Stop Microsoft users sending 'reactions' to email by adding a postfix header
neilzone.co.uk
neilzone.co.uk
Of course you can still send real thank you emails when you're genuinely thankful!
Is this just the sender looking for visibility? Does the recipient appreciate that yet another middle manager CC'd the whole company on a single sentence congratulatory message? Is there some other social function at work here that I don't understand?
Sometimes I think it would be funny if all the individual contributors coordinated to ALSO reply-all with "Congratulations", but it would become clear pretty quickly that something was up, and it's a mean thing to do to whomever is being congratulated.
Put differently, it's a kind of "parliamentary procedure": specific rules and expectations that everyone follows in the workplace to keep things "professional" and streamlined, in terms of interpersonal interactions. You may not actually give a damn about whatever the achievement is of Mx Boss, but you sending a congrats or party-popper emoji is sufficient to acknowledge that Mx Boss has done A Thing that Mx Boss thinks is significant.
You may not care at all, but saying that to Mx Boss would have consequences, even if it is the truth that everyone knows.
In the context of meetings-that-don't-have-to-be-meetings, it's (cynnically) a kind of show of fealty. I'm talking here about all-hands or whole department meetings, where every talking point is scripted ahead of time and you could pretty much just send the written up notes of what everyone is going to say already as an email and it'd save however many person-hours of work are wasted in just attending the meeting. For a while I was baffled about why our exec suite was okay with losing so much time across the org, then I realized in part it's an assurance of power: they declare a meeting, everyone shows up, and everyone is content within that social contract of how this stuff works. The point of the meeting isn't to convey information efficiently, it's to ensure that the power structure of the organization is functioning as expected.
Party: pooped.
But I wish the feature that you can write say @joel to get someone’s attention in large email threads with too many on CC would have been adopted by more mail clients.
I have never seen one myself.
Surely there’s an RFC for that right?
The ostensible "problem" with this would probably be "email is too slow for real-time chat!" But really, how much difference does a few seconds make? And modern connections/tech make me think that idea might have to be reevaluated anyways.
It wouldn't be a business but I could see such an app becoming quite popular.
I'm happy to report that this already exists: https://delta.chat/
Indeed, there is: https://datatracker.ietf.org/doc/html/rfc9078
It's used by Delta Chat (IM over email).
They should use Teams then, MS is in the process implementing untold horror into that one (my unfortunate friend works there) where they could live out nighmarish feature dreams and leave the email for actual content. Or keep all emails inside the company - send out the pigeons! - for f's sake then if they insist emoji based infantile communication.
In some cases, people would react to the reaction leading to some ridiculous chains of text.
I guess it was designed to be used with MSN servers, but people used them to connect to the "regular" ones as well.
We have finally found the root of all evil.
Google should really give a bit of credit (bonus levels) for advancing and maintaining existing applications. Their motivation structure is backwards.
[0] https://www.xda-developers.com/google-messages-imessage-reac...
The person sending the reaction sees it on the message like an iMessage, but everything in a non-iMessage group text (iOS, android, or other) gets sent the same “liked your message” text. Nowadays both Android and iOS parse those texts and apply the reaction as if it was sent normally, but that’s a decent change
Exchange converts all emails that go through it to HTML; it's wrapping the J in <font> tags to select Wingdings. Some Exchange installations do not provide plain-text copies of emails sent through them.
This was likely winmail.dat/TNEF[0], Microsoft's proprietary replacement of HTML.
[0] https://en.wikipedia.org/wiki/Transport_Neutral_Encapsulatio...
Granted HTML rendering of email is full of quirks in Outlook
We already have the 365 Web version for that.
Yet another example of Microsoft's failure to hire employees with native Windows development skills, now with the old guard that grew up with Win32 are reaching retirement age.
Took me years until I got a windows laptop and saw emails in outlook.
It may sound aloof, but I sometimes get hundreds of emails a day and it's a necessary filter.
This is no more artificial or disrespectful than a person looking across the room, catching your eye, and giving you a nod and a thumbs-up.
> "I do view it as disrespectful (more precisely, lacking discretion) that the sender decided to waste my time by making me look at and hit delete on their message."
If they'd replied with a full email saying "gotcha" or "understood" or "like it" you'd have to waste your time looking at it and hitting delete. (In Microsoft ecosystem, this doesn't happen, reactions appear on the original email like Github reactions appear on comments in issues - it's nice).
The text fallback also isn't great, I'd much rather they'd sendan email with nothing but the emoji than "person X reacted to your message". I suppose this is more corporate-compliant?
I like receiving reactions specifically because they save me time. When I get a typed message, I often take time to think through whether the sender is expecting some sort of acknowledgement (confirming that I got their message letting me know that I’m welcome to thank them again any time!) or whether it’s polite to terminate the exchange by not responding. Overall, I spend far more time than I’d like second-guessing my words. When I just get a reaction, it serves as a form of acknowledgement that doesn’t demand further response and frees me from an obligation.
Second, "reactions" are not part of the email culture or the standard email specs. It's unexpected and awkward.
So it's not being "anti-reaction" in general, it's being against some feature that only works in MS email apps, which then pops up in broken form elsewhere.
Gmail launched the feature in 2023[1].
I’ve personally not used the feature but emojis shouldn’t be images — they are after all Unicode code points. Not sure why MS’s implementation uses images at all.
[1] https://blog.google/products/gmail/gmail-emoji-reactions/
Another old dog refusing to learn a new trick. If you want to talk about not part of email culture, let's not forget images weren't part of the original. So people got over that. They embraced it just like they embraced HTML emails and everything else. You can stand in the middle of the stream, but it's just going to go right on by you
I still use a plain-text email client.
I can't think of a single person I know that uses a text-only email client (and yes, I know more than one of them) who isn't what you'd call a "power user".
All of them know that most people don't use a plain-text email client, they just prefer to use one themselves.
The idea of "reactions" to email is, to me, silly and unnecessary.
The context here is in how hot-garbage the feature is for people not using Outlook.
If the experience for all users was simply a thumbs up appearing in the subject line on sent emails, maybe I'd see the point you're trying to make. Instead, it's generating a complete email that occupies the same workflow as other actually important information.
If you think a thumbs up response to an e-mail carries the same importance as a written message being directly sent to the user, you are the person who is clearly in the minority.
oh, i'm in the minority alright but for totally different reasons than the incorrect nonsense you've just spouted. i think email is totally broken/useless at this point. to me, anyone still using email for personal communications is an ancient curmudgeon stuck in ways and unwilling to adapt to modern times. maybe corporate emails where it is only used for work purposes can remain some what clean. for personal email where everyone decided that it is okay to spam into oblivion any and every email address you've collected or bought with sales/promos/coupons/attention seeking/pr releases/etc without asking, email is ruined
At least I don't remember anything similar in IRC or ICQ.
Reactions make more sense to me in a kind of communication where you're doing real-time conversational stuff and brevity is important, such as instant messaging. In email, it doesn't really add anything.
Secondly: What does it mean to get a reaction to an email? Can I interpret your "thumbs up" as a sign to go ahead with something, or is it just an "acknowledge, I got your email". If we more or less agree at this point that email are "for serious business" then we must treat it as such and provide clear and precise communication. Reactions are the opposite of that, they are more easily interpreted wrong, depending on context, culture or mood or the recipient.
There might be a really good and reasonable use for "reactions" in emails, but I seriously doubt that Microsoft went all in and made their UI/UX experts do the research and those researchers came back with a clear answer that this was great and here's how to implement it. Given that this isn't even a standard I feel like it's something someone did on a Friday afternoon to show that you could done it. The piggy-backing on SMTP headers have all the hallmarks of a hack.
I've struggled my entire life trying to control my focus, and it's like the rest of humanity has decided to go all in on filling our shared environment with even more distractions. It sucks. It's manipulative. Let me turn it off. Leave me be. Better yet, help me pay attention to the things _I_ value.
/rant
However I really struggle with emojis as a general concept. I don't understand a lot of them, and it seems people put a lot of hidden meaning into them that I have to interpret, and I feel the cognitive load is a lot bigger than if people would just type out what the fuck they are trying to say.
When my friends use emojis, I know what they mean because I know them and the in-group slang we use. When a stranger or acquaintance uses emojis, the only way to know what they're trying to say is through context clues, like with any other unfamiliar slang.
So if someone I don't know well is using emojis, I just ignore the ones whose meaning aren't obvious given the context. If I think they're trying to say something important, I'll ask outright what they meant (and then get an understanding of how they use emojis and so are better able to understand them in the future.)
No, it's exactly the opposite. "Reaction" responses are irrelevant noise that clogs up the communication channel and wastes other people's time and attention.
People don't usually manually type these sorts of reactions because doing so is not often worth the effort. But this feature eliminates the friction, and encourages people to engage in behavior that is effectively spamming.
Usually on multiple devices. That said, they are fine when on certain mediums. Chat and slack type stuff where it’s sync conversations that’s fine (although on larger groups and at work it starts becoming an asynchronous bulletin board more than a chat). Email is asynchronous and I only need the final response if the conversation even requires one. If I’m asking you a question I don’t need an acknowledgment reaction followed by an answer 3 hours later. Just send the answer 3 hours later. If I asked you a question and said I needed an answer quickly, you could say I can’t get it to you for 3 hours or if you never say anything I’m going to assume you won’t be answering me (that’s ok!) but if you acknowledge it I will assume you’re working on the quick response I asked for and if o don’t get it I’ll probably be upset or wondering what the problem is.
Anyways, in general we’ve built up a lot of norms for various mediums. Email norms don’t need to follow chat norms.
I asked a question recently to someone and got a "like" back as the "answer". I then spent one minute pondering wtf this message even was (it was the first of its kind that I received) and another pondering wtf the respondent meant as the question was not of the form "do you like XXX?".
This "feature", at least in this case, was not a form of communication of but a disruptive form of confusion.
This "feature" is not just distasteful but actively harmful and anti to accepted email communication norms. On Slack it's annoying enough but it's a norm so it is merely distasteful.
In email, I see it as yet another EEE method from Microsoft.
Asking a question and getting "I like your question" as the sole response is incredibly obnoxious.
Replaces a lot of useless typing I had to do to sound polite when saying “fine, no further comment”. And then getting a notification from the other party acknowledging my acknowledgment… yuck.
These are shorter. And remind me of the voice chat program Roger Wilco too.
It's rarely obligatory (unless the sender literally requests an ack) but is more to offer a data point just in case it happens to be useful. In some cases it will definitely be useful, like to unblock something that can't proceed until the sender knows you've been briefed. For example, if I tell my kids they can stay out later than 10pm any night if I know about it, then even if they message me saying they'll be out late tonight, actually staying out late is blocked by my ack. Of course, this could just be turned into a yes/no question awaiting my answer, but that would be silly considering that I only say yes; they're not soliciting a decision from me, just acknowledgement.
Do you have an integration test for that in CI? That protocol sounds like the kind of thing which can break easily.
If I as a kid would hear that instruction "can stay out later than 10pm any night if parent knows about it" I would assume notification is necessary, but I would not return home early just because my parent did not ack it. And if my parents complained about that I would find them unreasonable.
Of course with your kids you might have been much more explicit about what you expect from them. Or who knows, maybe you are much more predictable in your response times than my parents were, so your kids might worry about you and call you if you don't respond anything.
They launched it last year[1].
[1] https://blog.google/products/gmail/gmail-emoji-reactions/
I replied thrice because I was surprised to read commenters replies as if Outlook introduced a new unheard-of feature. Including an interop related question I thought about later, because I still can’t understand why emojis involve loading images.
I found the thread too late to develop a well researched post and so commented as I read. Sorry if it offends your sensibilities. But why are you policing how I comment? :-)
Also, why Gmail is important: It’s a fairly major email provider so it sort of matters when they deployed this. Two market majors having a feature usually implies others will follow suit — eventually.
> is it supposed to be okay
This assumes so much. I’ll turn this around. Why’s it not okay? I can’t see any RFC that forbids this. It’s not a feature I’m interested in but I think it’s interesting that email is evolving to match what users are used to on Slack, Teams, WhatsApp etc. And (software) evolution is something I’m very interested in.
For example, if a sergeant says to a private "It's a nice day outside.", the private is obliged to respond, even if the statement is rhetorical. This leads to perfunctory responses, in this case it would be "Aye aye sergeant", or "Aye sergeant", or more casually "Er" or "Kill". You're not obliged to agree, just to acknowledge. Pretty similar to tapback responses in messenger apps and emails.
Sure, there's norms around how you talk on the radio, standing watch on the bridge, on chat channels for command and control, etc. But the rest of the time, the rest of the military talks to each other more or less like normal people with the addition of acknowledging relative rank.
You’ve replaced “read” in the first part with “agree” in the second, and those are not at all the same thing. I can’t let you know that I haven’t read your message.
If a parent says to the other “hey, I’m running late, I need you to pick up Tiny Tim from school”, an acknowledgement is paramount even if you read, understood, and agreed to the message.
> it's just The Generals Problem.
That problem is concerned with an unreliable communication channel, which does not apply to the situation.
The problem isn't the acknowledgement, it's the additional steps I have to make to acknowledge the acknowledgement to the email system.
Eg Whatsapp has 1 tick for sent, 2 for received, and turns them blue for opened.
Also, I think this philosophy that all software must be infinitely configurable, so that it can serve every whim of every possible user, and that if it has a clear idea of what it wants to do and how it wants to achieve that, and sometimes that way it is designed to be used, it's somehow unethical or abusive of the user or something, is the fundamental sickness at the heart of open-source software design. It turns programs into unclear bloated piles of buttons and switches that are overcomplicated to use and impossible to properly quality assure and impossible to design in a coherent way. For powerful professional creation tools (CAD software, publishing, programming, etc) that will be the primary software used for decades by experienced and educated professionals who will want to optimize their workflow and who have the time to invest in deeply learning that one specific tool, then I think that philosophy is fine, but for random chat apps and stuff, it's just frustrating.
Ultimately, this is something that I'd rather be handled at the carrier layer: I should be able to have my phone reject a text message and not pay for / receive it.
On the topic of configurability: Software should ultimately serve the end user. When a developer makes an undesirable (to a user) change to the software and provides the user no way to opt out of that change, it's serving the developer's interests, and it's doing a slightly worse job at serving the user.
I don't see how. All it means is that I won't see the reactions. That's my loss. I'm not forcing anyone else to do anything differently.
If it actually begins to interfere with communications too much, I can turn them back on.
> it's somehow unethical or abusive of the user or something
For me, that's not the thing at all. It's more that configuration options often make the difference between software being useful to me and not being useful to me. That's all.
When we think about email is that it is really explicit when there is something new to handle. There is a new email.
In Slack there are many channels with individual messages which can have reactions, and those individual messages can turn into threads which provides another place where you now need to actively scan to see if something is relevant to you.
This in general is something that bothers me with group communication that is non-linear. It's extremely hard to keep track of it all, and to catch up. Where do you start reading?
When we talk about email, it's much easier to filter for what is important. If your name is in 'To' or in 'CC' it's important enough.
Sidenote: the company I worked at encouraged people to put the group they're emailing into BCC, which makes discoverability as to which group the email was sent (and thus which group I am a member of) impossible to find out, as that information is purposefully hidden from me. But I digress.
In general I am a huge fan of purposeful communication, i.e. tagging someone when it's for them, vs throwing something out there and see who picks up on it or not.
Not to mention that I've seen cases where people get angry for you not having caught a message on Slack. If I wasn't tagged I might miss it. That's the reality of things if you're in so many channels.
Not to mention that leaving channels was frowned upon, as it is explicitly printed.
Signal, for some reason, notifies of reactions to your message on desktop but not on mobile (at least iOS).
We've had macros on computers since at least the 1980's. Just pick your standard acknowledgement and bind it to a hotkey, or a text expansion.
On macOS: Settings → Keyboard → Text Replacements…
:)
(gah, maybe it's dangerous to even joke about this)
https://www.cnet.com/tech/services-and-software/funny-or-ann...
Dunno why Microsoft decided to add the option to Outlook.
Additionally emojis serves a social purpose. Instead of chatting in the office kitchen there are all kind of humorous reactions.
I work on a chat component library (https://talkjs.com) and we support both emoji reactions and email notifications for missed chat messages. These emails can be replied to and they show up as chat messages in the conversation. It’d be very natural for us to add reaction support to the emails too, but I’d be reluctant to do so if it means a great UX for Outlook users but a terrible UX (overload of little reaction emails) for the rest.
Just copying what I said in the other thread: there actually is a standard for that already: https://datatracker.ietf.org/doc/html/rfc9078
Sadly, it seems it doesn't have broad adoption: https://bugzilla.mozilla.org/show_bug.cgi?id=1724363
Now to figure out how to stop the same thing happening in SMS.
<iPhone user> liked your message
It’s really the most clunky backfill that Apple chose, I suspect on purpose.
Welcome to the future.
All of this was because they wanted the critical mass of users and didn't want to work at establishing it themselves at the time.
(The funny part of it being that on the larger channels and networks, Comic chat was completely incapable of handling reasonably the large amount of chat volume in a channel)
Feels very similar where MS' entire philosophy is, if it works for us, we don't care if we spam non-MS people relentlessly.
Course it doesn't work that way, Sys-admins just end up banning/filtering or doing other work arounds to prune it.
And if that blocked message was a reaction I’d assume I can move on: they sent an email, I sent a response, they refused to accept it and then spammed me to boot.
There is a misunderstanding about how this works and how email works:
* Rejecting an email during the SMTP exchange results in the /sending/ server returning a response to the sender. This is NOT spam and it not under the control of the receiver.
* Rejecting an email post-acceptance and sending a reply is called /blowback/ and is typically considered poor practice and a type of spam. (But not UCE.)
* Adding a header to an outgoing message which causes the recipient's email server to possibly generate an email message to the recipient when they reply has nothing to do with the sender. The recipient's email provider is doing this, and the problem is you.
Let's recapitulate:
1) You use Microsoft as an ESP.
2) Somebody sends you a message, received by your ESP with this header.
3) Your ESP takes an action based on this header which results in you receiving feedback.
The whole thing is a shitshow, oblivious to externalities; but whatever.
So I turn them off as a matter of routine on all my accounts.
Obviously context dependent. For time-sensitive emails, it’s common to follow up with a call if receipt isn’t acknowledged.
Really?
https://support.microsoft.com/en-us/office/reactions-in-micr...
This reaction thing seems like a gift from Microsoft to spammers. E-mail recipients have a "like" button that instantly generates a reply, validating that the e-mail address is staffed.
It looks like this feature was removed eventually, but it's just one of those tasteless things that MS does every once in a while.
> (So if you want to opt out of Google Maps and Wi-Fi Sense at the same time, you must change your SSID of, say, myhouse to myhouse_optout_nomap. Technology is great.)
I knew about the Google Maps thing, I didn't know about this. That kind of stuff is so presumptuous and user hostile it's outrageous.
I’m happy for how easy it is to do this.
The system has to be resistant against bad actors taking someone else off the map who chooses to opt in.
It would be better if the syntax were simpler, like changing myhouse to __myhouse (double underscore == private identifier).
By setting SSID a certain way, you simultaneously show your intent to opt out and prove that you're the operator.
The "_optout_nomap" postfix atleast gives a veneer of privacy that your SSID doesn't need to be sent by every nearby device to a backend service to see if it's in the opt-out database and logged/read by some government backfeed etc.
Why not the same thing, but opt-in rather than opt-out?
Iirc the same optout method is used opting out of WiFi scanning.
Blanket behavior like this should always be opt in with explicit informed and uncoerced consent. A laughable proposition in this corporate world but a worthy aim nonetheless.
The proper way would be to design the protocol so that the identification information is useless in addition to disabling SSID broadcast.
That would of course mean that joining a device to network would be way harder unless you enabled at least network name broadcasting, which enables tracking again.
[1] under polish law, majority [2] of uses of received broadcast/shared public medium signal, is automatically legal. The only provision of privacy is encryption of said signal, because it's treated like shouting the information in public space.
Bypassing encryption is what turns it into unlawful violation of privacy.
[2] for historical reasons there's a mess involving radio&TV tax which was supposed to be paid per receiver, a bit like UK TV license.
Unfortunately nothing opt in ever gets wide adoption. So I expect to keep seeing these sort of infernal acts as people get bright but misguided ideas that require broad adoption to work. for example googles wifi cataloging does not work at all if to get cataloged you have to put "_cataloged" in your ssid.
Sharing your host's WiFi password with all your contacts should never get a wide adoption. It should never be an option anyway.
It shows Microsoft's astonishing ignorance of security.
If you have a contact, they are in their settings, and they're nearby and they can see your wifi network, a prompt will appear on your phone which asks if you would like to share wifi credentials with them.
There's some foolery going on to stop it popping up if you're using the device normally, like you have to be in settings or the home screen - or recently unlock your phone or something... But it's very explicitly: opt-in.
It's not opt-in for the owner of the network, who should really have a say in the matter.
I do use this feature from time to time, but it's typically on networks where either I'm the owner, or the owner's given me permission to share the creds.
This also opens up an attack surface (which I got to experience firsthand on a burner device at DEF CON 31), where someone spoofs an Apple device requesting network creds. The attack itself involves spamming share requests and catching you off guard, causing you to hit OK, or you just hit OK out of notification fatigue.
Computers were opt in.
Too fucking bad for them. This opt-out bullshit for everything like this, marketing emails, etc. is bullshit. I’m sick of it.
In the iOS way, your guest can share the creds privately to another person.
In the normal way, your guest can share the creds verbally to another person, which might be overheard by other unintended listeners.
I guess the ideal would be to allow the network owner (which would be determined by what method?) to share to guests with a flag set for no further sharing (and no viewing of credentials).
I usually try to use this feature when traveling: either I or my wife will add the new Wi-Fi and share with the other. It works roughly 3/4 times, but the remaining 1/4 is infuriating because there’s no button to manually start the sharing and no info about why it doesn’t work.
I would prefer a reliable button to AirDrop the Wi-Fi credentials instead of unreliable magic.
Or banish machines running Microsoft software from the network.
Take that background, and how MAPI essentially prioritises internal email capabilities, and slowly a perfect storm for creation of such misfeatures emerges.
Internally to a corporation, in Outlook/MAPI/Exchange way richer world, implementing such a feature is both simple and possibly easy more useful (less annoying emails to write when you want to just give a short reaction).
But then you hit two confounding factors - systems outside of corporate Exchange server - so instead of using a richer messaging feature you make it into extra text message - and systemd outside the corporate, where your message now leaks out.
This way you can start with reasonably well thought out user story, and end with crap like the way reactions work - and weird extra headers
I like this term, hopefully it enters dictionary. Stupidity doesn't buy yachts, vile malice does.
OP didn't mistake anything, this would be a public office clerk and they don't have private jets nor yachts.
If the former, you'll need to present an argument that Microsoft did not hold back the entire industry for 20 years with low quality products, severe user-hostility, and monopolistic practices.
If the latter, you should read up about the 1980s and 1990s and early 2000s.
Anti-market behavior is today completely normalized, so Gates is very much not alone. Malice is not an unusual phenomenon.
And they are everywhere.