How to make Word STOP rewriting your copy

[Spider Robinson]

I am a writer. I just got Office 2004 Tonight I tried to type a list of
characters' names in Word. I was not allowed to.

I typed the name Kathy. Word offered me a suggestion to follow it,
highlighted: the last name of a friend of mine. Rather startled that Word
had ever heard of her, I hit Return, telling Word that I did NOT want that
suggested last name to be typed, and I wished to start a new paragraph
instead.

Word defied me. It typed the unwanted last name for me, and did NOT start a
new paragraph.

Naturally I hit Delete. I intended to backspace, a character at a time,
until I had removed the offending last name.

Again, Word defied me. At the first touch of the Delete key, it selected,
highlighted, and deleted THE ENTIRE NAME, both first AND last names.

So I started over. I typed Kathy. Waited. Nothing. Then I typed a blank
space. At once, Word AGAIN typed, selected and highlighted the last name of
that friend of mine whose first name happens to be Kathy. AGAIN, I was not
permitted to correct Word's error. It knew better than I did what I wanted.

After experimenting, I found that the ONLY way to delete the unsolicited
last name was to place my insertion point one character BEFORE the last
character of the name--NOT after it--then hit Delete until everything BUT
the last character was gone. THEN AND ONLY THEN was I permitted to place the
insertion point AFTER that final character and delete it as well.

I also discovered by experiment that the same thing happened any time I
typed the first name of ANYONE WHOSE NAME WAS FOUND IN MY ENTOURAGE 2004
ADDRESS BOOK. Without consulting me, Word was consulting the rest of my
Microsoft software, before deciding what I would be allowed to type.

Between them, Word and Entourage had decided I was forbidden to write about
anybody but people I knew.

Such impertinence is intolerable, so I went looking through the Help
sections of BOTH programs for a way to disable this "feature." It is
nowhere mentioned, specifically or by inference. I went to Microsoft's
website and wandered there for hours. Nothing. I e-mailed Microsoft asking
for help, and I will wait for you to stop laughing before I proceed further.

There.

Finally I asked a friend who writes software. He said it acted like a
mistaken Spell check operation. So I looked at the Spelling and Grammar
menu item, and found an obscure Options button, and tried that, and found a
place where I was offered the opportunity to check a box that would force
Word to make spelling suggestions ONLY from its own main dictionary, rather
than from any Custom dictionaries.

Perhaps, I reasoned, the boneheads that wrote this mess thought the address
books of unrelated software were custom dictionaries, even though their own
company had written that software and so they must have known better. So I
checked that box, and looked to see if I had just solved my problem.

I had not.

Well, actually, I HAD. Really.

It just didn't LOOK like I had, you see. The reason it didn't look like it
was, Word lied to me, by omission. It failed to tell me that checking that
Options box WOULD NOT TAKE EFFECT UNTIL I HAD QUIT AND THEN REOPENED WORD.

I learned, totally by accident, that I HAD successfully debugged Word of its
Character-Name-Overwriter about two hours later, when I opened it up again
to show someone the unbelievably stupid planned-bug....and found it gone.

What imbecile at Microsoft decided to make it the DEFAULT CHOICE that users
may only write the names of people they exchange e-mail with? And worse,
WHY did he make it very nearly impossible to correct his spectacularly bad
judgment?

Worst of all, why did Microsoft release the product without any beta testing
at all? The bug would have been grossly obvious within the first day to
nearly every beta tester. It simply is not possible that I am the first
user who ever wanted to type a stranger's name.

In the same way, nobody working on Entourage ever noticed that it does not
appear to be possible to insert a hotlink into an e-mail. How often does
THAT ever come up, right? Only every other message....

It actually IS possible, but Microsoft's interface "designers" worked very
hard to make it seem impossible..... and then, in a real burst of humor,
fixed things so that if you DID accidentally stumble onto the secret (and
stupid) method of inserting a hotlink, IT LOOKS LIKE YOU DIDN'T. That is,
even after you do make typed text a hotlink in Entourage, it DOES NOT turn
blue and underlined. The recipient will see it that way, but you can't,
unless you cc yourself.

While we're at it, has anyone got even a wild-ass GUESS why the geniuses who
"designed" the Office package decreed that no user may Switch Identities in
Entourage (again, something that only comes up several times a day) WITHOUT
FIRST QUITTING WORD, POWERPOINT, EXCEL AND/OR MESSENGER?

That's right: you can't use more than one e-mail account without constantly
quitting and reopening THE ENTIRE OFFICE SUITE OF SOFTWARE. In what
imaginable circumstances would that NOT be a waste of time?

This is inexcusably bad interface. This is disgraceful. Shame on whoever
said Office 2004 was ready to ship. Microsoft Word, which has been rammed
down the throats of ALL users everywhere, is now unusable for any
writer..... and even if you guess how to fix it (which is the only way to
find out), it looks like you didn't.

--Spider Robinson
Author of THE CRAZY YEARS
http://www.spiderrobinson.com

 

[Michel Bintener]

Hello again,
Word's feature (though you do not seem to consider it that way) of
auto-completing names with entries from Entourage's address book is not a
Spelling & Grammar, but an AutoCorrect option. Have a look at
Tools>AutoCorrect. The third tab, labelled "AutoText", has a checkbox which
says "Show AutoComplete tip for AutoText, Contacts, and dates"; you'll
probably want to disable that one. And while you're already there, see what
other options you'd (not) like to have enabled.

The reason why the entire name disappears when you hit backspace is because
once you've marked an entry as a contact (by hitting the hard return or
Enter button after the suggestion has been shown), this bit of text will
turn into a field, I think. (Note the purple underlining, which shows you
that it's no longer "simple" text.) Your context menu will also offer new
options if you right-/ctrl-click the new address field. By hitting
backspace, you'll delete the entire field, which explains why all of the
text goes away. If you don't want to accept Word's AutoComplete suggestion,
DO NOT hit Enter when Word proposes you a name; hitting space will make it
go away, if I remember correctly.

As for your problems with Entourage, I've already agreed with you that the
omission of an Entourage-native way of inserting hyperlinks is a big and
obvious mistake. However, based on your problems with Word, you can
understand why you have to quit all Office programs before switching
identities: the AutoComplete feature which I've described in the paragraph
above relies on the address book of the current identity, so you'll have to
quit all programs relying on an Entourage identity before you can switch to
another one.

By the way, you CAN have several accounts in one single identity; just add
them in the Tools>Account menu. I currently have five or six e-mail
accounts, and I never use another identity apart from my main one. Try it,
and I think your frustration with Entourage (and Microsoft Office 2004 in
general, which, by the way, WAS thoroughly tested) will go away (mostly).

Hope this was helpful

Michel

 

[Daiya Mitchell]

Wow, masterpiece of patience, Michel.

Spider, if you had noticed the purple dots, the Help topic "What do the
underlines in my document mean?" would have pointed you toward a clue.
Otherwise, no, it's not very well documented, and even that answer is pretty
pitiful, and not exactly a solution, and I've passed along a complaint.

Also, by the way, hitting Return generally accepts a suggestion in most
computer situations, the return key tends to double as "enter" or "ok". Try
experimenting with ESC when you want to tell Word (or any program) "no,
don't do that!" That would have worked in this situation.

Spider, I'm not sure what version of Word you were using before, but perhaps
it was 5 and you are not so familiar with the modern versions? You'll find a
link to the excellent reference "Bend Word to Your Will" on this page as
well as pointers on basic control:
http://daiya.mvps.org/wordsetup.htm
(you've suggested a good edit for that page, thanks)

Please share other frustrations with Word, in case the answers are as simple
as they were in this case.

 

[Anybody]

Hello again,
Word's feature (though you do not seem to consider it that way)

<snip>

Word has a number of "helpful features", almost all of which are a
complete pain in the bottom that simply wastes a ton of my time trying
to get it to behave properly. I know what I want to do, Word does not,
so it should leave my text just how I entered it and not keep playing
with it no matter how many times I re-correct it's "correction". :-(

Plus the peculiarly Word "bugs" like backspacing from one line to the
previous one and the whole font changes for no reason at all. :-\

 

[Daiya Mitchell]

<snip>

Word has a number of "helpful features", almost all of which are a
complete pain in the bottom that simply wastes a ton of my time trying
to get it to behave properly. I know what I want to do, Word does not,
so it should leave my text just how I entered it and not keep playing
with it no matter how many times I re-correct it's "correction". :-(

Are you the same person to whom I already recommended links to control that,
or did you see them in my previous message on this thread? Please do--it's
really not that difficult to seize control from Word, but it does require
about 75 seconds to uncheck some boxes.

Plus the peculiarly Word "bugs" like backspacing from one line to the
previous one and the whole font changes for no reason at all. :-\

No, there is a reason, this is part of Word's design. Much formatting of the
text in a paragraph is stored in the paragraph mark at the end of the
paragraph. When you backspace from one paragraph to the previous one, you
delete the first paragraph mark, and leave the second one, so the formatting
in the second paragraph mark applies to all the text.

This is easier to see if you operate with nonprinting characters on so that
the paragraph marks show up as gray ¶s. Click ¶ on the standard toolbar.

I do not know enough about programming to know whether this is bad design or
not, but it isn't actually a bug.

 

[Anybody]

Are you the same person to whom I already recommended links to control that,
or did you see them in my previous message on this thread? Please do--it's
really not that difficult to seize control from Word, but it does require
about 75 seconds to uncheck some boxes.

I do have them all turned off. At least all the ones I can find. I
simply avoid using Word as much as possible, but sometimes you have to
use it.

No, there is a reason, this is part of Word's design. Much formatting of the
text in a paragraph is stored in the paragraph mark at the end of the
paragraph. When you backspace from one paragraph to the previous one, you
delete the first paragraph mark, and leave the second one, so the formatting
in the second paragraph mark applies to all the text.

This is easier to see if you operate with nonprinting characters on so that
the paragraph marks show up as gray ¶s. Click ¶ on the standard toolbar.

I do not know enough about programming to know whether this is bad design or
not, but it isn't actually a bug.

That would be fine, and obvious, but it seems to pick up all sorts of
weird fonts, styles, tabs, indents that aren't actually used anywhere
in the document. It could be the default font, but having already done
a Select All and changed the font for the entire document, it shouldn't
be changing it at all.

Word simply has some very peculiar and annoying "features" that
thankfully AppleWorks leaves out.

 

[Jeff Wiseman]

No, there is a reason, this is part of Word's design. Much formatting of the
text in a paragraph is stored in the paragraph mark at the end of the
paragraph. When you backspace from one paragraph to the previous one, you
delete the first paragraph mark, and leave the second one, so the formatting
in the second paragraph mark applies to all the text.

This is easier to see if you operate with nonprinting characters on so that
the paragraph marks show up as gray ¶s. Click ¶ on the standard toolbar.

I do not know enough about programming to know whether this is bad design or
not, but it isn't actually a bug.

No, this is not a "bug". In general, a bug is a mistake in the
programming such that it does not meet the requirements
specification that it was designed to or intended to function as.

This is much worse. It is a GUI design FLAW, pure and simple. In
fact, it likely was NEVER spec'ed by a GUI engineer as it has all
of the earmarks of a hack by a programmer. All of the formatting
is embedded inside of this "psuedo character" (i.e., the
paragraph mark) that is susceptable to all kinds of damage being
in the middle of the character stream. Since it is an basic
infrastructure concept (i.e., the implementation for paragraph
formatting), the original hack from way back when, got totally
buried by all of the years of evolutionary support code that has
gone into Word.

In short, it is a foundation oriented, Legacy flaw that would be
virtually impossible to correct in any cost effective fashion.

All because some programmer years ago probably had free-reign on
the GUI design of the text editor.

Don't expect this one to go away. If it did, Word would, in many
ways, cease to be Word

 

[JE McGimpsey]

This is much worse. It is a GUI design FLAW, pure and simple. In
fact, it likely was NEVER spec'ed by a GUI engineer as it has all
of the earmarks of a hack by a programmer. All of the formatting
is embedded inside of this "psuedo character" (i.e., the
paragraph mark) that is susceptable to all kinds of damage being
in the middle of the character stream. Since it is an basic
infrastructure concept (i.e., the implementation for paragraph
formatting), the original hack from way back when, got totally
buried by all of the years of evolutionary support code that has
gone into Word.

What you wrote might be true if you applied it to WordPerfect. However,
you're mistaken wrt Word.

Word documents are not "character streams". They are documents composed
of "stories", such as main, header, footer, etc. The main story is
composed of sections. Sections are composed of paragraphs. Paragraphs
are composed of words, which are composed of characters (I've left some
things out, but the concept remains).

The paragraph mark is a graphical representation of the properties of
the paragraph object, such as language, style, etc. Logically and
physically, it's not in the middle of anything. Text following the
paragraph mark may be in a completely different location in the file
since it belongs to a different paragraph object.

It's an incredibly powerful and efficient method of composing and
processing documents. But it's not as intuitive for most folks as WP or
TextEdit, which do store their text as character streams. And it's
hardly a "hack".

 

[Jeff Wiseman]

What you wrote might be true if you applied it to WordPerfect. However,
you're mistaken wrt Word.

Word documents are not "character streams". They are documents composed
of "stories", such as main, header, footer, etc. The main story is
composed of sections. Sections are composed of paragraphs. Paragraphs
are composed of words, which are composed of characters (I've left some
things out, but the concept remains).

Yes, the architectural (or conceptual) structuring aren't that
bad. It's the chosen design by which they are implemented that I
have a problem with (in this case, it's the choice whereby the
paragraph object's attributes are maintained)

The "character stream" that I was referring to is the GUI stream
and not a physical storage stream (although I still believe the
paragraph marker originally had it's roots there). The way any of
these entities is presented in the GUI determines just how
"intuitive" their handling is. For example, the presentation of a
header and a footer are such that it is intuitively obvious that
they are separate entities. Even the least experienced of users
would immediately assume (correctly) that changing the attributes
of one entity can safely be done in general without effecting the
other. Paragraphs on the other hand are not as has been repeatly
shown here. The list of characters in the Main are a continuum
the way they are presented. The format control is embedded in a
character in that stream (i.e., the paragraph mark) that only
masquerades as a character but behaves totally different.
Furthermore, it ALSO doubles as a hard return. What's worst, it's
typically hidden which allows it to be accidentally "tripped
over" resulting in total surpise behavior to all but the more
experienced users.

Take Interleaf for example. From the user's interface, there can
be no mistake that each paragraph is a totally separate entity.
There are no special hidden-by-default characters that you have
to tiptoe around to avoid surprises. Interleaf has its own
problems of course, but here is an example of a strength in
maintaining a fairly intuitive GUI main body text stream.

The paragraph mark is a graphical representation of the properties of
the paragraph object, such as language, style, etc. Logically and
physically, it's not in the middle of anything. Text following the
paragraph mark may be in a completely different location in the file
since it belongs to a different paragraph object.

Again, it's not the physical location of such a beast that screws
everyone up, it's the logical placement in the GUI that creates
the impact here. The paragraph marks most certainly and
intuitively appear "in the middle of" the body text of a
document. In a product where visibility to them is turned off by
default, the poor uninitiated are most certainly going to trip
over them repeatedly thinking that they are nothing more than
hard returns.

It's an incredibly powerful and efficient method of composing and
processing documents. But it's not as intuitive for most folks as WP or
TextEdit, which do store their text as character streams. And it's
hardly a "hack".

No flames intended here, however...

From the stories I've heard over the years as well as my own
exerience, I have seen more productive time totally lost using
this tool than most others I've seen due to its idiosyncratic and
unexpected behaviors (let alone its innumerable bugs and
susceptibility to viruses). I have seen numerous folks who have
had to totally redo papers because of a simple operation of the
tool that was assumed to behave in a straightforward way had some
serious unexpected side-effect for them. I have talked to more
than one person who had totally lost a grad thesis because of a
misbehavior of the save and backup mechanism of Word.

"Efficiency" implies an economy of effort to achieve something.
When you are trying to create a document, and you are spending
all of your time contending with aspects of the tool itself and
not accomplishing your primary goal, this is not an economically
desired result of using the tool.

This is *NOT* an "efficient" tool IMHO.

As far as the claim of "powerful" goes, I do not agree with the
way that term gets used either. It seems to always be used with
the connotation that it is "good". It is implied that just
because a tool can do so MANY things (assuming that you know the
associated black arts that are required), it is "better" even
though it does so at the cost of stability, ease of use,
intuitive interfaces, and user sanity. How "powerful" is a 40
bladed swiss army knife if nobody ever carries it because it's
too big? A D9 Caterpillar bulldozer is "powerful" but if you
designed it so that the operator had to lay on his side facing to
the rear of the machine using mirrors to see evrything he is
doing and being forced to use nothing but foot and mouth
controls, only a part of the machine's operation could be
considered "powerful" (and I'm sure that is the only feature that
the marketing department for Cat would be able to push on such a
beast!). Hence Word is declared by all to be "powerful".

To me, the "power" of a tool can only be measured by how much
PROODUCTIVE work can be done AT A GIVEN COST. Even the
engineering definition for power is related to how much work can
actually be done. If you must only hire people that are expensive
experts with the tool to use it, that reduces its "power". If
people have to constantly redo things, waste time figuring out
idiosyncracies that are not intuitive, and constantly have to
explain to others how to do things as well, the "power" of the
tool becomes pretty dismal in my mind.

But then, all of this is only IMHO, of course

 

[Daiya Mitchell]

Paragraphs on the other hand are not as has been repeatly
shown here. The list of characters in the Main are a continuum
the way they are presented. The format control is embedded in a
character in that stream (i.e., the paragraph mark) that only
masquerades as a character but behaves totally different.
Furthermore, it ALSO doubles as a hard return. What's worst, it's
typically hidden which allows it to be accidentally "tripped
over" resulting in total surpise behavior to all but the more
experienced users.

Take Interleaf for example. From the user's interface, there can
be no mistake that each paragraph is a totally separate entity.
There are no special hidden-by-default characters that you have
to tiptoe around to avoid surprises. Interleaf has its own
problems of course, but here is an example of a strength in
maintaining a fairly intuitive GUI main body text stream.

By the way, my original explanation that started this conversation was
actually wrong. Or backwards or something. The general concept was correct,
but the details as I explained them were wrong.

Anyhow, here's another analogy for you--I generally find that in
Dreamweaver--where the Code view shows the characters that hold the
formatting commands and the Design view does not--I frequently have to go
into Code View to deal with getting the paragraph marks in the right place.

Nevertheless, I have few problems manipulating the hidden paragraph marks in
Word and getting them to behave, even without showing them. Go figure.

Daiya

 

[John McGhie [MVP - Word and Word Macintosh]]

Hi Jeff:

What you write is interesting, but some of it would give a misleading
impression to people who do not understand Word well.

The "Paragraph mark" actually has no relationship to the physical storage at
all. It's actually a stylised graphical representation of the mark
subeditors used to make to show typesetters where to end a paragraph.

Its placement on the screen was intended to show users where scope of effect
of the properties of that paragraph end. It has no relationship with where
the properties actually are, or the codes in the file that delimit the
paragraph. In hexadecimal, each paragraph is delimited by 0D 0A (two
characters) unless your system is using Unix line-endings (0A only...).

The properties of the paragraph, including the formatting, are nowhere near
where the paragraph mark appears on screen. The paragraph mark does not
actually exist at all in the file: it's purely a display artefact.

The formatting properties are held in a series of tables of style data
stored as part of the header structure at the bottom of the document, below
the last paragraph mark. The only thing that exists in the paragraph is a
label that identifies the row in the first table that applies to that
paragraph. The other cells in the indicated row provide the rows in other
tables that contain the various formatting properties.

It's all smoke and mirrors (binary pointers, actually, but you get the
idea...) There are several reasons why they did it, but speed and
compactness were the main ones. It also means that internally the file is
fully object-oriented, which has substantial advantages if you wish to do
clever things using computers.

I believe you have a point in that it is difficult for users who know
nothing of electronic publishing to interpret what they see Word displaying
to them, and that there could be substantial improvements made in this
display.

However, I make my living in Word all day, using it for very long documents
(in the region of a thousand pages plus). The reason I use Word almost
exclusively is because I find it very much faster and more powerful than its
competitors.

My productivity in Word is three or four times what it would be in
FrameMaker, and ten times what I was able to achieve in Interleaf. Those
products were state of their art in their day, but I consider Word to be a
far more powerful product these days.

However, and here's the thing... If I am doing the kind of job that I would
be considering Interleaf or FrameMaker for, then obviously what I am doing
is very large and very complex. There would be no point in considering
those products if it weren't. Learning to reliably publish a 2,500 page file
on time and under budget is not something you are going to do in a weekend.
It's a large and complex job of work.

Those of us who, like me, have spent the past 20 or 30 years doing this for
a living have few problems with Word. No more problems than we have with
Interleaf or FrameMaker But it's not easy, and one doesn't know how,
you will have problems and lose data.

I liken it to flying a jumbo jet. I can fly a light aircraft (well, I
haven't killed myself yet...). But if you find yourself sharing an airliner
with 400 other souls, I suggest that you would prefer to have a qualified
747 captain at the controls. Word tries to be simple enough to enable
untrained users to create short documents easily. But when you are dealing
with long complex documents such as theses, you need a very powerful tool.
Word is also that, but like any other power tool, in untrained hands it
enables one to make a very costly mistake very quickly

We hang around in this group, and we look forward to passing our knowledge
on to those who want it. But we never pretended it was easy. The only
thing *I* will say is that when I do this, I am glad I have Word to do it
in, because it becomes a whole lot harder and slower in those other two
products

Cheers

Yes, the architectural (or conceptual) structuring aren't that
bad. It's the chosen design by which they are implemented that I
have a problem with (in this case, it's the choice whereby the
paragraph object's attributes are maintained)

The "character stream" that I was referring to is the GUI stream
and not a physical storage stream (although I still believe the
paragraph marker originally had it's roots there). The way any of
these entities is presented in the GUI determines just how
"intuitive" their handling is. For example, the presentation of a
header and a footer are such that it is intuitively obvious that
they are separate entities. Even the least experienced of users
would immediately assume (correctly) that changing the attributes
of one entity can safely be done in general without effecting the
other. Paragraphs on the other hand are not as has been repeatly
shown here. The list of characters in the Main are a continuum
the way they are presented. The format control is embedded in a
character in that stream (i.e., the paragraph mark) that only
masquerades as a character but behaves totally different.
Furthermore, it ALSO doubles as a hard return. What's worst, it's
typically hidden which allows it to be accidentally "tripped
over" resulting in total surpise behavior to all but the more
experienced users.

Take Interleaf for example. From the user's interface, there can
be no mistake that each paragraph is a totally separate entity.
There are no special hidden-by-default characters that you have
to tiptoe around to avoid surprises. Interleaf has its own
problems of course, but here is an example of a strength in
maintaining a fairly intuitive GUI main body text stream.




Again, it's not the physical location of such a beast that screws
everyone up, it's the logical placement in the GUI that creates
the impact here. The paragraph marks most certainly and
intuitively appear "in the middle of" the body text of a
document. In a product where visibility to them is turned off by
default, the poor uninitiated are most certainly going to trip
over them repeatedly thinking that they are nothing more than
hard returns.




No flames intended here, however...

From the stories I've heard over the years as well as my own
exerience, I have seen more productive time totally lost using
this tool than most others I've seen due to its idiosyncratic and
unexpected behaviors (let alone its innumerable bugs and
susceptibility to viruses). I have seen numerous folks who have
had to totally redo papers because of a simple operation of the
tool that was assumed to behave in a straightforward way had some
serious unexpected side-effect for them. I have talked to more
than one person who had totally lost a grad thesis because of a
misbehavior of the save and backup mechanism of Word.

"Efficiency" implies an economy of effort to achieve something.
When you are trying to create a document, and you are spending
all of your time contending with aspects of the tool itself and
not accomplishing your primary goal, this is not an economically
desired result of using the tool.

This is *NOT* an "efficient" tool IMHO.

As far as the claim of "powerful" goes, I do not agree with the
way that term gets used either. It seems to always be used with
the connotation that it is "good". It is implied that just
because a tool can do so MANY things (assuming that you know the
associated black arts that are required), it is "better" even
though it does so at the cost of stability, ease of use,
intuitive interfaces, and user sanity. How "powerful" is a 40
bladed swiss army knife if nobody ever carries it because it's
too big? A D9 Caterpillar bulldozer is "powerful" but if you
designed it so that the operator had to lay on his side facing to
the rear of the machine using mirrors to see evrything he is
doing and being forced to use nothing but foot and mouth
controls, only a part of the machine's operation could be
considered "powerful" (and I'm sure that is the only feature that
the marketing department for Cat would be able to push on such a
beast!). Hence Word is declared by all to be "powerful".

To me, the "power" of a tool can only be measured by how much
PROODUCTIVE work can be done AT A GIVEN COST. Even the
engineering definition for power is related to how much work can
actually be done. If you must only hire people that are expensive
experts with the tool to use it, that reduces its "power". If
people have to constantly redo things, waste time figuring out
idiosyncracies that are not intuitive, and constantly have to
explain to others how to do things as well, the "power" of the
tool becomes pretty dismal in my mind.

But then, all of this is only IMHO, of course

--

Please reply to the newsgroup to maintain the thread. Please do not email
me unless I ask you to.

John McGhie <[email protected]>
Microsoft MVP, Word and Word for Macintosh. Consultant Technical Writer
Sydney, Australia +61 4 1209 1410

 

[Jeff Wiseman]

By the way, my original explanation that started this conversation was
actually wrong. Or backwards or something. The general concept was correct,
but the details as I explained them were wrong.

I think I understand what you are getting at, althought issues
with the very ability to describe what's going on tends to prove
my point IMHO

Anyhow, here's another analogy for you--I generally find that in
Dreamweaver--where the Code view shows the characters that hold the
formatting commands and the Design view does not--I frequently have to go
into Code View to deal with getting the paragraph marks in the right place.

This is one of the issues I've talked a bit on. Which is the
"physical" reality and which is "virtual"? I do not know
Dreamweaver but it sort of sounds like the Code view is a type of
physical reality--that's why its called a code view. It contains
a representation of a type of physical layout of the text and its
controls. The old markup languages did this. As the stream was
processed a control would be hit and the mode of processing the
stream would change at that point. HTML works like this. It's one
reasonable way to hanld this since the intuitive nature of
handling a stream of text is closely related to a useful physical
implementation of it. However, when you start getting into the
more complexities of layout control, obviously this could start
to become onerous.

Nevertheless, I have few problems manipulating the hidden paragraph marks in
Word and getting them to behave, even without showing them. Go figure.

And you are here because you are in the minority catagory of
"experts" so that the rest of us can benefit!

What I was suggesting was that for you to sucessfully do what you
do, somewhere along the line you had to UNLEARN what would have
been intuitive to you in order to learn the actual way Word's GUI
functioned. Once you had wrestled with it and understood all of
its detailed idiosyncrasies, it becomes second nature.

The GUI engineer needs to design to meet the human factor. The
human mind seems to comprehend in terms of things or "objects"
and their relation to each other (does the term "object orient"
sound familiar When one of these objects presented to us is
not very cohesive, or the relationship between things we are
trying to understand has been reduced to its most complex form,
we stumble over them and get confused. Their presentation to us
doesn't map well into our thinking structure so we are forced to
change how we think about it into a less efficient way.

It doesn't mean we can't do it, only that it takes more effort
and once your done, you'll find plenty of others that can benefit
from the effort you made. GUIs that are structured well tend to
not need this as much. For example, the original Apple Tool Box
used by programmers had a text editing function. All application
programmers we required to use it. Because of this structure,
once a person knew that double-clicking on a word would highlite
the entire word, this worked on EVERYTHING. In the PC world,
these rules were different on every application and even varied
within a given application.

You can put either a good or a bad GUI on a well structure
application or system. You can only put a poor GUI on a poorly
structured system because on a poorly structured system you are
forced to structure the GUI around the internal archtectural
nuances of the code. When you have a poorly structured product,
you are forced to literally learn aspects of the code's
strangeness in order to operate it since those idiosyncracies are
forced to exist in the user's interface! That is why when there
is a hack of some sort put in a product, it typically shows up in
some impact to the user interface.

A good systems engineer/analyst can tell you a lot about the
structure of the code of a product by simply looking at its
interface and how it behaves. E.g., on the original Macintosh, if
there was a bug in the text editing behavior (say, the double
click on a word trick), that bug would show up EVERYWHERE basic
text editing functions were used (which, by the way is an example
of why properly structured code is EASY to find bugs in and correct).

The bottom line is that given the time and hands on experience,
you can learn a product's function but with a good structure and
it's associated GUI, learn the product is always much easier with
fewer irritating surprises.

 

[Jeff Wiseman]

Hi Jeff:

What you write is interesting, but some of it would give a misleading
impression to people who do not understand Word well.

The "Paragraph mark" actually has no relationship to the physical storage at
all. It's actually a stylised graphical representation of the mark
subeditors used to make to show typesetters where to end a paragraph.

<<stuff deleted>>

Thanks for pointing that out. I thought that I had emphasized
that I was speaking of the virtual structure of the Graphic User
Interface and not the internal storage structure, but apparently
I may not have conveyed that very well.

It's all smoke and mirrors (binary pointers, actually, but you get the
idea...) There are several reasons why they did it, but speed and
compactness were the main ones. It also means that internally the file is
fully object-oriented, which has substantial advantages if you wish to do
clever things using computers.

All very true. I doubt that they could have gotten it as far as
they have without this. However, building a product in an
object-oriented fashion still does not eliminate problems caused
by improperly fragmented objects or objects with very low
cohesion (typical of legacy systems that were not originally
structured very well).

However, I make my living in Word all day, using it for very long documents
(in the region of a thousand pages plus). The reason I use Word almost
exclusively is because I find it very much faster and more powerful than its
competitors.

Ah, but you have templates designed to avoid the Word pitfalls,
you know all the rules to keep from tripping over idiotic design
decisions in the product, you have a handle on which products
don't work for certain behaviors on certain platforms. You don't
represent the masses who are using word for no other reason than
the director of their division at work has a PC at home and wants
everyone in his division to use the same so that he doesn't have
to learn anything new

For you, your statement is obviously true, but I'm not yet
convinced that it is necessarily true for the bulk of the people
force to use Word. OR even if you were forced to share
maintenance of a large document set with other less experienced
writers--especially if any of them decided that wanted to start
formatting things differently, etc.

When you say competitors, first of all I assume that you are
including most "markup" type processors. I agree whole heartedly.
As far as other full formatting type tools, I assume that you are
specifically referring to things like Frame and Interleaf which
you mentioned. Comments on this below:

My productivity in Word is three or four times what it would be in
FrameMaker, and ten times what I was able to achieve in Interleaf. Those
products were state of their art in their day, but I consider Word to be a
far more powerful product these days.

Actually, my experience so far is the opposite, however I've not
yet been required to build a formally organized 1000 page
document set with a team of writers so with due respect, in need
to reserve final judgment.

Note that I believe Interleaf is in the same cataegory as Word
when it comes to the sheer number of bugs and faults in the
software. It is entirely likely that Interleaf far exceeded Word
in this respect. However, due to Interleaf's functional
structure, it is possible for a system administrator that knows
Interleaf to protect the end user from many of these. Word seems
to have had a lot more things that could cause greater damage IMHO.

However, and here's the thing... If I am doing the kind of job that I would
be considering Interleaf or FrameMaker for, then obviously what I am doing
is very large and very complex. There would be no point in considering
those products if it weren't. Learning to reliably publish a 2,500 page file
on time and under budget is not something you are going to do in a weekend.
It's a large and complex job of work.

Your point is well taken but I do disagree a bit with some of
this. If you are working on large projects (I've been involved
with 20,000-35,000 page document sets), then Word is a total
impossibility. I've seen it tried. Real BAD idea! But once the
"support" is in place for something like Interleaf, doing
one-offs such as letters or memos is IMHO far less problematic
than Word ever was.

Let me explain. You and others here have the skills to develop
templates and follow a series of rules that allow you to
accomplish what you must for your documents. What do others do
who do not have these advanced skills and who just want to make a
simple autonumbered list in a memo when the numbers keep changing
on them due to the very issues that have been discussed here in
recent times? Even if there are folks on staff to provide
templates for such things, the users still need to learn all of
the "gotchas" in order to do general work. In something like
Interleaf, all those rules and templates can be locked up in a
way that the casual user doesn't have to worry about. In general,
there is not a lot of special rules a person needs to learn.

With Word, if a user doesn't like the templates they've been
handed, they can tweak to their heart's delight and the next
person to have to come maintain their document will be totally
lost not understanding what the previous person was trying to do
with it formatting and layout wise. I've found that to a large
degree Interleaf just doesn't present those problems.

However, obviously the success of both depends on knowledgable
people setting up the "rules". If those people aren't there, Word
has the strength that each person is for theirself and those with
more knowledge can excel and help others. With Interleaf, your
stuck with the corporate "rules" regardless of how ambiguous or
screwy they may be. IF they are reasonable though, it just seems
that the average joe trying to type up a proposale to distribute
has less problems using Interleaf than Word, depending on thier
background experience, of course.

For individual projects where you are the only one controlling
your documents, I believe that you may be correct--Word might
normally be a better choice. But for almost any type of corporate
function (ignoring licensing costs), something like Interleaf can
function better in a lot of cases--even if the documents are small.

Again, a lot of my feelings on this are based on the fact that
the bulk of people out there being forced to use these tools
either don't have the advanced skilles to do averything "right"
or don't have the inclination (or time) to learn. It's not for
situations where given that with a person expert in the tool,
which should he use...

untrained users to create short documents easily. But when you are dealing
with long complex documents such as theses, you need a very powerful tool.
Word is also that, but like any other power tool, in untrained hands it
enables one to make a very costly mistake very quickly

Agreed, and again my point is that the bulk of people forced to
use this as a tool to some degree or another have "untrained
hands" and so costly mistakes keep being made many of which have
nothing to do with the "power" of the tool as they have with some
nonintuitive or buggy behavior of it. A tool like Interleaf
allows someone who IS highly trained to set things up in a way
that untrained folks don't have to learn much to accomplish much.
Again, though, that assumes there is someone around to set it up
for them, a big IF.

Don't misunderstand me. I am NOT saying Interleaf is better than
Word (nope, not gonna go there Interleaf's licensing was
ridiculously expensive and when I look at the documented list of
known bugs in Interleaf 6 there were like 1200 listed (can't
compare to Microsoft's list because they won't show the whole
thing to anyone but there are some very distinct FUNCTIONAL
and ease of use type advantages to that product that have nothing
to do with document size.

Note, I have a love-hate relationship with Interleaf since it has
helped provide me with a little job security on occasion. I have
been able to set things up on the systems in ways that users
could be protected from many issues without even having to read
or implement any kinds of rules at all. It's a little easier to
to this kind of thing with Interleaf from what I've seen.

We hang around in this group, and we look forward to passing our knowledge
on to those who want it. But we never pretended it was easy. The only

When you go through so much pain to learn something, I've found
it quite gratifying to be able to share what you learned with
others to help them avoid such trouble. You know..."if only
someone had told ME that before..."

You're here passing it on BECAUSE it wasn't easy, right?!

thing *I* will say is that when I do this, I am glad I have Word to do it
in, because it becomes a whole lot harder and slower in those other two
products

I do believe that If I had the same level of knowledge on Word as
I have on other tools, then I would agree mostly with this. It is
only solidified for me though due to the purchase price and
interoperability with other tools. That is why I watch this board
now.

My biggest gripe with Word at the present is that so many of the
simplest type operations have so many significant side effects
that an inexperienced user must learn to use them safely without
surprised. Microsoft's policy for having so many of these
"side-effects" turned on by default doesn't help.

 

[John McGhie [MVP - Word and Word Macintosh]]

Hi Jeff, Goodbye Weekend...

Thanks for pointing that out. I thought that I had emphasized
that I was speaking of the virtual structure of the Graphic User
Interface and not the internal storage structure, but apparently
I may not have conveyed that very well.

Either than or I missed it. I have to read two or three hundred posts in a
session, I have been known to miss the odd nuance here and there

All very true. I doubt that they could have gotten it as far as
they have without this. However, building a product in an
object-oriented fashion still does not eliminate problems caused
by improperly fragmented objects or objects with very low
cohesion (typical of legacy systems that were not originally
structured very well).

If that's the same as saying that parts of the Word UI sux, then I would
have to agree. Sure: If they were doing it today, they would do it better
They are using Longhorn (sorry: "Windows Vista") as an excuse to do
some major and overdue changes.

As with anything else in computing, in Word there's a zillion things they
would "like" to do better. But getting management to part with the
development funding to enable that is a matter of making a business case to
demonstrate that they would make more profit if they made the changes. When
you have a 90 per cent market share already, it's really difficult to
demonstrate "extra sales" Far be it from me to reveal the actual
political chicanery that goes on within the Microsoft product divisions, but
suffice to say they are like any other major corporation, a foetid
sequestration of fiefdoms all busily engaged in mutual genocide. Don't we
just love corporations? Right at the moment, Office Product Group has this
golden excuse: "Well, we gotta rewrite it because it's not 64-bit, and it's
gotta be 64 bit to work in Windows Vista..." Marketing doesn't know enough
to to spot the fallacy in this argument, so we're getting a whole raft of
badly needed and long-overdue rework being done in the name of making Office
"Vista Compatible". Now, I have no intention of letting MS Marketing in on
the secret, and I and the rest of the Microsoft Office users of the world
would deeply appreciate it if you didn't tell them either... Not until
Office 14 ships -- by that time we should have gotten most of the goodies we
have been waiting for. Ummm... The next version to ship is 12, so nobody
wake Marketing until three versions from now, OK?

Ah, but you have templates designed to avoid the Word pitfalls,
you know all the rules to keep from tripping over idiotic design
decisions in the product, you have a handle on which products
don't work for certain behaviors on certain platforms.

Yes. I do.

For you, your statement is obviously true, but I'm not yet
convinced that it is necessarily true for the bulk of the people
force to use Word.

It's not. That is the core of YOUR argument, and I happen to strongly agree
with it. Word works very well as a tradesman's tool for industry
professionals with years of experience. It does NOT work well for the
average corporate user.

OR even if you were forced to share
maintenance of a large document set with other less experienced
writers--especially if any of them decided that wanted to start
formatting things differently, etc.

I am. That *IS* what I do for a living. It's one of the reasons I specify
Word: I need to be able to suck in text from the rest of the corporation.

When you say competitors, first of all I assume that you are
including most "markup" type processors. I agree whole heartedly.
As far as other full formatting type tools, I assume that you are
specifically referring to things like Frame and Interleaf which
you mentioned.

By Competitors to Word, I really mean FrameMaker, WordPerfect and that's
about it. I thought Interleaf had largely disappeared? I certainly haven't
seen it used for a long while. FrameMaker is also disappearing. There's a
series of SGML or XML-based products appearing now (ArborText has been
around for a while...) but Word now offers XML.

Actually, my experience so far is the opposite, however I've not
yet been required to build a formally organized 1000 page
document set with a team of writers so with due respect, in need
to reserve final judgment.

Your point above applies exactly: Word is fine if you set your project up
properly before you begin. If you don't know what you are doing, you will
get into just as much trouble as you will with FrameMaker or Interleaf

Your point is well taken but I do disagree a bit with some of
this. If you are working on large projects (I've been involved
with 20,000-35,000 page document sets), then Word is a total
impossibility. I've seen it tried. Real BAD idea!

Did one of those last year, and one the year before. It's no big problem.
Once you get above a thousand pages, you have to use a raft of
"professional" techniques. You end up driving Word very much the same way
as you would drive FrameMaker on a project of that size. And if you do, and
take the time to customise it properly and build your templates to be
robust, Word will hum sweetly along with projects of unlimited size.

The project I worked on at the beginning of this year was 2,500 pages, 500
documents, 40 completely untrained authors (subject matter experts with no
idea how to use Word). We published directly to PDF from Word and printed
in four-colour.

Let me explain. You and others here have the skills to develop
templates and follow a series of rules that allow you to
accomplish what you must for your documents. What do others do
who do not have these advanced skills and who just want to make a
simple autonumbered list in a memo when the numbers keep changing
on them due to the very issues that have been discussed here in
recent times?

Good question. I shall be in Redmond Seattle in a month's time asking this
very question of the people who made Word. They will be doing their best to
avoid me. My name is mud in Redmond and they have been trying to avoid me
now for several years I have a sadly undeserved reputation for
apoplectic table-thumping and corridor karate in Redmond. Completely
unjustified, as anyone here would hasten to assure you!! But we do have a
substantive fix to one of the most worrisome issues in Word in the pipeline
for the next version. Regrettably I am under NDA not to tell you what it
is, but this is one we have been waiting for for a long time ...

In something like
Interleaf, all those rules and templates can be locked up in a
way that the casual user doesn't have to worry about.

Yeah. That arrived in the last version in PC Word. They stuffed it up so
it doesn't work right. I shall "mention" this to them...

However, obviously the success of both depends on knowledgable
people setting up the "rules". If those people aren't there, Word
has the strength that each person is for theirself and those with
more knowledge can excel and help others. With Interleaf, your
stuck with the corporate "rules" regardless of how ambiguous or
screwy they may be. IF they are reasonable though, it just seems
that the average joe trying to type up a proposale to distribute
has less problems using Interleaf than Word, depending on thier
background experience, of course.

I wanna finish this this weekend, so I am not going to bite... Yes I am...
This is a philosophical issue. I can lock down Word exactly as you
describe. If I do, users don't like it. Neither do I. It's a trade-off.
If it is utterly critical, I *do* lock Word down that hard, but I haven't
done a project that needed that for a long time. These days I tend to allow
a far greater degree of freedom, but there are times when I will lock down,
for example, the styles in use.

For individual projects where you are the only one controlling
your documents, I believe that you may be correct--Word might
normally be a better choice. But for almost any type of corporate
function (ignoring licensing costs), something like Interleaf can
function better in a lot of cases--even if the documents are small.

Yes, it can. But the implementation cost is horrendous, and the licensing
cost is frightening. At the end of the day, I found you needed an army of
very expensive people to run Interleaf successfully (experienced LISP
programmers are not cheap...)

Again, a lot of my feelings on this are based on the fact that
the bulk of people out there being forced to use these tools
either don't have the advanced skilles to do averything "right"
or don't have the inclination (or time) to learn.

Now: The fact that Word Processors have turned First-Rate Managers into
third-rate authors and fourth-rate typesetters is indeed a philosophical
point that deserves debate. I back out of that argument by saying the tool
(Word) has the power to be good. If the corporation chooses to expect
people with neither the time nor the interest to learn it to use it for
professional publication, then they get what they deserve.

A tool like Interleaf
allows someone who IS highly trained to set things up in a way
that untrained folks don't have to learn much to accomplish much.

As does Word. When companies buy Interleaf, they HAVE to spend the money to
implement it properly, otherwise it doesn't work at all When they buy
Word, they never bother implementing it at all. Most corporate templates I
see are so bad they are a liability

Don't misunderstand me. I am NOT saying Interleaf is better than
Word (nope, not gonna go there Interleaf's licensing was
ridiculously expensive and when I look at the documented list of
known bugs in Interleaf 6 there were like 1200 listed (can't
compare to Microsoft's list because they won't show the whole
thing to anyone

Including themselves Microsoft's list is a database, and depending on
how you query it, you "might" get most of the bugs we would consider to be
"Word" bugs. But with Word, 80 per cent of the work is done by things that
are bits of Office or bits of Windows, so it becomes quite a challenge to
determine which are really Word bugs, as opposed to bugs that hit Word. I
would be surprised if the Word bug list was as small as 1200, though. Word
is an older and larger piece of software than Interleaf ...

Note, I have a love-hate relationship with Interleaf since it has
helped provide me with a little job security on occasion.

Hell, my cat would have starved to death long ago if it hadn't been for
Word's orneriness

You're here passing it on BECAUSE it wasn't easy, right?!

I can't speak for the others, but after a week of being paid to produce **it
for idiots, it's nice to come in here and do something useful

My biggest gripe with Word at the present is that so many of the
simplest type operations have so many significant side effects
that an inexperienced user must learn to use them safely without
surprised. Microsoft's policy for having so many of these
"side-effects" turned on by default doesn't help.

Umm... Yes... I have mentioned this too, once or twice... I am currently
compiling the "Word Wish List" for us to take to Redmond for this year, and
guess what... That particular hardy perennial is at the top of the list yet
again.

Cheers

--

Please reply to the newsgroup to maintain the thread. Please do not email
me unless I ask you to.

John McGhie <[email protected]>
Microsoft MVP, Word and Word for Macintosh. Consultant Technical Writer
Sydney, Australia +61 4 1209 1410

 

[Jeff Wiseman]

I just realized that I hijacked this thread. It no longer has
anything to do with Word rewriting copies. Sorry about that.

However, for those masochists that are interested in some
documentation tool analysis and phylosophy, the following MIGHT
be of interest to you

If so, read on...

By Competitors to Word, I really mean FrameMaker, WordPerfect and that's
about it. I thought Interleaf had largely disappeared? I certainly haven't
seen it used for a long while. FrameMaker is also disappearing. There's a
series of SGML or XML-based products appearing now (ArborText has been
around for a while...) but Word now offers XML.

Just FYI, Interleaf was sold to a different company something
like 5 or 6 years ago (Interleaf verson 7 timeframe). I forget
their name but they put Interleaf through a major conversion and
pulled nearly all of the LISP core out replacing it with C++
programming or somesuch. This alone eliminated a large amount of
faults that existed simply due to the LISP and LISP compiler
problems. The end result also tended to run much faster. The name
was changed to "Quicksilver" and unlike Interleaf, is only
supported on the PC I believe.

Due to my employment situation, I've been out of the loop for a
couple years now but I believe that Quicksilver is still very
much alive and being used by a large number of corporations.

Framemaker is rather interesting. My understanding is that they
want to get off of some platforms (e.g., UNIX), but when they
try, huge corporations that have the tool so heavily embedded in
their documentation base (e.g. Nortel and Ericson) threaten to
sue and Framemaker continues to release.

I want to like Framemaker as it's GUI logical behaviors leads me
to believe it has been fairly well structured internally right
from the start. However, the mechanism chosen to support mulitple
document consistency and Books tends to be a manual operation
that requires more attention and understanding by the tech writer
than I feel should be necessary. However, due to their effective
structuring, I believe that a lot of these could be improved
easily, if they had the direction laid out for them.

Did one of those last year, and one the year before. It's no big problem.
Once you get above a thousand pages, you have to use a raft of
"professional" techniques. You end up driving Word very much the same way
as you would drive FrameMaker on a project of that size. And if you do, and
take the time to customise it properly and build your templates to be
robust, Word will hum sweetly along with projects of unlimited size.

The project I worked on at the beginning of this year was 2,500 pages, 500
documents, 40 completely untrained authors (subject matter experts with no
idea how to use Word). We published directly to PDF from Word and printed
in four-colour.

Well, now you've further peaked my interest It certainly
sounds as though as with Interleaf, once you have reached a
"critical mass" of knowledge, you can easily(?) set up a document
development environment whereby untrained authors can easily
create their documents and share them with others with a minimum
of training (one of the main "strengths" of Interleaf IMHO).

If this is really true, then my hope swells My problem is
that with the number of illogical infrastructure issues that are
so apparent at the basic levels of use with Word, I always had my
doubts about the support of higher level concepts. These types of
functions tend to be used less (i.e., they are understood by
fewer people and therefore not used as much) and therefore get
tested less and have fewer problem reports back to MS, and
therefore typically have more logical type bugs in them. This is
a rule of thumb that has always seemed consistent across all
software products I've ever worked with.

But if there really is a possibility that Word has these
capabilities, then I shall press on

Good question. I shall be in Redmond Seattle in a month's time asking this
very question of the people who made Word. They will be doing their best to
avoid me. My name is mud in Redmond and they have been trying to avoid me
now for several years I have a sadly undeserved reputation for
apoplectic table-thumping and corridor karate in Redmond. Completely
unjustified, as anyone here would hasten to assure you!! But we do have a

A man after my own heart!

substantive fix to one of the most worrisome issues in Word in the pipeline
for the next version. Regrettably I am under NDA not to tell you what it
is, but this is one we have been waiting for for a long time ...

Free Upgrades, Right!?!

Seriously though, IMHO the only way to effectively support the
common light-use document writer with a product that can ALSO
handle very complex issues, is to have the logical infrasturcture
modelled correctly in the first place so that the simplifed
abstractions of that architecture that must be presented to the
casual user are a clean subset of the overall. I beleive MS made
some mistakes in critical areas during the evolution of Word's
infrastructure that prevents them from doing that now. As you've
pointed out, it is a massive legacy product so it's now a matter
of patching up where you can. This usually has to be done in an
"add-on" type fashion though.

Yeah. That arrived in the last version in PC Word. They stuffed it up so
it doesn't work right. I shall "mention" this to them...

Again, what they are trying to accomplish with that is very
difficult since they are trying to implement what should be an
infrastructure item as an add-on to a legacy product. It is a
good (correct) concept with many forced compromises in the
implementation. Expect them to struggle with that one for a good
little while (but hope for their success).

I wanna finish this this weekend, so I am not going to bite... Yes I am...
This is a philosophical issue. I can lock down Word exactly as you
describe. If I do, users don't like it. Neither do I. It's a trade-off.
If it is utterly critical, I *do* lock Word down that hard, but I haven't
done a project that needed that for a long time. These days I tend to allow
a far greater degree of freedom, but there are times when I will lock down,
for example, the styles in use.

Actually, it's more an application or management issue I think
but it IS a trade-off as you said. The trick is, when you are
totally locked down, what can the basic user still do? In
Interleaf for example, they can still generate other styles if
they want but it's real easy to identify everyplace that someone
has "cheated" and change them if necessary. Other tools hog-tie
you so much you can't move or if changes are allowed, they are
hard to identify. I have assumed this (possibly incorrectly)
about Word.

Again, I'm playing the Devil's advocate here since I'm still
trying to come up to a power user's level on Word so hearing of
such possibilities is encouraging.

Yes, it can. But the implementation cost is horrendous, and the licensing
cost is frightening. At the end of the day, I found you needed an army of
very expensive people to run Interleaf successfully (experienced LISP
programmers are not cheap...)

Oh yea! However, unless you are getting into automated handling
of documents, you can get away without the LISP programmers to a
large extent. It's the licensing cost that eventually started
driving a lot of corporations to drop it.


Thanks for all of the insight you've share here, John! I look
forward to trying to master enough of Word to do a real
comparison someday.

 

[Elliott Roper]

I just realized that I hijacked this thread. It no longer has
anything to do with Word rewriting copies. Sorry about that.

However, for those masochists that are interested in some
documentation tool analysis and phylosophy, the following MIGHT
be of interest to you

If so, read on...

Just in case you are worried about boring us peasants. Don't be.
A question on the side:-
Have either of you used LaTeX for a really big document/project?

 

[Jeff Wiseman]

Just in case you are worried about boring us peasants. Don't be.

Yes but the right thing would be to start a new thread. I feared
that I was cutting into John's weekend too much so I figured we'd
try and let this one die here

A question on the side:-
Have either of you used LaTeX for a really big document/project?

Sorry but I have not used LaTeX for any kind of documentation
(although it is useful when performing other household chores

 

[Elliott Roper]

Yes but the right thing would be to start a new thread. I feared
that I was cutting into John's weekend too much so I figured we'd
try and let this one die here

It was very good while it lasted. I'll bet that John found it useful
for his discussions with Microsoft. Both sides (err) resonated with my
views on what is wrong and why.

Sorry but I have not used LaTeX for any kind of documentation
(although it is useful when performing other household chores

not to mention clothing. I'm sure its how it got its name from TeX.
which was of course written in a MIX (I keep slippin' them in) of Sail
and Lisp by Don Knuth aka God himself.

 

[John McGhie [MVP - Word and Word Macintosh]]

Hi Jeff:

I am pointedly ignoring our European Wine Connoisseur who tends to do
unmentionable things with rubber products...

Just FYI, Interleaf was sold to a different company something
like 5 or 6 years ago (Interleaf verson 7 timeframe).

Strewth! (Elliott will translate...) I *am* getting behind the times.

pulled nearly all of the LISP core out replacing it with C++
programming or somesuch. This alone eliminated a large amount of
faults that existed simply due to the LISP and LISP compiler
problems.

Oh good. Replaced them with C++'s uninitialised pointers instead, did we?
Now that *is* an advance

The end result also tended to run much faster.

Yeah, it would. C++ is bloody efficient if you know what you're doing. A
developer told me that if you know it well, you can just about 'feel' the
assembler macros under it...

Framemaker is rather interesting. My understanding is that they
want to get off of some platforms (e.g., UNIX), but when they
try, huge corporations that have the tool so heavily embedded in
their documentation base (e.g. Nortel and Ericson) threaten to
sue and Framemaker continues to release.

From observing their lack of meaningful enhancements, I came to the
conclusion that they wanted to drop the product altogether. I thought
InDesign was being pitched as a replacement for it, then I learned that
InDesign doesn't actually replace FrameMaker...

I want to like Framemaker as it's GUI logical behaviors leads me
to believe it has been fairly well structured internally right
from the start.

Oh yes! Me too Using FrameMaker for the first time, I shortly
commented to one of my co-workers that it was a really spiritual experience.
For the first time, I got the overwhelming feeling that whatever I wanted to
do, wherever I wanted to go, the application designers had been there before
me and sourced their information from someone who does MY JOB for a living!

To a Tech Writer, FrameMaker is how we live and breath. You don't have to
fight it, force it, or do any cruel and unusual things to it to make it
work. Pity it's such a bitch to live with in the modern office

However, the mechanism chosen to support mulitple
document consistency and Books tends to be a manual operation
that requires more attention and understanding by the tech writer
than I feel should be necessary.

He's talking about "Import Formats", a.k.a "How to completely and
irretrievably trash the formatting of your entire book with one
finger-slip..."

Well, now you've further peaked my interest It certainly
sounds as though as with Interleaf, once you have reached a
"critical mass" of knowledge, you can easily(?) set up a document
development environment whereby untrained authors can easily
create their documents and share them with others with a minimum
of training (one of the main "strengths" of Interleaf IMHO).

It's a bit like Interleaf... Anything in Word is 'easy' if you know how.
Learning how can be a challenge.

If this is really true, then my hope swells My problem is
that with the number of illogical infrastructure issues that are
so apparent at the basic levels of use with Word, I always had my
doubts about the support of higher level concepts. These types of
functions tend to be used less (i.e., they are understood by
fewer people and therefore not used as much) and therefore get
tested less and have fewer problem reports back to MS, and
therefore typically have more logical type bugs in them. This is
a rule of thumb that has always seemed consistent across all
software products I've ever worked with.

I find Word to be the opposite. The more arcane and rarely-used the
feature, the more likely it was designed by someone who actually understood
the problem, and the less likely it is that "Marketing" has castrated it in
an attempt to make it "easy".

Take the Bullets and Numbering functions. Please... Take them away! No,
seriously... Once you understand the mechanism, it's breathtakingly
powerful and quite eloquent. Unfortunately, Marketing has been at the user
interface. It is nearly impossible to understand it without doing a HEAP of
experimentation in VBA (down in the depths where Marketing didn't dare go.
Most of what you read in the Help is actually a very bad idea if you want to
use numbering. The user interface conspires against you. The Help system
is so simplistic it prevents the user ever achieving a reliable mind-map of
how the thing works. And the IntelliSense "Automatically stuff up your
document" "features" have a hundred ways to break it.

That 2,500-page document I automatically assembled? It contained numbered
headings and literally hundreds of numbered lists. And they all worked.

Free Upgrades, Right!?!

You get free upgrades now. And NO, this one will NOT be 'free'

I beleive MS made
some mistakes in critical areas during the evolution of Word's
infrastructure that prevents them from doing that now. As you've
pointed out, it is a massive legacy product so it's now a matter
of patching up where you can. This usually has to be done in an
"add-on" type fashion though.

You understand perfectly. It wasn't even "mistakes". It's simply the fact
that the architects of Word were busy in 1978. The product they created
then was well ahead of its time, and contained many eloquent responses to
the problems of the day. Those problems are no longer with us (Word was
initially designed to run well in 640 kb of memory on an 8 MHz
processor...). Sometimes when I want to really mash a huge document, I haul
out my copy of Word 6 for DOS. That thing was designed for a 30 MHz 80286.
Have you ANY idea how quick it is on a 3.8 GHz dual Zeon??

Again, what they are trying to accomplish with that is very
difficult since they are trying to implement what should be an
infrastructure item as an add-on to a legacy product. It is a
good (correct) concept with many forced compromises in the
implementation. Expect them to struggle with that one for a good
little while (but hope for their success).

It's not so bad. The main problem is that it's not backwards compatible.
You have quite flexible lock-down on Word 2003, but as soon as someone opens
the file on Word 2000 they can do anything they like!!

Actually, it's more an application or management issue I think
but it IS a trade-off as you said. The trick is, when you are
totally locked down, what can the basic user still do?

Type

Word has various levels of lock-down. The tightest of them restricts the
user to a defined list of styles, and will not allow changes to those
styles. The user can, however, apply direct formatting. If they do, you
can find it and remove it.

But if you get stuck in with VBA, you can lock it up tighter than a duck's
bum! If you do, you need to be very sure that you're analysis of the user's
needs is correct, because you can lock it up totally so they have no way
round you. Doing this is a lot of work: at the bank I work at, there is one
such application. It's actually Microsoft Word 2003, but neither of us
would recognise it as such: the user interface has been almost totally
replaced.

Again, I'm playing the Devil's advocate here since I'm still
trying to come up to a power user's level on Word so hearing of
such possibilities is encouraging.

Close your eyes... Now, pretend you are using FrameMaker, and do everything
you would in Frame. There you go: instant Word power user! When people ask
me "What's the best book I can get to learn to use Word?" I tell them "The
Adobe FrameMaker Handbook". Read it closely. Do everything it recommends.
Assume that everything you need is in there somewhere, and go look for it.
(Conditional Text is a little more difficult, but it can be done...)

Oh yea! However, unless you are getting into automated handling
of documents, you can get away without the LISP programmers to a
large extent. It's the licensing cost that eventually started
driving a lot of corporations to drop it.

Hmmm... C++ programmers are not "that" much more available. Sure, you can
always go buy some. The problem is the "salesman" says "Sure, you can
customise our product: it uses standard C++, and since you are a software
development company, you have a bunch of those around every water cooler."
The Pointy-Haired Boss in Purchasing nods wisely to hide his incomprehension
and signs the cheque... Some months later, the Dilbert Project Manager
fires off a memo that suggests in intemperate language that if he had any
"spare" C++ programmers, they would be coding features in the PRODUCT, not
futzing around with {Magic Documentation Solution} for the Tech Writers

VBA I can handle. It may be offensive to the sensibilities of the
Object-Oriented Mavens amongst us. And I will grant you it *is* slow
compared to the 3rd generation languages. But man is it ever powerful...

They tell me that VBA dot-Net is even nicer, but it's mind-blowingly complex
to get started, so I am hoping to retire before I am forced to learn it.
The big improvements are that it's secure, it has the entire system and
every application in its object model, you can lock it down properly, and it
compiles to object code so there's no speed penalty. Very much like,
ummm... AppleScript

Thanks for all of the insight you've share here, John! I look
forward to trying to master enough of Word to do a real
comparison someday.

Get your nose into that VBA Editor, and we'll see you back here in six
months... Actually, if you are going to investigate VBA, do it in Virtual
PC with Word 2003. The Visual Basic Development Environment in Mac Word is
totally castrated: you would never learn it in there. On the PC, it will
prompt you through, and the help is far more complete.

Key fact number one: if you are looking for DETAILED technical information
about Word in the Help, fire up the VBA editor and ask your question in
there. As soon as you switch contexts you are reading a different help
file. It's MUCH larger and MUCH more detailed.

Cheers

--

Please reply to the newsgroup to maintain the thread. Please do not email
me unless I ask you to.

John McGhie <[email protected]>
Microsoft MVP, Word and Word for Macintosh. Consultant Technical Writer
Sydney, Australia +61 4 1209 1410

 

[Jeff Wiseman]

Oh yes! Me too Using FrameMaker for the first time, I shortly
commented to one of my co-workers that it was a really spiritual experience.
For the first time, I got the overwhelming feeling that whatever I wanted to
do, wherever I wanted to go, the application designers had been there before
me and sourced their information from someone who does MY JOB for a living!

To a Tech Writer, FrameMaker is how we live and breath. You don't have to
fight it, force it, or do any cruel and unusual things to it to make it
work. Pity it's such a bitch to live with in the modern office

So you DO like it but still consider Word better in an overall view?

The thing that is sort of sad is that from the GUI behavior you
can pretty well surmise that the underlaying logical architecture
is very well structured. That means that a GUI to solve many of
the existing problems would be far easier to develop and would be
much more stable than on many other current text based products.
The developers just don't seem to be going after any of that.

He's talking about "Import Formats", a.k.a "How to completely and
irretrievably trash the formatting of your entire book with one
finger-slip..."

Heh! You got it!

Also, because the logical operation of format inheritance never
evolved past the "manual operation" type stage, there are some
inconsitancies I've found in the attribute sets that likely would
not have existed if they had gone to an automatic or
semi-automatic type inheritance. The programmers would have
immediately seen that mismatch between the local document and
global assigned attribute maps since they would have been defined
in the same data structure definitions and not in multiple
places. Bugs would have been found simply when adding more code.
THAT is one of the great advantages of a cohesively structured
product.

It's a bit like Interleaf... Anything in Word is 'easy' if you know how.
Learning how can be a challenge.

Right! But again, my belief is that when you DIDN'T know how, the
learning curve can be much steeper (better) in a supported
Interleaf environment for the basics than in a similarly
supported Word environemnt. To get the basic uninformed user to a
level of productivity can be much easier/faster in Interleaf in
many ways.

My impression is that there are far more "gotchas" when learning
Word basics than Interleaf basics. There have just seemed to
always be far more "exceptions to the rule" that a novice Word
user has to learn and guard against to avoid his thesis being
trashed in unexpected ways.

Regardless of the tool used, certain typographic logical (i.e.,
essential) functions must be learned. E.g., styles, hanging
indents, variables, etc. However, if the UI presentation of such
functions is structured similar to the logical form of the
function, it is very much easier for the uninitiated to conceive
of and learn the implimentation of the function. E.g., the
Presentation of a paragraph in Interleaf literally uses the style
name as the paragraph entity's "handle" allowing the paragraph to
be manipulated as a true single entity. To do that in Word, you
have to carefully select the entire paragraph including all
invisibles *AND* the invisible paragraph marker at the end of the
thing.

This further emphasizes the point I made way back at the start of
this thread that one of the MOST commonly manipulated entities in
Word (i.e., the paragraph thing) is *NOT* presented through the
UI as a single cohesive object. It's up to the writer to be
careful that he doesn't scatter the object's attribute across all
creation

I find Word to be the opposite. The more arcane and rarely-used the
feature, the more likely it was designed by someone who actually understood
the problem, and the less likely it is that "Marketing" has castrated it in
an attempt to make it "easy".

Usually what is happening there is that Marketing is dictating
not essential features, but rather design decisions that are not
their business. When allowed to do so, this ties the software up
in knots! Problem is, the designers don't know how to counteract
this other than sandbagging their development estimates 500%. The
solution is having good systems analysts that have excellant
object analysis and modelling skills. When you have folks like
that to spec the requirements on your systems, what usually
happens is any knee-jerk design decisions can either be blasted
out of the sky immediately with a mass of information as to the
ramifications of putting such a change in, or an immediate
alternate (read "better") solution to the same need can be
identified that is compatible with the existing essential
structures of the product--sometimes it is even discovered that a
capability is already in the product that just needs to be
documented with an application note (as opposed to adding a bunch
of new software).

However, many companies don't want the cost of such skilled
individuals. They choose to let the features get hammered out
between product line management and Design--*BAD* idea as it puts
the developers into a conflict of interest type position where
the emphasis will be on the immediately delievered
product--code--at the cost of logical stability (i.e. "flaws" and
not "bugs")

(And as an aside, this is one of the reasons that I am out of
work right now. But that's a different story

You get free upgrades now. And NO, this one will NOT be 'free'

I said "Upgrades", not "Updates". MS seemes to put out very few
Updates since they can't make any money off of them (although
this seems to have changed some in the years I didn't buy MS).
It's easier to add a new feature to a package of essential bug
fixes to then be able to charge for the release...

<<interesting stuff on earlier Word releases deleted>>

Sounds like we need a Microsoft History Channel. In the mean
time, John, you are doing great!