CONTACTS 
0201 - Dynamic Alternatives                         562 802 1639
020101 - Mr. Garold L. Johnson
020102 - dynalt@dynalt.com

SUBJECTS
HTML Improvements SDS Standard Procedure Need Validator Program
SDS Problem One Long Stream of Text
HTML Improvements Validator Shows SDS Code Can be Improved 070405 012
HTML Conversion Not Displaying Correctly in Opra Browser String of Te
Problem HTML Conversion Not Displaying Correctly in Opra Browser, Gar
Validator Program Shows HTML Code Defective, Gary Johnson
Johnson, Gary Provides Examples of Specific Problems
HTML Standards W3C for Coding Web Pages

1510 -    ..
1511 - Summary/Objective
1512 -
151201 - Follow up ref SDS 34 5Y3F, ref SDS 28 0001.
151202 -
151203 - Gary reported good news finding documentation for configuring the
151204 - ws_ftp program to support SDS. ref SDS 0 MZ4H  He submitted an SDS
151205 - record that was not working properly to create a new task in the
151206 - Schedule.  Review led to discovery of a subtle format issue.  The code
151207 - was modified to avoid future problems. ref SDS 0 IF3N  Made progress
151208 - reviewing format issues for HTML in documents and SDS records on the
151209 - Internet.  Main problem is nesting, which relates to the order of
151210 - entering commands. ref SDS 0 9F9K  A big issue for us is the p
151211 - command.  Information from W3C says that a closing command is
151212 - optional. ref SDS 0 3H5F  Gary's work typically uses a /p param.
151213 - ref SDS 0 IE7H  I was able to change the code to meet some of the
151214 - issues Gary describes. ref SDS 0 K97V  Would like to get more input on
151215 - questions identified as Action Items. ref SDS 0 AI  Gary reports new
151216 - information showing similar issue is shaping up on DOS4 compliance
151217 - with MS standards that impact performance of SDS functions to archive
151218 - pointers and to draw reports. ref SDS 0 YX7I
151219 -
151220 -  ..
151221 - Put an SDS update on the website to incorproate improvements and
151222 - changes. ref SDS 0 X16G  Sent Gary notice of progress via email linked
151223 - to this record.
151224 -
151225 -     [On 021106 received samples files showing corrections needed.
151226 -     ref SDS 39 TW6F
151227 -
151228 -
151229 -
151230 -
151231 -
151232 -  ..
1513 -
1514 -
1515 - Progress
1516 -
151601 - Improve Conversion of SDS Records to HTML
151602 -
151603 - Follow up ref SDS 34 5Y3F, ref SDS 28 0001.
151604 -
151605 - Received ref DRT 1 0001 from Gary with four (4) files attached in a
151606 - zip file...
151607 -
151608 -      1.  Letter to Gary on 021028, ref DIP 1 0001,
151609 -
151610 -          Error log generated by validator program. ref DRT 1 PPWX
151611 -
151612 -           ..
151613 -      2.  SDS record on work sending Gary the letter. ref SDS 36 0001
151614 -
151615 -          Error log generated by validator program. ref DRT 1 J86L
151616 -
151617 -  ..
151618 - Background
151619 -
151620 -    On 021025 reviewed ideas submitted by Gary Johnson in a letter
151621 -    submitted earlier on 020823 for improving HTML in SDS records.
151622 -    ref SDS 34 5Y3F  At that time, on 021025 could not locate the
151623 -    specific examples of errors in the files Gary submitted.
151624 -    ref SDS 34 N45N Since SDS records and other work product in
151625 -    document files that are converted into HTML render satisfactorily
151626 -    in IE and Netscape browsers, the goal is to more closely comply
151627 -    with published standards in order for a wider range of browsers to
151628 -    display SDS work product.  Even though IE and Netscape are popular
151629 -    throughout the world,  it is important to work on standards
151630 -    compliance for future viability.
151631 -
151632 -        [...below, more immediate issue of industry standards arises on
151633 -        using SDS operations that rely on MS DOS. ref SDS 0 YX7I
151634 -
151635 -     ..
151636 -    Yesterday, I asked Gary to send updated examples that can serve as
151637 -    a template to try modifying conversion operations so that the HTML
151638 -    code conforms with industry standards.
151639 -
151640 -     [On 021106 received samples files showing corrections needed.
151641 -     ref SDS 39 TW6F
151642 -
151643 -     [On 021114 Gary reports SDS records and documents display in Opera
151644 -     and have closer compliance with HTML standards. ref SDS 41 0001
151645 -
151646 -  ..
151647 - Gary's letter today says...
151648 -
151649 -    I removed most of the lines about deprecated tags as that is an
151650 -    issue for later. This is why there are messages missing in the
151651 -    error files. ref DRT 1 IP5Y
151652 -
151653 -        What is a depricated tag?
151654 -
151655 -           Research at....
151656 -
151657 -           http://www.w3.org/TR/REC-html40/conform.html#deprecated
151658 -
151659 -           Shows...
151660 -
151661 -           A deprecated element or attribute is one that has been
151662 -           outdated by newer constructs. Deprecated elements are
151663 -           defined in the reference manual in appropriate locations,
151664 -           but are clearly marked as deprecated. Deprecated elements
151665 -           may become obsolete in future versions of HTML.
151666 -        ..
151667 -        On 990618 there is a location for HTML procedures.
151668 -        ref SDS 23 XU9M  Later on 991231 there is another site with a
151669 -        different location for the same thing. ref SDS 25 5P5N
151670 -
151671 -  ..
151672 - Gary continues...
151673 -
151674 -    The basic issue continues to be violation of HTML nesting rules:
151675 -
151676 -         •  HTML tags should be closed. This means every tag that has
151677 -            an end tag should have one. Many browsers will allow you to
151678 -            leave out some of them, but the rule makes the HTML a lot
151679 -            easier to read and to debug. ref DRT 1 PPSR
151680 -
151681 -                See research below on end tag for para is optional.
151682 -                ref SDS 0 3H5F
151683 -
151684 -             ..
151685 -         •  HTML tags should nest correctly. All tags should be closed
151686 -            in the order in which they are opened. Think of it as an
151687 -            outline. ref DRT 1 Q149
151688 -
151689 -
151690 -     ..
151691 -    Example:
151692 -
151693 -    {b>{p>This is a paragraph{/b>{/p> -- The tags are closed in the
151694 -    wrong order. This should be {p>{b>This is a paragraph{/b>{/p> with
151695 -    the bold inside the paragraph or {b>{p>This is a paragraph{/p>{/b>
151696 -    with the paragraph inside the bold. ref DRT 1 IE7H
151697 -
151698 -
151699 -
151700 -
1518 -

SUBJECTS
P Commands

1804 -
180401 -  ..
180402 - Para Commands Permit Optional Ending Command
180403 -
180404 - Gary's letter today focuses attention on the para command, and need to
180405 - avoid nesting problems by sequencing commands correctly. ref SDS 0
180406 - IE7H
180407 -
180408 - There is a threshold issue of an end tag for a paragraph.  Without the
180409 - tag, the validator program reports nesting errors.  We want to omit
180410 - the /p tag because it is hard to figure out code for ending a
180411 - paragraph consistently, i.e, sometimes the end of a paragraph is not
180412 - clear.
180413 -
180414 -     [On 021103 Gary notes this is optional and can be address later.
180415 -     ref SDS 37 P15I
180416 - ..
180417 - We could try the feature that has worked with the font tag of
180418 - always starting with an ending font tag, but I suspect this will have
180419 - difficulties as well.
180420 -
180421 -  ..
180422 - There is some evidence that an end tag for the p command is optional,
180423 - as shown in para 9.3.1 of the W3C spec for version 4.01...
180424 -
180425 -          http://www.w3.org/TR/REC-html40/struct/text.html#h-9.3.1
180426 -
180427 - Therefore, it is helpful in this case to exercise the option.
180428 -
180429 - W3C seems to adopt the option of avoiding the ending /p command, as
180430 - shown at the following location explaining the font command...
180431 -
180432 -     http://www.w3.org/TR/REC-html40/present/graphics.html#h-15.2
180433 -
180434 - W3C begins a para with a p command, and omits a closing command, so
180435 - we may not be that bad.
180436 -
180437 -  ..
180438 - It may be possible to throw in the /p command in SDS records because
180439 - it uses the pre command, and so formatting is fixed.  There is very
180440 - little need for /p.
180441 -
180442 - Another solution is to do the entire thing with BR commands and
180443 - forget about P commands, since BR commands do not seem to require an
180444 - ending command.
180445 -
180446 - Suppose we change the code and instead of using P, use BR BR, which
180447 - seems to do the same thing, and does not cause the nesting problem.
180448 -
180449 -  ..
180450 - Actually, in an SDS record there is only one P command at the top
180451 - between the letterhead and Summary.  Everything else is done with br
180452 - commands.
180453 -
180454 -      [On 021111 changed code in 0122 to br br and changed the HTML
180455 -      profile so that F5 enters br br instead of p, since the effect
180456 -      seems to be the same. ref SDS 40 ZE4F
180457 -
180458 -
180459 -
180460 -
180461 -
180462 -
180463 -
180464 -
180465 -
1805 -

SUBJECTS
Introduction to Structure of HTML Document
Doctype Command First Line in HTML File Shows Comment on HTML Version

2005 -
200501 -  ..
200502 - Document Type Declaration Missing from SDS but Not from Documents
200503 -
200504 - The error log for the SDS record says in part....
200505 -
200506 -  1.  Warning: [24] A document type declaration should appear as the
200507 -      first line (line 1) of every HTML document. For example, for HTML
200508 -      4.01 Strict documents...
200509 -
200510 -      {!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
200511 -
200512 -      ...should be the first line. For HTML 4.01 Transitional
200513 -      documents, the first line should be...
200514 -
200515 -      {!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">.
200516 -      ..
200517 -      For HTML 4.01 Frameset documents, the first line should
200518 -      be...
200519 -
200520 -      <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Frameset//EN" "http://www.w3.org/TR/html4/frameset.dtd">.
200521 -
200522 -       ..
200523 -      Although HTML recommendations generally require this line, most
200524 -      browsers probably ignore it. If you are using HTML Validator's
200525 -      integrated editor, then you can add this from the Tags menu
200526 -      and/or from the Tag Inserter. ref DRT 1 04U5
200527 -
200528 -          Document type in this explanation seems to be a comment, so
200529 -          why is a comment an essential part of a display?  Would like
200530 -          to avoid delcaring we are meeting any particular standard,
200531 -          and simply create the formats we need to do the work.
200532 -
200533 -          There is a reference at....
200534 -
200535 -            http://www.w3.org/TR/REC-html40/struct/global.html#h-7.1
200536 -
200537 -          ...for para 7.2 that says...
200538 -
200539 -           ..
200540 -          A valid HTML document declares what version of HTML is used
200541 -          in the document. The document type declaration names the
200542 -          document type definition (DTD) in use for the document
200543 -          (see [ISO8879]).
200544 -
200545 -           ..
200546 -          The reference to ISO 8879 at...
200547 -
200548 -             http://www.w3.org/TR/REC-html40/references.html#ref-ISO8879
200549 -
200550 -          ...says...
200551 -
200552 -          "Information Processing -- Text and Office Systems --
200553 -          Standard Generalized Markup Language (SGML)", ISO 8879:1986.
200554 -          Please consult...
200555 -
200556 -             http://www.iso.ch/cate/d16387.html
200557 -
200558 -          ...for information about the standard.
200559 -
200560 -          That location is a website that offers to sell people ISO
200561 -          documents, products and services.
200562 -
200563 -           ..
200564 -          Could not find anywhere that explains what use is made of the
200565 -          doc type, other than to support W3C.
200566 -
200567 -           ..
200568 -          I am curious about when this spec was promulgated, since the
200569 -          book we used to develop the conversion tools does not say
200570 -          anything about this comment?
200571 -
200572 -          If the comment is added, how does the browser respond, and
200573 -          if it is omitted, what happens?
200574 -
200575 -
200576 -
200577 -
200578 -
2006 -

SUBJECTS
Font Command

2104 -
210401 -  ..
210402 - Font Command Deprecated Requires Ending Command
210403 -
210404 -
210405 - Analysis from the Validator opration in Gary's letter says...
210406 -
210407 -  3.  Error in line 6 at character 5: The closing tag for "font" was
210408 -      found, but the tag was never opened, has been closed too many
210409 -      times, or cannot have an end tag. ref DRT 1 04WD
210410 -
210411 -          The SDS record on 021028, ref SDS 36 0001, when converted to
210412 -          HTML shows on line 6 char 5...
210413 -
210414 -              {/font>{FONT FACE="Times Roman" SIZE=2>{B>{center>
210415 -
210416 -  ..
210417 - Line 1130, ref OF 2 624K, -label 02 in 070405 about 100 lines below.
210418 -
210419 -    ins_text!{center>{font face="Times Roman" SIZE=2>{B>!
210420 -
210421 -       Changed this to omit the initial /font command, and moved center
210422 -       in front of the font command, to reduce the nesting concern.
210423 -
210424 - Gary's report continues...
210425 -
210426 -  8.  Warning in line 11 at character 1: [30] The "P" tag is contained
210427 -      in a "FONT" tag (which was opened in line 6). This may be
210428 -      acceptable for some browsers (such as Internet Explorer and
210429 -      Netscape), but HTML 4.01 does not allow this. It may also cause
210430 -      problems or unexpected page rendering for more compliant browsers
210431 -      such as Opera. Possible solutions: 1. Close the "FONT" tag before
210432 -      using the "P" tag. Depending on the correct usage of the closed
210433 -      tag, you may be able to reopen it after using the "P" tag; 2.
210434 -      Eliminate the "P" tag; or 3. Reorder the "P" and "FONT" tags.
210435 -      ref DRT 1 UZ8L
210436 -
210437 -
210438 -  ..
210439 - Line 1180, ref OF 2 CZ8O, -label lpltrd in 070405 about 20 lines
210440 - below.
210441 -
210442 -    ins_text!{/B>{/font>!
210443 -
210444 -       This is the code that ends the letterhead op.
210445 -
210446 -       Changed this to so the ending font command occurs before the p
210447 -       command.
210448 -
210449 -  ..
210450 - Line 1720, ref OF 2 TF3L, -label lpIs in 070405
210451 -
210452 -    ins_text!{/font>{font FACE="Arial,Helvetica" SIZE=4 COLOR="black">{B>SUBJECTS{/B>!
210453 -
210454 -       Started to remove /font from beginning of this command string
210455 -       that formats the "Subject" field, but seems like we need it to
210456 -       turn off the prior font command that set the Courier font.
210457 -
210458 - Line 1750, ref OF 2 QX6K, -label bslist in 070405
210459 -
210460 -    ins_text!{/FONT>{FONT FACE="Courier" SIZE=2>!
210461 -
210462 -       Will try leaving this /font, since it ends the command that
210463 -       changed the font for Subjects.
210464 -
210465 -  ..
210466 - The font issue in SDS records is not too difficult, but in regular
210467 - documents I am concerned this will be hugely difficult to find a
210468 - consistent process.  We adopted the current method of beginning font
210469 - commands with a closing command, becausd a lot of trial and error
210470 - showed this is the only method that worked consistently.
210471 -
210472 -  ..
210473 - Possibly, Gary could modify the letter he submitted with the list of
210474 - errors, and produce it in a corrected format. ref DIP 1 0001  That
210475 - might provide a template for modifying 0122.
210476 -
210477 -     [On 021106 received samples files showing corrections needed.
210478 -     ref SDS 39 TW6F
210479 -
210480 - A template showing what works in IE, Netscape and Opra, avoids
210481 - guessing on interpreting standards and error reports.
210482 -
210483 -     [On 021103 Gary reports that font command sequencing with other
210484 -     commands is critical; still need template to guide improvement to
210485 -     SDS code. ref SDS 37 TW6F
210486 -
210487 -     [On 021105 Gary plans to submit template files to guid
210488 -     improvements to HTML for SDS records and documents. ref SDS 38
210489 -     TW6F
210490 -
210491 -
210492 -
210493 -
210494 -
210495 -
2105 -

SUBJECTS
Bold
Nested Commands Must be Entered in Consistent Sequence
Center Not Expressly Specified May be Deprecated Which Means Being Ph

2406 -
240601 -  ..
240602 - Bold Command Within Center Command and End Bold Before End Center
240603 -
240604 -
240605 -
240606 -  4.  Warning in line 6 at character 6: [28] The "center" tag is
240607 -      contained in a "B" tag (which was opened in line 6). This may be
240608 -      acceptable for some browsers (such as Internet Explorer and
240609 -      Netscape), but HTML 4.01 does not allow this. It may also cause
240610 -      problems or unexpected page rendering for more compliant browsers
240611 -      such as Opera. Possible solutions: 1. Close the "B" tag before
240612 -      using the "center" tag. Depending on the correct usage of the
240613 -      closed tag, you may be able to reopen it after using the "center"
240614 -      tag; 2. Eliminate the "center" tag; or 3. Reorder the "center"
240615 -      and "B" tags. ref DRT 1 8H7H
240616 -
240617 -       ..
240618 -  7.  Error in line 10 at character 2: The "center" tag (opened in line
240619 -      should be closed before the "B" tag is closed (nesting error).
240620 -      ref DRT 1 04ZG
240621 -
240622 -          Cannot find any authority for sequencing bold and center.
240623 -
240624 -          Actually, there is a note at....
240625 -
240626 -             http://www.w3.org/TR/REC-html40/present/graphics.html#h-15.2.1
240627 -
240628 -          ...that says...
240629 -
240630 -             Font style elements must be properly nested. Rendering of
240631 -             nested font style elements depends on the user agent.
240632 -          ..
240633 -          Center seems to be deprecated, i.e., it does not seem
240634 -          to be a stand alone command in 4.01, it is only shown as part
240635 -          of other commands, for example, at...
240636 -
240637 -             http://www.w3.org/TR/REC-html40/present/graphics.html#h-15.1.2
240638 -
240639 -          ...which we don't want to use, because center seems fast and
240640 -          easy.
240641 -
240642 -           ..
240643 -          The nesting issue seems to arise where we start with one
240644 -          command sequence and end with another, rather with the
240645 -          initial sequence.
240646 -
240647 -  ..
240648 - Line 1130, ref OF 2 624K, -label 02 in 070405 about 100 lines below.
240649 -
240650 -          Changed this code so that center comes before bold.
240651 -
240652 -
240653 -
240654 -
2407 -

SUBJECTS
Subject Description

2504 -
250401 -  ..
250402 - Subjects for Control Fields Should Not Be Bold
250403 -
250404 - Follow up ref SDS 34 5H9F.
250405 -
250406 - On 021025 review at that time indicated that subject descriptions are
250407 - now showing as bold, and it was decided this is satisfactory.
250408 - ref SDS 34 5H9F
250409 -
250410 - However, it seems that the descriptions are in fact bold in IE, but
250411 - not bold in Netscape.  We want consistant display.
250412 -
250413 - The code causing this is...
250414 -
250415 -    {/font>{font FACE="Arial,Helvetica" SIZE=4 COLOR="black">{B>SUBJECTS{/B>{/font>{font FACE="Arial,Helvetica" SIZE=3 COLOR="black">
250416 -    HTML Improvements SDS Standard Procedure Need Validator Program
250417 -    Johnson, Gary Provides Examples of Specific Problems
250418 -    HTML Standards W3C for Coding Web Pages
250419 -    {/FONT>{FONT FACE="Courier" SIZE=2>
250420 -
250421 -  ..
250422 - Nominally, the rendering should not be bold, but it is bold in IE, so
250423 - Netscape seems to have the correct display.
250424 -
250425 - Suppose we remove the /font?
250426 -
250427 - Didn't make a dent.
250428 -
250429 - Actually, there are a lot of differences in the way IE shows stuff,
250430 - from the display by Netscape for the same code.  The characters are
250431 - uniformly larger in IE.  Possibly we have a setting that is not
250432 - evident in either IE or Netscape that causes a difference in the
250433 - display.
250434 -
250435 -  ..
250436 - If we put a bold command in there....
250437 -
250438 - We get another mess.  IE is too big, and Netscape looks okay.
250439 -
250440 -  ..
250441 - If we change the size of the font from 3 to 2, then IE is great, but
250442 - can't read the thing in Netscape, because the print is too small.
250443 -
250444 - So, on balance, it seems the code we have is the best that can be done
250445 - to get something that is readable in both IE and Netscape.
250446 -
250447 -
250448 -
250449 -
250450 -
250451 -
250452 -
2505 -

SUBJECTS
Transferring Files
Temp Transfer Dir on Local Drive, 070405, 0122
Empty Temp HTML Directory
Maintain Internet Transfer Directory from Local Disk to Server by Del
WS FTP Pro Setup Parameters for Uploading, Downloading and Deleting F
Ws_ftp Log Establishing Transfer Has Occurred Not Showing Up in Direc
DOS Copy Command Parameters W98 Autoexec.bat Path SD 01 and 01 02 for
Johnson, Gary Ws_ftp Log Establishing Transfer Has Occurred Not Showi

3811 -
381101 -  ..
381102 - Maintain Internet Transfer Directory Macro 235
381103 - Ws_ftp Pro Log File Reported on File Transfers to Web for Maintenance
381104 -
381105 - Follow up ref SDS 33 MZ4H, ref SDS 29 1R4H.
381106 -
381107 - Received ref DRT 2 0001 from Gary reporting progress on getting a log
381108 - file to appear in directories where files have been transferred to
381109 - the Internet.  He says...
381110 -
381111 -       Question/Problem: I used to get a log file created in the local
381112 -       destination directory named ws_ftp.log. Since I upgraded to
381113 -       version 7.0, I no longer get this file. I want this file for
381114 -       tracking purposes. Can I get the program to create this file
381115 -       again? ref DRT 2 8Y5L
381116 -
381117 -        ..
381118 -       Answer/Solution: In previous versions of WS_FTP Pro, the default
381119 -       location for the log file was the active local system directory.
381120 -       This location is now a configurable option. By default, the log
381121 -       file is created in the users profile directory. In order to
381122 -       change this so that the file logs to the active local system
381123 -       directory, go to Options->Logging and change the path for
381124 -       Transfer Logging to a dot and a forward slash, i.e. "./",
381125 -       ref DRT 2 LY6H
381126 -
381127 -        ..
381128 -       I made this change and got a log in the same directory as the
381129 -       single file I uploaded. Is this the result you wanted?
381130 -       ref DRT 2 DZ7F
381131 -
381132 -          [On 021125 Gary reports the format for maintaining the
381133 -          transfer directory is not working. ref SDS 42 XM3F
381134 -
381135 -       I converted another SDS record to HTML, and the directory
381136 -       structure was cleaned up, so it looks like this problem is
381137 -       toast. ref DRT 2 5Z7L
381138 -
381139 -  ..
381140 -
381141 - We need to have Gary's DOS4 program set to show all filenames as upper
381142 - case, so the transfer directory can be maintained, since it is looking
381143 - for WS_FTP.LOG, per work on 021028. ref SDS 34 IM4M
381144 -
381145 -     [On 021103 Gary says he can set switches to launch 4DOS so it
381146 -     provides same formats as MS DOS. ref SDS 37 MZ4H
381147 -
381148 -
381149 -
381150 -
381151 -
381152 -
381153 -
3812 -

SUBJECTS
Format Issues Impact Performance Corrected and Prepared Explanation t
New Task in Schedule F1 Shift F1 Failed on Record Dated 021022 Possib
Procedures
Convert Diary to Schedule for New Task
Follow Up Anchor 0001
Anchors on Headings

4509 -
450901 -  ..
450902 - Creating New Task in Schedule Use F1 Shift F1 in Diary Record
450903 -
450904 - Received ref DRT 3 0001 from Gary transmitting an SDS record dated
450905 - 021022, ref DRT 3 L67L, which is not producing a new scheduled task in
450906 - the Schedule when he uses F1 Shift F1.
450907 -
450908 - Review showed a subtle format problem. ref SDS 0 UC5O  There are some
450909 - manual solutions.  Also modified 004 op to avoid future problems.
450910 - ref SDS 0 6X7F
450911 -
450912 -  ..
450913 - I looked at the record and it seems okay.
450914 -
450915 -     A minor issue of formatting that does not affect anyting is that
450916 -     "Progress" as the first bunch of stuff after Summary/Objective,
450917 -     can (but does not have to) be indented like this...
450918 -
450919 -         Actually, this turned out to the cause of the problem, per
450920 -         below. ref SDS 0 UC5O
450921 -
450922 -  ..
4510 -
4511 -
4512 - Progress
4513 -
451301 - However, Gary's record has "Progress" as an underlined headline only
451302 - a few lines below the Summary/Objective, leaving no room to write up
451303 - the summary.
451304 -
451305 -      [On 021103 there are quite a few records that do not handle the
451306 -      format for the first headline in the manner planned for SDS,
451307 -      although they should not crash the program. ref SDS 37 QA9L
451308 -
451309 - Using F1 F8 creates a time stamped outline guy, with an anchor, and
451310 - the time entry can be changed for "Progress," at least that is how I
451311 - do it.  But it is a matter of style, so others can adopt any
451312 - convenient format.
451313 -
451314 -  ..
451315 - To analyse this problem...
451316 -
451317 - Loaded it "as is" in my SDS records for 021022, and try to create a
451318 - scheduled task. ref SDS 32 0001
451319 -
451320 -     1.  Created a new blank task for 021022.
451321 -
451322 -     2.  Closed SDS and launched to archive everything and make it a
451323 -         diary record.
451324 -         ..
451325 -     3.  Opened record for 021022.
451326 -
451327 -         Used tl 0 at command line to turn titles off.
451328 -
451329 -         Used F6 to make a few blank lines below the line 1 header,
451330 -         then used get command to read in Gary's record, which I names
451331 -         glj.
451332 -
451333 -          ..
451334 -         Scrolled to the bottom of the record and deleted the lins
451335 -         from the initial blank record.
451336 -
451337 -     4.  Pressed F2 to save it.
451338 -
451339 -         No error messages.
451340 -
451341 -          ..
451342 -     5.  Pressed F1 Shift F1.
451343 -
451344 -         New task appeared in the Schedule.
451345 -
451346 -  ..
451347 - Seems to work perfectly.
451348 -
451349 - Could Gary have inadvertently closed the Schedule so there was no
451350 - where to put it.  I think this is covered, but not sure.  If we are
451351 - using ecur, instead of saving the initial launch configuration to get
451352 - back to the Schedule, then that could be one explanation, but is
451353 - pretty far fetched.
451354 -
451355 - It should not have anything to do with DOS4 issue, because everything
451356 - is internal to SDS code.
451357 -
451358 - Seems like another mystery.
451359 -
451360 -  ..
451361 - Maybe I spoke too soon.
451362 -
451363 - Just tired to open the new task in the Schedule and the program froze.
451364 -
451365 - That is a positive sign.
451366 -
451367 -  ..
451368 - Closed the SDS DOS session, and opened it again.  Tried opening the
451369 - new task again, and the system froze again.
451370 -
451371 - Opened the new task with the editor, and it shows that all of the
451372 - control stuff from the record ID on line 1 to the bottom of the
451373 - control field descriptions has been deleted.  That's a problem.
451374 -
451375 -  ..
451376 - Steppted through op to make a new task.
451377 -
451378 -     1.  During the process that creates follow up links from the new
451379 -         task to the headlines for the original record, the system
451380 -         jumped out of operation.
451381 -
451382 -     2.  This occurred because of the "Progress" issue noted above,
451383 -         ref SDS 0 YB4K, although it still isn't necessary to use the
451384 -         indented format shown above, i.e., Gary's format is okay, but
451385 -         we discovered a quirk that requires attention.
451386 -
451387 -          ..
451388 -         The reason for the crash, is that when the code finds a
451389 -         headline, it looks for an anchor above it.  If there is an
451390 -         anchor within 6 or 10 lines not sure which, it accepts that
451391 -         anchor.  If there is not an anchor within that proximity, it
451392 -         builds an anchor immediately above the top line of a series of
451393 -         headlines, if there are more than one.  In this case, the 0001
451394 -         default anchor is within the required range, so it did not
451395 -         build an anchor.  But the code handles the 0001 anchor in a
451396 -         special way.  After the process of adding anchors to all
451397 -         headlines, the follow up lines are processed, and first one is
451398 -         the 0001 default anchor, and this is removed.
451399 -
451400 -          ..
451401 -         It is later build back at the end, since it was not there, and
451402 -         since the headline Gary created relied on it, when during the
451403 -         process to update follow up links for the headline, it did not
451404 -         find that anchor because it had been removed, so...
451405 -
451406 -                     The fox ate the hen.
451407 -
451408 -          ..
451409 -         The code started doing things to the wrong parts of the
451410 -         record, resulting in deleting the top control struff.
451411 -
451412 -          ..
451413 -     3.  One solution is not to put a headline so near the
451414 -         Summary/Objective line.  If there is no need for a narrative
451415 -         to sumarize the record, then use the main record description
451416 -         as the first headline.
451417 -
451418 -         If the first headline is farther than 10 lines from
451419 -         Summary/Objective the then code will build an anchor, if one
451420 -         is not provided by the User.
451421 -
451422 -          ..
451423 -         Another solution is to use F1 F8 to create an outline level
451424 -         to begin the narrative below the Summary/Objective section,
451425 -         since F1 F8 automatically creates an anchor.
451426 -
451427 -          ..
451428 -     4.  Another solution is to change the code, so the 0001 anchor is
451429 -         not removed.
451430 -
451431 -         Did this below. ref SDS 0 6X7F
451432 -
451433 -         Testing on Gary's original record shows these code changes
451434 -         seem to handle the format Gary used.
451435 -
451436 -
451437 -
451438 -  ..
451439 - Line 1110, ref OF 1 3692, -label lnkthm in 004 about 70 lines below
451440 -
451441 -    ins_text!{0001 Follow up!
451442 -    join
451443 -    loc_cur 0 34
451444 -
451445 -       Changed this code to include 0001 anchor to fix problem reported
451446 -       above. ref SDS 0 UC5O
451447 -
451448 -       Changed loc_cur 0 28 to loc_cur 0 34 to account for 6 positions
451449 -       added for the anchor.
451450 -
451451 -  ..
451452 - Line 1180, ref OF 1 ZZ8M, -label 0000ov in 004
451453 -
451454 -    setlen 1 0
451455 -    -if @1 < 42 -goto o1sRf
451456 -
451457 -       Get length of line and if it is not long enough for there to be
451458 -       a second default follow up citation, then jump to next step.
451459 -
451460 -
451461 -  ..
451462 - Line 1090, ref OF 1 R74H, -label lnkthm in 004 about 50 lines below.
451463 -
451464 -    z 9 51
451465 -
451466 -       Changed this from 9 45 to 9 51 to accomodate the added line
451467 -       length from adding the anchor string.
451468 -
451469 -       Tested this on Gary's original record format, and it seems to
451470 -       have created a new task.
451471 -
451472 -
451473 -  ..
451474 - Line 1000, ref OF 1 A55H, -label start in 004 about 50 lines below
451475 -
451476 -    This code is adding in an 0001 anchor, so it may be redundant, or
451477 -    it may be there is reason we cannot do what we want to do in the
451478 -    subroutine above.
451479 -
451480 -    Testing seems to indicate we can unplug this stuff.
451481 -
451482 -
451483 -
451484 -
451485 -
4515 -

SUBJECTS
25.Bat and 001 DIR Command Different Format for Windows 98 Not Assemb
DIR /Z Command DOS4 /Z Switch Requires Revising SDS Code for Archivin
Pointers Global Using OTS for Sort, Select
Subject, 06112
Archive and Reports Use DIR /Z Switch for DOS4 Macros 001 and 0011 Bo
25.bat Global Pointers Archive SDS Records

5309 -
530901 -  ..
530902 - DOS4 Requires Special Switches for Compliance with MS DOS
530903 - Subject Reports and Archiving Working Better But Not Correctly
530904 - 25.bat Archive Op and Subject Reports Working Better But Not Correctly
530905 -
530906 - Gary has discovered in recent days that adding a /z param to the dir
530907 - command using the DOS4 shell to replace command.com in the MS DOS
530908 - program, makes progress on getting subject reports and the archive op
530909 - to run.
530910 -
530911 - Adding /z to dir commands in 25.bat results in getting the archiving
530912 - operation to take the approximate time that indicates it is processing
530913 - the data base, however, he is only getting the 00005 file, which is
530914 - the contacts.
530915 -
530916 -  ..
530917 - 25.bat should produce the following...
530918 -
530919 -              e: sd 08 00000 06 00001.......... 12 MB
530920 -              e: sd 08 00000 06 00003.......... 11 MB
530921 -              e: sd 08 00000 06 00004..........  3 MB
530922 -              e: sd 08 00000 06 00005..........  1 MB
530923 -
530924 -  ..
530925 - Gary indicates progress today by modifying the dir command in 001.
530926 -
530927 -     [On 021105 Gary got this list of files after switching from 4DOS
530928 -     to MS DOS. ref SDS 38 JW4F
530929 -
530930 - Changing "dir" to "dir /z" changes the structure of the command
530931 - strings.  Since these command strings are modified during program
530932 - operation, the change from dir to dir z causes the code to fail.  The
530933 - same occurs in the 0011 macro and to other macro files that control
530934 - subject reports.  This may explain some of the problems with failed
530935 - performance.
530936 -
530937 -  ..
530938 - In some respects this is similar to the HTML compliance issue that
530939 - Gary has brought forward.  Today, SDS code was modified to more
530940 - closely comply with HTML standards, even though the prior code was
530941 - rendering satisfactorily, per above. ref SDS 0 HE4U
530942 -
530943 - We need to think through a solution to the DOS4 problem in order to
530944 - enable full SDS performance on Gary's computer system.
530945 -
530946 -     [On 021105 Gary got this list of files after switching from 4DOS
530947 -     to MS DOS. ref SDS 38 JW4F
530948 -
530949 -
530950 -
530951 -
530952 -
530953 -
5310 -

SUBJECTS
Zip Update SDS with Improvements on HTML Format and Scheduling a New

5404 -
540401 -  ..
540402 - Update SDS Program on Gary's Computer
540403 -
540404 - Follow up ref SDS 36 0001
540405 -
540406 - Put an SDS update on the Internet at...
540407 -
540408 -       http://www.welchco.com/09
540409 -
540410 - ...with these changes since the last update on 021028. ref SDS 36 0001
540411 -
540412 - Improvements are described above. ref SDS 0 0001
540413 -
540414 -    [On 021103 Gary reports that he received and installed this program
540415 -    code update; SDS failed to launch. ref SDS 37 OF6F
540416 -
540417 -
540418 -
540419 -
540420 -
540421 -
540422 -
540423 -
540424 -
540425 -
540426 -
540427 -
5405 -