I don't see "issue" as a weasel word, although I can see why he does. To me, it's just a general term that captures "problem" and "fault" and "bug". I don't think a lot of people read it and think to themselves, "oh, I thought there was a fault here, but it turns out it's just an issue". "Issue" is equally negative.

Bug trackers become "issue trackers" in part because people started to use them to track more than bugs.

I think "issues" gain weasel points when combined with "may". It comes across as distancing yourself from whatever is going on.

The statement "Users may be having issues accessing Twitter" says to me that there may not be an issue, and if there is, it may be someone else's fault.

Speaking factually is the better course, IMO. "Some users are reporting problems accessing Twitter. Our engineering team is investigating the issue." No implications there.

People are used to hearing nonsense and discount everything you say once you start using it. I worked in a building where the building management tried to dress up a small fire as a drill by repeatedly declaring that all was well, even as a half dozen ladder trucks pulled up. Now a smoldering cigarette in a stairwell triggers building evacuations.

I always understood "may be having issues accessing Twitter" to simply mean Twitter is made up of large distributed services that rarely all go down. Instead service A in location C down might be down which means you might not be able to use the site in one of the many different ways you interact with it.

Yep, their wording now doesn't bother me much, but I find a lack of clear explanation after the fact particularly irksome (what was affected, where, why).

I once had a co-worker, LS, who insisted on never saying the word "problem"; everything was a "challenge". Always. If the data center exploded in a fiery ball, we were experiencing massive system challenges.

I bet our VP of Operations a hundred bucks that he couldn't get LS to say the word "problem". Without missing a beat, he turned to LS and said "I'll pay you fifty bucks to say the word 'problem'." And he did.

So while I support your quest for transparency, I really can't participate - it's proven expensive for me.

This is proper old English, was your coworker Indian?

Same for "kindly <do something>", interesting phraseologisms that are not used by modern Brits, but are still in place in ex-colonies.

I experienced the exact same thing from a consultant at a previous job. It was maddening.

Using the passive voice is a long tradition of almost everyone who's ever made a mistake: http://en.wikipedia.org/wiki/Mistakes_were_made

Often, I wish publicists would say things more tersely.

Blaming the passive voice is a long tradition of almost everyone who doesn't read Language Log.

You can write weaselly, evasive cop-outs without using the passive, and you can write sentences that use the passive while still putting the blame squarely where it lies.

Oh, but it's much easier with the passive, as the classics of the genre attest. That doesn't mean the passive voice isn't sometimes to be preferred, which is all that the Language Log people are saying.

Well, they're also saying that 1) it's very frequently blamed when it doesn't actually diffuse responsibility ("Mistakes were made by John. He's been sacked."); and 2) it's very frequently blamed where it doesn't even occur. This is not to say that it cannot be used in this way, but simple avoidance or blame of the passive is neither necessary nor sufficient.

"Users may be experiencing issues" is not the passive voice (not that you explicitly said it was, but IMAO you implied it), it's the continuous aspect, modified by whatever an auxiliary like "may" does. (this is where my grammar vocabulary starts to run out. conditional continuous maybe?)

passive would be "Issues may be being experienced by users".

While we are making the language more straightforward, it's not even "experiencing problems accessing Twitter." Like what problems, dry mouth and nausea? The problem isn't some incidental experience, the problem is that they can't access Twitter. Because it is not responsive, connections do not complete adequately, the intended page is not delivered. Twitter isn't working right. And this isn't a perfectly appropriate time to be puffing up its employees with honorifics like "engineer" (are they certified?)

Why not this: "Oops, the site is down because something isn't working right on our end. We're working on getting the site back up as soon as possible." (BTW, isn't working right is a euphemism for "broken")

Seriously, why not? Twitter is a great big company, why does it have to use language which puts a mere individual at a disadvantage? Is the position of this large, wealthy corporation so desperate that it cannot afford an ounce of humility at the moment it has made a mistake? Does it need to get the upper hand over little individuals looking at its web page?

You know the maid and the cook have to say big, frankly-worded mea culpas when something is off. "Diners may be experiencing issues digesting the Royal Egg Salad. Our Gastronomists are currently conducting an in-depth probe to resolve the issue." "We understand that diners are experiencing problems chewing."

Twitter doesn't owe me anything. They don't owe me uptime, they don't owe me an explanation of downtime, and they don't owe me apologies.

Obviously, the site is down. They needn't say more than that.

Populist counter-viewpoint: Twitter is literally nothing without the users. No content, no revenue, no value. They owe their users everything.

edit: I deserved that. :)

Figuratively nothing ;)

Why can't Twitter say it in straight-forward language:

"The site is down. We're working on it."

That would be so refreshing.

Perhaps it might not be true? I suppose if you were reading the error then by definition it was down but places like Amazon's EC2 have to be careful about saying "EC2 is down for 10% of the customers in the east cost availability zone #2" or some such.

There are many moving pieces, hence the rise of sites like downforme.org

Thanks for mentioning that. I had been using isup.me but ran in issues when checking while some images failed to load on a site I volunteer for - it would check the responsiveness on the URL of the server itself, which gave a 404 as it wasn't supposed to be accessed by public and isup.me would interpret that as the server always being down... downforme.org seems to deal with that better, thanks!

Exactly. The lowest bar we can possibly set is the only one that matters.

I don't mind being intentionally vague because you don't know what the problem is. What bugs me is when companies that make money off of your data, and provide open-source APIs so other people can also make money off of your data, don't post a public root cause analysis.

The RCA you show everyone else doesn't have to be as detailed as the RCA you maintain for your own records. However, you should let people know what went wrong and how your organization is learning from it. It helps your organization grow; it maintains a philosophy of openness, and it provides closure for all the poor people traumatized from being unable to tweet.

"For once you start down the path of the mealy-mouthed forever will it dominate your vocabulary."

It already dominates the world.

How is 'problem' or 'issue' really that different?

And 'may be' is technically accurate as well - Twitter has been working fine for me, so there was no 'problem' with my access.

That seems to indicate it was an issue after all?

To be fair sometimes when you can't tweet you get the message "sorry we did something wrong. try again later"

Welcome to California.

People will stop doing it when it stops being rewarding to do so.

I would love to see a documented A/B test of a failure message. But I have the suspicion that this is one of those "industry practices" based on presuppositions and anxiety rather than any demonstrated track record.