That bad cosmic joke again

And another day where the frustrations mount to the point that I feel like screaming (and did, for a very valid reason, cry a little)—and, no, I do not write about even half of these feel-like-screaming days:

I woke up in the morning to find a server hosted at Hetzner inaccessible. From the last time around, several months ago, I knew that the web/html interface for resets no longer worked in my browsers,* something that had back then cost me more than an hour of trial and error, until I had stumbled upon a web-service API that could be used from the command line. I now tried to find the commands that I used back then—but found no trace.**

*Barring later fixes. As to the reason, I can only speculate as no proper error message was given. Hetzner is yet another case of a service provider that gradually makes the interfaces less usable.

**Again, I can only speculate on the reason. Maybe this was something that was lost between backups before a notebook crash. Cf. earlier texts.

I went on the Internet to search for the API again—and found nothing. (Search engines are not what they used to be.)

Before moving on to the provider’s website,* I decided to check my emails. This, in part, because I already felt my annoyance growing, fueled by prior poor experiences with Hetzner,** and knew that a break was a good idea; in part, because I should have received email notifications about the server being down, and thought that these might contain a direct link to the right documentation. (They did not.)

*Known to be poorly structured and filled with “Buy now!” messages. There is a reason that I began with a search engine.

**Including the aforementioned issues, several others issues, emails to the support that either go unanswered or are answered weeks later, and an actual letter to complain about e.g. the unanswered support emails, which is still, it self, unanswered—after several months.

Among my emails, I found a message that my step-grandmother, one of the most lovable women that I have ever met, and as dear to me as my regular grandparents, had died. Day ruined.

I spent the next half hour writing a short email, just a few lines, to my step-father, hindered by how hard it was to find suitable words, my own sorrow, and a mind that kept wandering, especially to how different things turn out for different persons: she made it to an amazing 107, while (among many other losses) my mother died at 67, my maternal grandfather at 61 or 62, depending on the months involved, when I was 7, and my paternal grandfather at, maybe, 69, when I was 1 or 2.

Email done, I then went to the providers website. While it was as bad as I remembered, I soon found the information, and with a bit of puzzling, I managed to recreate the right commands in just a few minutes.

Things seemed to, within what was possible, be looking up a little. I went back to my emails, to clean up all the “server down”, “reset requested”, “server up” messages—and found that the half-hour email to my step-father had been unilaterally rejected by Gmail as alleged spam. Kicked when I am already down.

Now truly at the point of wanting to scream, I moved on to this text to get some pressure relief.

Excursion on Gmail:
There are a great many reasons not to use Gmail, including major privacy and security concerns. This mishandling of spam filtering is yet another.

Apart from the above misclassification being absurd in light of the contents, format, and sending address, there is not one line in the return email on how to remedy the situation. Moreover, email providers simple should not reject the delivery of emails, except in the most blatant cases: they might well classify an email as spam, but the email should still be forwarded to the recipient, so that he has the ability to override the decision. Spam-filtering is and must be a user* decision.

*I originally wrote “client-side”, which would be the typical case. However, a server-side intervention under the user’s control is equally valid. Moreover, there is the possibility of a third-party, e.g. an employer, forcing client-side filters outside the user’s control, which is at best disputable, at worst as bad as what Gmail does.

This is the more absurd, as an email to my father, who also uses Gmail, was not rejected as spam—or, if it was, I received no such notification.

Advertisement

Follow-up: Notes on my recent flight with Finnair

Three follow-ups to my complaints about a recent flight:

Firstly, my chronology* in Stockholm appears to be off: Add half-an-hour to 12:15 and we might still have taken off more-or-less on time. However, we arrived in Helsinki with a delay of, possibly, twenty minutes. Exactly when and where these twenty minutes were lost, I can no longer reconstruct; however, chances are that they took place in Stockholm, as airline schedules tend to have time buffers, and they might well have taken place as I waited to board. In that case, the time waiting and standing must be increased; if not, then some other time interval.

*I only began to takes notes once on board the second-leg airplane and went by memory for the earlier phases.

Secondly, the overall delay, relative my travel time, was shorter than it felt: about an hour in the air; about an hour lost due to the re-scheduled meal. With hindsight, I wrote as if I had lost twice that amount of time, and might have been more forgiving, had I written the same text today. However, as with a working day, every extra hour hits home that much harder the longer the day has already been. Moreover, the circumstances, especially the screaming babies, made it that much worse—and, indeed, much of the earlier text was directed at the circumstances and the mishandling by Finnair (as opposed to just the delay).

Thirdly, Finnair has continued to give a poor impression, even after the flight: I sent a link to my text as a courtesy information. In order to do that, I had to visit the imprint to find a usable and even semi-relevant email address—the official contact channels pushed those user-hostile contact forms. It then took more than a week before I had any type of reaction* and the reaction was a complete disaster: Half the (short) answer was in German, informing me that email support was only available in English—despite my having used an email address for Germany and/or Germans. (And what about Finnish?) The other half claimed that

*I did not necessarily expect, let alone require, a reply, but given that one was sent, I would have expected something much more professional.

Could you please attach a file with your notes? Due to security reasons we do not have access to this webpage. As soon as you send it to us, I will forward your Email to our responsable department.

This made clear that English was not a strength of customer support either. Moreover, the approach to online access is inexcusable and amateurish (if, admittedly, not uncommon). Now, if Finnair wants to deny their “first-level support” Internet access, it is of no concern to me. But: if so, Finnair must take measure to ensure that e.g. blog entries and similar linked-to material can be read when needed, e.g. by providing a means to request retrieval by staff members with greater rights. To push the extra effort onto the customers is ridiculous. (And, no, I am not going to comply.) Moreover, either this woman is too daft to just forward the link to “our responsable department” or there are restrictions on Internet access well beyond first-level support. And: what if my “notes” had been in German too? (And: what if some customer does not speak English at all?) Either Finnair would now have demanded a translation, which would be far too much effort, or the original claim that support was only available in English would be proved incorrect.

Too many emails

Unethical, annoying, intrusive, customer hostile, whatnot, sending of email is not limited to spam*. Consider e.g. my recent flight booking with EuroWings: As of now, I have received a total of five (!) emails as a result, two which should definitely not have been sent, one which is disputable, and two that are acceptable—and this not counting the “please rate your flight” that I expect to receive a few days after the actual journey. The result is a waste of my time, a risk that I accidentally overlook something important among the unimportant, and a recurring feeling of “not those idiots again”. Five emails from one sender might still be tolerable—if it was only that sender (but it is not). While I am comparatively passive in the eCommerce and whatnot field, even I occasionally have two or three businesses sending such unwanted emails in the same time period; those more active might have even more, possibly supplemented by a range of newsletters and similar messages. (In both cases, obviously, this is on top of regular spam.) This is just not in order.

*With the reservation that I tend to think of more-or-less any unwanted email as spam and might use the word “spam” in that sense on occasion.

To look at these emails more in detail:

The acceptable ones are the confirmation of booking and the confirmation of payment-received (contingent on the fact that I payed by invoice; another payment method might not have allowed this email).

The disputable is a notification about online check-in. There is some legitimate value here, but in my case, and the case of most repeat customers, it would have been better if the email had never been sent: firstly, the information and link to check-in would better have been included in the confirmation, the one justification in the delay being that online check-in is only available within three days of departure*. However, I knew about the three-day rule, it is mentioned elsewhere**, and a pre-mature call of the included link could simply lead to an error message of “please try again on or after the Xth”. Moreover, the current implementation is a definite “value subtracted” one compared to a manual visit by the customer: It is possible to check-in with just the booking number, but the link still leads to a page which insists on a log-in or new registration—almost certainly for the unethical reason of tricking unwary unregistered users to register, regardless of whether they consider this in their own best interest. Even for the “wary” this is a negative, because additional steps are required to find the right page for an account-less check-in.

*It self possibly disputable, but off-topic. I suspect that this is related to choice of plane, that the exact model, seats available, etc., are not finalized earlier than this. However, in a worst case, an explicit choice of seat could be replaced by more abstract criteria, e.g. window/middle/aisle, close to exit/faraway from engine, whatnot. Then again, cf. below, an earlier seat-choice than check-in appears to be possible …

**I have not re-checked exactly where, but I do know that I noted it during my booking. If it is not present in e.g. the booking confirmation, it would be easy to add.

Moreover, this is exactly the type of email that could be imitated and abused for phishing, and the prevalence of which lowers the sensitivity about phishing in the general population. (Indeed, even I did not reflect on the risk until I had already called the link—but on no point did I enter any data that could be of use to a phisher.)

The unacceptable ones: Firstly, a patronizing checklist with (the German equivalent of) “Have you thought of everything?”—pure idiocy and, if at all needed, it should have been provided together with the confirmation information. Secondly, a request that I choose my preferred seat. Notably, the choice of seat came at a time when check-in was not yet possible, implying that I would need to visit EuroWings website twice (once to choose seat; once, a few days later, to check in), were I interested in this offer. In as far there is some value here, it is limited and not worth the bother in most cases. So, I have a greater chance at finding my preferred seat by choosing before the time-limited check-in, but the rules are the same for everyone and the difference is likely to be small even for those keen on specific seats.* In contrast, if the ability to choose seat (or even check-in) was available at the time of booking—that would be good!**

*I suspect that most people are not that interested to begin with, especially as information on the more important criteria, like annoying or four-hundred-pound seat-neighbors, loud near-by children, and similar, are not available in advance …

**But note that restrictions as in the above footnote on the three days might apply.

As to the constant “rate this-and-that” emails, they are an inexcusable intrusion upon the customer and a poor way of getting feedback.* In fact, I suspect, it is less a matter of getting true feedback and more of aggregating statistics, which, while of some value, is less useful than more specific feedback. Firstly, any forms and whatnots for feedback are better given with a confirmation email than after the fact, so that the customer can chose when and if to give feedback. Secondly, if I want to give feedback, I have no interest in forms and whatnots—I write an email! (And, notably, this email has usually already been sent when the harassing request for feedback comes …)

*Possibly excepting some strongly reputation driven fields, e.g. Uber-style services with regard to the individual driver. However, even here it would seem reasonable to only give a rating when something was sufficiently above or below par that it was noteworthy. Certainly, the scales must be normalized to have an average performance imply 3-out-5, not the current “anything less than 5-out-of-5 is an insult”.

Worse: if the customer does not give feedback, chances are that one or two reminders are sent, further wasting the customer’s time and showing a complete disrespect for his decision to not give feedback.

Of course, this type of email is another potential in-road for phishing attacks.

As a counter-measure, I strongly encourage businesses (websites, organizations, whatnot) to adhere to a strict rule about email parsimony; indeed, I see them as under an ethical obligation to do so: If an automatic email is not obviously beneficial to the recipient (not the sender!) and reasonably* expected in context, it should not be sent. Moreover, it is better to send one longer email covering several sub-topics than several shorter with a sub-topic each. For instance, a booking confirmation is both beneficial and reasonably expected. A stand-alone unsolicited checklist is usually not beneficial and it is certainly not reasonably expected, but it might be OK if included in an already legitimate email (e.g. a booking confirmation). If there is any other email that might seem worth sending, it should be sent manually to reduce the risk of abuse and in order to err on the side of too little.**

*As in e.g. “what would a reasonable person with little prior exposure reasonably expect”—not as in “what would a reasonable person consider likely based on prior experiences”. Note that there is a dependency on circumstances, e.g. in that I would not normally expect a “your flight has been canceled due to a storm” email, but that this hinges on my not expecting a storm. If a storm has occurred and left my flight canceled, we have a different situation.

**As an aside, the idiotic German legal fiction that if someone already is a customer, then he is expected to be interested in new offers, and businesses are now allowed to send unsolicited advertising emails/letters/whatnot, fails largely on allowing automatic offers. If this was restricted strictly to manual communications, it would be within the plausible, but, as is, businesses just spam every single customer automatically, causing a very poor ratio of interest and a lot of annoyance, barely better than spam to complete strangers. (But this is improving due to sharper laws.)

To this a possible exception exists in that users might be given a list of choices for what emails they want to receive, e.g. booking confirmation (pre-selected), check-list (de-selected), …, to which the business must then adhere—deliberate choice by the user trumps parsimony. This would have the additional advantage of reducing unethical practices like hiding an “agreement” to this-or-that in the Terms-and-Conditions or claims likes “you agree to this-and-that, but can retract your agreement at any time by writing a letter to our customer service”.

Excursion on the customer/user side:
I strongly recommend that as many of these emails as possible be ignored. This with the three-fold idea to not waste own time, to reduce exposure to phishing attacks, and to not encourage misbehavior.

To the last point, I note e.g. that if no-one ever calls up the feedback forms, then businesses will eventually be discouraged and stop sending emails.

To phishing, I recommend more specifically never to enter any type of data over a link sent in an email or through an automatic email request (and to be very cautious with any manual request). For instance, for an online check-in above, it is better to manually go to the website and find the right entry point there (even the aforementioned attempt to court registrations aside).

Excursion on contractual obligations:
A business-to-consumer contract should work according to the simple principle that the business provides a service and receives money in return, the money being the almost* sole obligation of the customer and contingent on the service being provided adequately**. The result should be rights for the customer and obligations for the business. In current reality, it is often the other way around: yes, the customer still pays, but the rights are given to the business and the obligations put upon the customer. Pick up a typical business-to-consumer contract or Terms-and-Conditions and note how much is said about what the customer must or must not do. Note the freedoms businesses presume to take, e.g. with email addresses. Note how customers are increasingly seen as obliged to give feedback and ratings—often with only five-star ratings being acceptable. Etc.

*Exceptions include general, common sense, and usually not-necessary-to-state restrictions like that a rented item must not be damaged, as well as some situation-dependent that might reasonably apply, e.g. that a rented item must be returned at a certain location no later than a certain time.

**At least in Germany, this is a widely ignored condition: the typical attitude is that a contract is a one-sided obligation for the customer to pay, with the service being provided on a “if nothing goes wrong basis”.

A particular annoying behavior, at least in Germany, is to forbid certain uses—not warn against them as dangers, not describe them as warranty invalidating, or similar. This is an inexcusable presumption: if a certain use is not illegal, it is entirely* up to the buyer how he uses the product, including what risks he takes—end of story.

*Under normal circumstances. Exceptions might exist in special cases, e.g. that buying a DVD and then making and distributing copies for personal profit is not allowed. I am, however, hard pressed to come up with an example that does not involve a potential damage to the seller’s or producer’s business opportunities and/or a use of a non-private kind.

Email addresses and the abomination of a display name

A strong case can be made that various Internet standards created before the Eternal September, the commercialization of the Internet, the (once) dominance of Internet Explorer, … where superior to what came later.* One interesting counter-point is the “display name” of an email address, which has annoyed me for ages. This idiocy appears** to be present as far back as RFC 822 in 1982, possibly even earlier, depending on what implementations predated this document. (My own history on the Internet “only” goes back to 1994.)

*Of course, allowing for deficiencies due to a smaller amount of practical experience and a changing world.

**RFC 822 gives “mailbox” as “addr-spec” (i.e. a proper email address) or “phrase route-addr”, which seems to match the idiocy under discussion. Its 2001 replacement, RFC 2822, actually uses “display name” in its descriptions.

In effect, instead of using an email address like “john.smith@example.com”, senders are allowed to use e.g. “donald.trump@whitehouse.gov <john.smith@example.com>”.* Here “Donald.Trump@whitehouse.gov” is the display name, which has no actual impact on the email handling. Of course, John could equally use “Trump”, “Hillary Clinton”, “mermaid lover”, or “info”.

*There might be cases where additional escaping or use of quotations marks is needed. I have not investigated this in detail, and I deliberately do not wish to include quotation marks in the examples, even at the risk of a slight inaccuracy, due to incompetent handling by WordPress.

Here we see the first problem: The display name is highly unreliable. Not only can it be used to try to fool the email-illiterate user into making incorrect assumptions, but major confusion can arise when one party switches* display name between two emails, or when several parties use the same** display name. This problem is made the worse, because some email clients rely very strongly on the display name, e.g. in that only the display name is present in overviews of a mailbox (at all or per default). Indeed, even when the email address is displayed, the display name can make it less accessible, e.g. through pushing relevant parts out of sight.

*Say that someone started with “John” and switched to “John Smith”, because there were other Johns; or that someone went with “John Smith” professionally and “Johnny Boy” when writing his family, and that some recipient was both a family member and a professional contact.

**With “info” being the paramount example. In contrast, the email address proper is unique. (But note potential complications like equally ill-advised attempts to allow generic Unicode characters, which might cause e.g. an apparent Latin “A” to be exactly that in one address and and a Greek capital alpha in another.)

A particular complication is mailing lists: Because the sender determines the display name used for the mailing-list address, the eventual recipients can receive dozens of display names for the same mailing list. I can still recall trying to automatically put emails from one or two mailing lists into different folders at work—we were stuck with Outlook, Outlook only allowed filters* based on display name, and even with half-a-dozen alternative display names appended to the filter I regularly found emails that by-passed the filter… Of course, if emails to two different mail-lists used the same display name, filtering would be done incorrectly… (But, in all fairness, these were more Outlook issues than email-specification issues.)

*With reservations for terminology. I have not used Outlook in a good long while.

Another issue is that this feature is mis-designed (even its existence aside): now parsers need to handle two inconsistent formats, writers of emails need to understand two formats, etc. Indeed, because the display name can be empty, a parser needs to handle both “john.smith@example.com” and “<john.smith@example.com>”—and if faced with “john.smith@example.com”, it can only conclude that this actually is an email address (not a display name) after having noted that a “<>” expression does not follow. Absolutely amateurish… A better solution would be to put the display name in angle brackets, which allows for easier and more consistent parsing, and is less likely to cause misunderstandings (i.e. “<Johnny Boy> john.smith@example.com”, not “Johnny Boy <john.smith@example.com>”).

A minor potential advantage is the ability to replace a non-descript email address with something easier for the recipient to recognize. I note e.g. that my own first email address (provided by my college and based on my user name in its systems) was something like “f94-per@nada.kth.se”*. However, the advantage was very minor even back then, very few are stuck with such addresses today,** and that “john.smith@example.com” is named “John Smith” does not need additional mention. If worst come to worst, the (claimed) identity should be clear from the email body, even if at some loss of comfort. Schemes like allowing several people to use the same email account with different pseudo-identities are highly disputable, and it is better to either give them separate accounts or to not use pseudo-identities with the one account at all, because they are likely to do more harm than good. (As an example, a customer-service department should not use the names of individual co-workers as display names for one common account.)

*Do not try it. Chances are that I misremember the details, my last log-in was likely in the late 1990s, and I have no idea what my password might be—even should the account still be functional… The local part, should anyone be interested, comprises a program-of-study identificator (“f”, in my case), the year of enrollment, (a hyphen,) and some letters from the student’s name.

**Or they have only themselves to blame for registering using the likes of “ahf38js” (instead of e.g. “john.smith384”) when a name is already taken.

Unfortunately, display names see heavy use and I very often even receive email back, where my sender address has been abused as a faked display name (i.e. if someone uses just the email address proper, e.g. “john.smith@example.com”, he receives a “john.smith@example.com <john.smith@example.com>” back). This is utterly pointless on two counts: firstly, it adds no information compared to just using the address proper; secondly, it unnecessarily forces the use of one format when the other would have done just fine. Outlook, as already mentioned, seems to consider the display name more important than the actual email address. Certainly, the display names picked by the sender, especially in a commercial context, are often quite poor—as with “info” (why not include the company name?!?) and other very generic phrases.

Eternal September? I wish! (And some thoughts on email)

One of the most unfortunate trends of the Internet is that erstwhile standard procedures, behaviors, whatnot are forced out by inferior alternatives, as an extension of the Eternal September. Indeed, the point where even the Eternal September can be viewed with some nostalgia has long been reached:

The name arose through a combination of how, every September, the Internet would see a sudden burst of college freshmen, who still needed to learn how to handle themselves and who were an annoyance to older users until they had done so; and how the popularization of the Internet outside of college caused this inflow of unversed users to take place over the entire year. Even so, in the early days, the new users tended to be of over-average intelligence, tech affinity, and/or willingness to adapt—and many could still continuously be made to leave their newbie status behind. The problem with the Eternal September was its Hydra character: Cut of one head and it grew two new.

Today’s situation is far, far worse: There is no filtering* of who uses the Internet, be it with regard to intelligence, technical understanding, willingness to learn from more senior users, …; and, by now, the vast majority of all users are stuck in a constant newbie state. Indeed, we have long reached a point where those who have been on the Internet since before the problems became overwhelming** are viewed as weirdos for doing things the right way***. Worse: Websites are usually made for the “lowest common denominator”, with regard to content, language****, and interface, making them far less interesting than they could be to the old guard. This is paralleled in a number of negative phenomena on the Internet (and, unfortunately, IT in general): Consider e.g. how much less profitable it would be to spam a collective of STEM students than a random selection of the overall population, or how much less successful an Internet-based virus among the tech savvy.

*A formal filter, a legal restriction, an equivalent of a driver’s license, or similar, was not in place before the Eternal September either. However, Internet access outside of higher education was reasonably rare, and even within higher education it was far more common in STEM areas than in e.g. the social sciences. Correspondingly, there was an implicit filter that made it far more likely for e.g. a math major to have Internet access than for e.g. a high-school drop-out.

**The linked-to Wikipedia page puts 1993 as the start date in the U.S., but other countries like trailed somewhat. I started college in 1994 and the situation was reasonable for a few years more, before the Internet boom really started—after which is has been downhill all the way.

***Note that while there is some arbitrariness to all rules and there is usually more than one legitimate way to handle things, there is at least one important difference between the “old ways” and the “new ways” (even aside from the benefit of continuity and consistency, which would have been present with the old rules): The old ways were thought-out by highly intelligent people and/or developed by trial-and-error to a point where they worked quite well—the new are a mixture of complete arbitrariness; ideas by less intelligent and less experienced users, or even managers of some software company; attempts to apply unsuitable “real-world” concepts to the online world; … To this must be added the technical side: Someone who understands it, all other factors equal, is objectively better off that someone who does not—and less of a burden to others.

****Even Wikipedia, once an exemplary source of good writing, has gone downhill considerably, with regard to both grammar and style. (Notably, the “encyclopedic writing” aspect is increasingly disappearing in favor of a more journalistic or magazine style. I have long had plans for a more detailed post on Wikipedia, including topics like an infestation with unencyclopedic propaganda, but have yet to get around to it.)

A particularly depressing aspect, but great illustration of the more general problems, is the (ab-)use of email by many businesses, government institutions, and similar, who simply do not understand the medium and how to use it properly. No, I am not talking about spam here (although spam is a gross violation)—I am talking about everyday use.

Consider e.g.:

1. That many businesses and virtually all (German) government institutions fail to quote the original text when replying and replace the subject line with something to the effect of “Your message from [date]”.

   The former makes it harder to process the message, in particular when a re-reply is needed, often forcing the user to open several other messages to check for past contents; the latter makes it much harder to keep messages together that belong together*, to find the right reply to an email, identify why a reply was sent**, etc. To boot, these problems likely contribute to poor customer service through creating similar issues on the other end, e.g. through a member of customer support not having all the information present without looking through old emails or some ticket system: Even if the information is there, access will be slower, more resources will be wasted, and there is a major risk that important information will still be missed.

   *Not only manually, but also through giving automatic “threading” mechanisms in email clients an unexpected obstacle.

   **When the original text is not included, this becomes even harder. In a worst-case scenario, when several emails were sent to the same counter-part on the same day (rare but happens), it might not even be possible to make the correct match—or only through comparing various IDs in the message headers. The latter is not only much more effort than just looking at subject lines, it also requires that all involved software components have treated them correctly, that the counter-part has used them correctly, and that the user knows that they exist…

   The explanation for this absolutely amateurish and destructive behavior is almost certainly that they have never bothered to learn how to handle email, and just unreflectingly apply methods that they used with “snail mail”* in the past. This is the more absurd since going in the other direction, and altering some snail mail procedures in light of experiences with email, would be more beneficial.

   *This phrase gives another example of how the world can change: Twenty years ago, I could use the phrase and simply assume that the vast majority of all readers would either know that I meant “physical mail sent by the post”—or be willing both to find out the meaning on their own and to learn something new. Today, while typing the phrase, I am suddenly unsure whether it will be understood—and I know that very many modern Internet users will not be willing to learn. I might be willing to give the disappearance of the phrase a pass: We can neither expect every phrase ever popular to continue to be so in the long term, nor the phrases of any group to be known in all other groups. However, the attitude towards learning and own effort is a different matter entirely.

2. When messages are quoted, established rules are usually ignored entirely, especially with regard to putting the quote ahead of the answer and to intermingle quote and reply, which makes an enormous difference in the ease of processing the reply. Some tools, notably MS Outlook, more-or-less force a rule violation on the users… When quote and reply are intermingled it is usually not done in the established manner, with separating new lines and use resp. non-use of a “> ” prefix; instead, the new text is simply written straight into the old and separated only by a highly problematic* use of a different color.

   *Among the problems: The colors are not standardized. The result becomes confusing as to who wrote what in what order after just a few back-and-forths, to the point of making a lengthier email discussion almost impossible (whereas it was one of the main uses of email in the days of yore). It forces the use of HTML emails (cf. below). There is no guarantee that the colors will be visible after printing or a copy-and-paste action to another tool (notably a stand-alone editor). Not all email clients will be able to render the colors correctly (and they are not at fault, seeing that HTML is not a part of the email specifications). Generally, color should not be used to make a semantic differentiation—only to highlight it. (For instance, in the example below, an email client could automatically detect the various “>” levels and apply colors appropriately; however, the “>” signs must remain as the actual carriers of meaning.)

   To give a (simplistic and fictional) example of correct quoting:

   >>> Have you seen the latest “Star Wars” movie?
   >> No.
   > Why not?
   The one before that was too disappointing.

3. Ubiquitous use of “no-reply” addresses: Anyone who sends an email has a positive duty to ensure that the recipient can reply to this email. This includes event-generated automatic messages (e.g. to the effect of “we have received your email” or “your package has just been sent”) and news letters. Either make sure that there is someone human able to read the replies or do not send the email at all.* This is not only an ethical duty towards the recipient, it is also a near must for a responsible sender, who will be interested in e.g. tracking resulting failures.

   *The exact role of this human (or group of humans) will depend on the circumstances; often it will be someone in customer service or a technical administrator.

4. Abuse of email as just a cost-saver relative snail mail: There is nothing wrong with sending relevant attachments in addition to an email text, and in some cases, e.g. when sending an invoice*, even a more-or-less contentless email and an attachment can be OK (provided that the recipient has consented). However, some take this to an absurd extreme, notably the outrageously incompetent German insurance company HUK, and write a PDF letter containing nothing that could not be put in an email, attach the resulting file to the email, and use a boiler-plate email text amounting to “please open the attachment to read our message”. This, obviously, is extremely reader hostile, seeing that the reader now has to go through several additional steps just to get at the main message** and it ruins the normal reply and quote mechanisms of email entirely. To boot, it blows up the size of the message to many times what it should be*** and increases the risk of some type of malware infection.

   *This especially if the contents of the invoice are to some degree duplicated in the email proper, including total amount, time period, and due date (but more data is better). Writing an invoice entirely as a plain-text email is possible, and then the attachment would be unnecessary; however, there can be legitimate reasons to prefer e.g. PDF, including a reduced risk of manipulation, a more convincing and consistent visual appearance if a hard-copy has to be presented to the IRS, and an easier differentiation between the invoice proper and an accompanying message. (There might or might not be additional legal restrictions in some jurisdictions.)

   **Note that it is not just a matter of one extra click to open that one attachment that one time. Consider e.g. a scenario of skimming through a dozen emails from the same sender, from two years back, in order to find those dealing with a specific issue, and then to extract the relevant information to clarify some new development: If we compare a set of regular emails and a set of emails-used-to-carry-PDFs, the time and effort needed can be several orders larger for the latter. Or consider how the ability to use a search mechanism in the email client is reduced through this abuse of email.

   ***This is, admittedly, less of an issue today than in the past (but HUK has been doing this for a very long time…). Still there are situations where this could be a problem, e.g. when a mailbox has an outdated size limit. It is also a performance issue with regard to other email users: The slow-down and increase in resource use for any individual email will be relatively small; however, in the sum, the difference could be massive. What if every message was blown-up by a factor of 10, 100, 1000, …? What would the effects on the overall performance be and what amount of band-width and processing power (especially if spam or virus filters are applied) would be needed? For instance, the two emails at the top of my current mailbox are, respectively, an outgoing message from me at 1522 Byte and the reply to said message at 190 Kilo(!)byte—roughly 125 times as much. The lion’s part of the difference? A two-page PDF file…

5. Use of HTML as an email format: Such use should, on the outside, be limited to recipients known both to handle the emails in a compatible manner and to be consenting: HTML is not supported by all email clients, and not in the same manner by all that do. It poses an additional security and privacy risk to the recipient. It bloats the message to several-to-many times the size it should be. It makes offline storage of the email more complicated. It makes it harder to use standard reply and quoting mechanisms. The risk of distortion on the way to the recipient is larger. … Notably, it also, very often, makes the email harder to read, through poor design.

   To boot, the reason for the use is usually very dubious to begin with, including the wish to include non-informative images (e.g. a company logo), to try to unethically track the recipients behavior (e.g. through including external images and seeing what images is retrieved when), or to make the message more aesthetic*. A particular distasteful abuse is some newsletters that try emulate the chaotic design of a commercial flyer or catalog, which often deliberately try to confuse the readers—either the newsletter senders are incompetent or they try to achieve something incompatible with the purpose of a newsletter**.

   *This is simply not the job of the sender: The sender should send his contents in a neutral form and the rendering should be done according to the will of the recipient and/or his email client—not the sender. Efforts to change this usually do more harm than good.

   **Most likely, but not necessarily, to use it as advertising. I note that while newsletters are often unwelcome and while the usual automatic addition of any and all customers to the list of recipients is despicable, the abuse of a newsletter for advertising is inexcusable: Many will consent to being or deliberately register as recipients because they are interested in news about or from the sender; and it is a gross violation of the trust placed in the sender to instead send them advertising.

   There are legitimate cases where a plain-text email is not enough to fulfill a certain use-case; however, they are rare and usually restricted to company-internal emails. For instance, one of the rare cases when I use HTML emails is when I want to send the tabular result of a database query to a colleague without having to use e.g. an Excel attachment—and even this is a border-line spurious use: In the days of yore, with some reservations for the exact contents, this could have been done equally well in plain-text. Today it cannot, because almost all email readers use a proportional font and because some email clients take inexcusable liberties with the contents*.

   *For instance, Outlook per default removes “unnecessary” line-breaks—and does so making assumptions that severely restrict the ability to format a plain-text document so that it actually is readable for the recipient.

   Of course, even assuming a legitimate use-case, it can be disputed whether specifically HTML is a good idea: Most likely, the use arose out of convenience or the misguided belief that HTML was a generic Internet format (rather than originating as a special purpose language for the Web). It would have been better to extend email with greater formatting capabilities in an ordered, centralized, and special-purpose* manner, as has been done with so many other Internet related technologies (cf. a great number of RFCs).

   *Which is not automatically to say that something should be developed from scratch or without regard for other areas—merely that it should be made to suit the intended purpose well and only the intended purpose. Some member (or variation of a member) of the ROFF-family might have been suitable, seeing that they are much closer to plain-text to begin with.

6. A particularly idiotic mistreatment of emails is exemplified by Daimler and in recent discussion for another large auto-maker (which one, I do not recall):

   If an email is sent to an employee at the wrong time, e.g. during vacation, the email is simply deleted…

   The motivation given is absurd and shows a complete lack of understanding of the medium: This way, the private time of the employees would be protected. To make matters worse, the “threat” comes not from the outside but from a (real or imagined) pressure from within the company to always be available. In effect, Daimler has created a problem, and now tries to solve this problem through pushing the responsibility and consequences onto innocent third parties.

   Email is by its nature an asynchronous means of communication; and one of its greatest advantage is that the sender knows that he can send a message now, even outside of office hours or during vacation periods, and have it handled on the other end later. He does not have to consider issues like whether the recipient (if a business) is open or (if a person) is at home with his computer on. Moreover, the “later” is, with some reservations for common courtesy and stated dead-lines, determined by the recipient: He can chose to handle the email in the middle of his vacation—or he can chose to wait until he is back in the office. Whichever choice he makes, it is his choice; and if he chooses the former against his own best interests, well, then he only has himself to blame.

   By this utterly ridiculous rule, one of the greatest advantages of email is destroyed. To boot, it does this by putting an unfair burden on the sender, who is now not only required to re-send at a later and less convenient time—but who can see a number of additional disadvantages. Assume e.g. that the sender is about to head for his vacation, sends an important and urgent email, goes of the grid for two weeks, and comes back to see that his email has not even been read. Or take someone who writes a lengthy email and loses* any own copy after sending—should he now be required to re-type the entire thing, because of a grossly negligent policy of the recipient’s? Or what happens when employees in very different time zones or with very different vacation habits try to communicate with each other? Should the one work during his normal off-hours or vacation so that the other can receive the email during his time in the office? What happens if the notification** of “please send again” from company A is it self deleted by company B?

   *Disk crashes and accidental deletes happen; I have worked with email clients that do not automatically save sent emails; and, in the spirit of this post, not all users actually know how to retrieve sent emails that are saved…

   **Daimler apparently at least has the decency to send such notifications. I would not count on all copy-cats to follow suit.

   Want to keep your employees from reading company emails in their spare time? Do not give them email access from home or do cut it off during those times when no access is wanted! The way chosen by Daimler turns the reasonable way of handling things on its head—to the gross detriment of others. (This even assuming that the intended goal is legitimate: These are adults. We could let them chose for themselves…)

Yahoo tries to pull a Facebook?

I have long had an email account with Yahoo for various and sundry (including as a backup, in case the addresses that come with my domain are temporarily not reachable). So far, this has been an extremely frustrating experience: Yahoo is one of the worst thought-through and poorly programmed websites that I am aware off. Whoever is guilty of this atrocity should be lead to the next wall and put before a firing squad.

Today, however, the worst of many, many user-hostile idiocies from Yahoo came to my attention: They appear to have “pulled a Facebook”, and installed various publications and notifications behind the back of the users—utterly ruining any remaining credibility.

To make matters worse, when I try to disable one of these settings, I am met with an error message—and the setting returns to the share state…

My advice to anyone using Yahoo for anything non-trivial: Do not. Reduce your account activity to near zero, remove all contacts, etc. Instead find yourself a good ISP with your own email addresses or, if that is too costly, a decent independent and pure email service. (For those in Germany: My own ISP, bytecamp^e, has so far been very satisfactory at < 10 Euro/month, including domain, webserver, and unlimited email addresses.)

The same advice applies, obviously, to other services of a similar character, including Gmail and Facebook. The latter already has a large number of “delete your account” recommendations here on WordPress.