[sclug] TopPosting Was: Helping others into FOSS/Linux

[Roland Turner SCLUG]

I have been around the 'net long enough to see the top-posting argument
come up more times than I can remember, and the pro/con arguments are
starting to seem a little moot to me. I am starting to consider
complaining about top-posting to be more harmful than actually
top-posting.

Tim Sutton wrote:


> Are those (usenet netiquette) people serious? I personally hate it when


There are plenty of people who derive pleasure out of telling other what
to do; a plausible justification merely encourages this.

> people  reply at the bottom of an email and I have to scroll down to
> read it. If I  want to know the context of the reply I just scroll down
> to get it, but in  most cases I dont so I save myself a whole heap 'o
> RSI from excessive mouse  button twiddling :-)


Before I worked in corporate contexts where top-posting was the norm, I
also found it offensive. However, having worked in such contexts for
several years, I'm now inclined to agree with you; in almost all cases I
will already have seen the context and would much rather just see the
meat. The availability of the context for the benefit of those who don't
know it is helpful, but I'm starting to think _better_ achieved through
top-posting (a) because posters aren't wasting time editing for the
marginal benefit a few/no readers and (b) because in almost all cases
readers are saved having to wade through quoted material.

It is worth commenting that netiquette was devised in a time when the
email audience and the machine resources (CPU, RAM, disk space, network
bandwidth) available were all orders of magnitude smaller than they are
today. It is hardly surprising that alternate norms of conduct are
arising.

Alex Butcher wrote:


> Reply inline, quote selectively, and everyone will be happy.


An ideal, and one that I tend to conform to, but I'm starting to think
that expending time and effort telling others to do so is
counter-productive.

Hamlesh Motah wrote:


> Both Neil's and your views expressed how I felt when I was asked not to
> top post (once I read what it was).  I must not be understanding, if its
> such a bad practise why hasn't it formed a greater more weightier part
> of the business community email etiquette, whom to be fair are starting
> to use email more than the average geeks.


The average geek apparently assumes that the average business user is
stupid, rude or controlled by Microsoft. Whilst each of these stereotypes
contains germs of truth, they do not hold for everyone. My feeling is that
this near-universal behaviour of business users does indicate that there
is some value in top-posting; if there were not then at least some
for-profit organisations would willingly have taken measures to stamp it
out. I've never heard of one (non-technical) organisation that has
attempted to do so.

It also occurs to me that corporate email practices arise in an
environment where a lot of collaboration is performed by regular email
(rather than mailing lists or newsgroups; bear in mind that Netiquette
arose in the context of newsgroup communication). Perhaps the problem is
simply that the environments are sufficiently different that different
behavioural norms are sensible. The question (addressed below) becomes one
about what to do when two worlds collide.

Simon Huggins wrote:


> A. Because it breaks the logical sequence of discussion
> Q. Why is top posting bad?


Ah yes, reasoning by humour :-) More seriously, Simon's presentation draws
the different norms into sharp contrast. In the context of a newsgroup or
mailing list archive that you are searching (rather than reading
sequentially), or even following a newsgroup where there are posting
delays measured in days (as was the case in the days when the Netiquette
arose)  so that different participants will see the messages in different
sequences, it will frequently be the case that you'll read a message
without having recently read its context and that you'll often be wanting
to refer to (rather than recall from memory) that context. Top-posting
corporate email users are, by contrast, accustomed to propogation delays
of minutes and, more importantly, of having email messages be seen in
substantially the same sequence by all recipients, on which Neil Owens
wrote:

> But this doesn't hold true if the question and answer were sent in two
> separate e-mails.  EVERYONE, Microsoft/Lotus users included would read the
> oldest message first (the question) then the newest message (the answer).


Which is rather the point, and quite probably why heavy corporate users
(a) don't clutter their email with closely edited quoting and (b) are
puzzled when the geeks start bitching about top-posting.

Patrick Kirk alleges:


> The reason Microsoft and commercial email clients top post is that
> usability studies show this is how people like it.


I'd be intrigued to see documentary support for this claim. I'm willing to
accept that users reared on Outlook will see Outlook as good.

> No-one will buy
> something they don't like.  The interesting question is why a standard
> was set without doing a usability study first.


I am reminded of the utter astonishment that arose on the part of my
German colleagues when I tried to explain how little correlation there was
between Australia's constitutional arrangements and the manner in which
the nation actually functions. Whilst it may be difficult for the yonger
generation to believe it, there was a time when software, protocols,
networks and behavioural norms formed without the aid of usability
studies, focus groups, lifestyle consultants or telephone sanitisers :-)

Darren Davison replied:


> There is a significant number of people who certainly don't like it that
> way at all.


You can't please all of the people all of the time. Better to please most
of the people most of the time which, unlike the situation when the
Netiquette arose, now means top-posting. As far as I can tell, the
anti-top-posting lobby is now a tiny minority.

> The simple fact is, that *combined with prudent trimming*
> to provide context, such responses are more inclusive for the wider
> audience.


As are responses provided in 20 languages, which is exactly why the
European parliament does it for its official documentation, however this
does not come free of charge. (My point: when advocating a behavioral
change, it is not enough merely to assess the presumed benefits; the
likely costs must also assessed, and on a comparable basis.)

For the vast majority of email users, the marginal benefit of not
distressing a few people who want to tell them how to write is greatly
outweighed by the time/effort that "prudent trimming" would cost.
Inclusivity is a desirable objective, but there are practical limits.

> Top-posting and/or not trimming quoting alienates a much larger
> proportion of that audience.


I am not entirely convinced that this is correct in the typical email use
scenario for a corporate user; mixing up what you're saying with remnants
of what others have said may well alienate more of your audience in that
environment than top-posting would.

The question then becomes, as before (and addressed below), what happens
when two worlds collide?

>  That's before considering the additional cost for those who pay for
> bandwidth and therefore pay maybe a dozen times to receive the same
> 50Kb of text with one sentence added at the top each time.  Not such
> an issue on the desktop these days of course,


This was an issue in the distant past. With _typical_ total bandwidth
availability being dozens, hundreds or even thousands of times what it was
just 15 years ago, this issue is now moot.

> but certainly an issue for most mobile device users.


In this special, but increasingly common, case, the reality is quite the
reverse. Top-posting plays _much_ better with mobile devices than
interspersing quotes and comments does. Most mobile email clients deal
with their limited bandwidth by only fetching the first 'n' bytes of any
message largely in order to avoid swamping by attachments. A convenient
side effect for top-posters is that even if n is set particularly low
(you're travelling to a distant country, the roaming rates are
extortionate), the message still gets through. If, instead, the message is
mixed with (rather than followed by) quotes that the recipient has most
likely already seen, then the risk of the message not being seen increases
dramatically.

> The more interesting question to me is why do [top-posters|non-trimmers]
> compose email responses in a way that they would never dream of
> composing any other form of written communication?


Presumably because email is not like any other form of written communication.






Astute readers will have noticed that while rebuffing deprecation of
top-posting, and even supporting it to some extent, I myself tend to
selectively edit and to reply in context. Like anybody else, my initial
experience with email was in a particular environment with a particular
characteristics, an experience which has shaped my habits and what I feel
comfortable with. I do suspect, however, that those characteristics were
severely more constrained than those against which most (but not all)
SCLUG members cut their teeth in.

I am now going to indulge, briefly, in that most revolting of public
spectacles when elders (not that I'm old...) talk about "wen ah war a
lud...". My email experience started with half-duplex text terminals (an
environment so unpleasant that I'll refrain from offending tender ears
with the specifics; suffice it to say that vi, at least in its "visual"
mode, was beyond our reach; we had to use "ex") and 300 baud modems. The
entire university (20 000 students) shared a _single_ 9600 bit per second
(V.32) modem which provided a UUCP feed to a nearby university. None of
this wasteful IP rubbish... That's about 0.5 bits per second each; my
phone gives me 20 000 times that, my desktop 1 000 000 times that. When I
talked about "orders of magnitude" above, I wasn't kidding. Even attaching
signatures to email messages was explicitly frowned upon. You might think
that this would lead to very bandwidth-miserly habits indeed (and it does,
in some contexts), but I suspect that it also provides me with a broader
view than that held by a lot of people who appear to be parroting 15 year
old arguments in a current resource-rich context that no longer supports
the same arguments.

Section 3.2 of RFC 791 (the Internet Protocol itself) counsels us as
follows (and in these resource-rich days, I wastefully quote at length):

> The implementation of a protocol must be robust.  Each implementation
> must expect to interoperate with others created by different
> individuals.  While the goal of this specification is to be explicit
> about the protocol there is the possibility of differing
> interpretations.  In general, an implementation must be conservative
> in its sending behavior, and liberal in its receiving behavior.  That
> is, it must be careful to send well-formed datagrams, but must accept
> any datagram that it can interpret (e.g., not object to technical
> errors where the meaning is still clear).


This so-called conservative/liberal principle applies to many activities
where differing views and assumptions come into contact (or, as I framed
it a couple of times above, where worlds collide). While it is worthwhile
to mind your Ps and Qs, it is also worthwhile to deal gracefully with the
failure of others to do likewise. The principle does not go on to say that
one should bitch and moan about the failure of others to be conservative
in their sending behaviour, quite the contrary, it advocates granting a
liberal reception whenever the meaning is clear. Needless to say, this is
an ideal but it is nonetheless widely practiced, and not only in the
network communication sphere. I have been the grateful beneficiary of such
grace as a long-term visitor in four different cities (Reading being the
only one in which I speak the native language) in three different
countries over the last half-dozen years. This liberal acceptance on the
part of a great many people in society at large enables movements and
collaborations that would not be possible in an environment where
Netiquette Nazis rule (nor indeed where actual Nazis rule).

I suspect that we've reached the point where top-posting is here to stay,
not merely as some toxic by-product of Microsoft's success, but as a
positive behavioural trait for corporate email users and that, therefore,
it is probably going to be beneficial all round for those who eschew
top-posting, even for those who don't accept _any_ of my arguments for the
benefits of top-posting, to simply adopt a more liberal attitude; to learn
to live with it, if not love it. I really do believe that publically
bitching about it, and worse, brow-beating those who do it, has become
actively harmful.

So Hamlesh, if you're still reading, don't apologise for raising this. I
think that it's an important and difficult question and one on which
attitudes still have some distance to shift.

- Raz