16 Apr
2020
16 Apr
'20
8:52 a.m.
I recently received a bunch of hardware/software documents to scan. I am
only roughly familiar with the systems covered by the docs. I have
encountered a couple situations that I would like guidance on.
1. One document is a software installation manual in a loose leaf binder
with other documents. It has a title page, tables of contents, etc.,
several chapters, and then it gets interesting. It has several appendix
sections (starting at A), an index, then more appendix sections
(starting at A as well), and then another index. The document title and
its font match of the second set of appendix sections and second index
matches the table of contents and chapters. The first set of appendix
sections and index are a little different. The topics covered in the
sets of appendix sections are not the same.
Should I create two different pdfs with different appendix sections or
create a single pdf with both sets?
2. One document is missing the title page and table of contents. Should
the pdf just be what I have or should I create those pages for the pdf?
Thanks,
alan
16 Apr
16 Apr
9:32 p.m.
On Thursday, April 16, 2020 at 8:52, Alan Perry via cctech wrote:
1. One document is a software installation manual in a
loose leaf
binder with other documents. It has a title page, tables of contents,
etc., several chapters, and then it gets interesting. It has several
appendix sections (starting at A), an index, then more appendix
sections (starting at A as well), and then another index. The document
title and its font match of the second set of appendix sections and
second index matches the table of contents and chapters.
I've scanned roughly 450 manuals. What you describe might be the result of
a manual update. Some updates include replacement pages, with the intent
that the replaced pages are discarded. I've encountered manuals, though,
where both the old and new pages were kept, perhaps to retain a record of
the changes.
Should I create two different pdfs with different
appendix sections or
create a single pdf with both sets?
Where both old and new pages were present, and where they could be
differentiated clearly, I made a separate PDF for each manual printing.
That is, I'd have two PDFs with the same part number with different print
dates -- one containing the old (original) pages, and the other containing
the new (replacement) pages. See, for example:
/pdf/hp/64000/hardware/64161-90901_Jan-1984.pdf
/pdf/hp/64000/hardware/64161-90901_May-1984.pdf
and the update ("Manual Change Sheet"):
/pdf/hp/64000/hardware/64161-90901-MCS_May-1984.pdf
...from which the later manual was created at Bitsavers.
2. One document is missing the title page and table of
contents.
Should the pdf just be what I have or should I create those pages for
the pdf?
If you know how the title page and TOC should appear, I would add them. If
you wish, you could add a note, such as "(Reconstructed)", as a page footer
on the added pages. I've reconstructed missing front and back cover pages
where the manual is part of a family of manuals that all share the same
basic cover and title page designs.
I think PDFs, ideally, should allow one to reprint an extinct manual in its
entirety. So I include the blank pages, covers, inserts, etc. from the
original when I make mine.
Thanks,
You're welcome.
-- Dave
11:18 p.m.
On Apr 16, 2020, at 21:32, J. David Bryan via cctech wrote:
Sometimes I have come across shrink-wrapped manuals and later updates, and
scanned them as found. I wouldn?t want to deny other people the opportunity
to apply updates to manuals, you know?
-Frank McConnell
On Thursday, April 16, 2020 at 8:52, Alan Perry via cctech wrote:
TRVTH. I used to do exactly this when HP sent updates. I put replaced
pages at the back of the manual, and usually did not refer to them thereafter.
When the binder filled up, that?s when I might consider discarding them.
HP had the habit of printing the update date and sometimes update number near
the bottom of the updated pages.
1. One document is a software installation manual
in a loose leaf
binder with other documents. It has a title page, tables of contents,
etc., several chapters, and then it gets interesting. It has several
appendix sections (starting at A), an index, then more appendix
sections (starting at A as well), and then another index. The document
title and its font match of the second set of appendix sections and
second index matches the table of contents and chapters.
I've scanned roughly 450 manuals. What you describe might be the result of
a manual update. Some updates include replacement pages, with the intent
that the replaced pages are discarded. I've encountered manuals, though,
where both the old and new pages were kept, perhaps to retain a record of
the changes. Should I
create two different pdfs with different appendix sections or
create a single pdf with both sets?
Where both old and new pages were present, and where they could be
differentiated clearly, I made a separate PDF for each manual printing.
That is, I'd have two PDFs with the same part number with different print
dates -- one containing the old (original) pages, and the other containing
the new (replacement) pages. See, for example:
11:37 p.m.
On Thursday, April 16, 2020 at 23:18, Frank McConnell wrote:
Sometimes I have come across shrink-wrapped manuals
and later updates,
and scanned them as found. I wouldn?t want to deny other people the
opportunity to apply updates to manuals, you know?
Oh yes, especially the ones that would say, "Insert the six paragraphs
below between the third and fourth lines of page 23." :-)
I used to despise HP when they would send updates that consisted of
instructions to modify existing manual pages instead of sending replacement
pages. I recall one update, maybe for the HP 64000 logic station mainframe
service manual, that instructed me to change a dozen or so schematics --
simple things, like "replace feedback resistor R23 with the active filter
circuit shown below." That one got stuck in the front of the outdated
manual, as I just couldn't bring myself to butcher the pages as they
required....
When it came to manual updates, the Logic Systems Division was aptly named.
-- Dave
11:53 p.m.
On Apr 16, 2020, at 23:37, J. David Bryan wrote:
On Thursday, April 16, 2020 at 23:18, Frank McConnell
wrote:
The folks who did the 3000 manuals didn?t do that very often. What we
got were almost always update packets that simply replaced pages in their
manuals. Once I think I remember getting a sticker to be stuck over the
replaced text on a page.
Sometimes I have come across shrink-wrapped
manuals and later updates,
and scanned them as found. I wouldn?t want to deny other people the
opportunity to apply updates to manuals, you know?
Oh yes, especially the ones that would say, "Insert the six paragraphs
below between the third and fourth lines of page 23." :-) I used to despise HP when they would send updates that
consisted of
instructions to modify existing manual pages instead of sending replacement
pages. I recall one update, maybe for the HP 64000 logic station mainframe
service manual, that instructed me to change a dozen or so schematics --
simple things, like "replace feedback resistor R23 with the active filter
circuit shown below." That one got stuck in the front of the outdated
manual, as I just couldn't bring myself to butcher the pages as they
required....
When it came to manual updates, the Logic Systems Division was aptly named.
The 3000 folks I think had to make some effort to get beyond paste-up
to where they were doing their manuals and updates in something, and it didn?t
happen until the not-so-early 1980s. Sometimes I thought the thing was
TDP/3000 and the camera-ready copy came out of a 2680A or 2688A printer. I
don?t really know how they got there or how it worked, and by the end of the
1980s I think they had moved away from TDP/3000 to something else.
-Frank McConnell
17 Apr
17 Apr
1:06 p.m.
On Thursday, April 16, 2020 at 23:53, Frank McConnell wrote:
The folks who did the 3000 manuals didn?t do that very
often. What we
got were almost always update packets that simply replaced pages in
their manuals.
To be fair, the 1000 (DSD) manuals were updated by replacement pages also.
What was odd and frustrating about the LSD folks was that a given update
might have two replacement pages and fifteen instructions to modify
existing pages.
Interesting that each division seemed to have its own rules.
The 3000 folks I think had to make some effort to get
beyond paste-up
to where they were doing their manuals and updates in something, and
it didn?t happen until the not-so-early 1980s.
The pre-1980 1000 software manuals, and pretty much all of the hardware
manuals, appeared to be typeset. Starting around 1980 through 1986 or so,
the software manuals had typeset headings and chain-line-printed body text.
After that, they appear to have been mastered on a laser printer.
Sometimes I thought the thing was TDP/3000 and the
camera-ready copy
came out of a 2680A or 2688A printer.
I recall receiving one manual that was actually printed on a 2680...on
fanfold blank letter-size paper. Had to de-perf it and punch holes for a
binder.
-- Dave
9:02 a.m.
On 4/16/20 9:32 PM, J. David Bryan via cctech wrote:
Looking at the document, my case looks more interesting.
As I noted, the two sets of appendices are mostly completely different.
There are 9 appendices that look like they are from the primary document
and 5 that look like they are from inserted pages. Both sets start
numbering from "A". There is one title shared between the two sets.
But the really interesting thing is that the copyright date on the
"primary" document is AFTER the date in the footers of the "inserted"
pages.
FYI - the document is Computervision CADDStation System Software
Installation.
So, any recommendations on what I should do?
Also, any recommendations on a software tool to slice and dice PDFs that
is inexpensive?
alan
On Thursday, April 16, 2020 at 8:52, Alan Perry via
cctech wrote:
Should I create two different pdfs with different
appendix sections or
create a single pdf with both sets?
Where both old and new pages were present, and where they could be
differentiated clearly, I made a separate PDF for each manual printing.
That is, I'd have two PDFs with the same part number with different print
dates -- one containing the old (original) pages, and the other containing
the new (replacement) pages. See, for example:
/pdf/hp/64000/hardware/64161-90901_Jan-1984.pdf
/pdf/hp/64000/hardware/64161-90901_May-1984.pdf
and the update ("Manual Change Sheet"):
/pdf/hp/64000/hardware/64161-90901-MCS_May-1984.pdf
...from which the later manual was created at Bitsavers.
10:06 a.m.
On 17/04/2020 17:02, Alan Perry via cctalk wrote:
Looking at the document, my case looks more interesting.
As I noted, the two sets of appendices are mostly completely
different. There are 9 appendices that look like they are from the
primary document and 5 that look like they are from inserted pages.
Both sets start numbering from "A". There is one title shared between
the two sets.
But the really interesting thing is that the copyright date on the
"primary" document is AFTER the date in the footers of the "inserted"
pages.
FYI - the document is Computervision CADDStation System Software
Installation.
So, any recommendations on what I should do?
I have no particular recommendation as such, other than to scan every
page and to keep them all together: if they started life as one folder
then I'd suggest that they should continue that way.
If you do that and a "better organisation" or "standard" appears
later,
it should be easy enough to post-process your PDF to arrange the pages
to meet that new standard.
Also, any recommendations on a software tool to slice and dice PDFs
that is inexpensive?
pdftk works well for managing pages and PDFs. I've used pdfshuffler too,
when I want a "visual" tool.
Antonio
--
Antonio Carlini
antonio at acarlini.com
1:41 p.m.
On Friday, April 17, 2020 at 18:06, Antonio Carlini via cctech wrote:
pdftk works well for managing pages and PDFs.
I'll second that. It's been invaluable for combining PDFs (such as when a
supplier insists on delivering its general catalog in individual sections),
as well as extracting subsets of pages.
https://www.pdflabs.com/tools/pdftk-the-pdf-toolkit/
I find that the free one works fine for my applications.
-- Dave
1:42 p.m.
On Friday, April 17, 2020 at 9:02, Alan Perry via cctech wrote:
But the really interesting thing is that the copyright
date on the
"primary" document is AFTER the date in the footers of the
"inserted"
pages.
Sometimes an update contains a replacement title page (containing the
update date), and sometimes updates supersede other updates for the same
manual. I've seen ones that have different change dates at different
points in the manual, with the title page reflecting the date of the last
update. Maybe you have something like that.
So, any recommendations on what I should do?
If you can't clearly determine whether the pages you have belong to one
manual print date or two, then it certainly wouldn't hurt simply to include
both sets of pages in one PDF. My personal goals for scanning, in
decreasing priority, are:
1. Preserve the exact representation of a manual.
2. Preserve the information in a manual in a usable format.
3. Preserve the pages of a manual.
Even if you have to fall back to #3, something might turn up later that
would allow you to go back and reorganize the pages into a more useful
arrangement or into an original format.
-- Dave
2:42 p.m.
On Apr 17, 2020, at 9:02, Alan Perry via cctalk wrote:
As I noted, the two sets of appendices are mostly
completely different. There are 9 appendices that look like they are from the primary
document and 5 that look like they are from inserted pages. Both sets start numbering from
"A". There is one title shared between the two sets.
But the really interesting thing is that the copyright date on the "primary"
document is AFTER the date in the footers of the "inserted" pages.
Do not underestimate the powers of humans to make errors when creating or applying
updates.
FYI - the document is Computervision CADDStation
System Software Installation.
So, any recommendations on what I should do?
Also, any recommendations on a software tool to slice and dice PDFs that is inexpensive?
Depends what you want to do. On a Mac, I use Preview which is the default PDF viewer; it
allows me to cut, copy and paste scanned pages from one PDF to another.
I also do some things with Perl and CAM::PDF which makes writing filter-like programs that
do page-level operations fairly easy, e.g. I have one that deletes the odd-numbered pages
from a stdin-supplied PDF and writes the resulting PDF to stdout and I think it?s about 12
lines total. Anyone care to guess what my use case for this is?
-Frank McConnell
1493
days inactive
1494
days old
10 comments
4 participants
participants (4)
-
a.carlini＠ntlworld.com -
aperry＠snowmoose.com -
fmc＠reanimators.org -
jdbryan＠acm.org