Re: Larger than 32bit signed?

Hi.

The issue "XML-RPC doesn't do what I need, can we add x to it?" has come  
up so many times.  Really, search the archives and you will find literally  
hundreds of lively well thought out discussions about just this.

The owner of the spec (Dave Winer) is loath to change it, because of all  
the interoperating implementations and I think at this point just about  
every library owner shares that view, irrespective of how good they think  
any particular change would be.  This applies not just to the official  
spec itself, but to any spec purporting to apply to "XML-RPC".

I remember one solution that seemed to make sense, and I'm fairly certain  
Dave approved of the approach as well, if he's still lurking on this list  
he could confirm.

The solution was to create a spec for a language and call it "most  
definitely not XML-RPC, call it anything else you want".  In the spec, say  
"this is XML-RPC plus (or minus) feature X".  Make sure the  
implementations you use on both ends support the new protocol and you are  
off and running.  The assertion was that the more widely needed the new  
feature was, the more widely implemented it would be by the owners of the  
various XML-RPC libraries.

I'm fairly certain this has been followed a number of times but I know of  
no case where any new features have been implemented beyond what the  
original requestor needed to have to get his job done.  99% of people  
everywhere are using plain vanilla XML-RPC and loving it.  (as long as  
they don't care about dates, locales, error codes or database  
interoperability :)

HTH

----->Nathan


On Wed, 08 Feb 2006 08:10:24 -0800, Mark Ellzey <[email protected]> wrote:

>>
>>
>> The strength of XML-RPC is the interoperability. The spec has been
>> stable for years now and I see no real chance that it will get
>> changed now.
>>
>
> This is an unfortunate way to see this technology.
>
>> Why doesn't using strings "scale"?
>
> Because the app or user must have previous knowledge of this string or
> datatype to be converted. This seems like one of the issues that
> xml-rpc was trying to fix, I may be wrong in that notion.
>
>>
>> You can use double, of course, which gives you a continuous Integer
>> range of +- 2**52.
>>
>>
>> Just how big are the numbers you want to send?
>
> greater than or equal to 4,294,967,295. I would love to be able to use
> 64 bit numbers.
>
>
> It sounds like the best bet here is to use a structure defining what
> the data-type actually is but use strings. Thanks for the reply.
>
>
> Yahoo! Groups Links
>
>
>
>
>
>



-- 





---
(([^/]+)/([^/]+)){0,1}/*(([^/]+)/([^/]+)){0,1}/*(([^/]+)/([^/]+)){0,1}/*
(([^/]+)/([^/]+)){0,1}/*(([^/]+)/([^/]+)){0,1}/*(([^/]+)/([^/]+)){0,1}/*
---



Nathan Young
N. C. Young Design
(530)629-4176
http://ncyoung.com


 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/xml-rpc/

<*> To unsubscribe from this group, send an email to:
    [email protected]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/