OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

sdd message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [sdd] Alternate definition of artifact




Julia McCarthy wrote:
> It sounds like you two (Debra and James) define payload as the bits in 
> the package that end up copied as the deployed resources. That is what I 
> was calling the package content that is closely similar to the deployed 
> resources. (A copied bit is not the same bit.) Not all deployed 
> resources are copied from bits in the solution package. Some deployed 
> bits are calculated during deployement. Would you consider the inputs to 
> those calculations to be payload?

Can you give a concrete example here?  I think that the inputs used to
calculate or create dynamic deployed bits are really configuration
parameters, and not part of the payload itself.  The only things that
I consider payloads are things that exist in the solution package that
are copied to the hosting environment and are not modified by the install
or deployment runtime.  Anything that exists in the solution
package that is modified or changed before being copied,
based on configuration or other inputs, I would consider an artifact.


> I believe that the bits that end up copied to the deployed solution are 
> part of the artifact as well as bits that are used to calculate contents 
> or provide other instructions for deploying content. 

OK, I think I may see your point.   A solution package exists on machine A.
The end user wants to deploy it to machine B.  So, something is first copied
from A to B.  This is the artifact (for example, an FTP transfer of 
setup.exe).  Then the actual copying from the artifact
blob to their final destination on machine B occurs.  I'm fine with calling
that intermediate blob an "artifact".  It is short-lived (i.e. the artifact
is deleted from machine B's temporary directory once the payload bits are
copied out of it and into their final resting place, and all other dynamic
content based on things in the blob are created, such as config files).
Is that right?

> I still don't see 
> the value in distinguishing between bits that are directly copied vs. 
> bits that are used to calculate or provide instructions. 

Well, now I might disagree.  Bits that are directly copied are the payload.
Bits used to calculate dynamic values to me sound like configuration values
(aka artifacts) which are used to lay down dynamic data
(in config files, or whatever).

> Even if there 
> is value in distinguishing (I imagine I'm alone in not seeing the 
> value), I think that both concepts are part of artifact. i.e. What you 
> call payload can be an artifact... or payload plus metadata... or 
> payload alone.

Will wait for your reply on my above A vs. B example before responding.

-jhf-


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]