Why top posting has won (2018)
solipsys.co.uk
solipsys.co.uk
So the key is "the thread needs to be distilled" as OP puts it. And often that's more work than just finding the right sentence to quote.
My approach these days in my email is:
99% of the time I quote nothing and delete everything my client puts into the compose window, relying on the default thread-view in most email clients to supply the context for my readers.
1% of the time I need to quote, and I quote liberally, treating words as a wiki, and editing/sculpting the text in '> ' just as much as my own reply below it.
> Top posting vs replying inline feels like an incomplete dichotomy.
There are just so many options.
It's kind of alarming to me that the default in say gmail is to constantly re-send the original message chain back and forth. I guess it then gives the whole chain to newcomers to the thread?
I use past tense because I haven't used any of this Microsoft stuff for over 10 years, but I assume this is still the case.
This sort of moves the question to "why did Outlook only support top-posting?" I don't have a clear answer to that, but if I look at the general state of Microsoft and email at the time then it's probably a combination of the Not-Invented-Here (and maybe some "EEE") attitude at Microsoft at the time combined with a general apathy and ignorance towards email.
Microsoft Email/Exchange/Outlook as best I can remember have always defaulted to quoting the message and leaving the cursor at the top. And they didn't prefix the quoted text in any way, so to do inline replies you'd have to make it clear which text was quoted and which text was new. People would use colors or different fonts when "rich" text became supported but it was more work and most people just took the easy path.
[1] Very early on, MUAs did not (at least by default) quote the original message. This was because most people were using teletypes or slow (300 baud) terminals and possibly a line-oriented editor such as 'ed' so quoting the original message could add significant extra time needed to compose a reply. Also, with a mail client that does proper threading, you can see which message is being replied to and you don't really need to include that again in the reply.
Top posting did not win anything. Lazyness did.
In hindsight, all of that was a mistake because all my coworkers top-posted. Top-posting might be bad (IMHO), but a mixture of bottom and top is even worse.
I now just top-post. I'm still not a fan but it is what it is.
Such as in-reply-to (no need to quote the whole mail at the bottom of yours) and obviously the sender (no need for an explicit signature), and you could use x-face if you needed something fancier or, if your company really really needs to remind everyone how social they are at every message, you could have added an x-coprorate-blurb custom header, etc.
Whatever, emails these days are nothing more than an archaic last resort password reset protocol.
In addition to what the sibling comment already points out, most people don't even know what "top posting" or "bottom posting" is because they just do what the email client does by default (which in Outlook's case: strongly favours top posting).
The only "lazy" thing here is this kind of smug judgemental attitude instead of looking for the actual reasons.
It is, $job's outlook install still "top posts" by default, just like always.
[1]: Likely done for reasons of Outlook compatibility/familiarity.
That. Reading an email thread in Outlook is an exercise in masochism.
Some interesting aphorisms from the master La Rochefoucauld:
- "In the misfortune of even our best friends, there is something that does not displease us."
- "Old men delight in giving good advice as a consolation for the fact that they can no longer set bad examples"
- "No persons are more frequently wrong than those who will not admit they are wrong."
- "We promise according to our hopes and perform according to our fears."
- "A refusal of praise is a desire to be praised twice."
- "We only confess our little faults to persuade people that we have no big ones."
- "Almost all our faults are more pardonable than the methods we resort to to hide them."
My theory is that a good aphorism merely gives us "2+2", which prompts us to test it and conclude "4". This is much more persuasive than spoon-feeding us "2+2=4" in the first place, because we naturally trust our own conclusions more than the conclusions of others.
There’s something about being told things directly that pushes people away, but a clever aphorism disarms us because it makes us feel smart for “getting it”.
There’s something inside us that prefers indirect communication.
This may come from culture or neurodiversity. Some of us greatly prefer explicit communication.
It's an engineer-heavy company, and he's a marketing guy. He was brought in to give them some sales direction, some strategy. He comes up with some standard strategy, presents it. No, reasons. Meetings and presentations and essentially repeated NOs later, months and months of this, he eventually says to himself:
This is industry standard stuff. They're rejecting it because they don't understand. I'll do a big presentation, come up with loads of metaphors and catchy phrases and graphs, and present it.
He does this, result: oh, brilliant stuff, let's get this all implemented! Wonderful. It's the exact same stuff as before.
The story was told to me in terms of - "ah, isn't it great that everyone finally understood!", and my response was based on exactly your observation. They understood roughly as much as before, but if you don't get the metaphors, you look silly!
Top posting takes more effort to do well, but makes it much easier to follow an email chain.
> It takes so much effort to read the email to figure out who said what
Email clients used to color each quoting level, and mails started with something like:
John Doe wrote on $date:
> Jane Smith wrote on $date:
> > Joe Shmoe wrote on $date:
> > > what they wrote
> > what they wrote
> what they wrote
> while also keeping state of all the different threads.
That’s why we had threaded views that showed messages in a tree, and showed when the subject (line) changed. People were encouraged to edit the subject line to reflect topic drift, so that you had a good overview when looking at the thread visualization, which would omit repetitions of the same subject text and only show where it changed.
Example (threaded view at top, colored quotes at bottom): https://robot.unipv.it/clipedia/images/mutt-threads.png
Scroll a couple of screens in the message text and you see something like:
> > > > I think that ...
> > No, I remember that ...
> On the contrary, I believe that ...
Unfortunately I have not seen mutt coloring the index with the same colors as the quoted text, which would help a lot (if you have a view that shows the index along with the message).
E-mail readers have people switching from document to document and starting anew with each. When viewing documents we start at the top, and the reply is the most prescient thing to see.
Even in a back-and-forth between 2 people, there are some people who'll reply a second time to the same email, not to replace their earlier response, but to add more information.
(Aside: Also, it seems to be rare for people forwarded a top-posted thread context to actually read it, unless they're looking to assign blame.)
In every company, I end up having to reconstruct context for someone, or do archaeology on some interaction, and the top-quoting both helps and hurts.
Downsides of lazy top-posting for these purposes include having to break out all those messages, and fill in the gaps from maybe multiple copies at different points.
Upsides include when, say, a single email is the only record, and the fact that a bunch of people top-posted means you have more raw information, in the quotes, than if that single email was in a history of people doing proper Netiquette with minimal quoting of the parts they were responding to.
There are a whole bunch of reasons and implications for how things are done in corporate practice, and IME most of them come down to one of: poor tools, poor use of the tools (including by "digital natives"), and corporate misalignment + politics.
I'm imagining a cognitive prosthesis that would help you understand the structure and dependencies of someone's argument and would give you a "wavy red line" or list of points if what you're saying is irrelevant, incoherent or doesn't address the actual arguments being made.
A big thing I value about inline posters is that I know they read the whole message. Doing the work of summarising and maintaining context signals to me that what they have to say might be worth reading, even if I disagree. It also signals they adhere to conventions around reasoning and discourse.
Would providing computer tools for argument defeat the purpose though? Would it just help people pump out simulacra of thoughtful responses? Creating a tool to be weaponised by pedants and "motivated reasoners"? It might actually make it _harder_ to find signal in the noise.
I feel like most people would start at the conclusion first, then use the tools to make the strongest version of their argument. A kind of "google maps for thinking" where the destination is already punched in. The computer would just be used to determine the most acceptable route to get there.
In a perfect world, tools for thinking would help everyone undertand the territory better rather than being something that just helps your tanks drive over it :/
> But when the thread is long, and when your reply is to a point deeper inside, suddenly the receiver has to do a lot of work. You, the sender, have saved yourself time, but the receiver has to do more work than you have saved. Potentially a lot more work.
The problem is that readers are also lazy, and increasingly people don't have time to follow long threads at all. With inline responses, it's far too easy for threads to branch off in a million directions. Once that happens you'll often lose the forest for the trees. It's reasonable for usenet-style internet discussions where participation is voluntary and asynchronous by definition, but in work collaboration it's a strong anti-pattern. In a situation where there is a goal and desired outcome, top-posting is a reasonable way to keep a thread linear and outcome-focused. It can still break down for complex or controversial issues, but in that case email is the wrong medium, instead jump on a call or write a proposal.
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?Top Posting is how you assert dominance and gaslight an entire thread while ignoring a nuanced point of view. That complexity is not present in tabs vs spaces.
RFC 1855 [1] makes it clear that the correct way is to quote relevant text and then reply.
In a normal office environment e-mail moves much faster, people open threads with the context already in mind, and expect to see the relevant new info at the top. Bottom posting would either make everyone sigh heavily when they see an e-mail from that one person or would earn someone a quick coaching session to adopt the prevailing style of the office environment.
But if you go on to a mailing list that wants bottom posting, that’s what you do. It’s a courtesy.
I wish there was some good way to silently communicate that preference to likeminded. Then I would know right away when to switch to proper netiquette without the other person having to go first. And how often was I in an email exchange with someone else that also prefer to not top-post, but both did that anyway since that is just the expectation now?
I am not sure if there exists a good tool for threaded discussions with multiple different focus areas - something like git but for conversations?
But I and colleagues use it regularly, and it absolutely hits the spot for non-linear discussions that are intended to find conclusions.
It’s still better than a teams call.
Why Top Posting Has Won (2018) - https://news.ycombinator.com/item?id=22801233 - April 2020 (140 comments)
Why Top Posting Has Won - https://news.ycombinator.com/item?id=17302808 - June 2018 (3 comments)
Unfortunately, GMail doesn't do it any better.
Outlook did not use to make it easy to follow threads at all.
I wonder what the carbon footprint is for poorly threaded email clients that encourage including the entire thread in every reply.
Official version is: "because there is a whole history of correspondence and it is convenient to forward it to new participants".
In reality? It doesn't matter. Almost no-one reads, neither top- or bottom-posted mails. But there is a drawback in top-posting and I mean a "my comments inside original post, in color/bold/with indent/randomly inserted between two phrases". There is no standard of citing in top-posting - thus sometimes original mail gone. Edited, re-edited and commented in various, inconsistent and often unreadable ways.
I agree the difficulty quoting sucks, but that's mostly because of the switch from top posting to bottom posting. When people copy-and-paste the bit they are replying to and stay in the top posting paradigm things aren't so bad.
You're not meant to take a proverb and apply it literally in all cases. It's a piece of wisdom you pull out when you deem it appropriate. The analysis and individual judgement of the human being is implied.
If you included more it wouldn't be a proverb.
Incidentally: Bloggers, please put dates in your actual blog post templates. The date is an important part of the context of any article.
[EDIT: The author of the article has already added the date to their template. Thanks!]
However, I don't find that to be the case for most of the written content I consume and knowing where some content was created temporally is important context more often than not(at least in my experience). Now, I can't nor would I want to dictate what others do on their own hosted content(where they generously share their hard meticulous work).
Comments on HN and other places lamenting lack of dates however should be fair game to desire such additional context. But then again, you commenting back is just participating in the same dialogue so I guess my entire message has been just meandering.
The date is more important than you're giving it credit for. Every creative work has a context and an implicit perspective that comes from that context. This is especially true for nonfiction articles, where the date tells you things like:
* Whether the author is reacting to something that just happened or whether they have the benefit of historical hindsight.
* Where the article fits in the author's overall body of work (including works they haven't written yet!).
* What, if any, recent events may have prompted the author to write the article.
* The prevailing intellectual climate, which carries focuses and blind spots that may be very different from what we have now.
* Whether the article is about something immediately useful or whether it's more likely to be of historical interest.
You can already tell, for instance, that "Thoughts on Software Development (1998)" is going to be talking about very different things than "Thoughts on Software Development (2012)" or "Thoughts on Software Development (2025)". An article like "Better C Programming (2020)" probably contains some useful advice; whereas "Better C Programming (1991)" should be taken with a large grain of salt.
Instead of making readers ask questions like "Why is the person talking about operating system monopolies while saying nothing about LLM model ownership?", it's easier and more helpful to just put the date at the top.
The idea was that some content is, or benefits from being perceived as, "evergreen". Always relevant.
Maybe -- but it's still deceptive, I think.
On the other hand, even here on HN we see people talking about "unmaintained" GitHub repos where the last commit was more than a few months ago. So, recency-bias is a real thing, and marketers certainly don't want to be penalized for honesty. :)
This was a while ago. Nowadays, the obvious opposite extreme is common. Blogspam that is "updated" with a current date, but no changes to content.
> the inventor
Aside: email didn't really have "an inventor". Shiva Ayyadurai does not count.
> should have made it so the previous content should have been in a special segment
So how would you do this then, where you reply to a specific part?
When specific context is needed, besides “the whole message” just do exactly as you and I are both doing right now. But in general, quoting the whole message is pointless unless you’re adding a new recipient on, since we all have the threads in our email clients anyway. Whenever I can, I nuke the entire quoted part out of my replies.