Replying before/after quoted text
Matthew Thomas
moment -- `Linux: Default quote pref should be set to place caret above
quoted text'.
The reported problem is that somehow the Linux build of Mozilla is
defaulting to begin message replies below the quoted text, when other
platforms are defaulting to begin above. This is actually 4xp, but it
was inevitably going to lead to a big cake-throwing session about what
the default should be. A while ago, Akkana Peck suggested that I raise
the issue in this group, and Ben Bucksch suggested the same thing in bug
20966 just now. So, here we go.
First, a little bit of history, as far as I know it. (I stand
correctable on any of this.)
Previous versions of Netscape, up to Communicator 4.03, did not even
have a preference for whether to begin replies before/after the quoted
text (which shall henceforth be referred to as `TQT'). The cursor was
always placed below TQT, and if you wanted to move the cursor up above
TQT, you did it yourself.
Just before Communicator 4.03 was shipped, a preference was introduced
for beginning replies either above or below TQT -- and, what's more, the
default setting was to begin replies above. Akkana Peck was, ah,
slightly annoyed at this, and managed to get the default for the Unix
version changed to below TQT, while the defaults on Windows and MacOS
stayed at above TQT. And that's how it's stayed until now.
Now, the big question: What should the default be? And should it even be
a preference at all?
My personal opinion is that the default should be below the quoted text,
and that this should not be a pref. I'm probably unrealistic in seeking
the latter, but I'd really like to see the former.
There is a bug open <http://bugzilla.mozilla.org/show_bug.cgi?id=12699>
on points required for Mozilla to achieve the `Good Net-Keeping Seal of
Approval' <http://www.xs4all.nl/~js/gnksa/>, a basic standard for
well-behaved Usenet clients.
The default of beginning replies above TQT was one of the points raised
when Communicator 4.5 failed to qualify for the GNKSA in 1998
<http://www.xs4all.nl/~js/gnksa/Evaluations/netscape-communicator-4.5.txt>:
|
| Netscape has adopted a extremely bad quoting convention as the
| default: instead of encouraging the highly desirable `phone
| conversation' style of interlining at the appropriate places (and
| removing irrelevant material), it puts the insertion point above
| the quoted text. Apparently this is done in conformance to existing
| other software (of the `groupware' type, based on the mistake of
| mapping concepts onto mail/Usenet that do not translate well at all).
| On Usenet, other needs have given rise to standards in reply message
| composition that are flagrantly violated by this ill-behaved followup
| creation strategy.
|
| This broken quoting convention causes trouble: confusion (it is
| unclear to which part of the quoted text the new material applies
| to), excessive quoting (people aren't made to understand that editing
| out the irrelevant material from the quoted material is important),
| and irritation (because of aggravation because of this bad behaviour,
| causing more confusion as well as more irritation in its turn).
It's my experience that this applies not only to Usenet (where concise
quoting is important for following discussions where some individual
messages may not reach your server), but also for e-mail. The more
e-mail you deal with each day, the more likely you are to need reminding
what an individual conversation is about, and having to read the bottom
half of a message before the top half can be rather tiresome.
It's worth noting that Outlook Express 5.0 for Windows defaults to
replying above TQT, but Outlook Express 5.0 for MacOS (which was, I
think, released after the Windows version) defaults to below TQT.
Let the flamewar begin ...
--
Matthew `mpt' Thomas, usability weenie
The Revolution will not start before the quoted text
Ben Bucksch
>
> <http://bugzilla.mozilla.org/show_bug.cgi?id=20966> is open at the
> moment -- `Linux: Default quote pref should be set to place caret above
> quoted text'.
Following the feedback, it has been changed to "Default should be to
begin reply after quoted text" and fixed.
However, there were objections from the UI leage and the bug was
reopened. Please read the recent comments in the bug report.
Answering lake:
"Quote before new text" is not at all about Unix* and not about news. In
news, it is an absolute requirement, but also makes a whole lot of sense
in email IMO.
What you said mostly boils down to "users don't care at all about
quotes". So, it's better to default to "don't quote" than "quote after
new text", as akk sugested. (But I personally do *not* think, not
quoting is a good solution, because I'd often had to open 2 msgs in
possibly 2 different folders for reading one msg.)
> Encouraging them to do so is making them work harder.
Not encouraging them is making the readers work harder.
> Eudora, Outlook, Outlook Express, CC:mail, & Microsoft Mail
> habituated them to this
Why should Netscape users be used to Outlook's (completely broken)
behaviour?
> There is no bad behavior for users only bad interfaces.
And "quote after new text" makes them do the wrong thing.
*It was pp in 4.x, but I don't think, this is the best solution.
J. Moreno
<mozill...@bucksch.org> wrote:
> Matthew Thomas wrote:
> >
> > <http://bugzilla.mozilla.org/show_bug.cgi?id=20966> is open at the
> > moment -- `Linux: Default quote pref should be set to place caret above
> > quoted text'.
>
> Following the feedback, it has been changed to "Default should be to
> begin reply after quoted text" and fixed.
Which is how it should be, IMO.
> However, there were objections from the UI leage and the bug was
> reopened. Please read the recent comments in the bug report.
First off, this is not a unix, window, mac, be, amiga, acorn, or any
other OS problem
The OS is totally irrelevant.
> Answering lake:
>
> "Quote before new text" is not at all about Unix* and not about news. In
> news, it is an absolute requirement, but also makes a whole lot of sense
> in email IMO.
>
> What you said mostly boils down to "users don't care at all about
> quotes". So, it's better to default to "don't quote" than "quote after
> new text", as akk sugested. (But I personally do *not* think, not
> quoting is a good solution, because I'd often had to open 2 msgs in
> possibly 2 different folders for reading one msg.)
More importantly (especially for mail, most particularly for
businesses) is that there are /old/ messages which you have to refer
back to -- a message that says "That sounds like a good idea" isn't
very useful more than a day latter.
> > Encouraging them to do so is making them work harder.
>
> Not encouraging them is making the readers work harder.
Absolutely, and no matter what is done, a certain amount of work HAS to
be put into clear communication.
> > Eudora, Outlook, Outlook Express, CC:mail, & Microsoft Mail
> > habituated them to this
>
> Why should Netscape users be used to Outlook's (completely broken)
> behaviour?
There exist a problem with quoted text -- namely that you might have
just finished reading the original 2 seconds ago, but instead of
copying broken behavior that attempts to solves one problem by creating
an even larger one, the original problem should be addressed.
I say attempts because what it actually does is reverse existing
convention in the hopes that OTHERS will copy the behavior and that
this will then benefit the current user when he reads their replies.
At best it's a long term solution that works if most others eventually
follow his lead. At worst (and I think this is the situation as it
exists today) it simply makes things worse by confusing people and
generating conflict between the two camps.
IMO, bug 35929 is a good start on a solution to the original problem,
and does so without any change by anyone else, the user sees an
immediate benefit -- IMO this removes any need to try to change the
existing convention, which is working just fine for the most part as it
is.
> > There is no bad behavior for users only bad interfaces.
>
> And "quote after new text" makes them do the wrong thing.
Yes, and the opposite /does/ the opposite -- it encourages the person
sending the message to do the work that is /necessary/ in order to keep
the communication going smoothly.
--
J. B. Moreno
Pessimist: Half empty Optimist: Half full Engineer: Wrong size glass
Greg Miller
> In article <38FBB1BF...@bucksch.org>, Ben Bucksch
> <mozill...@bucksch.org> wrote:
> > However, there were objections from the UI leage and the bug was
> > reopened. Please read the recent comments in the bug report.
I'm really pretty stunned that anyone would ever want the reply to
appear above the text being replied to. How is the reader supposed to
figure out what the message is about? That's a massive useability
problem.
--
http://www.classic-games.com/
History does not repeat itself, but it does rhyme.
*** Please limit .sigs to four lines and avoid HTML mail or posts. ***
Steve Morrison
If people are reading messages threaded, the presumably they've already
read the previous messages. If a message has 20 replies, and I read all of
them, why should I have to scan/scroll through all the text just to see the
comment?
-Steve
Chuck Simmons
>
> If people are reading messages threaded, the presumably they've already
> read the previous messages. If a message has 20 replies, and I read all of
> them, why should I have to scan/scroll through all the text just to see the
> comment?
>
> -Steve
You are not familiar with expiration? The news server I use expires text
groups in about a week and binaries in two or three days. This neans
that the whole thread is almost never available. Moreover, though I use
a threaded view for most newsgroups, I do not use threading for mail. In
mail, threading gets confusing because many people will hit reply and
then change the subject. Finally, there does not seem to be a perfect
threading algorithm. Whatever is done, there will be outliers and out of
order posts. Quoting a bit serves as a reminder of how the conversation
got to the point which a post or mail message addresses. I often want
the reminder before I read the new message. New after quoted retains
chronology and makes an incomplete thread easier to follow. This is
especially helpful if you read a large number of groups.
Chuck
--
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons chr...@webaccess.net
J. Moreno
(Text relocated below).
> Greg Miller wrote:
> >
> > J. Moreno wrote:
> >
> > > In article <38FBB1BF...@bucksch.org>, Ben Bucksch
> > > <mozill...@bucksch.org> wrote:
> > > > However, there were objections from the UI leage and the bug was
> > > > reopened. Please read the recent comments in the bug report.
> >
> > I'm really pretty stunned that anyone would ever want the reply to
> > appear above the text being replied to. How is the reader supposed to
> > figure out what the message is about? That's a massive useability
> > problem.
> If people are reading messages threaded, the presumably they've
> already read the previous messages. If a message has 20 replies, and
> I read all of them, why should I have to scan/scroll through all the
> text just to see the comment?
Because just as often as you come in and read 20 sequential replies,
you'll come in and find a reply that is 3 days old to a message that is
a week old. Hell, just 20 minutes latter makes a huge difference in
being able to read it without referring to the previous message, and it
encourages good trimming practices (note: above you could have trimmed
out my name and Ben's comments without loosing anything essential).
But you're right, that is or can be a problem -- but again, posting
against convention doesn't solve that problem for the people that are
complaining about it.
As I said in my previous message, there is a solution to that problem
and it's this: <http://bugzilla.mozilla.org/show_bug.cgi?id=35929> (if
I understand Ben right this works at the moment, there's just no UI for
it, is that right Ben?).
Ben Bucksch
>
> As I said in my previous message, there is a solution to that problem
>
> and it's this: <http://bugzilla.mozilla.org/show_bug.cgi?id=35929> (if
>
> I understand Ben right this works at the moment, there's just no UI for
>
> it, is that right Ben?).
No. However, I will try to make it work (and I have not much reason to
believe, that I will fail), especially if it helps with setting the
caret to the end by default.
Ben Bucksch
Ops, I should have checked the bug. The bug you mention (35929, "Hide
quotes") is working in my tree modulo UI, yes.
I was speaking about bug 35930
<http://bugzilla.mozilla.org/show_bug.cgi?id=35930> "Scroll to
non-quoted text" (or "Skip Quotes").
Greg Miller
> If people are reading messages threaded, the presumably they've already
> read the previous messages. If a message has 20 replies, and I read all of
> them, why should I have to scan/scroll through all the text just to see the
> comment?
Not everyone reads messages threaded, and I don't remember every
message I've read during my lifetime.
Ben Bucksch
whole msg.
Lakespur Roca wrote:
>
> Ben Bucksch wrote:
>
> > Answering lake:
> >
> > "Quote before new text" is not at all about Unix* and not about news. In
> > news, it is an absolute requirement, but also makes a whole lot of sense
> > in email IMO.
> >
> > What you said mostly boils down to "users don't care at all about
> > quotes". So, it's better to default to "don't quote" than "quote after
> > new text", as akk sugested. (But I personally do *not* think, not
> > quoting is a good solution, because I'd often had to open 2 msgs in
> > possibly 2 different folders for reading one msg.)
>
> This is not so. I did not say that they do not care about quotes. What
> I said is that they use quotes in email differently than may people
> responding to this bug seem to expect. And yes this in part because of
> the way 4.x and others work. But it is primarily because users have
> two qualities that are not going to change even with the quote at the
> bottom. These two things are a.) they will do the minimum work
> required even if it means the reciever of the message has to work
> harder, and b.) they will resist change
>
> > > Encouraging them to do so is making them work harder.
> >
> > Not encouraging them is making the readers work harder.
>
> This also I feel is incorrect for the majority of users. Given that
> the message writer doesn't edit and quote as you would like and even
> if you but the carrot at the end this won't change, the reader has to
> work harder to find the actual reply message.
Are you saying "We won't change users' behaviour (quoting unedited), so
the reader will have to work harder, whatever wo do."? I don't think,
that's right (see below).
> (Yes there will be a
> few who do change their editing habbits but they will not be the
> majority as evidenced by numerous psychological studies on human
> habituation, reactions to change and behavior studies.)
>
> > > Eudora, Outlook, Outlook Express, CC:mail, & Microsoft Mail
> > > habituated them to this
> >
> > Why should Netscape users be used to Outlook's (completely broken)
> > behaviour?
>
> Are you saying that the 4.X doesn't put the carrot at the beginnning
> like Outlook?
> Or we didn't lose customers to competators. And we don't want them
> back? I don't understand your point here.
I just didn't assume, we lost users to the competition. May I should?
Anyway,
- There're hundreds of mailers, which quote with the conversational
style.
- I don't think, "they do it the same way" is a good argument.
> To you the bahavior may be broken now but to the user this isn't. They
> don't care about "conversational style" as we see it they have
> conversational style from a more temporal point of view.
> Sure they
> want the originating material referenced to refresh their memory but
> it should not get in the way of the conversation as they experience
> it.
Is it that, what the users really want? I can well imagine so, and I
think, bug 35930
<http://bugzilla.mozilla.org/show_bug.cgi?id=35930> "Scroll to
non-quoted text" (or "Skip Quotes") being fixed and enabled by default
is, what *really* helps here: The quote doesn't get in the way, but is
still there for reference on request (read after scrolling).
The quote *does* get in the way, if positioned at the bottom, because
- Unless there's an obvious end of the msg (like the author's name), I
have to scroll down to see, if there's more text after the quote.
- I see no way, software can help me with this (other than hiding quotes
completely, what is not, what I want).
> As it is now in Navigator, the origional message is sent to some one.
> The origionator knows what they sent, they wrote it and to those that
> are copied and sent to they will read the origional message (we are
> talking Mail here) In replying do they need a repeat of what was just
> said? Given that we know most messages are not as lengthy as this or
> as detailed propably not they remember what the origional message is
> about. Now when the reply is recieved now the origionator has to wade
> through the message he wrote again and the others that were copied, if
> Reply ALL was used have to read what they just read.
See above.
> If another
> reciever was added to the replied message in your case they would get
> the thing from the beginning . That's not a bad thing but in that
> case the sender will preface the message as they might a forwarded
> message.
Hopefully.
> > And "quote after new text" makes them do the wrong thing.
>
> When you say it makes them do the wrong thing you should really ask
> yourself why you think it is wrong and why the users would think it's
> wrong. I really think from the users perspective that they would not
> agree.
Why "quote below new text" is bad:
- See above
- It makes following longer conversations impossible. You can't
reconstruct, what is a reply to what, so the quote is useless. (And
don't tell me, normal users never reply to a reply.)
Maybe somebody else can eloborate this better?
As a more general note: Why are you assuming, Netscape users were
newbies? If you look at Netcenter's statistics, most of they are
business users and propably use the software quite frequently. Let
alone, most newbies will (unfortunately) propably use Outlook Express
now, as it is "already there". Do you have contrary info?
And business users will most likely have discussions (like this one) via
email in a rapidly increasing amount.
Akkana Peck
> > Given that
> > the message writer doesn't edit and quote as you would like and even
> > if you but the carrot at the end this won't change, the reader has to
> > work harder to find the actual reply message.
You never replied to my question in the bug as to why we auto-quote at
all. Lots of users prefer not to auto-quote (I know I've heard that
from several AOL users, who prefer to copy/paste and put in their own
attributions) and I've wondered whether our defaulting of auto-quote to
on is a remnant of the 4.x enterprise focus, but I haven't heard from
anyone who actually knows.
Ben Bucksch:
> <http://bugzilla.mozilla.org/show_bug.cgi?id=35930> "Scroll to
> non-quoted text" (or "Skip Quotes") being fixed and enabled by default
> is, what *really* helps here: The quote doesn't get in the way, but is
> still there for reference on request (read after scrolling).
Ew. Then the first block of quoted material won't be visible, but all
the others will? That sounds very odd, and I expect we'd get complaints
if we made it the default.
> > As it is now in Navigator, the origional message is sent to some one.
> > The origionator knows what they sent, they wrote it and to those that
> > are copied and sent to they will read the origional message (we are
> > talking Mail here) In replying do they need a repeat of what was just
> > said? Given that we know most messages are not as lengthy as this or
> > as detailed propably not they remember what the origional message is
> > about. Now when the reply is recieved now the origionator has to wade
> > through the message he wrote again and the others that were copied, if
> > Reply ALL was used have to read what they just read.
Maybe I got lost in the attributions here, but if this is still Lake, it
seems as if you're now arguing the other side. That's exactly the point
that many of us are trying to make -- that the recipient doesn't want or
need the quoted stuff, and defaulting to putting it first (or not at
all) encourages the sender to eliminate the quoted material or at least
trim it as small as possible.
> > When you say it makes them do the wrong thing you should really ask
> > yourself why you think it is wrong and why the users would think it's
> > wrong. I really think from the users perspective that they would not
> > agree.
Why unedited quoting is wrong:
1. It makes message length expand with each message. If everyone did
this, the resulting mailbox needing to be downloaded expands as the
square of the number of messages in the conversation, instead of just
being linear. People who have slow or expensive connections, or small
disk quotas on their ISP, tend to be bothered by this and consider it
rude.
2. It is much harder to read than interleaved quoting: if the reader
wants to use this context he's spent time downloading, he has to scroll
several pages down to find the context, then scroll back up and somehow
find his previous position, then scroll down again and find the last
context then search down from there the next time he needs context, and
so forth. Mostly this is too difficult, so most people don't bother, so
it would have been better had the reply not been there at all.
...Akkana
Simon Fraser
<akk...@netscape.com> wrote:
> > Lakespur Roca wrote:
> > > Given that
> > > the message writer doesn't edit and quote as you would like and even
> > > if you but the carrot at the end this won't change, the reader has to
> > > work harder to find the actual reply message.
>
> You never replied to my question in the bug as to why we auto-quote at
> all. Lots of users prefer not to auto-quote (I know I've heard that
> from several AOL users, who prefer to copy/paste and put in their own
> attributions) and I've wondered whether our defaulting of auto-quote to
> on is a remnant of the 4.x enterprise focus, but I haven't heard from
> anyone who actually knows.
I think auto-quoting is a hang-over from traditional email/news
clients that do it by default. For example, my news client just
auto-quoted your entire message, and I deleted most of it.
By describing it as a hang-over, I'm not saying that it's a bad thing;
in fact, I think we should continue to auto-quote. There is, however,
one behaviour which can improve matters; if there is a selection,
just auto-quote the selection. A number of Mac news clients do
this, and it's a well-received feature. I believe we have a bug
on this already.
Simon
--
Simon Fraser Entomologist
sfr...@netscape.com http://people.netscape.com/sfraser/
Ben Bucksch
>
> one behaviour which can improve matters; if there is a selection,
>
> just auto-quote the selection. A number of Mac news clients do
>
> this, and it's a well-received feature. I believe we have a bug
>
> on this already.
>
*Bug#:*23394 <http://bugzilla.mozilla.org/show_bug.cgi?id=23394>
J. Moreno
Fraser <sfr...@netscape.com> wrote:
> By describing it as a hang-over, I'm not saying that it's a bad thing;
> in fact, I think we should continue to auto-quote. There is, however,
> one behaviour which can improve matters; if there is a selection,
> just auto-quote the selection. A number of Mac news clients do
> this, and it's a well-received feature. I believe we have a bug
> on this already.
Yes, <http://bugzilla.mozilla.org/show_bug.cgi?id=23394>. Also, there
are only two Mac newsreaders besides Netscape that do NOT do this
(Nuntius and rnMac) both are 3+ years old and were unusual for in not
doing this even at that time.
And in general the same thing applies with mail (although there I'm not
up on what all of them do, so I can't give specifics).
Greg Miller
>
> Maybe I got lost in the attributions here, but if this is still Lake, it
> seems as if you're now arguing the other side. That's exactly the point
> that many of us are trying to make -- that the recipient doesn't want or
> need the quoted stuff, and defaulting to putting it first (or not at
> all) encourages the sender to eliminate the quoted material or at least
> trim it as small as possible.
Well, why not add something for mind-reading recipients like Moz
has now for HTML-friendly recipients?
--
http://www.classic-games.com/
Smarter than God? No, but I'm smarter than he was when he was my age.
J. Moreno
<gmi...@classic-games.com> wrote:
> Akkana Peck wrote:
>
> >
> > That's exactly the point that many of us are trying to make -- that
> > the recipient doesn't want or need the quoted stuff, and defaulting
> > to putting it first (or not at all) encourages the sender to
> > eliminate the quoted material or at least trim it as small as
> > possible.
>
> Well, why not add something for mind-reading recipients like Moz
> has now for HTML-friendly recipients?
Because there's no way to figure out how much or little to include when
the sender does think the recipient(s) need to be able to read what has
gone before.
It could do the "mind-reading" about whether to put new text before or
after quoted text, but that would be a chancy thing (particularly in
newsgroups) and still doesn't address the real problem -- because it's
the recipient that needs to decide whether or not quote material is
necessary which again goes back to bug 35929. Now that could be
"smarted" up in all kinds of ways, but there's no rush to do so. I
think adding it as a command, maybe giving it a keyboard and/or button
shortcut, would be just fine for the moment -- additions can always be
made latter.
Greg Miller
> It could do the "mind-reading" about whether to put new text before or
> after quoted text, but that would be a chancy thing (particularly in
> newsgroups) and still doesn't address the real problem -- because it's
> the recipient that needs to decide whether or not quote material is
> necessary which again goes back to bug 35929. Now that could be
> "smarted" up in all kinds of ways, but there's no rush to do so. I
> think adding it as a command, maybe giving it a keyboard and/or button
> shortcut, would be just fine for the moment -- additions can always be
> made latter.
I was really being sarcastic when I suggested a mind reader flag.
The fundamental problem stems from deficiencies in email text
formats and protocols. One mail reader can't unilaterally fix that.
Frankly, I think this is all being blown out of proportion over a
matter of subjective opinion. I think mozilla.org should hold online
polls to settle this and similarly minor issues and just live with
the majority's choice for a default. Maybe that should be a BugZilla
feature... allow the owner of a bug to add a poll for it. :)
Matthew Thomas
>...
> <mozill...@bucksch.org> wrote:
> >
>... [reply-after-quoted-text encouraging senders to practise snippage]
> > >
> > > Encouraging them to do so is making them work harder.
> >
> > Not encouraging them is making the readers work harder.
>
> to be put into clear communication.
This is the key point, I think.
As time goes on, it's a pretty safe bet to say that the proportion of
human *conversations* -- as opposed to just orders or memos -- which are
conducted entirely in the RFC-822 medium (whether that be e-mail or
Usenet) is going to increase.
This is going to result, over time, in a continual worsening of every
problem caused by replying before the quoted text:
* difficulty for the reader in understanding the context of the message
(as they receive messages at an increasing rate, and find it harder to
remember the exact content of previous messages);
* difficulty for the reader in following the conversation in the
message (as threads start to get longer and longer, and individual
messages start quoting more and more previous messages each);
* an increase in the bandwidth and time required to download and store
messages which unnecessarily quote the entire previous message (the
same reason as above);
* an increase in the time required for the reader to page through the
entire quoted previous message, just in case the sender has actually
inserted their own comments after a particular section (since starting
the reply before the quoted text fails to alert the sender to the
desirability of snipping irrelevant material).
So yes, we should expect that the sender of a reply will snip irrelevant
quoted material. Just as we should expect that the sender of a postal
package will put enough stamps on it to cover the cost, so that the
recipient isn't left with a bill to pay. And just as we should expect
that a stranger calling us on the telephone will introduce themselves
first, so we won't be left wondering who it is.
In the postal and telephone systems for most countries, neither of those
two other behaviors I've just cited are compulsory -- they're just
social norms which help the communication system run smoothly. And just
as there are social norms for postage and telephony, so there are social
norms which the world needs to establish in order for e-mail and Usenet
to continue to work properly over the next half-century or so. This is
one of them -- treating messages like conversations, with replies
interlocuted after relevant portions of the original.
Mozilla shouldn't necessarily *force* something which is `desired
behavior' on the Internet, but where possible it should *encourage* it.
It's not just the usability of Mozilla we're talking about here, it's
the usability of the Internet as a whole.
So starting replies after the quoted text should be the default. If a
particular company or other organization finds that the majority of its
e-mail is internal, in the form of single messages or very short
conversations, then it doesn't need to play nicely with the rest of the
Internet; it can set the default to begin before the quoted text in its
own organization-specific distribution of Mozilla, and the rest of the
world need not suffer for their habits.
If snipping and interlocuting in replies is `too hard', then we should
make it easier. Why isn't there a mini-toolbox in the message
composition window, with two tools -- a text tool for normal editing,
and a snipping tool for single-drag deletion of bits of quoted text? Why
can't I press Shift+Return in the middle of a quoted line in a
plain-text composition window (to make a comment in the middle of a
paragraph of the original message), and have the second half of the line
automatically indented with spaces to restore its alignment? Why doesn't
Mozilla insert the `...'s for me whenever I snip some of the original
message? And so on.
--
Matthew `mpt' Thomas, usability weenie
http://critique.net.nz/project/mozilla/
Jennifer Glick
certain default reply behavior and now we want to change that.
The big legacy email clients that many users first learned email with defaulted
to the top when replying to a mail message. Email packages such as cc:mail and
Microsoft mail. (this doesn't apply to unix mail apps who's users are more
tech savy). And the top big email apps of today (Eudora, Netscape and Outlook
and Outlook Express) taught people that the carot defaults to the top when
replying. Look at mail replies that you get from you less tech savy friends
and relatives, they mostly tend to reply to the mail message directly on top of
the original message.
If email was still a new application to people, I see no problem defaulting to
reply at the bottom. But I am worried that email apps have already taught users
one behavior and that changing that expected behavior will cause confusion.
There would still be a pref for users to change the carot to default to the
bottom. But for more novice users (who will never find the pref if we default
it to the bottom), the default would be set to what they are already used to
and expect to see.
Matthew Thomas wrote:
> <http://bugzilla.mozilla.org/show_bug.cgi?id=20966> is open at the
> moment -- `Linux: Default quote pref should be set to place caret above
> quoted text'.
>
> --
> Matthew `mpt' Thomas, usability weenie
Dannie M Stanley
threads. My 2 major complaints with netscape mail are:
1. My signature is below the quoted text.
2. I can't see who the message was to in my inbox view (I have many
aliases).
The first thing i do when i reply to a piece of mail is scroll to the
bottom of the message cut my signature scroll back to the top and paste
it (using the annoying alt+x and alt+v rather than my KDE shortcut keys
because Netscape seems to negate them).
I hate to receive mail that has all of the quoted text above the
message. At least give us the option!
DS
--
Dannie M Stanley
Webmaster/Developer
SpinWeb Net Designs, Inc
http://www.spinweb.net
--
Dannie M Stanley
Webmaster/Developer
SpinWeb Net Designs, Inc
http://www.spinweb.net
Ben Bucksch
> I guess my big concern is that email applications have ALREADY trained users a
> certain default reply behavior and now we want to change that.
Even, if that means some pain in the short term, there's no need to
follow the mass, if it is wrong.
> But I am worried that email apps have already taught users
> one behavior and that changing that expected behavior will cause confusion.
Do you seriously think, our users won't be able to use the software
anymore, just because the caret sits at another place by default?
I guess, the confusion will hold on for only some seconds - a minor
price for the "confusion", that all those bad postings and emails and
the reactions to them cause.
> There would still be a pref for users to change the carot to default to the
> bottom. But for more novice users (who will never find the pref if we default
> it to the bottom), the default would be set to what they are already used to
> and expect to see.
I know, that powerusers can change the pref, that is not the question.
The problem is, that we would encourage bad net behaviour, from which
everybody suffers.
> Matthew Thomas wrote:
> > It's worth noting that [...] Outlook Express 5.0 for MacOS
> [...] defaults to below TQT.
... which is a change to the previous version, which sat the caret to
the top. => OEMac did exactly, what is proposed for Mozilla Mailnews.
Brian Smith
>Jennifer Glick wrote:
>
>> There would still be a pref for users to change the carot to default to
the
>> bottom. But for more novice users (who will never find the pref if we
default
>> it to the bottom), the default would be set to what they are already used
to
>> and expect to see.
>
>The problem is, that we would encourage bad net behaviour, from which
>everybody suffers.
I think that the mode that is best for the author should take precedence
over the mode that is best for the readers. I don't see the logic in saying
"we're making it harder for you to type your message so that it will be
easier for other people to read it" to a potential customer/user.
Why doesn't the newsreader just show you the message in reply mode? I see no
point in opening up a new window with the same information, it is less work
for the user, it doesn't seem like it would be bothersome for people that
don't want to respond, and you wouldn't have to worry about where to put the
caret.
- Brian
Ben Bucksch
> Lakespur Roca wrote:
> > Sure they
> > want the originating material referenced to refresh their memory but
> > it should not get in the way of the conversation as they experience
> > it.
>
> Is it that, what the users really want? I can well imagine so, and I
> think, bug 35930
... and bug 35929 being fixed ...
> is, what *really* helps here: The quote doesn't get in the way, but is
> still there for reference on request (read after scrolling).
>
> The quote *does* get in the way, if positioned at the bottom, because
> - Unless there's an obvious end of the msg (like the author's name), I
> have to scroll down to see, if there's more text after the quote.
> - I see no way, software can help me with this (other than hiding quotes
> completely, what is not, what I want).
As I said in the bug: Don't do tradeoffs - find better solutions.
My last post might be read as request to do a tradeoff towards better
net keeping. I am convinced, this tradeoff should be made, if no better
solution can be found. But I still hope, we don't have to.
Everybody, please think about ways to solve this problem to the
confidence of all parties. If I didn't miss something, the requirements
are
- Referenced text has be be above the refering text
- No unnecessary quotes
- A way to lookup the relevant text of the previous msg*
- Fast conversation, i.e. "it should not get in the way
of the conversation as they experience it."
- No dependance on a certain mail/news reader (but composer, of course)
*This is garanteed in general by the "References"/"In-Teply-To" headers.
Bugs 29653 and 37209 and a 4.5-like "References" line in the UI might
help our users here, not sure about other readers.
Ben Bucksch
> I think that the mode that is best for the author should take precedence
> over the mode that is best for the readers. I don't see the logic in saying
> "we're making it harder for you to type your message so that it will be
> easier for other people to read it" to a potential customer/user.
Can you build up your own car and just drive it on the streets? Maybe
even without a drivers licence? In most (all?) countries you can't (or
it is forbidden at least). There are requirements, cars have meet in
order to ensure the usability of the streets and the lives of
participants. Even more, there are requirements for drivers of these
cars with the same justification.
The net has no laws -- it has standards and a netiquette. GNKSA (Good
Net-Keeping Seal of Approval) <http://www.xs4all.nl/~js/gnksa> lists
requirements, news user agents (most make sense for mail user agents as
well) should meet. Most requirements target at helping users to follow
the netiquette. "Default caret before quote" constantly makes the
evaluators bashing the UA.
What would you think about a car, whose blinkers* are optional?
*I don't know the correct english word. Maybe "direction of travel
indicators" or "turn signals"?
Dan Mosedale
Jennifer Glick <jgl...@netscape.com> wrote:
> I guess my big concern is that email applications have ALREADY
> trained users a certain default reply behavior and now we want to
> change that.
>
> The big legacy email clients that many users first learned email
> with defaulted to the top when replying to a mail message. Email
> packages such as cc:mail and Microsoft mail.
As I understand it, Mozilla is currently aiming at the average net
user. Although some business users certainly learned email with
legacy corporate systems, do you believe that the today's average
net.user learned email working in such a company at a time when such
systems were popular?
As far as training goes, in what way is there a learning curve
associated with whatever placement of the caret? The user is
presented with a caret one place or the either. The only possible
choices of things to do are
a) start typing
b) click somewhere else in the window to move the caret
No matter where the caret starts, aren't both these actions
instrinically obvious?
Dan
--
Jennifer Glick
Ben Bucksch wrote:
> Jennifer Glick wrote:
> > I guess my big concern is that email applications have ALREADY trained users a
> > certain default reply behavior and now we want to change that.
>
> Even, if that means some pain in the short term, there's no need to
> follow the mass, if it is wrong.
>
> > But I am worried that email apps have already taught users
> > one behavior and that changing that expected behavior will cause confusion.
>
> Do you seriously think, our users won't be able to use the software
> anymore, just because the caret sits at another place by default?
Cause confusion, not disable users.
>
>
> I guess, the confusion will hold on for only some seconds - a minor
> price for the "confusion", that all those bad postings and emails and
> the reactions to them cause.
I guess I don't see why this is "bad" behavior. If this is how some users prefer to
email, what is wrong with that?
>
>
> > There would still be a pref for users to change the carot to default to the
> > bottom. But for more novice users (who will never find the pref if we default
> > it to the bottom), the default would be set to what they are already used to
> > and expect to see.
>
> I know, that powerusers can change the pref, that is not the question.
> The problem is, that we would encourage bad net behaviour, from which
> everybody suffers.
If you look at how novice/average folks use email, you see a definite pattern. It
is a back and forth exchange between 2 or more folks with the newest stuff always
being added to the top. NOT because this is CORRECT or INCORRECT behavior, but
because this is how they learned to use email. If we suddenly default the carot to
the bottom of the email message, do you think all these folks will suddenly start
typing their responses below? I expect they will simply scroll back to the top and
start their response there, because that is what they are used to doing.
J. Moreno
<Smi...@mail.medicine.uiowa.edu> wrote:
> Ben Bucksch wrote
> >I know, that powerusers can change the pref, that is not the
> >question. The problem is, that we would encourage bad net behaviour,
> >from which everybody suffers.
>
> I think that the mode that is best for the author should take
> precedence over the mode that is best for the readers. I don't see
> the logic in saying "we're making it harder for you to type your
> message so that it will be easier for other people to read it" to a
> potential customer/user.
This is a seriously daft idea -- the program is a tool that is supposed
to be designed to aid the user. The goal of this transaction is to
communicate with another person, the design should facilitate that over
ease of use for the user (and even more so for ease of adoption).
Doing otherwise is like making a "lighter" sledge-hammer --- sure it's
easier to pick up, but it sure doesn't do much for breaking the rocks
or driving a spike.
And you don't tell the customer that it is harder to pick up and swing,
you tell them that it works better for breaking up rocks or driving a
spike.
Because what you are currently doing is telling them "Hi, we make it
real easy to post in a way that will get you thanked by a tiny majority
(and ignored by some, insulted by others, and considered either rude or
an idiot by the majority)".
Greg Miller
> > Do you seriously think, our users won't be able to use the software
> > anymore, just because the caret sits at another place by default?
>
> Cause confusion, not disable users.
Then what confusion are you talking about?
> > I guess, the confusion will hold on for only some seconds - a minor
> > price for the "confusion", that all those bad postings and emails and
> > the reactions to them cause.
>
> I guess I don't see why this is "bad" behavior. If this is how some users prefer to
> email, what is wrong with that?
Because it's hard on the reader/victim.
> If you look at how novice/average folks use email, you see a definite pattern. It
> is a back and forth exchange between 2 or more folks with the newest stuff always
> being added to the top. NOT because this is CORRECT or INCORRECT behavior, but
> because this is how they learned to use email. If we suddenly default the carot to
> the bottom of the email message, do you think all these folks will suddenly start
> typing their responses below? I expect they will simply scroll back to the top and
> start their response there, because that is what they are used to doing.
So the worst case is that nothing changes? If most users continue to
do what they're doing now and some switch to a better way of doing
things, that's a clear win.
Ben Bucksch
>
> Ben Bucksch wrote:
> I guess I don't see why this is "bad" behavior.
See ms post <news://new.mozilla.org/38FCF4A...@bucksch.org> and
akk's followup <news://new.mozilla.org/38FE5852...@netscape.com>.
Just imagine this post, if everybody just typed everything at the top,
not making clear, what text it refers to. It would be unreadable.
> If this is how some users prefer to email, what is wrong with that?
Nothing, as long as both parties like it. That's what the pref is for.
But you should assume, that the other party does no like it, because it
is considered bad in general (see above).
If somebody sends me such a mail/post, I think,
- he is a bloody newbie (nothing wrong with that)
and propably has a mail client with broken defaults
- he has an inferiour mail client
or
- he is rude (he thinks "eat my shit")
I take the freedom to react accordingly.
> If you look at how novice/average folks use email, you see a definite pattern. It
> is a back and forth exchange between 2 or more folks with the newest stuff always
> being added to the top. NOT because this is CORRECT or INCORRECT behavior, but
> because this is how they learned to use email.
So, the reason for this problem is not convience, but just the broken
defaults? This would be a clear reason to default to "caret on bottom".
Real novices (there're still a lot to come) are not yet trained to any
behaviour and should be trained to good ones. Average folks propably
know the preferences and can change the default, if they really want.
> If we suddenly default the carot to
> the bottom of the email message, do you think all these folks will suddenly start
> typing their responses below? I expect they will simply scroll back to the top and
> start their response there, because that is what they are used to doing.
Maybe, we should look at OEMac 5.0 users? (See below.) Do they really
still write comments on top, moviung the caret manually up in every
reply?
We should assist in two ways:
- Get UI for the "(don't) auto-quote" pref
- Figure out, how to encourage breaking the blockquote in the HTML
editor and interleave the comments. Ideally, a bloody newbie should
know, that he should write the comments this way, but I have no idea how
to achieve this. Maybe some editor folks? (ccing .editor)
Ben Bucksch
> My 2 major complaints with netscape mail are:
> 1. My signature is below the quoted text.
There's no other correct way. Good user agents cut the sig (and
everything below, of course) in quotes. This means, that the recipient
would not have the option to include the text you quoted in his reply
(or it would be more work for him).
> I hate to receive mail that has all of the quoted text above the
> message. At least give us the option!
The pref "caret on top" will stay (at least in the near future), we're
just discussing about the default.
Brian Smith
> Brian Smith wrote:
> > I think that the mode that is best for the author should take precedence
> > over the mode that is best for the readers. I don't see the logic in saying
>
> Can you build up your own car and just drive it on the streets? Maybe
I knew that this kind of reply was coming, and I completely agree with
lot of the logic behind it. But I think it really is a matter of
prioriies. What is most important for Netscape Mail/News: helping people
migrate from the "improper" newsreaders that have the largest market
share? letting people use the sofware in a way that is intuitive for
them? helping people follow netiquette?
If user adoption is the #1 concern, then perhaps it belongs at the top.
If netiquette is #1, then maybe it should be on the bottom. It depends.
Maybe Mail/News should have a "netiquette tip of the day" feature? Maybe
it should warn users about typing in messages that are all-caps? I am
guessing that AOL/Netscape's priority is probably user adoption, though.
Has anybody done user-group testing to see how the people prefer to read
and write responeses? Just because GNKSA and newsgroup-gurus recommend
one way doesn't mean that way is best. Are people more likely to edit
out irrelevent content when the caret is at the top or the bottom? Which
placement is has been shown _empirically_ to be most likely in helping
posters follow netiqqette? What if corporate people prefer email to have
replies at the top (the case in my company) and usenet people prefer
replies at the bottom or in the "conversation" style? How many end-users
even care (or know about) GNKSA ratings? Probably not many.
- Brian
Brian Smith
"Why doesn't the newsreader just show you the message in reply mode? I see
no
point in opening up a new window with the same information, it is less work
for the user, it doesn't seem like it would be bothersome for people that
don't want to respond, and you wouldn't have to worry about where to put the
caret."
I was disappointed that nobody replied to the comment above (which I have
copied-and-pasted it from my other post). I think this was much more
significant than the first part of that post. Is there a reason that this is
a bad idea?
- Brian
Ben Bucksch
> I said:
> "Why doesn't the newsreader just show you the message in reply mode?
> I see no point in opening up a new window with the same information,
> it is less work for the user, it doesn't seem like it would be
> bothersome for people that don't want to respond, and you wouldn't
> have to worry about where to put the caret."
>
>
It makes sense in /some/ way: Inituitive software usually doesn't
require an extra command or even mode for editing, but makes editing
"in-place" possible. Modes are generally considered bad.
I don't completely understand, what you suggest. I see 2 possibilities:
* Go directly to compose mode, never "show" the msg.
* This makes several optimizations for reading impossible, e.g.
* The "hide quotes" enhancement etc.
* Add email address to address book
* Open URL on click
* How do you access attachments?
* How do you delete mails? A composed mail is a new one.
* Go to compose mode after a click into the msg (or after a command)
* This creates an in part modeless UI again, because there are
different commands for reading (see above) and writing mail
2 more reasons, why both suggestions might not be desirable:
* We have more than one reply/forward command.
Which one do you want to favor? I bet, GNKSA will bash us on this
one then - and they would be right.
* Last, but not least, the mail you read is another one than the one
you compose.
* This is the reason for some problems above and there're
propably more.
* It invalidates the argument about intuitivity at the
beginning. "In-place editing" would mean to alter the msg
you recieved and this is not, what you do
J. Moreno
<Smi...@mail.medicine.uiowa.edu> wrote:
> I said:
> "Why doesn't the newsreader just show you the message in reply mode?
> I see no point in opening up a new window with the same information,
> it is less work for the user, it doesn't seem like it would be
> bothersome for people that don't want to respond, and you wouldn't
> have to worry about where to put the caret."
>
> I was disappointed that nobody replied to the comment above (which I
> have copied-and-pasted it from my other post). I think this was much
> more significant than the first part of that post. Is there a reason
> that this is a bad idea?
Basically because then you confuse two different things -- their
message and your reply to it. I would think it'd be near impossible to
do this in an intuitive way -- even if it wasn't totally different from
every other mail/news program in existence.
Also -- Mozilla's current "reply" UI makes it difficult enough to
switch from one type of *reply to another as it is, making it even more
difficult or confusing would definitely NOT be a "win".
*Followup, Reply, Followup&Reply, Forward (Inline, Quoted, Attachment).
Dan Mosedale
Jennifer Glick <jgl...@netscape.com> wrote:
>
> If we suddenly default the carot to the bottom of the email message,
> do you think all these folks will suddenly start typing their
> responses below? I expect they will simply scroll back to the top and
> start their response there, because that is what they are used to doing.
>
The mailer in Netscape versions 2 and 3 did default to putting the
caret at the bottom of the mail. In Netscape version 4, the change
was made to default the caret to the top. At that time, I noticed
that many Netscape 4 users with whom I regularly corresponded who had
previously always started the replies at the bottom began typing
replies at the top directly as a result of this change in defaults.
Presumably, they changed because the path of least resistance/effort
had changed (I even did it occasionally for exactly this reason).
On what do you base your expectation that people are likely to bother
moving the caret away from its default?
--
Matthew Thomas
>
> I guess my big concern is that email applications have ALREADY trained
> users a certain default reply behavior and now we want to change that.
Yes, because changing to the more usable behavior now will be a whole
lot less painful than changing later, when Mozilla has many more users
than it does now.
> The big legacy email clients that many users first learned email with
> defaulted to the top when replying to a mail message. Email packages
> such as cc:mail and Microsoft mail.
And now they are (if OE 5 for MacOS is any guide) starting to realize
their mistake and change back, for which we can be truly grateful.
> (this doesn't apply to unix mail
> apps who's users are more tech savy).
So, are you suggesting that (a) today's non-tech-savvy users are never
going to become tech-savvy, or (b) Unix mail apps will eventually change
to have the same default setting as cc:mail and Microsoft Mail, rather
than the other way around? (Excuse me while I dip my thermometer into
Hell to see how cold it is ...)
> And the top big email apps of
> today (Eudora, Netscape and Outlook and Outlook Express) taught people
> that the carot defaults to the top when replying.
Defaults are not something you teach. Defaults are something you *use*
to teach people good behavior without forcing it on them.
If you change the default back to how it was in Netscape 1.x, 2.x, 3.x,
4.0, 4.01, and 4.02, people may take a while to readjust, but they'll
learn. They'll learn to reply at the bottom, and they'll learn to snip
unnecessary parts of the quoted text. And it will definitely be worth it
for the readers of their messages.
> Look at mail
> replies that you get from you less tech savy friends and relatives,
> they mostly tend to reply to the mail message directly on top of the
> original message.
Yes, but *only* because it's the default and they don't know any better.
And readers suffer as a result.
> If email was still a new application to people, I see no problem
> defaulting to reply at the bottom. But I am worried that email apps
> have already taught users one behavior and that changing that expected
> behavior will cause confusion.
The number of current Internet users is very small, compared to the
number of Internet users we will have in ten or even five years time.
> There would still be a pref for users to change the carot to default
> to the bottom. But for more novice users (who will never find the
> pref if we default it to the bottom), the default would be set to what
> they are already used to and expect to see.
Let me tell you a story.
Once upon a time, the Continental Congress of the United States got
together and decided that the US should have a postal service. So they
appointed Benjamin Franklin as Postmaster General, and set up a Post
Office Department. The year was 1775.
From 1775 to 1847, if you wanted to send a letter using the US postal
service, you just folded and sealed your letter, and wrote the address
of the recipient on the outside. The postal service delivered the
letter, and expected the recipient to pay for the received mail.
This was obviously the wrong thing to do. Often the recipient couldn't
be found, or was unwilling or unable to pay for it. But the postal
service still had to carry the cost of taking the letter to its
destination, so they ended up losing money. The only way they could
really recoup this money was to charge higher fees to those recipients
who *did* pay. And of course the postman's job was slowed down
considerably by the fact they had to collect money from each recipient.
In short, it was a usability nightmare ... about as bad as trying to
scan a multi-quoted e-mail message for context when replies have been
made above the quoted text instead of below it.
So in 1847, postage stamps were introduced. Sure, it was not what the
users were `used to and expected to see' (your words). In the first five
years of postage stamps in the USS, only two percent of letters were
sent using them. But people learned, and got used to using stamps. And
in 1855 they were made compulsory -- you couldn't send a letter without
putting a stamp on it.
And so the US postal service has continued ever since. In England, where
postage stamps were introduced by Rowland Hill in 1840, the option to
send mail without a stamp remained, but it was not the default -- it was
charged twice as much as pre-payment.
Just think about that for a bit. For *seventy-two years*, the default
pref for sending postal mail in the United States was wrong. But then it
was changed. And did the US postal service suffer huge usability
problems and die? I don't think so. In fact, if it had remained as it
was, in the twentieth century the large volume of unpaid-for junk mail
would have made the system costly and inefficient to a massive degree.
The default pref for sending e-mail with Mozilla has only been wrong
since version 4.03 -- about three years. We *can* still change it now,
to begin replies after the quoted text, and Internet users of the future
will thank us for it.
If we do not, however, those directly responsible for the decision will
(I think) be recorded in the history books alongside Lieutenant Colonel
W. L. Maberly of the British Post Office, who derided Rowland Hill's
original ideas for pre-payment of postage as `a most preposterous plan,
utterly unsupported by facts, and resting entirely on assumption'.
We now return you to your regularly scheduled programming.
--
Matthew `mpt' Thomas, Mozilla user interface QA
<http://critique.net.nz/project/mozilla/>
Matthew Thomas
>...
> I think that the mode that is best for the author should take
> precedence over the mode that is best for the readers. I don't see the
> so that it will be easier for other people to read it" to a potential
> customer/user.
Here's the logic:
For any given message, whether it be an e-mail message or a Usenet
message, the expected value (i.e. average) of the number of recipients
of the message will always be greater than 1.
That is, you can't send a message to fewer than 1 person, but you can
send it to more than 1 person (a *lot* more than 1 person).
Given that, the interests of the recipients necessarily outweigh the
interests of the senders -- since there are more recipients than there
are senders.
So as much as is practical, you should make life easier for the readers
before you make life easier for the writers.
And since messages take much more time to write than they do to read
anyway, a small addition to the length of time they take to write (i.e.
encouraging snippage) will have much less of a proportional effect on
the writer than an equal addition to the length of time they take to
read (from wrong-ordered quoting) will have on the reader.
Matthew Thomas
>...
> Maybe I knew that this kind of reply was coming, and I completely
> agree with lot of the logic behind it. But I think it really is a
> matter of prioriies. What is most important for Netscape Mail/News:
> helping people migrate from the "improper" newsreaders that have the
> largest market share? letting people use the sofware in a way that is
> intuitive for them? helping people follow netiquette?
I don't know about you, but I started contributing to Mozilla to try and
make the Internet a better place: one where standards-based software
such as Mozilla is dominant, and one where Net communication is
practised well -- i.e. where Netiquette is followed. And if that
involves a little bit more learning on the part of Mozilla users, that's
just fine with me.
> If user adoption is the #1 concern, then perhaps it belongs at the
> top. If netiquette is #1, then maybe it should be on the bottom. It
> depends. Maybe Mail/News should have a "netiquette tip of the day"
> feature?
`Tip #37: E-mail and Usenet messages are generally easier to read if you
begin your reply after the quoted text, rather than before. This is the
opposite of the default setting in Mozilla. To change it ...'
Do as I say, not as I do. Yeah, I'm sure that would go down *very* well.
> Maybe it should warn users about typing in messages that are
> all-caps?
Not a bad idea, that.
> I am guessing that AOL/Netscape's priority is probably user
> adoption, though.
Doesn't that include helping to get rid of the reputation of AOL users
as `clueless newbies'?
J
All of the BIG email/news apps in the beginning *all* defaulted to
bottom posting a reply, ie., Lynx. Didn't have anything whatsoever to do
with being *tech savvy*, it was the *only* way. I don't remember who was
the first to introduce *top* posting, I think it was Microsoft
(figures). Even Forte Agent defaults to top posting a reply.
The convention desired by *most* is bottom-posting simply because it
retains thread continuity and logical reply ordering in a multi-reply
thread. I've never seen a school exam where you answer questions at the
top of an exam paper nor have I seen inter-office memos with replies at
the top, makes no sense. Usenet is mostly responsible for top-posting
because that's where most users get their start with applications that
default to top posting.
Bottom posting by default is the more sensible choice and should be the
default. If users want to top post because of group *tradition* then a
simple CTRL+HOME will get you to the top rather quickly. The same is
true to zip you to the bottom with CTRL+END ....
J
J
BRAVO !!!
J
Ben Bucksch
[long, uncutted quote]
>
> BRAVO !!!
>
> J
- Please do not quote unedited.
- Please do not post anonymously.
(Personally, I have no problem, if you use a pseudonym,
as long as
- it has an identity, i.e.
- it is reasonable unique
- you use it often
- the email address is valid .)
- In gernal, please read the
charter <http://www.mozilla.org/community.html>
before posting and follow it.
Thanks.
Selected quotes from the charter
<http://www.mozilla.org/community.html>:
To 1)
Edit your followups.
Do not quote the entire content of the message to which you are
replying. Include only as much as is necessary for context. Remember
that if someone wants to read the original message, they can; it is
easily accessible. A good rule of thumb is, don't include more quoted
text than new text.
To 2)
No anonymous messages.
If you have something that you are offering for others to read, you
should be willing to attach your name to it, and to give people the
ability to reply to you privately about it, in case they feel that a
public reply would be inappropriate.
So-called ``address munging'' is frowned upon. Your return address
should be replyable. If you want to avoid spam, use mail filters, or
don't post. Other alternatives are not welcome here.
Chuck Simmons
>
> J <J...@J.org.netscape.com> wrote:
> [long, uncutted quote]
> >
> > BRAVO !!!
> >
> > J
>
> - Please do not quote unedited.
> - Please do not post anonymously.
Good grief! If you don't know who J is, there is no help. Personally, I
use my real name and my real address but I conclude from the small
amount of spam I get that my provider has filters. Not everyone has this
nice situation.
Chuck
--
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons chr...@webaccess.net
Jay Garcia
>
> J <J...@J.org.netscape.com> wrote:
> [long, uncutted quote]
> >
> > BRAVO !!!
> >
> > J
>
> - Please do not quote unedited.
> - Please do not post anonymously.
I'll make you a deal Mr Bucksch !!!
I'll post according to the Mozilla Guidelines if, and only IF, when you
come across an [OFFTOPIC] post regarding Communicator that you will
recommend them to re-post their question in
snews://secnews.netscape.com/netscape.communicator where they can get
the best and most efficient help available as has been available for the
last 5 years via the Netscape Champion Program. Posts in n.communicator,
n.unix, n.macintosh and n.navigator usually get answered in a matter of
minutes. The Champions that are replying in these groups account for
well into the thousands of replies entered on a monthly basis. Simply
sending the poster to a *help* page is usually more confusing to the
user and non-productive. If we Champions deem it necessary then WE will
make the decision to off-load the user to another URL or FAQ .....
Thank You
Jay Garcia
Netscape Champion
--
The UFAQ is Here => http://WWW.UFAQ.ORG
*** REPLY TO "GROUP ONLY" - DO NOT EMAIL ***
Jay Garcia
>
> Ben Bucksch wrote:
> >
> > J <J...@J.org.netscape.com> wrote:
> > [long, uncutted quote]
> > >
> > > BRAVO !!!
> > >
> > > J
> >
> > - Please do not quote unedited.
> > - Please do not post anonymously.
>
> Good grief! If you don't know who J is, there is no help. Personally, I
> use my real name and my real address but I conclude from the small
> amount of spam I get that my provider has filters. Not everyone has this
> nice situation.
>
> Chuck
> --
> ... The times have been,
> That, when the brains were out,
> the man would die. ... Macbeth
> Chuck Simmons chr...@webaccess.net
I plead the 5th ... Think I'll go get one for the weekend .......... :o)
Jay
Chuck Simmons
>
> Chuck Simmons wrote:
> >
> > Ben Bucksch wrote:
> > >
> > > J <J...@J.org.netscape.com> wrote:
> > > [long, uncutted quote]
> > > >
> > > > BRAVO !!!
> > > >
> > > > J
> > >
> > > - Please do not quote unedited.
> > > - Please do not post anonymously.
> > > ...
> >
> > Good grief! If you don't know who J is, there is no help. Personally, I
> > use my real name and my real address but I conclude from the small
> > amount of spam I get that my provider has filters. Not everyone has this
> > nice situation.
> >
> > Chuck
>
>
> Jay
Heck. I pegged you quite a while back. Drop by and I'll buy you a Dog or
three. The emporium is just across the parking lot and there hasn't been
a shooting incident in the parking lot in nearly a year. The
neighborhood is getting dull so I probably better move before I die of
boredom. Ben can come too but he can't complain about the accommidation.
I'll put him up in the Doubletree. The bar there has not had a shooting
in almost 2 years.
Jay Garcia
Certain posting habits tend to get quite revealing over the years as
well as spotting an NNTP-Posting-Host: xxx.xxx.xx ... Not rocket
science by any stretch.
The last time I was in Colorado I gave a Chipmunk a peanut, he brought
it to his cave in the rocks and he came out with a nickel and dropped it
at my feet .....
Jay
Greg Miller
> Brian Smith wrote:
> > I am guessing that AOL/Netscape's priority is probably user
> > adoption, though.
>
> Doesn't that include helping to get rid of the reputation of AOL users
> as `clueless newbies'?
But how much does user adoption have to do with the default position
of the caret? The last Gartner Group (is was them, wasn't it?) survey
of browser usage indicated that very close to 50% of all MSIE users
used it because AOL gave it to them. Another chunk used it because
Microsoft or some other company gave it to them. Netscape/Mozilla will
likely become the #1 browser again eventually, but not because of any
default settings.
J. Moreno
<gmi...@classic-games.com> wrote:
> mp...@student.canterbury.ac.nz (Matthew Thomas) wrote:
>
> > Brian Smith wrote:
> > > I am guessing that AOL/Netscape's priority is probably user
> > > adoption, though.
> >
> > Doesn't that include helping to get rid of the reputation of AOL users
> > as `clueless newbies'?
>
> But how much does user adoption have to do with the default position
> of the caret? The last Gartner Group (is was them, wasn't it?) survey
> of browser usage indicated that very close to 50% of all MSIE users
> used it because AOL gave it to them. Another chunk used it because
> Microsoft or some other company gave it to them. Netscape/Mozilla will
> likely become the #1 browser again eventually, but not because of any
> default settings.
If that's true, then why not do what's best for most people?
Greg Miller
Exactly. That's why I'm pushing to have the caret default to below
the quoted text. That's what's going to benefit the most people.
Mark Evans
> Steve Morrison wrote:
>>
>> If people are reading messages threaded, the presumably they've already
>> read the previous messages. If a message has 20 replies, and I read all of
>> them, why should I have to scan/scroll through all the text just to see the
>> comment?
>>
>> -Steve
> You are not familiar with expiration? The news server I use expires text
> groups in about a week and binaries in two or three days. This neans
> that the whole thread is almost never available. Moreover, though I use
As well as it being possible for someone to have the followup available
before the original post. Even the best threaded newsreader cannot
produce a post which isn't yet available to it.
--
Mark Evans
St. Peter's CofE High School
Phone: +44 1392 204764 X109
Fax: +44 1392 204763
Mark Evans
> Ben Bucksch wrote in message <39067619...@bucksch.com>...
>>Jennifer Glick wrote:
>>
>>> There would still be a pref for users to change the carot to default to
> the
>>> bottom. But for more novice users (who will never find the pref if we
> default
>>> it to the bottom), the default would be set to what they are already used
> to
>>> and expect to see.
>>
>>The problem is, that we would encourage bad net behaviour, from which
>>everybody suffers.
> I think that the mode that is best for the author should take precedence
> over the mode that is best for the readers. I don't see the logic in saying
> "we're making it harder for you to type your message so that it will be
> easier for other people to read it" to a potential customer/user.
Except that what matters to the reader is appropriate formatting of
the reply/follow up. They have no way of knowing (or for that matter
caring) where the writers software put its default insert point.
The stronger way of putting it is "If you create something
difficult for people to read and understand then you are simply
wasting your time. If you expect people to bother reading it then
make it easier for them to read".
Mark Evans
> Jennifer Glick wrote:
>> But I am worried that email apps have already taught users
>> one behavior and that changing that expected behavior will cause confusion.
> Do you seriously think, our users won't be able to use the software
> anymore, just because the caret sits at another place by default?
Interestingly enough those same users have no problem using
"word processors" sensibly. Virtually every one of these
pieces of software, on opening an existing document, puts
the cursor at the very top.
This isn't a software problem, it's a wetware issue.
Maybe the solution is to have the software randomly select
a point each time.
Mark Evans
> Brian Smith wrote:
>> If user adoption is the #1 concern, then perhaps it belongs at the
>> top. If netiquette is #1, then maybe it should be on the bottom. It
>> depends. Maybe Mail/News should have a "netiquette tip of the day"
>> feature?
> `Tip #37: E-mail and Usenet messages are generally easier to read if you
> begin your reply after the quoted text, rather than before. This is the
> opposite of the default setting in Mozilla. To change it ...'
They are even easier to read if the reply follows the specific point
(be that paragraph, sentence, even a single word) which the reply addresses.
Unfortunatly there is no software solution to help with this. Natural
Language Understanding is unlikely to be part of the software for a long
time :)
>> Maybe it should warn users about typing in messages that are
>> all-caps?
> Not a bad idea, that.
As well as a whole host of other things which can be detected
by an algorithm. (Which definitly includes poor quotation.)
Mark Evans
> Jennifer Glick wrote:
>> > I guess, the confusion will hold on for only some seconds - a minor
>> > price for the "confusion", that all those bad postings and emails and
>> > the reactions to them cause.
>>
>> I guess I don't see why this is "bad" behavior. If this is how some users prefer to
>> email, what is wrong with that?
> Because it's hard on the reader/victim.
Due to that person being used to a language which reads from the top
of the page downwards.
Which is why footnotes are used instead of headnotes on paper.
No-one would even think of adding new material at the beginning
if they were using a wordprocessor. It would thus be interesting to
know where the "add at the top" behaviour actually comes from.
If you had a paper mail and were to use the same sheet of paper
to send a reply back where would you write your message?
Mark Evans
> In article <390727C1...@netscape.com>,
> Jennifer Glick <jgl...@netscape.com> wrote:
>>
>> If we suddenly default the carot to the bottom of the email message,
>> do you think all these folks will suddenly start typing their
>> responses below? I expect they will simply scroll back to the top and
>> start their response there, because that is what they are used to doing.
>>
> The mailer in Netscape versions 2 and 3 did default to putting the
> caret at the bottom of the mail. In Netscape version 4, the change
> was made to default the caret to the top. At that time, I noticed
> that many Netscape 4 users with whom I regularly corresponded who had
> previously always started the replies at the bottom began typing
> replies at the top directly as a result of this change in defaults.
Though there is then the possibility of no snippage being done with
this default this is surely the lesser of two evils. Compared with
generating something which is difficult to understand. Becuase the
reader needs to start in the middle, read to the bottom, then return
to the top.
> Presumably, they changed because the path of least resistance/effort
> had changed (I even did it occasionally for exactly this reason).
The real problem is how to get people to reply to each part of the
original message. Which is even easier to understand (but not something
software can really help much with.)
Mark Evans
> The convention desired by *most* is bottom-posting simply because it
> retains thread continuity and logical reply ordering in a multi-reply
> thread. I've never seen a school exam where you answer questions at the
> top of an exam paper nor have I seen inter-office memos with replies at
> the top, makes no sense. Usenet is mostly responsible for top-posting
The only places you are likely to see documents where it is usual to
add above the original text is where the *language* used is one which
reads from bottom to top.
Putting the answers/comments/etc *after* the original is a natural
language convention. Common to all literate people.
> because that's where most users get their start with applications that
> default to top posting.
How do they manage with wordprocessing and text exiting programs though...
> Bottom posting by default is the more sensible choice and should be the
> default. If users want to top post because of group *tradition* then a
> simple CTRL+HOME will get you to the top rather quickly. The same is
> true to zip you to the bottom with CTRL+END ....
Except that those people concerned probably think everything is done
with the mouse and that middle block of keys is their for decoration :)