I can't find it now, and I may have misread it, but I was responding to what I thought was a suggestion to use SPEC 1-* C2X 1.
Melinda's approach has the advantages of being well documented, extensively tested, and already performance tuned. She used a uuencode technique. See her paper entitled "Streamlining Your Pipelines." -----Original Message----- From: CMSTSO Pipelines Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Rob van der Heij Sent: Wednesday, August 01, 2007 6:28 PM To: [email protected] Subject: Re: [CMS-PIPELINES] Checksums? CRC? (and hex display?) On 8/1/07, Paul Gilmartin <[EMAIL PROTECTED]> wrote: > > I believe that Melinda discusses a method for transmitting data thru > > hostile gateways in her paper on Rita. > > > Not enough information to find it. Is it significantly better than a > PACK stage? The "hostile gateways" in that context are those that translated EBCDIC to ASCII and back (depends on what EBCDIC and what ASCII, but rarely gets you identical result) which she solved by using base-64 as far as I recall. It's helpful to carry your 8-bit bytes through a narrow connection. But the spreading is not optimal in such a way that it would work as error correcting code. I think the "digest" stage is the way to go. And because it's common standards you can actually compute the digest on both platforms and compare. Rob
