish their data quickly and easily using one of the 
simple TAP service tools that are available.

If they have to modify their database system just to meet the minimal 
specification, it may prevent them from doing it.

Another common scenario is for a shared MySQL database system to be 
managed by a third party who would probably be unwilling to install a 
custom extension just for one user.

My own preference would be to set the criteria for a mandatory feature 
fairly high, so that the bar to deploying an entry level system for an 
inexperience user is a low as possible.

However, I agree that in general, having as much as possible in the 
mandatory set is a good thing to aim from.

> 
> In that vein, I assume all of you know http://sqlfiddle.com ?

Not seen that before.
Very useful, thank you.

Cheers,
Dave

--------
Dave Morris
Senior Troll
Wide Field Astronomy Unit
Institute for Astronomy
University of Edinburgh
--------