Results of X3T10 Letter Ballots (95-014 -- 95-024)

Lohmeyer, John JLOHMEYE at cosmpdaero.ftcollinsco.ncr.com
Sat Jun 17 21:50:00 PDT 1995


[[ 95-034R0.TXT : 5470 in 95-034R0.TXT ]]

 --
John Lohmeyer             E-Mail:  john.lohmeyer at symbios.com
Symbios Logic Inc.         Voice:  719-573-3362
1635 Aeroplaza Dr.           Fax:  719-573-3037
Colo Spgs, CO 80916     SCSI BBS:  719-574-0424 300--14400 baud

                                                                  X3T10/95-034
Accredited Standards Committee*
X3, Information Technology
                                                   Doc. No.: X3T10/95-034
                                                       Date: June 17, 1995
                                                    Project:
                                                  Ref. Doc.: 95-014 -- 95-024
                                                   Reply to: Mr. John Lohmeyer
                                                            Symbios Logic, Inc.
                                                            1635 Aeroplaza Dr.
                                                            Colo Spgs, CO 80916
                                                            (719) 573-3362

To:         X3T10 Membership
From:       John Lohmeyer, X3T10 Chair
Subject:    Results of X3T10 Letter Ballots 95-014 through 95-024

The results of the eleven letter ballots were (Yes:No:Abstain:No Response):

37:12:4:6    95-014  Recommendation to form new ATA TC
48:4:0:7     95-015  ATA+PI Project Proposal
52:0:0:7     95-016  SPI-2 Project Proposal
52:0:0:7     95-017  ESPC Project Proposal
52:0:0:7     95-018  FCP-2 Project Proposal
51:0:0:8     95-019  SSA-PH1 Project Proposal
52:0:0:7     95-020  SSA-PH2 Project Proposal
51:0:0:8     95-021  SSA-TL1 Project Proposal
52:0:0:7     95-022  SSA-TL2 Project Proposal
50:1:0:8     95-023  SSA-S2P Project Proposal
52:0:0:7     95-024  SSA-S3P Project Proposal

All ballots passed.  The first (95-014) only requires a simple majority.  I
will forward the voting results to X3/OMC.  The remaining ballots all received
at least one comment.  Comment resolution for ballots 95-015 through 95-018
will be on the X3T10 agenda for the July meeting.  Comment resolution for
ballots 95-019 through 95-024 will be referred to X3T10.1 for their June
meeting.

The following table gives the voting details:

Organization           Voting Member         14 15 16 17 18  19 20 21 22 23 24 Comments
3M Company             Mr. Alan R. Olson     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Adaptec, Inc.          Mr. Norm Harris             Y  Y   Y                    
Adaptec, Inc.          Mr. Lawrence Lamers   Y  Y            Y  Y  Y  Y  Y  Y  
                       (A)
Advanced Micro Devices Mr. Ron Apt           Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Amdahl Corp.           Mr. Edward Fong       Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
AMP, Inc.              Mr. Charles Brill     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Amphenol               Mr. Michael Wingard   Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Ancot Corp.            Mr. Jan V. Dedek      Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Apple Computer         Mr. Ron Roberts (A)   Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
AT&T                   Mr. Joe Lawlor        -  -  -  -   -  -  -  -  -  -  -  
Burr-Brown Corp.       Mr. Dennis R. Haynes  Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  All IV
BusLogic               Mr. Clifford E.       -  -  -  -   -  -  -  -  -  -  -  
                       Strang Jr.
Ciprico Inc.           Mr. Gerry Johnsen     N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Circuit Assembly Corp. Mr. Ian Morrell       -  -  -  -   -  -  -  -  -  -  -  
Cirrus Logic Inc.      Mr. Joe Chen          Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
CMD Technology         Mr. Edward Haske      Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Congruent Software,    Mr. Peter Johansson   N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Inc.
Conner Peripherals     Mr. Michael Bryan     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Dallas Semiconductor   Mr. Louis Grantham    Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Digital Equipment      Mr. Charles Monia     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  IV on 22 only
Corp.
Eastman Kodak Co.      Mr. Robert Reiseh     N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
ENDL                   Mr. I. Dal Allan      Y  N  Y  YC  Y  Y  Y  Y  Y  Y  Y  
Exabyte Corp.          Mr. Edward Lappin     N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  IV on all
FSI Consulting         Mr. Gary R. Stephens  Y  n  Y  Y   Y  n  Y  n  Y  n  Y  no cmnts rcvd
Services
Fujitsu                Mr. Robert Liu        Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Future Domain Corp.    Mr. Kevin J. Calvert  Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
                       (A)
Hewlett Packard Co.    Mr. Stephen Holmstead N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Hitachi                Mr. S. Nadershahi     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Honda Connectors       Mr. David McFadden    Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
IBM Corp.              Mr. George Penokie    Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
IIX Consulting         Mr. Duncan Penman     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Interphase Corp.       Mr. David Lawson      -  -  -  -   -  -  -  -  -  -  -  
Iomega Corp.           Mr. Geoffrey Barton   -  -  -  -   -  -  -  -  -  -  -  
Linfinity Micro        Mr. Dean Wallace      N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Madison Cable Corp.    Mr. Robert Bellino    Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Maxoptix Corp.         Ms. Donna Pope        -  -  -  -   -  -  -  -  -  -  -  
Maxtor Corp.           Mr. Pete McLean       Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Methode Electronics,   Mr. Bob Masterson     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Inc.
Molex Inc.             Mr. Joe Dambach       A  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
NEC                    Mr. Chris D'Iorio     Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Oak Technology, Inc.   Mr. Peter Brown       Y  Y  n  n   n  n  n  n  n  n  n  no cmnts rcvd
Panasonic              Mr. Stephen F. Heil   N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
QLogic Corp.           Mr. Skip Jones        N  N  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Quantum Corp.          Mr. James McGrath     A  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Seagate Technology     Mr. Gene Milligan     N  N  YC YC YC  YC YC YC YC N  YC 16,21,22,23,2
                                                                               4IV
Sequoia                Mr. Thomas 'Rick'     N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
                       Tewell
Silicon Systems, Inc.  Mr. Stephen G. Finch  Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Sony                   Mr. Scott Smyers      Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Storage Technology     Mr. Erich Oetting     A  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Corp.
Sun Microsystems, Inc. Mr. Robert N. Snively N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Symbios Logic, Inc.    Mr. John Lohmeyer     A  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  14,15IV
SyQuest Technology     Mr. Patrick Mercer    N  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Tandem Computers       Mr. John Moy          Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Thomas & Betts         Mr. Harvey            Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
                       Waltersdorf
Trimm Technologies     Mr. Gary M. Watson    Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
UNISYS Corporation     Mr. David Hudson      Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Unitrode               Mr. Paul D. Aloisi    Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Western Digital        Mr. Jeff Stai               Y  Y   Y  Y  Y  Y  Y  Y  Y  
Corporation
Western Digital        Mr. Thomas Hanan (A)  Y  N                              
Corporation
Woven Electronics      Mr. Doug Piper        Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  
Zadian Technologies    Mr. Dennis P. Moore   Y  Y  Y  Y   Y  Y  Y  Y  Y  Y  Y  

Key:
Y     Yes vote
YC    Yes vote with comments
N     No vote with comments
n     No vote without comments (treated the same as No Response)
A     Abstain
-     No Response (did not return ballot)
IV    Marked `Individual Vote' for specified ballots

Comments received for ballot X3T10/95-014 (Recommendation to form new ATA TC):

Ciprico Inc. (Gerry Johnsen):

Ciprico votes no with the following comments:

1)  Both SCSI and ATA are primarily specifications for the attachment of
storage peripherals to small computer systems and therefore have many areas of
overlap both in terms of technical specifications and the companies involved.
Splitting the interfaces into multiple subgroups in and of itself will not
improve either.

2)  Creating multiple standards bodies with different meeting sites and dates
will make participation by smaller companies more difficult and expensive.

Congruent Software, Inc. (Peter Johansson):

This memorandum accompanies the NO vote I submitted in response to X3 Subgroup
Letter Ballot X3T10/95-014r0.  The reasons for Congruent Software's opposition
to the formation of a new Technical Committee under X3 to be assigned the ATA
projects 791-M, 948-D, 1120-D and 2008-D are given below.
At this point in the development of SCSI and ATA, the common command set
utilized represents an immense investment in systems software.  Any move that
tends to lessen close coordination and cooperation between SCSI and ATA
standards development efforts would inevitably tend to lessen the
interoperability of this shared command set base.  Congruent Software
considers any such move regressive and not in the best interests of the
industry as a whole.

A second, important ground for objection is procedural.  At the Harrisburg
plenary it was clear that a significant number of the advocates of a split
between ATA and SCSI wish to concentrate the meetings in California and hold
plenary sessions more often than bimonthly.  This appears to be a
contravention of the Rules and Procedures of X3, X3/SD-2 sections 4.3.1 and
4.3.2.  The reason for the procedures, as I understand them, is to avoid
placing undue burdens of travel expense and commitment of personnel time upon
participants.  Although Congruent Software itself would be advantaged by
meetings concentrated in California, others would be penalized.  With respect
to increased frequency of meetings and the need to cover both ATA and SCSI
meetings at different locations and different times, any participating company
(not just small companies such as ourselves) would have to bear a great
increase in cost and time, to little benefit.

Eastman Kodak Co. (Robert Reiseh):

The ATA projects are involved in resolving standards issues for computer
storage peripheral devices that have the same fundamental issues found in SCSI
devices.  Therefore the ATA group should remain a part of X3T10.

Exabyte Corp. (Edward Lappin):

I am voting NO on the recommendation that X3 establish a new Technical
Committee to be assigned the ATA projects.  A working group vote of 9-4, while
better than 2 to 1, is not a strong consensus.  Without that strong desire to
form a new TC, even from all that were present for the particular vote, I do
not see a very good chance for success.  Therefore, I feel I must vote no at
this time.  However, a stronger positive consensus could alleviate my
reservations about recommending the formation of a new TC.

Hewlett Packard (Stephen Holmstead):

Splitting the groups means twice as much overhead, including twice as many
plenaries.  There are many areas of common interest (e.g., MMC) which would
also incur a larger burden in trying to keep the two groups in sync.
The proposed reason of encouraging more participation is vague; splitting the
groups will discourage some participation that now exists.

Linfinity Micro (Dean Wallace):

We voted no because we would like the ATA working group to be held the same
week the X3T10 meetings are held.  This allows us to attend the ATA meetings.
If the groups are separated we won't attend the ATA technical committee.

Molex Inc. (Joe Dambach):

No interest.

Panasonic (Stephen Heil):

In resolving the NO vote from Panasonic Technologies, Inc. regarding Letter
Ballot X3T10/95-014r0, please consider the following comments as reasons for
the NO vote. Panasonic Technologies, Inc., is opposed to the formation of a
new Technical Committee under X3 to be assigned the ATA projects 791-M, 948-D,
1120-D and 2008-D) for two primary reasons.

1)   Separating the work of ATA and SCSI (both presently assigned to X3T10)
will not be conducive to industry convergence. Both ATA and SCSI are
considered Low-level interface for computer peripheral devices and are
therefore appropriately assigned to X3T10 given its charter of Low-level
interfaces.

     a)   Recently, there has been a trend to bring some commonality between
ATA and SCSI. This is very apparent in the more advanced ATA projects such as
ATA-2, ATA-3, and ATAPI which all "borrow" significantly from SCSI. This
commonality is very important at many areas in the industry especially in the
area of operating systems and device drivers.  ATA efforts have adopted the
CDB structure of SCSI making it easier for developers and system integrators
to make implementations which share development investments. This is a benefit
to both manufacturers and users.

     b)   X3T10 has formed a Systems Issues study group which has been
investing significant effort toward addressing ATA and SCSI convergence issues
such as the SCSI command structure in ATAPI and is now investigating command
overlap for ATA. If ATA were moved to a separate TC we would loose a important
opportunity to have some level(s) of compatibility between SCSI and ATA and
instead we would be encouraging the inefficient practice of duplication of
effort.

2)   Segregating ATA to a separate TC will significantly hinder the
opportunity for Panasonic Technologies, Inc. and other companies to
participate in the process of developing these standards.

     a)   At the May meeting of X3T10 in Harrisburg, PA, it was very apparent
that a primary reason to form a new TC for ATA was to limit the meeting
locations to the California area. This places member companies not located in
that geographic area at a significant disadvantage because of the increased
travel cost and time investment necessary to participate.

     b)   It was also apparent at the Harrisburg meeting of X3T10 that there
was an intent to increase the frequency pf the ATA meetings. This too places
member companies not located in that geographic area at a significant
disadvantage.

     c)   Limiting meeting location and having meetings more frequently that
six times per year appears to be direct violations of the X3/SD-2,
Organization, Rules and Procedures of X3, sections 4.3.2 and 4.3.1
respectively.

QLogic Corp. (Skip Jones):

I voted no these two ballots because I believe the commonalties between ATA,
ATAPI, and SCSI do not appropriately facilitate segregating their committees
|from X3T10.

Included amongst these commonalties are functionality and logistics.

ATAPI borrows SCSI commands, SCSI multi-thread models, and attempts to operate
over the same host drivers designed for SCSI.  Separate committees only
ensures further alienation between the protocols and their implementations,
thereby increasing market fragmentation unnecessarily.

Additionally, the companies and their representatives which have a vested
interest in one of these protocols (that is, ATA and SCSI) typically have a
vested interest with the other protocol.  Separate meeting locations at
separate times typically serve only the one or two companies and/or
individuals that attempt to manipulate and steer the standards process for
their own self-serving gratification.  Rarely does segregation facilitate the
standards process over a fair cross-section of industry participants.

Quantum Corp. (Jim McGrath):

I will send you email [none was received] but briefly the creation of a new TC
does not appear to offer benefits comparible to its cost (more meetings,
possible divergence from X3T1 activities).

Seagate Technology, Inc. (Gene Milligan):

The following comments accompany Seagate's NO vote on the recommendation to
form a new Technical Committee for ATA projects:

1) An important function standards bodies should provide is intelligent
management of alternatives to bring order to the market place served by the
standards group. This function is weakened by creation of too many Technical
Committees. The fact of this as an issue is attested to by the reactions of
the X3T10 membership to the formation of IEEE P1285.

2) The ATA and SCSI projects have benefited from having a wide variety of
company participants. It is difficult to provide consistent participation in
standards meetings more often than once per month. For many of the
participants vitally concerned with ATA projects, the existence of X3T10 and
X3T11 already require one plenary per month. Creation of an additional
peripheral interface Technical Committee will result in a 50% overload factor.

3) In general the peripheral interface Technical Committees have suffered from
an imbalance between the component, peripheral device, and host system
manufacturers and certainly with end users. Creation of a third Technical
Committee to define peripheral interfaces undoubtedly will aggravate this
imbalance.

4) For the smaller systems (as opposed to mainframes and very largesubsystems)
X3T10 has established an operating environment that has served the market
well. The dynamics of X3T10 has avoided conflict between standards
alternatives such as SCSI, ATA, and ATAPI.

5) The more vocal advocates of creating a new Technical Committee have alleged
that the creation of the new Technical Committee will free the ATA projects of
the cumbersome X3T10 procedures. Their suggestion must result from one of the
following cases:

 a) The advocates have not yet participated enough in the X3T10 process to
understand that they are not shackled, but are participating in an open
standards process which operates in accordance with X3 procedures which have
been accredited by ANSI.

 b) The advocates have an expectation that they will be given cart blanche as
a new Technical Committee to operate as they did as a closed specification
development association.

Sequoia (Thomas `Rick' Tewell):

The direction of evolving ATA into a separate entity from the family known as
"SCSI" is misguided. In a period when companies seek to combine technologies
rather than separate them, it does not seem reasonable for ATA to "separate".
This seems to be much more a "political" effort rather than a technical one
and we are opposed to it.

Storage Technology Corp. (Erich Oetting):

Storage Technology Corporation has abstained from voting to establish a new
Technical Committee for ATA projects.

We feel the decision on forming a new Technical Committee should be left to
those organizations with more direct involvement in ATA technology.

Sun Microsystems, Inc. (Robert N. Snively):

I find that I am forced to vote against the approval of recommending that X3
establish a new Technical Committee to be assigned the ATA projects (791-M,
948-D, 11200-D, 2008-D).

1)  I believe that the convergence of the command sets for ATAPI and SCSI,
especially in the area of CD-ROM commands, is essential to provide a wide and
common software support base for CD-ROM devices.  I believe that this
convergence will be more difficult to achieve if ATAPI and SCSI do not
exchange their ideas in a close committee relationship.  In particular, it is
very important that the new ATAPI concepts be brought into the SCSI CD-ROM
command set.

2)  There are many examples of successful and timely standard developments
having quite different technologies and architectural bases within a single
committee.  Such examples include HIPPI, IPI, ESCON, and Fibre Channel within
X3T11 (formerly X3T9.2).  I believe that this model is an appropriate model
for the ATA and SCSI projects within X3T10.  I had believed that this model
was essentially what had been happening within X3T10 and am surprised that
this question has even been raised.  If the ATA participants feel that the
working environment within X3T10 is somehow inadequate, it is in their
interest and X3T10's interest to improve the relationship, rather than destroy
the relationship.

3)  Working with two committees will force my company to spend more resources
to be sure that our interests (including convergence of the command sets and
functional capabilities of CD-ROMs) are properly met.  Most of the key
companies participating in the ATA activities are also participating in the
SCSI activities and vice versa.  For that reason, I believe that most X3T10
member companies will have the same problem.  I also believe that the
management of those companies would have a strong interest in coordinating and
advancing both the ATA and SCSI activities together, since they address
complementary marketplaces.

Symbios Logic (John Lohmeyer):

I have abstained on this ballot.  OMC and X3 will have to deal with this issue
and either outcome of splitting X3T10 or leaving it together will result in
problems.  There are factions of the ATA group that believe separation will
solve all their problems.  I think this opinion is naive, but they may need to
experience this for themselves.  I think separation will mostly result in
pushing conflict resolution to higher levels (e.g., Public Review and X3).

SyQuest Technology (Patrick Mercer):

I have voted NO on X3 Subgroup Letter Ballot X3T10/95-014r0 to establish a new
Technical Committee to be assigned the ATA projects.

I feel that the SCSI and ATA projects should be brought closer together,
rather than separated.  I am in favor of the SCSI and the ATA groups
maintaining a close communication and being well informed of each others
activites, which is enhanced by having both groups under X3T10 and would be
seriously jeopardized by splitting them up.  My personal feeling is that there
is already a subtle hostility between the SCSI and ATA groups, which I believe
is detrimental, which would be made worse by this proposal.

I am also concerned that the overall number of meetings will become excessive
and those wishing to atttend both project meetings would be forced into
choosing one or the other.

Comments received for ballot X3T10/95-015 (ATA+PI Project Proposal):

ENDL (Dal Allan):

I originally voted in favor of ATA+PI as it seemed like the right way to go
for long term merging of two important markets.

However, upon re-reading the proposal I was drawn to the repugnant implication
that the objective is to kill the existing ATA-3 and ATAPI projects by
replacing them with ATA+PI.

This negative can be converted to Yes by clearly stating that the objective is
not to slow down progress on completing ATA-3 and ATAPI, and is a follow- on
activity.

If the objective is to delay all standards progress on ATA-3 and ATAPI in an
effort to create a stronger alloy of ATA+PI forged in the fire of dispute and
acrimony then my ambivalent negative becomes an emphatic negative.

QLogic Corp. (Skip Jones):

Merging ATAPI and ATA while simultaneously splitting ATA off into it's own
committee effectively results in splitting ATAPI off into it's own committee.

I have no problem with merging ATAPI and ATA AS LONG AS ATA does not split
into it's own.  Therefore, until the split issue is resolved I'll be voting NO
to put anything into ATA.

Seagate Technology (Gene Milligan):

The following comments accompany Seagates NO vote on the proposal to combine
the existing ATA-3 and ATAPI projects into a single project and into a single
document:

1)  The existing projects are progressing well and have established
expectations in the market place that are being fulfilled. Diverting these
projects at this stage, to save host bus adapter manufacturers from purchasing
two ANSI standards is not an appropriate tradeoff in view of the delays that
are likely from changing in midstream.

2)  The goal of making the host bus adapter requirements clearer for the two
applications ATA and ATAPI would be better served by either a set of profiles
(e.g. a PCI Profile) and/or standards addressing host adapters.

3)  For disc drives, the ATA standard has facilitated the high volume market
place by staying free of a plethora of options. The lack of options avoids
confusion in the market place. The corollary is that for high volume the disc
drive may likely need to implement nearly all, if not all options. The
expectations for ATA-3 have been progressing with that in mind. It is likely
that having ATA and ATAPI in separate standards will not impinge too much on
this important aspect. However having an optional ATAPI mode in the same
standard will be highly confusing to the disc drive ATA market.

4)  For ATAPI CD-ROM and Tape applications combining ATA and ATAPI may delay
the availability of the initial ANSI standard and may also result in confusion
over the allowable options.

5)  The target date for a dpANS forwarded to X3 of June 1996 is reasonable for
the combined document and is illustrative of the delay referred to in the
above comments.

6)  The statement that there are no known legal considerations is not
accurate. At least two patents have already been identified in the ATA
project.

7)  The recommendations for close liaison assume the formation of a new
Technical Committee. We are opposed to the formation of a new Technical
Committee and note that the ATA and ATAPI projects are better served by the
direct participation of the impacted organizations within X3T10.

Western Digital Corp. (Thomas Hanan):

This item should be addressed in T13.

Comments received for ballot X3T10/95-016 (SPI-2 Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project SPI-2:

1) SPI-2 should be a new addition of SPI not a revision. Delete revise SPI to.

2) According to the balance of the proposal SPI-2 will be faster than Fast-20.
Delete and would incorporate Fast-20 into SPI-2 (it is currently a separate
draft standard) from the needs.

3) It has been the recent practice of X3T10 to leave connector issues to other
groups to conserve X3T10 resources for items that are more central to their
scope. Consequently 2.2. (b) should either be deleted or expanded to include
the SCA connectors.

4) Although the title of 3.7 indicates a duration, either for one day should
be deleted or made more flexible.

5) I think 4.4 is not correct. I think at least AMP has identified patents and
if SPI-2 includes SCA I know this to be the case.

Comments received for ballot X3T10/95-017 (ESPC Project Proposal):

ENDL (Dal Allan):

I vote in favor of the project and vote against the acronym.

ESPC can all too readily be confused with Enhanced/Extended SPC (SCSI-3
Primary Commands).

This project is unique to parallel SCSI, so parallel ought to be a
discrimination factor in the title and as this is a physical interface it
ought to have an I to imply 'interface' rather than a C to imply 'Commands'.

If we are to go to 4-letter acronyms then I suggest ESPI for Enhanced SPI
(SCSI-3 Parallel Interface)

Dropping back to 3-letter acronyms:

EPC for Enhanced Physical Configuration has less to dislike but it still does
not imply the physical interface.

My preference is EPI for Enhanced Parallel Interface.

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project for ESPC:

1) To provide a vector for document order entry, add SCSI to an appropriate
place in the title.

2) Delete the last paragraph of 2.1. It is not clear and misleading and
probably wrong.

3) For the same reason delete the first paragraph of 2.2. I suspect a high
degree of compatibility or my concept of interoperability is intended rather
than compatibility.

4) I can not agree to item (g) being included in the scope without knowing
much more about what this statement means.

5) A technical report does not document. Change 2.3 by deleting documenting
the and deleting however, the ESPC technical report would serve to document
these applications rather than invalidate them.

6) Change section 2.4 from a standard to a technical report.

7) Change section 3.3 to Not Applicable as the text.

8) Change 3.5 to ... the subject matter

9) Although the title includes duration, either delete for one day or make it
more flexible.

10) Change the text of 4.2 and 4.3 to Not Applicable.

11) I think 4.4 is misleading. I dont think the ANSI patent policy is
applicable to technical reports. I think the text should be changed to Not
Applicable.

12) Why is SCSI-2 a closely related standard?

13) I think 5.2 should be expanded and certainly SIP is a candidate.

Comments received for ballot X3T10/95-018 (FCP-2 Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project for FCP-2:

1) This is not a SCSI-2 project. Either change SCSI-2 to SCSI-3, SCSI-4, or
SCSI. Alternatively change the title to Fibre Channel Protocol - 2 for SCSI
(FCP-2).

2) In the needs delete and harmonize the document with other SCSI-3 standards.

3) The second paragraph in needs may be an appropriate thing to do during the
project but it should be deleted from the needs.

4) Item (a) in 2.2 may be an appropriate thing to do during the project but it
should be deleted from the candidates.

5) Although the title indicates duration either delete the durations or make
them more flexible.

6) In 5.1 and 5.5 delete SCSI-2.

7) In 5.2 add FC-AL, FC-PH2, and the SCSI-3 command sets.

Comments received for ballot X3T10/95-019 (SSA-PH1 Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project for the SSA Physical layer 1:

1) Line fault detector for links is not a clear goal.

2) It seems that the physical layer does not have anything to do with
preserving the capability to transport SCSI commands. I think section 2.4
should be changed by changing while preserving the capability to transport
SCSI
command and status information to which can transport SCSI command and status
information.

3) I see no reason that SCSI-2 is a closely related standard.

4) Much of the SSA project proposals share common text, see my comments on
letter ballot X3T10/95-024r0 for generic comments on the proposal.

Comments received for ballot X3T10/95-020 (SSA-PH2 Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project for the SSA Physical layer 2:

1) It seems that the physical layer does not have anything to do with
preserving the capability to transport SCSI commands. I think section 2.4
should be changed by deleting  while preserving the capability to transport
SCSI command and status information.

2) The title of  3.5 is peculiar. Everyone is competent in subject matter but
that is not pertinent. The important aspect is the subject matter rather than
subject matter.

3) SSA-PH2 from a standards point is not an initial implementation point. In
section 4.1 change provide an initial implementation point to be.

4) I see no reason that SCSI-2 is a closely related standard.

5) Much of the SSA project proposals share common text, see my comments on
letter ballot X3T10/95-024r0 for generic comments on the proposal.

Comments received for ballot X3T10/95-021 (SSA-TL1 Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project for the SSA Transport layer 1 :

1) It is not clear due to the cover page and the title being different which
layer this proposal is addressed to. The lack of clarity in the scopes of
these project proposals does not help this confusion.

2) It is not clear at all what a web topology is and therefore it is not clear
why their support is a need. Perhaps Web Topology should be addressed by
section 3.2.

3) Section 3.2 should also define the special term minize used in the goals.

4) It seems that the transport layer does not have anything to do with
preserving the capability to transport SCSI commands since this is an initial
set of SSA standards. I think section 2.4 should be changed from while
preserving the capability to transport SCSI command and status information to
which transports existing and anticipated SCSI command and status information.

5) It is not clear whether this proposal is expected to be a new project or
that it should have been submitted as a revision to Project 0989 D. Is it
expected that OMC will study the proposal closely enough that they assign it
Project 0989 D upon approval or that after some new number is assigned X3T10
will process a request to withdraw Project 0989 D?

6) The contents and title of section 5.3 seem to have been scrambled.

7) Much of the SSA project proposals share common text, see my comments on
letter ballot X3T10/95-024r0 for generic comments on the proposal.

Comments received for ballot X3T10/95-022 (SSA-TL2 Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project for a SSA Transport layer 2:

1) Web topology is not an understood term and is not clear as to why that is a
need. If it is a need web topology should be defined in section 3.2.

2) SCSI-2 is not a closely related standard.

3) Much of the SSA project proposals share common text, see my comments on
letter ballot X3T10/95-024r0 for generic comments on the proposal.

Comments received for ballot X3T10/95-023 (SSA-S2P Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligan's NO vote on the proposed new
project for mapping SCSI-2 command sets onto the SSA transport layer and
physical interface:

1) If this project proposal is approved it will create an unnecessary standard
which is redundant to the new project for mapping SCSI-3 command sets onto the
SSA transport layer and physical interface.  SCSI-2 command sets are a
compatible subset of the SCSI-3 command sets. Therefore the SCSI-3 mapping is
by definition compatible with SCSI-2.

2) Having two simultaneous standards to accomplish the same thing is fraught
with danger. Unless the documents are identical they will very likely result
in incompatible implementations due to using a different source document for
the implementation.

3) Much of the SSA project proposals share common text, see my comments on
letter ballot X3T10/95-024r0 for generic comments on the proposal.

Comments received for ballot X3T10/95-024 (SSA-S3P Project Proposal):

Seagate Technology (Gene Milligan):

The following comments accompany Gene Milligans YES vote on the proposed new
project for mapping SCSI-3 command sets onto the SSA transport layer and
physical interface:

1) It is not clear, and probably wrong that configurability results from a 10
meter length of cable. I assume the configurability would not be reduced by a
11 meter length.

2) There is no need for an implementation of SCSI-3 features on a serial
interface. X3T10 has already forwarded two such protocols for serialinterfaces
(FC and 1394). The last sentence in 2.1 should change a serial interface to
the SSA serial interface.

3) It seems to me SAM should be just a related standard. It is a requirement
not a goal and mapping does not apply.

4) Just as purely editorial comments, the goals should be stated in the same
form. In (b), which should be (a) replace define the with support for and
delete provide in the present (d) and (e).

5) Related to the untitled figure, these comments will be made without
accounting for the fact that the SSA-S2P project should not be authorized.
The figure is either unclear or the projects are not appropriate. The diagonal
lines imply that the SSA-PH2 will support only SSA-TL2 while SSA-PH1 will
support both transport layers and that SSA-S3P can be used only with SSA-TL2
while SSA-S2P can be used on either transport layer. If such restrictions
apply the restrictions should be the opposite as those shown. I suspect the
depiction represents the restrictions of someones implementation rather than
how the standards should be structured.

6) Having read the transport and physical layer project proposals I was unable
to form an opinion on what was covered by the scope of the physical layer and
what was covered by the scope of the transport layer. However the patent
statements gave a slight clue.

7) Referring to item (1) in section 3.4 it is hard to swallow that SSA-UIG is
an end-user organization. As far as I know it, or its successor is a consortia
of manufacturers without membership of organizations  like Mutual of Omaha or
the South Side Computer Club.

8) It would be more useful, I think, to give an estimate of the minimum useful
life rather than the maximum since the maximum is satisfied by no useful life.

9) In terms of a SCSI-3 commands mapping protocol SCSI-2 does not have
relevance. However I would not object to its inclusion if the SCSI-3 command
mapping protocol project is rejected and a statement is added to the SCSI-3
command mapping protocol project proposal that it will provide downward
compatibility (downward in my mind - perhaps upward in your mind) with SCSI-2.

10) The SSA project proposals do not have the document numbers referenced in
the letter ballots.






More information about the T10 mailing list