This message is from the T13 list server.
On Fri, 13 Dec 2002 09:41:55 -0800, Gary Laatsch wrote: >This message is from the T13 list server. >I have been silently thinking this since we first decided to split into >volumes. Remembering this early SCSI (SASI) < 100 page specification to the >"BACK IT UP....BEEEP BEEEP BEEEP" specification they have today (which every >vendors supports differently) and wondering....are we going there also.... Of course we are going there. The whole world is going there. Just look at the CD and DVD format mess. Just look at all the types of tape. Just look at all the types of removable storage used by cameras. Just look at all the types of cables and adapters we need today. Why does this happen? I take you to the office of the Chief Technical Officer of the XYZ Corp... CTO: But why do we need to develop our own device interface? Why can we use one of the standard device interfaces? IP Lawyer (Legal Dept): Because we don't own the IP used by the standard. We must have our own interface standard backed by our own IP. ... Get it? Customer be damned. Standards be damned. IP ownership is the only thing that matters. It is really unfortunate that this serial interface thing is called Serial ATA because it probably isn't ATA and in the end will look nothing like ATA. It is unfortunate that all those people that don't like T10 and the SCSI standards latched onto a "serial version of ATA" to attempt yet another end run around SCSI. I don't know how successful this SATA thing will be (I don't know anything about it - never read the spec). But based on things I hear it is complex and full of things that are based on various "hidden agendas". It sounds more and more like yet another 1394 - the difference is that 1394 was a great college grad student project back in the 1970s while SATA is yet another "SCSI replacement". I saw in person the extent to which these SATA people will go in their attempt to corrupt ATA into something different and "replace SCSI". This happened at a T13 meeting in Longmont last spring. The SATA people wanted (maybe they went ahead and did this anyway?) a way to use the ATA 48-bit register scheme to pass a SCSI CDB to a SATA device. My reaction at the time was that these people really don't care about ATA or ATAPI, they are living in their own little world... existing standards be damned! Will there be a ATA/ATAPI-x volume 3? Sure there will. But never forget that ANSI INCITS will publish any document as a standard, even documents that represent interfaces that will never be implemented - for example, see the T13 Tailgate standard. The reason why this happens is this: The people companies send to committee meetings can't go back to their company and say "Thanks for sending me to all those committee meetings but the committee just decided that all the 'work' that was done during the last year was a waste of time and there will be no published standard". Any committee activity that goes on for a significant period of time and results in a large document will get forwarded to INCITS to be published even if it represents something that no one is interested in. Again, I don't know if SATA will turn out this way but there is always the chance that SATA will turn out to be yet another 1394. Hale *** Hale Landis *** www.ata-atapi.com ***
