I think Serializer is appropriate because it implements the 
org.apache.pivot.serialization.Serializer interface (it doesn't extend 
java.io.Reader). Serializer doesn't require that both readObject() and 
writeObject() be implemented.

On Jul 1, 2010, at 8:45 AM, aappddeevv wrote:

> I like BXML because it reminds me its XML based. If you want to get rid of
> the two-concept you could call it BxmlLoader or BxmlReader. I think
> BxmlReader is more java like if I remember correctly and implies streaming
> semantics. If someone implements json or groovy reader then the name helps
> distinguish it better as well.
> 
> 
> -----Original Message-----
> From: Greg Brown [mailto:[email protected]] 
> Sent: Thursday, July 01, 2010 8:28 AM
> To: [email protected]
> Subject: BeanSerializer
> 
> Hi all,
> 
> You may have noticed that I renamed BeanSerializer to BXMLSerializer. It
> occurred to me that "BeanSerializer" fairly strongly implies a two-way bean
> serialization capability, which is not what BeanSerializer does.
> "BXMLSerializer" is clearer: it loads BXML files. It is also closer to
> WTKXSerializer, its predecessor - in fact, I found myself accidentally
> referring to it as BXMLSerializer more often than not.
> 
> Please let me know if you have any questions or concerns about this change.
> 
> Thanks,
> G
> 

Reply via email to