Hacker News Reader

I'd ask that we consider a different spectrum than good and bad, but naive and abused.

Look at LL Bean and their cancellation of their lifetime warranty [1]. That linked article talks about abuse of the warranty, but I'm fairly sure people were trolling through ebay + thrift shops and buying items to return and resell as new. That's abuse and they had to make changes to it.

Or consider how widely reviled PayPal is for how they "take" their customer's money and lock it up. I had a friend who worked there that described it as "a fraud detection company that occasionally takes payments". You can't even imagine the ingenious lengths people go to commit fraud and rip PayPal and other people off. So they have by necessity become this draconian organization.

When Google/Gmail was small it was easier, they were more naive. But now? I imagine every scammer in the world is trying to corrupt and steal gmail accounts as a stepping stone to greater fraud - and some significant portion are trying to do so by claiming that someone else's account was hacked and that ney need special access.

I don't think there's any easy answers to these thorny problems but forcing it into a false good/bad dichotomy isn't helping.

1 - https://slate.com/business/2018/02/l-l-bean-has-ended-unlimi...

Perhaps, but I think GP has a good first step:

gave even the most rudimentary of ways to speak with someone (livechat, helpdesk, whatever)

Otherwise, we are just guessing. Some will give them the benefit of the doubt. Some will assume the worst. The truth is simply somewhere in between.

The opposite of the LL Bean anecdote is a company that gave a lifetime warranty and made a name for itself in the nautical outerwear industry. People bought in part because of the warranty. After a few years they started having to make good on the warranty- not because of fraud, but because out wear doesn't last forever.

That's reasonable, but also consider that the Apple iCloud "hacking" from a few years back was largely social engineering Apple's tech support into providing them with enough information to get access.

>>When Google/Gmail was small it was easier, they were more naive. But now?

Poor Google. Even with ~ $100 Billion in yearly revenue and over $100 BILLION in cash they cannot hire customer support people?

10,000 new employees in a third world country at, say, $25,000 year (including other expenses) costs $250,000,000 a year. https://finance.yahoo.com/quote/GOOG/key-statistics?p=GOOG

> Yet there is no way to get ahold of google other than their product forums, posting and praying to not get a cheap canned response (which is what happened to me).

I believe the only way to get relief from Google is to know an insider, who can escalate to someone who can do something. All their public support channels are dead ends. Google sees you as a product, not a customer, and it's customer service skills are so bad that even if you're a real customer you can get the same treatment.

Stuff like this is why I switched to used Fastmail with my own domain.

> I used to LOVE google back in the early gmail/android days. They were the good guys. What happened?

I don't think anything happened. You just hit an important function that they're utterly terrible at.b Despite their success, they're clearly not omnicompetent.

Sometimes you can get that insider by becoming a customer of a different part of the company from the one that is giving you trouble.

We had an email problem with a major free email provider that also ran an advertising network (no, it was NOT Google or Microsoft...).

Our problem was that they kept deciding all the mail we sent to their users was spam. None of it was actually spam. The only things we were sending were:

• receipts for purchases

• installation instructions and license keys for whatever the user had just purchased

• if the user was on a subscription plan with a term longer than a month, a notice a few days before subscription billing that they were going to be billed

• responses to support tickets the user initiated

We'd complain to the louts running the email service, and they would fix it...and a month later it would happen again. Each time it took them longer and longer to fix it.

Then we realized we were spending a large amount buying ads on their ad network. Our guy in charge of ad buying called up our account rep at their advertising sales department, and asked why the heck we should keep paying them a large amount to drive customers to us, if they were going to make it so we could not communicate with those customers if those customers used their email service? It doesn't take a young Einstein to figure we shouldn't.

The account rep put our guy on hold a few minutes, then came back and had conference called in someone high up in their IT department, explained the problem. The IT department head conferenced in the person in charge of mail filtering and told him to put in a white list entry for us, and he wanted it done right away. About ten minutes later we were told it was done, and we never had another problem that I know of with email being misclassified as spam.

This disturbed me quite a bit. None of our mail was spam, so no one was harmed by that white list entry. But they had plenty of people using their ad network who were pushing sleazy products and who I greatly doubt only sent non-spam mail. Would they also be able to get white listed just be calling an ad rep and complaining?

We never had any spam classification problems with customers who used Google or Microsoft free email services, so have no idea if similar things work there, or if they keep a stronger separation between the ad network and the mail services.

In my more cynical moments I wonder if this sort of the thing might be the reason advertising companies run free email services. If they can amass a large number of email users, and they grant exceptions in the spam filters for their bigger advertisers, that provides an incentive to use their ad network over one with a smaller email service or without an email server, so you'll get more customers that you email market to at will.

We never had any spam classification problems with customers who used Google or Microsoft free email services, so have no idea if similar things work there, or if they keep a stronger separation between the ad network and the mail services.

My sister-in-law has a blog, email newsletter, and online store, all of which got blacklisted by Google after a number of her newsletter subscribers marked her letter as spam rather than unsubscribing. The only way she eventually fixed the problem was by finding a Googler contact who was able to get in touch with somebody who could whitelist her. A piss-poor way to run a business.

I was that guy many times (Former Spam, Abuse, Delivery SRE). You have to realize that %95 of the time when people reach out to report that mail that they are sending is getting canned its because they are sending SPAM and don't realize it. Seriously, all day, every day I got reports of "site x is unfairly being canned" only to go look up the data and see that the site is basically funneling hundreds of thousands of emails into the system and most users are making them as spam. We know the normal "I want to unsubscribe but I just click Spam instead" rates. We know an order of magnitude more than that as well, and most the time when people reach out they just don't understand email and how to have decent content people actually want.

And sometimes they are a sales rep that work with Pfizer, annoyed that regular correspondence about their top products were going to spam because they contained information about the price of Viagra..

We know an order of magnitude more than that as well, and most the time when people reach out they just don't understand email and how to have decent content people actually want.

The flip side of this is that many of us know the excuses big email services make about how bad a problem spam is and how our content might be spam and we just don't understand... when the content is the kind of reasonable and necessary communications that 'tzs mentioned further up in this thread.

I'm getting awfully bored of bounce messages from apparently credible organisations whose members have signed up, telling me that the legally required and personalised information we have just sent to our customer appears to be unsolicited bulk email. There has been a small wave of these recently, which leads me to think that someone screwed up a scoring system that multiple organisations are using to vet incoming traffic.

In any case, that's not a minor mistake, that's an automatic recognition system that is hopelessly inaccurate. It could even be dangerous, for example if a customer isn't getting notifications about payments and forgets they haven't cancelled a subscription, or if some action is required of them if they don't want to accept some change in terms for a service they use.

Oh I know.. trust me. I used to work in the team that handled that very problem and yet right now I can't get google to stop spam processing mails that fetches from my private IMAP server for my wife's gmail account. Everybody I know that was in that team has abandoned Google in the decade since I was there so I no longer have an easy in to figure out what is going on.

The problem is that for every single technical, competent person trying to reach google, there are 20 technical incompetent people who "know exactly what the problem is" (and don't), and 1,000 people who have a legit support problem that can be resolved with a google search, and 1,000,000 people who are just going to waste your time completely. =/

The moment you have direct support you have to respond to all of the above. There is no way to create a support team that just handles the clueful people.

I was on the dns contact email address which should be relatively hard to find right? Yea it was spam central with free energy devices, complaints about the government, etc. Its impossible to provide human support at the level that google is playing. =/

The moment you have direct support you have to respond to all of the above. There is no way to create a support team that just handles the clueful people.

Sure. But if my little businesses, with only a few people and a modest budget available to do literally everything, can manage to treat our customers with more attention and respect than one of the richest and most powerful businesses in the world, something is wrong.

Its impossible to provide human support at the level that google is playing. =/

Sorry, but I simply don't buy that. Google have grown huge and incomprehensibly rich thanks in part to technology but also in part to refusing to communicate with individuals when things go wrong (and, frankly, hoping they don't actually get called on it because it's not efficient for every little advertiser to sue for that $50 back when Google served their ads on obviously inappropriate sites or whatever). Like Facebook and other big organisations which have become highly influential through their scale and pseudo-monopoly status, Google now need to hire, train and support people -- lots of people -- in order to provide an acceptable level of communication if those dealing with them have problems.

If they can't do that then IMHO they should be sued, regulated or otherwise impaired until sufficient incentive/deterrent exists. The cost of letting these huge, influential organisations literally destroy businesses or wreck lives through their negligence is simply too high to allow the status quo to continue.

| Sure. But if my little businesses, with only a few people and a modest budget available to do literally everything, can manage to treat our customers with more attention and respect than one of the richest and most powerful businesses in the world, something is wrong.

2.2 Billion users. 74,000 employees. That's 300,000 users per employee. Converting Google's entire current operation into customer support, and assuming 1 ticket per year per customer, would require every Google employee to solve 1.7 tickets per minute (365/8) to keep up. Assuming a 5 minute ticket resolution (and still 1 ticket per year per user) would require 500,000 trained customer support personal to support that; assuming $20K per year, that equals to ~$10B. Google's profit was ~20B last year. This is not very likely to happen; and conditional it does, it would kill very large parts of the IT sector.

Can we kindly attempt to actually get to a solution that is actually mutually acceptable by all parties for this?

Clearly Google need to hire a lot more people if they are going to provide an acceptable level of support.

If that means their profits are significantly reduced, but they finally provide an acceptable standard of communications, I have no problem with that. How much of that $20B profit came from mistakes that hurt other people, which they only got away with because they didn't communicate and it wasn't worth suing them?

The current solution isn't acceptable either.

"There is no way to create a support team that just handles the clueful people."

How about the option of paying $20 for a person at Google to read and respond to an email query? I would love that option and many other people would too, if the response was competent and reasonable.

They need to think of a way to have a xkcd tech support(https://xkcd.com/806/) like system. With all those "think out side the box" mindsets, someone should be able to come up with a workable solution.

What about pay per support request? I'd think the people who are desperately trying to get back their email account legitimately are going to be much more willing to pay, say $10, than a free energy crank. Presumably they could set the price at a level they can afford even if everyone and their brother use the support service.

Under adversarial conditions (basically, any large-scale ops), this becomes the highest-leverage way for adversaries to aggressively DDoS the weakest point of such support queue.

Also, consider the conditions:

* You are a scammer

* You pay $10 to Google to "fix getting back into your account"

* Google knows that account does not, actually belongs to you, but now are in a position, where they either: eat the $10 with a WONTFIX, and pave the way to a mass lawsuit or refund the $10, and have a free way for scammers to exploit.

> There is no way to create a support team that just handles the clueful people

I've wished on occasion that someone would create a certification system, kind of like what we have for software engineers and system administrators, but for end users. Users could get certified in various aspect of operating and self-installing consumer technology, and when a certified user called tech support they could be directed past the tier of support that assumes the caller is an idiot.

You don't have to know an insider. Get your problem to the front page of a major news aggregator, and an insider will contact you promptly. HN seems to work very well, and I've heard good results from Reddit too.

I see an interesting parallel to malpractice liability for physicians. The well known and understood best way to avoid malpractice is not expertise or avoiding mistakes, it's having good beside manners and treating patients with respect. (Expertise and attention to detail are very, very important of course but have less impact on chances to get sued than these other soft factors, according to this advice.)

I've seen doctors miss cancers before and escape unscathed after frank discussions with the patient, while others that do everything technically correct are sued over a hangnail. Not saying missing cancer is "ok as long as your nice", I'm saying that people don't really sue over money, they sue over hurt feelings and seek retribution in monetary form. (There's a big exception to all this in the patients who are opportunistically seeking an opening to sue. They are simply dangerous to have as patients, but are usually quick to move on.)

Finally, the physician who has good bedside manners is also a better witness and more sympathetic to juries.

These can be hard lessons for techies who "aren't doing anything wrong".

> I used to LOVE google back in the early gmail/android days. They were the good guys. What happened?

By then they were already a very large company with lots of forces pulling them in different directions. Google were 'the good guys' right up until they became an advertising company, and it's been downhill from there.

Their chant of "Be less evil" was dead and buried 10+ years ago. They were the underdog in the beginning but absolute power corrupts absolutely. It was only a matter of time before they began to abuse their position as the world's leading search engine and positioned themselves as the internet gatekeepers.

They were never the good guys. Apple is not the bad guys. They are both huge corporations with tons of people with varying motivations and goals. You were and are creating a narrative about intentions that probably does not exist.

A single corporation has a few people with much greater power than most of the other people.

Pointing out that there are many people doesn't imply that intentions don't exist. Even if you work at Google, the people that you see and the intentions that you infer for them, might not truly reflect the intentions of the people higher up.

Yes but malintent is much rarer than oversight. Even for actions that hurt customers, are not truly a choice. Save for Google and Facebook, even CEOs that hurt the bottom line for a minor moral victory will be replaced with other people.

By the way, it is kind of interesting what drives someone to steal someones gmail account and than use it as his own - unless it's name was in some way unique and desired

It is, to own something like a 3 - 4 letters long username or email address is worth more to some people.

How do you get Gmail username with 5 characters or less?

Too late by now I think.

I got my Gmail account in late 2004 and there was a lower limit on username length then.

yep, i distinctly remember wanting just my first name and it not allowing it. i have one of the first 500~ accounts, and it wasn't taken

You have to hack an existing one and make it your own, apparently. :-)

Especially distinguishable words, and especially names.

Domains are a great example of this, too. I recently reached out, out of curiosity, and learned that the owner of the .net domain for my own four-letter name wants a quarter of a million dollars for it.

A shame, really. They're not doing anything with it, best I can tell. But if that's what someone would pay for it... You can bet someone would try to steal it.

At Google, the "user" and the "customer" are not the same entity.

That causes a problem of incentives.

Google and other companies don't want customers.

> Google and other companies don't want customers.

What do they want, by your reckoning?

your data please <insert user disabled but already checked check-box here> :)

That doesn't make any sense. They're not some supervillian who wants your data just to have it; they want to sell it (or products derived from it) to their real customers, which are big ad-buyers.

Google and Facebook want customers, ad-buying customers.

Indeed. And if you had an ad to market you want it targeted. Targeting can only occur if there is data.

Google has customers and users. One gets support, the other doesn't.

But you should take good care of your cattle otherwise you have nothing to sell to your customers.

Just in case anyone was curious, no nothing has been resolved, just the same runaround despite the info I have to provide. Also, if anyone is curious here is a link to the google group discussion: https://productforums.google.com/forum/?utm_medium=email&utm...

Companies are never the good guys. If I recall correctly Google never actually cared about customer service. They expected your questions to be resolved automatically by their software. If you aren't a paying customer to a service related to Google Ads you basically get shafted.

Worked there.. I cared. I responded on forums, dealt with customers large and small, bot direct and as a point of engineering contact.

Here is the problem with a company like Google: SPAM. Want to support the regular user? You are going to get people bitching that they can't find the "save" button and they are afraid of losing documents. You will get people asking you where their Nigerian prince went. You will have people who don't own a computer but need access to gmail. etc. Controlling the ways that support enters the system controls the amount of pure bullshit support requests you have to filter.

My favorites from the people that made far enough to find a way to contact us directly:

* A 30 minute long voicemail from a person talking about how he invented a system to produce energy from.. something.. we never did figure out what the model was.

* A person asking YouTube to send over camera to record some event they were having.

* A scammer who was annoyed that their scam emails were being filtered, who spent a lot of effort trying to convince US that they really were an estate lawyer in a country that didn't exist.

* Somebody telling us that they lost access to an email that didn't exist (and had never existed) and kept slipping up what their name really was.

* Piles, and piles of people trying to sell us stuff we didn't need or want.

Just make all calls to support paid per minute - it shortens the calls and removes the nutters. Adjust the price dynamically to meet demand.

Spammers would sign up for enterprise accounts, drop a couple hundred or more to get priority access, then spam the ever living hell out of the internet because they knew that we couldn't just disable the accounts outright. In the time it took to get the account shut down they more than made their money back. Eventually it would turn out that the card was stolen so the money would get reversed.

Dealing with credit cards is a nightmare, especially in support situations. What happens if the problem is totally on the users end? (Hi, I can't log into my gmail account, its bob@yahoo.com) Do you refund them since you can't fix the issue? If not they are going to be pissed. What if the problem is quite a lot complicated and requires several hours to fix? In the enterprise space we _constantly_ had to explain to mail carriers that the Cisco PIX device was mangling their SMTP EHLO line in a way that would break the SMTP protocol which is why mail was not going through. If an end user calls it can take hours to debug something like that. Will the user demand a refund when $200 later its ruled to be their shitty ISP's fault?

Their customers aren't really the end-user, but advertisers. They just want to appear to be the good guys.

Perhaps they should offer a paid customer support service just so that they're finally incentivized to help the end-user.

Google has 2bn users. How on earth would they provide support for that? I feel your pain but we have to be realistic. If support is of essence you don't sign in free services.

Paid support tbh. It can create perverse incentives, but having an option is better than no option.

> They were the good guys.

Villains who twirl their moustaches are easy to spot. Those who clothe themselves in good deeds are well camouflaged.

I like that. Very thought provoking. Thank you.

This is exactly why I'm teaching myself skills in it and computing.

I cannot rely on Google.

do no evil?