Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Using email wrong (arne.me)
184 points by abahlo on March 6, 2022 | hide | past | favorite | 182 comments


Another tip is that for your inbox (human sent email), treat it as "other people's todo lists". This means you put yourself first at work before you get to other people's work. Your strategy may vary depending on your organization.

When I changed my perspective on email in this light, I started to really see who were the delegators and who were the experts. As you might imagine, the delegators don't typically do much and the experts are usually doing it all.

You can then strategically ignore email and when experts come asking for help, you know it's something impactful. Same applies for delegators, but you typically view them as lower priority as they usually aren't impactful at all from your perspective.


100% -- This is the only thing from GTD that's stuck with me very well; Strongly separate your emails and texts from your own "inbox of stuff to do." Filtering one to the other AND NOT RESPONDING should be a discrete, extremely conscious, step in your flow -- only stopping if it fits the 5 minute rule.


I'm more of a fan of "Inbox Zero for Life": https://web.archive.org/web/20210311074347/https://xph.us/20...


See, why I prefer mine is; storage is (effectively) free, focused attention is not.

Inbox Zero requires you to seriously think and evaluate every single email, but my way allows me to subconsciously triage.

And I just realized why it's so effective and feels so effortless -- it uses a mental mechanism that we all must use everyday, unfortunately; it's the same filter that lets us ignore advertising (e.g. billboards), which we've all had to practice anyway.


In support of delegation:

(This was the philosophy at a Scandinavian firm I worked at).

A more-experienced employee might want to delegate some of their less important tasks to less-experienced people, so that the former can focus on more important higher-value tasks.

If a more-experienced employee with more pay is doing a task that can be done satisfactorily by a less-experienced employee with lesser pay, that means the organization is spending more than it should to get that task done.

Of course, there are some assumptions here:

* The less-experienced is able to do the work with quality.

* The less-experienced wouldn't take an insanely longer time up complete the task than the more-experienced senior.

(Now please don't start a war here saying that everybody should have identical pay and there must not be any hierarchy whatsoever. It's beyond my control.)


Read and flag (Thunderbird: numberkeys 1 to 5).

Only answer directly if it really burns (aka you are responsible) or it is answered in a sentence.

All other things come later.


In defense as a delegator ;

---

I have, through my career, managed, directed, employed, mentored MANY people that were WAY smarter than I am (Looking at you JDB) -- but I had "mortar skills" -- oor as i call them "Lego Skills"

I am talented in bringing disparate engineering disciplines together.

Ill give you a real world example ;

Brocade.

I was the TPM on the build of their new HQ in San Jose, I was responsible for literally the entire go-live of collapsing all offices and data-centers down to the new HQ (I have dont this with other companies, including ILM, Facebook, Lockheed)

Its kinda my jam....

anyway, there were teams Server, storage, network, app, HR, blah blah not to mention the relocation of ~60,000 devices from colos in the valley to the new 4-storey DC building made at Brocade's HQ campus...

I was brought in because I knew how to gather the data req'd to inform an actual migration strategy.

---

I met with each SME of every discipline, asked them of their needs, inputs, workproduct, team, etc...

I did this with each SME for each discipline separately.

I then formed a plan of action, both tactical and hypothetical.

Once I had a plan together, held a meeting and showed the cohesive plan, with every teams dependencies and I threw that up on the wall and had all the SMEs in a meeting and asked them

"Tell me what is wrong with this plan. Show me any deficiencies that your team may suffer, and what is also needed from the other teams to make YOUR team successful. Throw as many darts as you can at THE PROBLEM and not the people on the other team."

A flawless move of ~4,000 employees and ~60K devices to a new colo with ZERO downtime to regular employees.

---

So, just know that just because youre not an SME in a deep subject, delegators have value. In my case : I was not an expert in ANY of their fields, but I knew how to tie a team together in a way that was less divisive. Super successful and I allowed each SME to air their project issues and requirements in private, so as to keep emotions to a min. Esp considering prior to my joining the project, they had not been talking cross-teams at all.

Delegation is a skill -- you need to know enough about the problem (both project and team issues) to be effective, and really what it comes down to, is ensuring that your SME stays focused on being an SME for [SUBJECT] and remove the Human Emotional friction that can happen on teams of really smart people.


In this case I'm pretty sure GP talks about the informal delegators that just wants to push the problem to someone else. Don't take it personally :)


Yeah. No offense to TPMs, PMs, or EMs just doing their job. I meant people who pass the buck.


I am aware, just wanted to make sure this alternate view also was aired out there so people can have perspective and just not automatically bucket delegating managers into a single "passing the buck-bucket" :-)

Thanks for the posts


Quality project management is a joy to work for has been my experience.


Agreed. And it's SO RARE.


Many people are both delegators and experts.

Especially on teams.

In sports, a pass is a delegation.


A pass could equally be seen as delegation by the passee to the passer. “Run this ball up the field and deliver it to me when I need it.”


What pisses me off (and prevents me from using my own version of the system effectively proposed in the article) is companies that send you both actionable and a boatload of spammy emails via the same email address.

Github "confirmatory 2fa" is an example. If I could, I'd turn it off, but there's no option for that (There's only an option to turn it "fully" on. hah!).

I had been redirecting github emails to my preferred folders, but then when the 2fa email came up, I had to spend 5 minutes tracking it down, because only the inbox sends notifications on my phone. So I reluctantly removed the rules, and now it litters my inbox by design.

I hate 2fa with a passion.


Same with phone notifications.

You enable notifications from Uber/UberEats/whatever because it's useful to get notifications like "the driver is on the way", "the driver is here". But then they also send you spam notifications.

And at least on iPhone, I don't know of a fast way to toggle notifications for a given app. Endless-scrolling the list of all apps in the notifications config menu is annoying enough to where I never bother.

My carrier (Telcel MX) does this too. They send important messages about my subscription on the same channel they spam me with offers. They even send me warnings about phishing attacks from fake Telcel ads. Well, maybe people wouldn't be so susceptible to it if you didn't train them to get used to ad spam.


Next time you get a notification from a sender you want to silence, swipe left on the notification, then tap “manage notifications”. That should take you straight to the notification settings for the specific app.


That still doesn't address the problem - on iOS, notifications are "all or nothing" for any given app.


Yes, unfortunately there is only so much you can do.


> on iPhone, I don't know of a fast way to toggle notifications for a given app.

In the notification screen, drag the notification left and hit "Options". It'll give you options to mute, turn off, or fine-tune the settings.


You can turn those off for UberEats FYI. In the privacy settings.


An Australian bank I used has used their 2FA phone number to send me marketing messages about loans. I was absolutely furious about it. 2FA numbers should not be used for your marketing team’s sleazy trash.


I have the same general opinion around "categories of email noise from a company", but I find Github is possibly the best one I know of. It's not perfect. Jira is an example of a bad one (while also having a high volume of email if you set it up for such).

Specifically, for any who haven't noticed, there are several additional email addresses in most emails you'll receive if you're watching Github Repos, and you can filter on these:

https://docs.github.com/en/account-and-profile/managing-subs...

So I have a series of filters for several of those emails, like "If 'mentions' in CC'd put in important folder", plus some filters for "If '[reponame]' in title, put in dedicated folder". And then all other github emails go to some other folder, and never my inbox. (That all said, I may be completely misunderstanding you; as I don't think I deal with "comfirmatory 2fa emails)


There are also quite some headers GitHub sets which often can be used for filters. E.g. instead of checking for `[reponame]` in the title, you can check the `List-ID` to filter for a specific repo.


Rules are the answer. I don't agree with the idea of using separate email addresses for this reason. Instead, let it go to your main one and filter by sender & subject.


I agree but I also have one trashemail-address for all the give-email-to-get-something buttons and every 'serious' subscription-account gets a dedicated forward only email so I can replace them without worrying about which accounts are affected.


Use an Authenticator app (TOTP) as you ought to be doing anyway, and the email 2FA disappears.


Until you switch devices and then you lose access to your account or spend a morning wading through Authenticator transfer hell for each account. I learned that lesson last time I switched phone and now everything lives in my password manager + sms 2fa


There’s a middle ground with an out-of-band shared decrypt. Authy has the most user friendly TOTP sync that lives on device, but are other more open source ways to have your cake and eat it too in this regard.


Or OTP Auth in the App Store, uses iCloud sync. I use that one happily for years now.

https://apps.apple.com/de/app/otp-auth/id659877384?l=en


> lose access to your account

When you set up the device and GitHub showed you a handful of codes and said "don't lose these they're important" you kept them, right?

> Authenticator transfer hell

I've switched my primary TOTP device twice in the past few months and I have no idea what you're talking about.


Paid bitwarden has a cloud based totp that I've been using for this reason. Works perfectly.


There are multiple better solutions to that:

- Use a 2FA app with sync like Authy (even Google Authenticator has basic backup/export support nowadays, but it's still far too easy to lose your codes with it)

- Use a password manager with 2FA support like KeePassXC/1Password/Bitwarden

- Use a YubiKey (ideally have two in case you lose it)


The built-in iOS keychain will let you save your MFA tokens as well, and this can sync between devices. Or, you can use Raivo, and back up the tokens either manually or in iCloud.


If you switch phones often, a Yubikey might be better 2FA (it's definitely safer).


Authenticator added single-code transfer of everything now.


Meanwhile Amazon still sends me an email and a text message every time I log in to a new device on an account with TOTP 2FA enabled. At least those don't require any action from me, but it's still annoying.


Why don't you just filter those 2fa emails on the subject instead of sender, so they end up in your inbox ?


Report them as spam. Eventually, GitHub email will be blocked, and they will need to get their act together

https://glockapps.com/isp-feedback-loops/#:~:text=ISP%20Feed....


Do they provide distinct titles? Could filter based on that on top of domain.


I think if you set up TOTP the emails will stop.


I know none of y'all asked, but absolutely love my email system.

My inbox is my to-do list. If there is an email there, a follow-up action is required. This means that I'm militant about archiving any email that doesn't need to be in my inbox. I've been doing this for at least ten years.

Since last year, I also use Abine's Blur for creating virtual disposable email addresses. (Basically iCloud Hidden Emails before it was cool.) these email addresses forward to my actual email addresses and can be replied to. Everything that asks for emails that could become noisy subscriptions gets a Blur email. If the emails get insanely spammy, I delete the email address. Spam over.

I also have a few email aliases for things like bills, receipts, invoices, etc. Since approx 2.5 years ago, these forward to an email that forwards those emails to Expensify. I also have spend alerts set up with all of my credit card companies and bank accounts, so anything I don't get receipts for also gets sent to Expensify. Subsequently, I have a detailed record of all of my budgeted and unbudgeted expenses broken up by week.


That inbox approach is my approach as well. I've tried many times to shift to an actual TODO list - with the idea that using my inbox as a TODO list means extra overhead - every time I open my email I'm having to re-process my TODO list.

But it never works for me. I keep coming back to Inbox having three priorities:

  - unread (must be triaged immediately - scanned then archived, starred, or left as normal read)
  - starred
  - read
Although it's really two, because starred/read don't really have much difference in priority for me.

Most of the time I sit at somewhere between 0-5 emails. If I'm having difficulties or overwhelmed, that number may grow substantially temporarily.


Using mutt, I can edit e-mails in my Inbox. I sent one to myself with the subject "TODO" that I keep re-editing with any todo-list style stuff. It's been sitting in my inbox for a few months.


Disclaimer, 26 years using emails.

What I love about emails is that no ads are pushed into it, one of the few places where you can read information without Javascript and decide when and if to load external content.

I think innovations should be welcome, but they should have their own name and avoid considering a legacy system wrong by default.

Reinvent the wheel is hard,however I hope to see something great in the future (gpg for everyone for example).


What I like about email is that it's one of the last bastions that are still under your control and work for you, not against you. It's not siloed in some "app" who wants to "engage" you, it lets you set up rules even against other companies' will (I'm sure no company out there wants me to be able to automatically filter out their marketing spam - or worse, redirect it back to some high-level employee and waste their time) and all your data is still within your control and you can build custom software to deal with it if needed.

I pretty much always opt for email notifications as opposed to some proprietary system - this allows me to decide what to do with them (including turning them back into a push notification by using something like Pushover, while the official app's push is disabled because they mix in marketing spam with useful notifications) even when the sender would rather not have me do that.


> no ads are pushed into it

Believe it or not, I receive more spam than non-spam email each month.


Wanted to downvote, but instead I'll reply by saying, obviously parent was speaking of ads in a valid message, not spam.


Nowadays when signing up for valid services, some don't let you opt-out of the marketing emails until much later in the flow; the worst offenders ignore your opting out options by including a new category of marketing-type emails every couple of months that you have to go in and untick.

Wouldn't that be classified as spammy ads from a valid message (when done using dark patterns with no one-click unsubscribe option)?


Honestly, I'd classify those as spam, without any additional qualifications. Giving my email address to a service is permission to send emails related to that service. It is not permission to send advertisements for other services, nor is it permission for affiliates to send advertisements.

Recently, I gave my email address when having my car serviced. The next day, I received an email from Sirius XM, because the shop has a partnership trial deal with Sirius. I marked that email as spam, because it was entirely unrelated to any legitimate use for which I had provided the email address. That it was being sent by a legitimate company doesn't change the fact that it was spam.


I've actually had decent experiences in the last few years. The vast majority of services have a clear opt-out link, and pretty much everything else has an unsubscribe link in the emails that works. I think legit senders are terrified of getting spam blacklisted, so they have to offer than option.

The only exception I found was the democratic party mailing list. No idea how I even ended up on that as I don't live in the US, but NEVER give them your real/main email address. The spam is relentless and impossible to block because they are continually setting up new domain names.


Some "legit" senders only have a "snipe"-unsubscribe: you can only unsubscribe from a given "list", which often has just one message.

When a new ad campaign occurs, they create a new "list" and mail everyone again.

This is spam.


Did you read the linked article? It’s not proposing a new tool, just using rules/filtering to categorize into inbox (humans), paper trail, news letters.


That’s basically what Hey.com does. Inbox, Newsfeed and Receipts.


Yup, the article mentions that the author was inspired by Hey.com


I do inbox zero for many years now and almost get an aneurysma when I see how some people use their inbox.

I filter emails like the article suggests, just a bit more detail.

Newsletter, notification, and work are my main tags.

Then I have some specials like tax office, banks, crypto, academia, etc.

Most of it gets archived right when it arrives.

At bussy times, I have like 20 mails in my inbox for 1-2 days, but most of the time my inbox is empty or had under 5 mails.

I also treat these mails as tasks. Everything in my inbox needs to be tended to, and when it's done I send a reply or archive it.

Works like a charm for over a decade now.


Well, my inbox has tens of thousands of emails. Every January or so, I put last year's messages into an archive for that year. When I look for something, I use the search feature.

Works like a charm for over a decade now. (But I don't get aneurysms when other people handle it differently.)


Haha.

I also get one when I see people having open more than 5 browser tabs at a time.

Guess, I'm just not good at handling masses of information visually.


I have about 23 browser windows open with 3 to 40 tabs each :-)

Come to think of it, it is quite amazing that good software supports so different styles of usage...


The problem with this approach is that if any service/spambot does, somehow, get your personal email, you're stuck with crap.

The better solution is to make a whitelist, and only let that in. Then have a convenient means to let people put themselves on your whitelist. And an even easier way to kick them off your whitelist.

One solution: http://blog.nawaz.org/posts/2018/Sep/solving-my-email-proble...

HN discussion: https://news.ycombinator.com/item?id=18100807


And/Or give everyone a unique address. (Possibly with a unified one for "friends and family", if you want to be super strict even combined with the approach in your link)

Sieve filter to put it in folders derived from the name (you can do this dynamically, no need to put every address in there). Now you're playing with power.

For the first time in decades I feel in control of my e-mail again.


This was my original idea to solve the email problem, and I actually started coding it up. Then I realized a whitelist works as well and is much easier to implement.


I'm curious why you even started coding? All it takes is an email host where you have a catch-all and a sieve filter on a couple of lines.

I saw your whitelist approach before, got a bit inspired, and considered doing the same thing but never got around to it because it seemed like too much of a hassle (it needs to have not only a separate running process but also maintain persistent state). Funny how perceptions differ.

While both share some benefits I think they're somewhat orthogonal. Your approach mostly works for humans mailing you individually (as you noted) while the biggest benefit for mine is for group- mass- and automated e-mail.

IMO the golden thing would be to combine them, only using the whitelist for a subset of addresses you actually use for personal comms.


> I'm curious why you even started coding? All it takes is an email host where you have a catch-all and a sieve filter on a couple of lines.

Well my original idea was to have a unique email address for each person. So I needed a way to generate one on the fly. But now that you mention it, a catch all would have worked - I could generate the email even after giving the person his/her unique address.

Still, there would be some burden to actually create those custom email addresses every time I meet someone new.

The other side of the coding would be to ensure that when I email someone, the From/Reply-To addresses are the ones for that person. But what if I need to email 2 different people?

For me, having a single email address for family/friends wouldn't work - it always gets leaked somehow.


Yes, I use the Fastmail VIP feature for this. I makes a vast difference.

I’ve got tens of thousands of emails in my inbox, even after filtering, and, at my age, I simply don’t want to spend an hour a day pawing over email. I just let it recede into the past. If someone pings me about something I “missed”, I can always dig up their email.

I have no VIP email in my inbox.


Can you filter into folders based on VIP status or not?

Wait, you have "no VIP email in your inbox" -- where does the VIP mail go?


You can set a rule to filter into a folder based on VIP status.

I have no VIP mail in my inbox, because I processed it. And that’s nice.

Edit: I should add that FM has a switch to only show VIP mail in the inbox.


Interesting! Can I filter all non-VIP mail into something other than my inbox? :)


Yes, you can set a rule for when “sender is not a VIP”.


> The problem with this approach is that if any service/spambot does, somehow, get your personal email, you're stuck with crap.

Is this really a problem? I've had my email address for years, am active in the tech industry and have hundreds of accounts set up with that address, and yet I get near-zero spam.

The vast majority of services support opting out of marketing emails and respect that, those that don't are rare and can be dealt with an email rule, GDPR complaint, or both.

On average, I probably get a single email per day that actually lands in my inbox, and those are emails I actually want to receive.


Yes, it's a problem for some. I've had the same email address for almost 20 years.

For me, marketing emails is spam - even if I signed up for it. Often you need to sign up temporarily to access a service, and their signup option doesn't always let you specify what types of emails you want to receive.

> and can be dealt with an email rule, GDPR complaint, or both.

Both take more effort than a simple whitelist based system. I simply press a keystroke and the address is out of my whitelist.

I also can now sign up for anything if I need to and never have to put in any effort to prevent it from hitting my inbox. When you step back and think about it, it's hard to recommend "Really think whether you want to sign up, and later unsubscribe/set an email rule" vs "Sign up, and don't worry - it'll never go to your inbox."

For most of my email life, I followed your route: Be careful about what I signed up for, and spend time filtering/unsubscribing. When I switched to whitelists, it was just so much better. Of all the things one has to deal with in the world, spending time tending to the inbox is just wasteful.

In the last day, I received 14 emails. 2 were personal. A few were alerts I've set up (e.g. credit card spending exceeding X amount) - I wouldn't want to lose these, but I also wouldn't want them in my inbox. The rest were for things I signed up for because I needed to buy something but did not offer an opt out option while signing up. I'm not going to spend any time unsubscribing from them.


Why aren't there any email services that provide "Zero Trust" email? An email firewall would solve so many problems:

- All email goes directly to a PROCESS folder - Emails you have responded to or in your Contacts get processed by Rules/Enter Inbox - Allow/Deny context for individual or entire domain - Display sender domain prominently - Display TLD in alert before following any clicked link - Block link clicking, images, etc. from unapproved contacts

Now you are left with seeing only the things you approve of, unless you make a willing choice to go and view the Wild West of unapproved senders. Once trained, it's easy to see that you don't often get unsolicited messages you are interested in.

This should be easy to do, but no provider I've used has made it the default. I've had to cobble together my own solutions to make this work.


I have a rule in Thunderbird that puts all email that comes from senders not in my contacts in a special folder.


Same here. For simplicity, incoming emails from senders that aren't in my addressbook go automatically to the "Junk" IMAP folder, together with what the email hoster's spam filter moved there.

That rule has a few exclusions. For example, incoming emails from a handful of particular universities and companies I work with can go to "Inbox" because seeing them ASAP is likely important.


What happens when you ask someone to email you, but you don't send them an email first?


Then I might not see their message for a day or so, unless I'm actively checking more frequently for it, or unless I add them to my addressbook ahead of time.


That's why my rule doesn't put it in the junk folder. I put it in a sub-folder of the inbox so that it is easily seen.


Because this increases load on tech support for people who thought they approved everything but didn’t.


Isn't this what Hey (Basecamp folks) does with their "Screener" feature?

https://www.hey.com/features/the-screener/


I did this as well: http://blog.nawaz.org/posts/2018/Sep/solving-my-email-proble...

I believe hey.com provides this. Some years ago I did find a few other obscure providers with this service. Here's another one: https://www.spamarrest.com/features/


That's basically what hey.com (mentioned in OP) is doing, i think, although I haven't used it. it's gotten mixed reviews.


But if you sign up for services with papertrail@, what about when that service sends you something you need to take action on? (Update payment details, etc) Wouldn’t you miss it?


Yeah, that method won't work. Almost every service that I use sends all 3 types of emails: paper trail, news, and occasionally something important.

My approach is to make custom rules for the most frequent emails based on sender and subject (the paper trail type emails often have predictable subjects). What's left is between 5 and 10 emails a day that I can sort manually.


I do that too, but it ends up being an evergreen chore to setup new rules and every once in a while I give up :/


I typically just unsubscribe from the news.


that was my thought too


My personal anecdote:

When I worked at some MSP I forwarded an email to the team lead and walked to his desk to discuss some things about that email.

When I asked if he saw the email from me he just showed me his Outlook littered with bazillion messages and my email wasn't even on the first page, despite the time I need to walk to his desk was minuscule - we were sitting in the adjacent rooms.

After seeing that and the look of despair in his eyes I asked him to allow me to look what exactly was spamming his address so hard. It turned out what the bulk of it was the automated messages from a various monitoring tools and tons and tons of CCed messages, because corporate bureaucracy and the team lead "should be in the loop", duh.

So I took my time to write a proposal to reorganize all that shi^W mess.

We had a couple of pretty big clients which we supported and we had a couple of teams in our organization supporting different roles (server/hardware, SAN, virtualization, networking etc), so I proposed the creation of a bunch of mailing lists/groups (we were using Exchange from the parent company) with an address split by the role and the client abbreviated name (2-3 letters): $role.$client@msp.tld, ie:

hw.CL1@msp.tld, san.CL1@msp.tld, monitoring.CL1@msp.tld etc and

hw.2CL@msp.tld, san.2CL@msp.tld, monitoring.2CL@msp.tld and so on.

Of course it took some time to implement (especially changing the addresses in the various automation tools), but after that and with the help of some pretty basic Outlook rules to sort out the messages to the corresponding folders the usual 1100+ messages per day in the team lead's Inbox dropped to less than a 100.


Interesting. I just find it pretty easy to just never delete and use "new." If something's super spammy I'll filter for it, of course, but even dividing "papertrail" and "humans" seems like an extra step I don't need.*

*note, this requires a speedy client where you can see a good number of subject lines and main body at the same time and just kind of arrow down quickly; I switch variously between mutt and thunderbird. I don't know how people deal with stupid bloated behemoths like today's Outlook.


I've been using Thunderbird since early 2000's. I used to run my own mailserver; stuff happened, and I decided to switch over to my ISPs mail offering (which is the same as what I ran: Postfix, Dovecot, Spamassassin, and Squirrelmail/Roundcube if I need it).

They also provide a Sieve server. This is important.

If you use built-in Thunderbird filtering, then it only runs when mail is delivered to Thunderbird; and unless all your clients have the same filtering rules, then things could get messy. With Sieve, the filtering occurs on the delivery server; so it happens before you start your client.

If you use a webmail service with filtering, that also occurs on delivery, and server-side. But Sieve is great if you want to run a proper mail client.

The Sieve rules notation is a pain; but you can use Roundcube as a Sieve rules editor - it provides a nice interactive form-based rules editor, so you don't have to remember the rules syntax.


Funny, you're describing things that I've tried to do in the past "in theory," but after it all, I've found my own brain-filtering (on a big monitor screen) to be sufficient. Namely, even if it's like 40-50 emails a day, with maybe 2-3 from humans, it's not bad when in Thunderbird I'm not preloading html/javascript and I can see all the subject lines. I can just arrow down the "news" pretty fast.


> Namely, even if it's like 40-50 emails a day, with maybe 2-3 from humans, it's not bad when in Thunderbird I'm not preloading html/javascript and I can see all the subject lines. I can just arrow down the "news" pretty fast.

But it's still a time & attention tax you have to pay every day. Implementing rules based on sender & subject would be a one-off task and should cut that down dramatically and have a lasting impact.

In my case I probably have a hundred rules or so for any automated crap, and now I only get on average one email per day that makes it into my inbox and in the vast majority of cases it's an email I actually want to receive (and if not, it gets flagged and left there until I have time to set up a rule, unsubscribe from the list or send a GDPR complaint).


The trash/spam part of it literally takes less than 5 min a day? I'm not reading each email. Something something premature optimization? :)


Some interesting companies in this space:

- Paced Email (https://www.paced.email) - you make special email addresses that will collect all the emails you have, and release them at the specified frequency (e.g. once a week on Friday evenings). I use this to make newsletters not interrupt me throughout the week. I often will do one pass where I just triage all the stuff that shows up in my paced emails, and then actually consume that content/respond to it at leisure afterwads. - Mailman (https://mailmanhq.com/) - it withholds all emails except those from "VIPs" to be delivered at specified time slots, by default 3 times a day, so that you don't get interrupted by unimportant emails. Additionally, they're working on a Paced Email-style product to release emails at intervals.

A couple major downsides of Mailman vs Paced Email is that you need to trust Mailman with access to all your emails, and it only works with Gmail; whereas Paced Email is opt-in via special email addresses which you can configure via separate DNS, and isn't vendor-specific. But it is more all-inclusive regarding wholly controlling your email cadence.

Also, while somewhat expensive, I do find that Superhuman does make it easier/faster for me to both sort and burn through emails, which makes coping with the volume of emails much easier. I use filters to sort my emails into separate buckets to get around the issues that other commenters have with company email addresses being overloaded for both important and unimportant use cases.

I've come to view handling email as 80% about quickly triaging them - ignore/unsubscribe/silence anything that isn't important, snooze anything you can't/don't want to handle now, then handle everything else efficiently.


Another option is "inbox" , "p0", "p1", "p2 . inbox & p0 you check once per hour, p1 once a day and p2 twice a week.

Instead of categories, assign an attention budget.


Email is just a messaging service. Whatever messaging service you use, the core is the same: people send you messages you need to act upon. Chat apps are no different from email in this regard, they just mask the problem of a non-zero inbox by merging all messages from one sender into a single thread. This way, important things you'd want to keep unread in email sinks into oblivion until you forget about it. (And feel good like you did everything you had to)


You just made me realize that email would look better if it grouped messages by sender instead of just by subject.

It was a huge step forward for me when I switched from Outlook to GMail. Outlook originally did not group emails at all. Each message was a line item on its own, sorted by date sent. So even individual replies back and forth were scattered across my Inbox. Then GMail grouped all replies into a single line item, as a "thread" or "conversation", which I could then expand. Eventually Outlook followed suit, although at first imperfectly. It grouped by Subject instead of the more accurate email header, Message-ID. (Actually I believe this technique predates GMail, at least in one text-based email reader, Mutt.)

I think email should take this one step further, and group messages first of all by sender, then by thread.

  sender (or recipient group for emails with more than one recipient, like SMS clients do)
  |- thread
     |- message
This would help with companies that send you many messages, whether it's spam or just receipts and other notices. But it would still not be bad for emails from individuals too. Even though someone may write you many emails about many different things, each with their own Subject, it still makes sense to group them. That mimics real life, where a real person's many conversations that they have with you are still visually grouped into one person, one organic body, in your mind.

Facebook would also be better this way. Collapse posts by poster, so that people who post 5 times a day don't take up more of your news feed than those who post just once every now and then.


I was a heavy email user before Gmail advent, and was always using folders in my email clients to do just that. For me it turned out that most emails work better the gmail way. I'd benefit from a simple option to group all messages from/to a specific sender into one thread though


I've been doing the same for a bit over two years, but using the sender's email address and a screening process instrumented with some app I wrote (Screenr[1], shameless plug). That gives you something similar to the process in Hey.

I don't think I'm using email wrong.

Unrelated: what is controversial about Basecamp?

[1]https://github.com/cfe84/screenr/


The controversial part may be a reference to Basecamp's last year's ban on “societal and political discussions” at work, which was apparently motivated by the desire to end internal discussions about the company itself [1].

[1]: https://www.theverge.com/2021/4/27/22406673/basecamp-politic...


I'm curious to see a list of recommendations for vendors that pass the "non-controversial" test there: every employee likes what every other employee and manager is doing (and presumably, every non-employee should too).


Indeed, what is controversial about Basecamp? Have been reading their blog (and now blogs) for over 15 years. I can't speak as a customer - or as an employee - but they seem to be one of the most humane companies I've come across.


Kerfluffle last year where they shut down their DEI committee, adopted a politics-at-work policy similar to Coinbase’s, and offered six months’ severance to anyone who wished to leave. ~40% of the company took the offer and the story was all over tech twitter, HN, etc.


With over 121,000 unread emails in my “everything” gmail account I’ve come to the same conclusion.

It only becomes a major issue when I want to find an email from a real person. It’s amazing - every marketing email seems to come up on every search I do almost like they’re keyword stuffing.

If there’s a better way to do it that doesn’t involve obsessive inbox 0 protocol but what the OP describes is what I’ve mentally decided to do as well.


One thing I do is that if something comes in that I think I might want to refer to at some point, I'll give it an &archive label. (I used to have a more elaborate set of labels but these days I mostly just throw most things in one place and rely on search for the rest). And then just periodically delete older stuff that hasn't been files--especially if it's unread. I might take some time to skim things to see if there's anything I want to save.


Isn’t this kind of email easy to classify (on a large scale)? Gmail has a similar inbox system in its web reader.


Yeah, the tabs on my personal Gmail work pretty well in keeping down the clutter. Very little that ends up in Promotions needs to get looked at and I just delete the whole thing periodically. Updates grabs most of the rest.

I actually used to use a separate email address at Yahoo to handle all the ecommerce cruft but I find that Google does a good enough job of organizing that I don't bother any longer.

It's also not a bad idea to go on an unsubscribe mission every now and then.


It doesn't work that well. Why ask a computer to guess what I want when I could just tell it?


I do unsubscribe to things now and then and I find Gmail does a pretty job of keeping the large number of notices of sales, new products, etc. separated from my main inbox. And although I ignore most of this I do find some useful information now and then.


yes. this article reads like an ad.


What for? Hey where the author got the idea from? The author doesn't use it. Fastmail? The author's method works with any mail service that supports custom domains and email aliases. Seems a stretch to call it an ad when it's not clear what they're promoting.


My first impression was also that this was an ad for Hey. Maybe that's not what the author intended, but it reads that way to me and it's a bad first impression.


But a bad one with no case for unique value being made.


The most important concept when it comes to email is email bankruptcy.

Too many emails? Even in the "from a human, directly to me" category? Select all, mark as read, archive. If it's important enough it will come back.


A solution to avoid declaring bankruptcy would be to sort by sender. Turns out even in the fullest inboxes, the actual number of senders is manageable, and you can easily select all for a given sender and flag (for stuff that does require actioning), delete or archive.

Ideally, do this in conjunction with unsubscribing from the associated list and/or create rules so that automation takes care of future emails of that kind.

It might take you a few hours but in exchange you'll get back a usable inbox that will remain usable in the future thanks to the rules you set up.


Do we need to batch notify all senders that we are doing this?

How often are we allowed to do this before it becomes rude?


No, you don't need to notify.

At the point you "declare email bankruptcy", you are probably already missing and not responding to emails. Rude or not, it's already happening. email bankruptcy is an attempt to keep it from continuing to happen, like it will if you continue on the course you are on.


As a developer I get hundreds of automated notification emails and somewhere in there is an email that was actually typed up by a human. I tried to turn off all those notification emails but people keep adding me to automated lists that I cannot control. I just cannot beat the incoming flow of emails. So I have been ignoring my email for the last year and nobody really complained about it. If something is really important people will contact me by Teams/Slack, whatsapp or an old fashioned phone call. The means of contacting me also reflects the priority.


> I’m using Fastmail with a custom domain and have aliases with rules for the different destinations, for example papertrail@example.org or newsfeed@example.org

I want to try the custom domain thing eventually, but will have to be careful not to let my domain get into the wrong hands. So I will be turning on 2FA on my registrar, and stay away from ccTLDs that have reputation issues (like .ru for example). Also I will ensure the domain keeps getting renewed. The only caveat is if I DIE someone can reclaim my domain and get access to all my accounts, which is worrying.


I’ve been doing this for nearly 15 years and haven’t died.


This comment would be extra creepy if you had.


After hearing horror stories about Google disabling people’s accounts and them losing everything, I’ve started to do this. I still use gmail as a client, and I haven’t moved everything over, but I feel more comfortable about it all. The next ate (for me) is to start paying for my domain 10 years at a time.


I've been doing this for 10 or so years with no problem. I have a .org domain, auto renewal is set up, and I've enabled all email notifications around domain status with my registrar (which, yes, has 2FA enabled).

Hadn't thought about what happens to this setup when I die. Probably the simplest thing to do is give someone I trust access to the account. Alternatively I wonder if there's a way to keep credentials in escrow, which can only be legally released to a named beneficiary on death.


Maybe you can put in your will to have a lawyer go through and delete all your old accounts? Would be useful for a living will too; if you're in a coma for 10+ years, the domain would lapse then too.


> The only caveat is if I DIE someone can reclaim my domain and get access to all my accounts, which is worrying.

Nobody can hurt you once you are dead.


You might want to lock into using a registry that supports "registrar lock" as well. That will protect you from unwanted changes and deletions.


I’ve done the exact same thing, and in addition all email that’s not flagged as Spam goes to a folder called Screener … the workflow for the Screener is to add addresses to Groups (of Contacts) & then I have a small number of rules that file mail. So any email sent to me that’s from an address I’ve already received is getting filed into a Group’s folder. I never would’ve thought that without seeing Hey. I do wish the workflow could be that dragging an email from Inbox or Screener to a Group folder could add addresses to a Group.


I don't have all the aliases but I agree that inbox is for human emails and stuff that needs important/urgent action. Everything else gets unsubscribed or a custom gmail filter + label.


If you’re a regular user of a free email service, or don’t want to create multiple email addresses on your paid email service like Google Suite, you can always use aliases.

If your email is me@mine.com, me+[anythinghere]@mine.com is a valid email address and messages sent to it and will end up in your inbox.

I tend to use specific aliases for services that send a lot of notifications (like me+amazon@mine.com) and generic ones (like me+archive@mine.com) for everywhere else (specially tools I am signing up for to try out, and I hate drip email campaigns they trigger)

You can then set up rules on your email service provider to automatically route emails to a sub folder, spam or archive based on the to address.

I use G Suite — I mark all emails from Amazon as read automatically (I still want to be able to search for them if needed), move most email to spam automatically, and move newsletters I really want to read to the inbox instead of updates and star them.

Oh, and you can also automatically respond with “unsubscribe” to stop receiving emails from services. Most marketing email services support this.


I've been trying to use the "+" alias whenever I sign up for literally anything. At least half of the e-mail validation methods out there have no idea that "+" is a valid character. Quite annoying. Luckily I use a custom domain and everything at that domain forwards to my one user account, so I just use a new email name for fields that don't support "+".

Fastmail has a feature to automatically sort mail with a "+" into a dedicated folder, so you only need to add filter rules for those addresses that don't support the "+" character.

You still can't ignore the filtered folders, though, as sometimes you'll receive a notification out of the blue from a service or a person that is time-sensitive. So I went back to one giant inbox and I now sort mail that doesn't have a "+" sign into a dedicated "Personal" folder. Have a few more filters for jobs, receipts, and other emails that will reliably generate mail I don't ever need to read.


Nice. I have been following something similar but even more straightforward.

INBOX - Everything that I'm OK with, else they are either unsubscribed or blacklisted at my @oinam.com (humans and non-humans).

I have my primary email (since 2005), work emails, and a few other project emails all on the same Apple Mail Client[1]. I also look at just two panes -- the mail list and the message window -- no toolbox, no nothing. Took some time but have learned all the possible keyboard shortcuts.

I follow the -- reply if I can do it quickly and within a few minutes, else flagged for scheduled email-time.

Newsletters that I want to subscribe to; are subscribed with the known formula of brajeshwar+source@gmail.com. I use Gmail just for newsletters and beta sign-ups, waitlists, etc.

https://cdn.oinam.com/img/oinam/brajeshwar-apple-macos-mail-...


Wait wait slow down.

I can sign up for services with plus signs and use the plus signs to do things? Does this work on Gmail? What other services does it work on?

Is this a well-known fact???

I've never heard of this. My mind is about to blow here.

I worry that bad actors might strip the plus parts out, but in a basically cooperative ecosystem, this seems like a really big deal.


It works on most email services. Some bad actors (including large corporations) do strip the plus signs out (or don't accept them to begin with) but it's rare for anyone to bother.

I typically use the name of the service after the plus, because it gives an added bonus of knowing who is selling my address.


Support for plus addressing mainly depends on the online service you're signing up with. Your mail provider almost certainly supports it; their business is email. But many webservers have idiosyncratic notions about how to validate an email address (they could be using any damn library, and their business isn't handling email).

Most websites seem to accept plus addressing. But I haven't encountered an online banking service, for example, that accepts plus addressing.

Incidentally, it doesn't have to be a plus sign; other symbols can be used.


A more interesting thing with Gmail is that it ignores dots. Foo@gmail.com is the same as f.oo or f.o.o! That's less common behavior, so it's less common for spammers to strip out.


Yes. Any email service should allow it, as it's standardized and very old.

However, some services that you sign up for might reject email addresses with + signs in them.

With Gmail, one thing you can do is insert a period anywhere you want in your email address. For some reason they decided to ignore that character when matching incoming email to an account. That behavior isn't standard, but services you sign up for should never have a problem with it.


> Yes. Any email service should allow it, as it's standardized and very old.

As far as I know it’s not standardized at all; it’s not in the SMTP spec. It’s actually annoying that some people assume this is standard because they can break functionality, like assuming foo+bar@ is the same person as foo+qux@ when it’s not guaranteed to be.


It's not in the SMTP spec because SMTP doesn't need to know about it. Your objection is correct but in fact that's the desireable behavior -- senders can't assume that foo+bar and foo+qux are the same person. If they could make that assumption, subaddressing would be defeated as a useful filter signal. As it is, it's not just plus signs -- qmail has been using '-' as the subadress separator for decades.

Subaddressing is standardized in RFC 5233 as a filtering signal.


On gmail and fastmail and maybe some other providers, you can put a plus sign before the at sign, and the plus and anything in between are stripped out so you can effectively have infinite email addresses.

Not all providers do this, but it's easy enough to check if yours does, just send an email to yourself.


Note, on gmail you can't filter emails based on the +


I've definitely had filters set up based on this in the past. Did they change it in the last few years?


And some services refuse to accept these addresses.

I've found old school approaches such as https://www.spamgourmet.com/index.pl to solve this better.


Yup.

Another tidbit (and this one is gmail-specific): dots are insignificant. Mail sent to x.y@gmail.com will arrive at xy@gmail.com


but beware: there are a lot of services by overzealous colleagues who 'check' email addresses and reject '+' (for religious reasons?).

Email addresses can be quite complex (cough), what those may never have come across. See 6.1 in RFC822.


Yes, but not that well known. It’s part of the SMTP spec.


It's not part of the SMTP spec. SMTP treats the local part as an opaque identifier. It's somewhat common though and RFC 5233 extends the sieve filtering language to support it: https://datatracker.ietf.org/doc/html/rfc5233


Yes and yes.


I wont list my email tips, because they will cause your bank to flag up your online purchases as fraudulent when they are not, thus generating a shed load of extra phone calls and other activities in order to shop online.

But its a great way to check out what banks and card processors use as metrics and what they data share, and you can find out when someone gets hacked or ignored marketing preferences to stay off mailing list. Oh and spam becomes a thing of the past without needing any 3rd party anti spam provider.

Think out of the box.


^ hey everyone, this person set up a catch-all email inbox and routes mail from sites/vendors to site@theirdomain.com - eg Spotify goes to spotify@theidomain.com

Groundbreaking!


I just use completely different email accounts for different purposes. This also makes it easier to receive email notifications at a different rate and on different time schedules. So, notifications for some emails are displayed quickly, some once a day, some never (manual refresh only)...

Can you do this with aliases on the same email account? I guess it's possible, but less likely to be supported by a typical email client.


Some spammish senders will strip the + because they know people use it to segregate their mail into different buckets and they don't want this.


With many email services, you can set up unlimited aliases for your own domain. I do this with Fastmail. No need for the +


You don't even need to provide your own domain - Fastmail gives you a subdomain for free. For example, if your Fastmail address is myname@fastmail.com, then any_address@myname.fastmail.com will get to you.


Does this really need to be said in 2022? The advent of plus addressing was more or less driven because techies were already able to make custom addresses for their domain, and the + could be a standard thing that non-techies could use with their basic email account without needing to reconfigure things on the backend, or use a messy catch-all account.


> Does this really need to be said in 2022?

Something you have known for years may be new to someone else. Relevant XCKD: https://xkcd.com/1053/


And some clueless bigcorps just choke on email addresses with +


Which means their email will go straight into the trash.


The problem with adopting a strategy like that is it would take months or years to make a meaningful difference, unless you actually went back and switched over all the services you've subscribed to, to these new accounts.


Another little note is that Fastmail also allows you to mark all emails for a specific alias as "read".

I personally have a newsletters@example.com, buys@example.com and signups@example.com all automatically marked as read.


Funny story about using the alias as suggested, I have a couple of websites I can't unsubscribe because they can't process the character "+" on their forms.


it's essentially automated filtering by separate email addresses for human, papertrail and newsfeed, right?

Resonates quite a lot more than one per sender as I tried for some time.


Anyone have a suggestion for those sites that don't allow + or % in the email address? This is always the hangup for me when I try to organize like this.


I use a catchall on my domain, and every site gets its own email when I sign up.


Custom domain + cloudflare email forwarding service

E.g.:

* inbox@custom.me -> me+inbox@gmail.com


You can choose some other character for the same purpose? Like .-_!

(Assuming you are using a solution that gives you that level of control)


Outlook.com let's you make aliases and you can direct their messages into a folder. I.e myshopping123@outlook.com


Then you have to setup a more complicated filter on the client side based on sender, etc.


And if you’re getting enough mail from an organization to matter (and to be annoying in your inbox), then the cost of setting up a filter is tiny.


Outlook has aliases. I sign up for non-important stuff like newsletters with an alias, which filters into its own folder.


If you haven’t tried Hey from Basecamp it truly is a wonderful service and I can’t recommend it enough.


> Inbox — This is where all emails sent by humans end up. That’s it.

How do you filter on this? Do you use a CAPTCHA?


He has separate email addresses on a custom domain.

He then gives one of those separate emails to humans.

Companies he orders from gets another email address.

Social media companies get another one. Etc.


- a while back I tried to apply this principle in my gmail - filter everything not from a human out of my inbox using rules and filters (filtering anything with the word “unsubscribe” works well until someone forwards a thread to you that started with an email from a service).

It is a little frustrating though, I’m sure gmail and most modern email services could do a very good job on a filter like this if they tried.


No, it‘s not perfect — but if something slips through, I‘ll sort it out manually.


I use a separate address that I only give to people. Never use it for forms or sign-ins, and if anything marketing-related makes it there it's instantly spamboxed.


I'm a happy Hey user. It made my e-mails manageable again. It's so good!


If you're using gmail, just mark everything that isn't sent by a human as spam.


So you mark receipts from purchases, travel confirmations, and newsletters you signed up for at some point as spam? Honestly, if it's from a legit organization, it should be just as easy to unsubscribe from any periodic mailing.


I'm not the person you replied to but I do use the spam button for any unwanted emails I get. If it's a receipt or a shipping notification then no worries, thanks for letting me know but if you think me signing up for your service or buying something from you is me opting in to your email marketing then you're mistaken and to the spam box you'll go.


But it's not spam and you're messing with the filter


yes, and on my work email, this also includes. cold emails from recruiters and sales people.


It's actually far worse on my work email. I attend a lot of events (whether virtual or in-person), I download a fair number of docs some of which are gated, etc. The bottom line is that I get on a lot of email lists. Some I'm actually interested in but I also get a lot from someone who scanned me at a booth when I stopped to ask a question and so forth. The reality is that channels like these are where a B2B company gets a lot of their leads.


yep, my inbox is great as a result too.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: