hirchert@uxe.cso.uiuc.edu (12/15/89)
jlg@lambda.UUCP writes >The X3J3 response to my first round public review comments >finally arrived. It was dated 17 November, but was not mailed >until 27 November (postmark). This is irritating for two >reasons. First, there is a 15 day limit to reply to the >committee if I think their response didn't address my comments >fully - they used 10 days of that limit waiting to mail the >letter! Second, the response contained the addresses for >obtaining the second draft of the standard as well as the >address to send second round comments - but the second public >review ENDED three days BEFORE they mailed this information. > >Fortunately, I had ample warning of the second public review >from other sources. It is clear that the committee had no >strong intent to widely advertise it (although _some_ individual >members of the committee did). > >Finally, for the most part the response did _not_ address the >issues I raised in my first round commentary. For example (as >I predicted) the committee incorrectly cited my first round >comments as support for the new POINTER facility. > >Outrage would be a mild term for my opinion of this event. >This is especially true since there was nothing particularly >individual about the responses. The committee aparently >completed and cataloged all the responses months ago. The >response to me was a generic cover letter together with an >automatically mapped set of responses to my comments - how >long could that take? 1. As a general rule, the committe tried to generate letters postdated by several days in order to insure that respondents would have their full 15 days. Apparently, something went wrong in your particular case, but this shouldn't have been the norm and most certainly was not done intentionally. 2. The text of the responses was approved last May. At that time, the committee believe that all the responses would be mailed out within a month, so it believed that an enclosure with the responses would be sufficient. Had it realized the delays that would occur, I believe it would have arranged for a separate mailing. X3J3 _did_ arrange for this comment period to be 4 months rather than the customary 2 months. X3J3 _did_ send notices of the comment period to all the publications that were notified of the first comment period. In other words, X3J3 attempted to advertise this comment period just as strongly as the first comment period. With 20/20 hindsight, we can see that in some cases, these measures were not as effective as they were the first time, but there was no reason to expect this at the time the committee was taking action. What do you think the committee should have been doing additionally to advertise the second comment period? 3. I was not on the subcommittee that drafted the response to your comments on pointers, so out of curiosity, I looked up your comment: " Pointers are needed for many applications. Although ALLOCATE and DEALLOCATE serve most of the necessary functionality, linked lists and other indirect referencing schemes are still left out. " I know I certainly would read this as supporting the idea of adding a pointer capability to Fortran. What in this is supposed to tell the committee that you wouldn't accept the particular way they chose to add this capability? (One of the purposes of the second comment period is to allow you to express objections to the specifics of the changes X3J3 has made, but why do you feel it was an inappropriate response from the committee to assume from what you said that you would welcome its decision to add pointers to the language?) 4. As you surmise, the responses were generated by pasting together responses to individual points and that many of these responses were sent to multiple people. This does _not_ mean that there was nothing individual about the responses. Every point was read to determine which response was to be sent. (Most of the subgroups writing responses has more than possible response on a given topic.) In the case of the particular subgroup I was on, I would guesstimate that about half the responses we sent out were sent to only one commentor. (That's half if you count the "generic" responses repeatedly for each commentor it was sent to. If you go by distinct responses, its more like 90% that were sent to only one commentor.) This is not to say that every response is a perfect match for the comment it responds to, but the committee certainly attempted to make them at least reasonable. As I mentioned above, the responses and the mappings of responses to commentors were approved last May. At that time, the machine readable versions of the responses and mappings (in those cases where they existed at all) were not in the hands of the person who would eventually generate the letters, but in the hands of the subgroups that generated them. In addition, none of the subgroups used the same text processing system that was eventually to be used for the letter, so conversion had to take place. The transport and conversion of all this data took substantially longer than was anticipated. At this point, copies of all the responses were generated and were distributed to the the original subgroups for proofreading. (With 20/20 hindsight, I can suggest that most commentors would have been much less upset by occasional typos or even wrong responses than they were by the delays that this engendered, but the gentleman generating the letters was concerned that if there were errors in the letters, he would be accused of sabotaging the responses, and so he wanted it to be someone else that said the letters were OK to send.) The process of shipping reems of paper all over the country to get them proofread by all five subgroups consumed another couple of months. At this point, corrections were applied, most of the letters were regenerated and proofread one more time by one additional person, and the bulk of the letters were mailed out. However, it was discovered that a few of the letters had responses either missing or incorrect. Over the next several weeks these were corrected, but this work was all done by e-mail, and it can be time consuming to try to get a 40 person committee to agree on text via e-mail. (It really is much easier to stick us all in a room and have us thrash it all out in person, but our next meeting isn't scheduled until January and you can't get enough people to change their schedules to attend a previously unscheduled meeting.) As these problems were resolved, the remaining letters were mailed out. I would judge from the date you said yours was mailed that yours must have been one of the last letters mailed, perhaps even the very last one.
jlg@lambda.UUCP (Jim Giles) (12/16/89)
From article <50500174@uxe.cso.uiuc.edu>, by hirchert@uxe.cso.uiuc.edu: > > 1. As a general rule, the committe tried to generate letters postdated by > several days in order to insure that respondents would have their full > 15 days. Apparently, something went wrong in your particular case, but > this shouldn't have been the norm and most certainly was not done > intentionally. I did an informal survey of the people I know who received responses to their first roung comments. In all cases, the response letter was dated _exactly_ ten days before the postmark on the envelope which carried the responses. Mine was the only one to actually be _mailed_ after the close of the second review - but all of us had our 15-day reply interval curtailed. (Now, of course, a sample of 4 response letters in not adequate to prove a systematic pattern. But the exact ten day delay does seem to be a bit too coincidental.) > 3. I was not on the subcommittee that drafted the response to your comments on > pointers, so out of curiosity, I looked up your comment: > > " Pointers are needed for many applications. Although ALLOCATE and > DEALLOCATE serve most of the necessary functionality, linked lists > and other indirect referencing schemes are still left out. " > > I know I certainly would read this as supporting the idea of adding a > pointer capability to Fortran. What in this is supposed to tell the > committee that you wouldn't accept the particular way they chose to add > this capability? What I was asking for was linked lists and "other indirect referencing schemes" (like trees, DAGs, GRAPHS, etc.). There is nothing in my comment which even mentions using POINTERS as a poor (slow, inadequate) substitute for RANGE/IDENTIFY. The overwhelming majority of the current POINTER proposal has nothing to do with the recursive data structures I asked for. You say: "I know I certainly would read this as supporting the idea of adding a pointer capability to Fortran." Well, so would I. At the time I wrote that I had no objections to adding pointers to Fortran. You know - POINTERS? Those things that Pascal or Modula have? Or, perhaps the kind already found in many Fortran implementations (like Cray's)? Even C style pointers (with all that ugly pointer arithmetic) would have been preferable to what the committee chose to do. However, my present objection (as regards the public review responses) is the casual way that the response implies that the present form of the pointer proposal was what I asked for! As I pointed out before, it rather like getting a MACK truck when I asked for an economy car - and then being told that I was given what I requested. I suppose I _COULD_ use the truck to commute to work (just as I _could_ use the present pointers for something useful), the rest of its abilities are not only unneeded, but undesireable.
hirchert@ux1.cso.uiuc.edu (12/20/89)
1. Several additional examples have been provided of public comment responses that were mailed later than the date on their cover letter. Now I know that the people doing the physical preparation of these letters discussed postdating the letters to avoid this problem, and I know that the letters were in fact produced postdated, but apparently the time period allowed was not long enough to allow for everything that had to be done prior to mailing. Apparently the dating problem was more general than I realized (although apparently not generally as bad as the 10 days that J Giles experienced). I still believe that there was no systematic intent to deny commentors their right of appeal. (By the way, when we are talking about public comments that are so large it takes the technical committee more than a year to respond to them, I think 15 day limit on protests is a bit silly, that's ANSI's rule and X3J3 has no control over it.) 2. Presley Smith (psmith@mozart.uucp) writes >In fact, the rules called the SD-2, are fairly clear that all letters are >supposed to be sent prior to the start of the next review and that the >negatives are to be addressed with the commenters before the start of >the next review by response to those commentors. That is all part of >milestone 14 which must be completed before the return to milestone 12 >which is another public review cycle. As Presley well knows, this is what X3J3 believed to be true and the goal it attempted (unsuccessfully) to meet. When X3J3 inquired whether it was necessary to send out the responses before forwarding a revised draft for the second public comment period, they were told that it was not and that all that was required was that X3J3 approve the responses before the next review, not that they actually be mailed by then. In other words, the rules in SD-2 are not quite as clear as Presley is suggesting. Nevertheless, X3J3 has been revising its internal procedures for handling public comment to try to make certain that the kinds of delays that occurred in the sending of the responses to the first rounds do _not_ occur in the handling of the second round of comments. 3. Presley also says >On an appeal or protest, you will get a much higher level of attention >that what you get with the public review letter. The management groups >of the standards processes work the protests and appeals. Maybe so, but so far all they have done is send us the protest letters through the same channels they send us regular and late second round comments. So far X3J3 has been given no special instructions for handling these letters, and so they have been put into the same process as the second round comments. (The people involved in that process _have_ been asked to take special care in processing the protest letters, but that's as far as it goes (as of now).) Thus, it may be that the management groups intend to monitor X3J3's handling of these letters more closely, but that has had no immediate effect on how they will be handled technically. I suppose that since it is probably more than 15 days after even the last of the first round comments should have been received, this may be a moot issue by now, anyway. Kurt W. Hirchert hirchert@ncsa.uiuc.edu National Center for Supercomputing Applications
jerry@violet.berkeley.edu ( Jerry Berkman ) (12/20/89)
In article <6800003@ux1.cso.uiuc.edu> hirchert@ux1.cso.uiuc.edu writes: >I suppose that since it is probably more than 15 days after even the last of >the first round comments should have been received, this may be a moot issue >by now, anyway. > >Kurt W. Hirchert hirchert@ncsa.uiuc.edu >National Center for Supercomputing Applications It's not. The reply to my first round comments was dated Nov. 30 and postmarked Dec. 1. It states I have "15 working days of the date of this letter", so there are still a few of us with a few days left to mail in objections. However, after a wait of 1 3/4 years, it is really unfair to allow only 3 weeks for a reply. - Jerry Berkman, U.C. Berkeley
psmith@mozart.uucp (Presley Smith) (12/21/89)
In article <6800003@ux1.cso.uiuc.edu> hirchert@ux1.cso.uiuc.edu writes: > >2. Presley Smith (psmith@mozart.uucp) writes >>In fact, the rules called the SD-2, are fairly clear that all letters are >>supposed to be sent prior to the start of the next review and that the >>negatives are to be addressed with the commenters before the start of >>the next review by response to those commentors. That is all part of >>milestone 14 which must be completed before the return to milestone 12 >>which is another public review cycle. > > As Presley well knows, this is what X3J3 believed to be true and the goal it > attempted (unsuccessfully) to meet. When X3J3 inquired whether it was > necessary to send out the responses before forwarding a revised draft for > the second public comment period, they were told that it was not and that > all that was required was that X3J3 approve the responses before the next > review, not that they actually be mailed by then. In other words, the > rules in SD-2 are not quite as clear as Presley is suggesting. > Nevertheless, X3J3 has been revising its internal procedures for handling > public comment to try to make certain that the kinds of delays that occurred > in the sending of the responses to the first rounds do _not_ occur in the > handling of the second round of comments. > The SD-2 is, in fact, VERY clear on the processing order. In fact, the X3J3 was supposed to have ALL the letters MAILED by the time the 2nd public review was STARTED not after the public review was 1/2 over or finished. This issue is not dead. It will have to be addressed by SMC and the BSR because of the problems that have occurred. Pressure from the international organizations have caused problems in the ANSI processing rules. This is a set of issues that need to be addressed by ANSI and ISO to obtain a better consensus on what the rules are. The two organizations handle standards development such different ways, that the conflict occurs as to which rules are being followed. If this becomes the ISO standard, which I expect to happen, it must still be approved as an ANSI standard under the ANSI rules for the acceptance of other standards. It is unclear as to how this situation will effect the final processing of this standard as an ANSI standard. It's very likely that an additional public review will be required in this country due to the rules violations that have occured before that standard can ever become an ANSI standard. >3. Presley also says >>On an appeal or protest, you will get a much higher level of attention >>that what you get with the public review letter. The management groups >>of the standards processes work the protests and appeals. > > Maybe so, but so far all they have done is send us the protest letters > through the same channels they send us regular and late second round > comments. So far X3J3 has been given no special instructions for > handling these letters, and so they have been put into the same process > as the second round comments. (The people involved in that process _have_ > been asked to take special care in processing the protest letters, but > that's as far as it goes (as of now).) Thus, it may be that the management > groups intend to monitor X3J3's handling of these letters more closely, but > that has had no immediate effect on how they will be handled technically. > >I suppose that since it is probably more than 15 days after even the last of >the first round comments should have been received, this may be a moot issue >by now, anyway. A protest letter that states you do no agree with the committee action on this issue or that issue will be handled by the committee in a manner similar to public review comments. The committee will review it again, take action on it in some way (maybe not do anything??) and will tell you what they did. X3 will certainly note the number of protest letters that are generated from the committee's responses. It's a measure of satisfaction with the responses. That is much different to an appeal. There is a section of the SD-2 that defines the appeal process. Appeals are done on rules violations, procedural issues, etc. They typically are not done on technical issues. If you write an appeal, you will, in fact, get attention at a much higher level than X3J3. X3 may request X3J3's input on the appeal, but X3, BSR, etc. will have to work the appeal directly. An appeal is in fact the process of going over the technical committees head to protest to the higher level committees. You have the right on an appeal to appeal to X3 and if rejected there appeal to the ANSI Board of Standards Review (BSR). On appeal, you have the right to present your case directly to those bodies and work with those bodies to resolve the problem. There was an appeal on the ANSI C standard that kept that standard from being finalized for many months. If appeals start happening on Fortran 8x, it could delay approval of the proposed standard for some period of time. X3 in the debate on whether to keep FORTRAN 77 as an active standard, several times stopped the debate, got clarification on the rules and procedures, and was extremely careful to do everything by the book fo be sure that no appeals could be filed on rules violations. They have become more sensitive to those issues as a result of the problems with ANSI C.
khb@chiba.kbierman@sun.com (Keith Bierman - SPD Advanced Languages) (12/21/89)
In article <4161@convex.UUCP> psmith@mozart.uucp (Presley Smith) writes:
..... rules discussion...
was STARTED not after the public review was 1/2 over or finished. This
issue is not dead. It will have to be addressed by SMC and the BSR because
of the problems that have occurred.
If memory serves, at the Long Island meeting, Bill R. as the X3 rep
was specifically asked questions about this procedure. He asserted
that this was legit. I believe that the chair followed up with X3. If
there were rules violations it isn't because there was a lack of
effort.
cheers all.
--
Keith H. Bierman |*My thoughts are my own. !! kbierman@sun.com
It's Not My Fault | MTS --Only my work belongs to Sun*
I Voted for Bill & | Advanced Languages/Floating Point Group
Opus | "When the going gets Weird .. the Weird turn PRO"
"There is NO defense against the attack of the KILLER MICROS!"
Eugene Brookspsmith@mozart.uucp (Presley Smith) (12/22/89)
In article <KHB.89Dec20175012@chiba.kbierman@sun.com> khb@chiba.kbierman@sun.com (Keith Bierman - SPD Advanced Languages) writes: > >In article <4161@convex.UUCP> psmith@mozart.uucp (Presley Smith) writes: > >..... rules discussion... > > was STARTED not after the public review was 1/2 over or finished. This > issue is not dead. It will have to be addressed by SMC and the BSR because > of the problems that have occurred. > >If memory serves, at the Long Island meeting, Bill R. as the X3 rep >was specifically asked questions about this procedure. He asserted >that this was legit. I believe that the chair followed up with X3. If >there were rules violations it isn't because there was a lack of >effort. 1. I did NOT accuse X3J3 of "lack of effort." In fact, X3J3 did a LOT of extra effort to get these public review comments out in a timely manner. Part of the problem was pressure from the international community and the differences in the rules. ANSI, SMC, and BSR is going to have to come to grips with those issues. X3J3 was caught between the proverbal "rock and hard place." X3J3 did what they had to and tried to do it right. 2. Bill Reinhuls was representing SPARC and X3 at the Long Island meeting. It has been the practice with other committees also to put some parallelism in the processing to shorten the time. X3 and ANSI are certainly interested in timely completions of standards efforts (although some international groups don't believe that...). Officially, the committee cannot forward the standard to X3 for further processing until the previous letters are completed and the people have either protested or withdrawn. Instead, the standard was forwarded to X3 so the public review would be started sooner with the understanding that these letters from the first public review would be in the hands of the reviewers by the time the standard was available again for review. That's not what happened. And that set of issues will have to be reviewed by the standards management groups. I don't have any insight as to what will happen... maybe nothing. Also, Bill Reinhuls or any INDIVIDUAL on X3 or SPARC does not have the right or the authority to bend or break the rules. What Bill said was strictly legit by previous practice that had been approved. What has happened in fact is questionable at best.