![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
There are times I think Google are somewhat competent -- and then I try to use Google Contacts
(Rant mode on)
Seriously -- WTF, Google? Google Contacts is perhaps the most unusable piece of software I am forced to use, and it exemplifies everything that's wrong with Google as a company. They forced me into this idiotic "Contacts Preview" UI, and years later it still fails to fulfill the most basic functions of a contact list.
(Yes, I could abandon Google Contacts -- all I have to do is abandon Gmail. Suffice it to say, that's a tall order at this point, although I may eventually be pushed into it.)
The one that always burns my butt (which I just hit again, which inspired this particular rant) is the fact that there is no way to say which email address to use for someone in a group. Google loves nothing more than to combine your contacts, so that instead of having four contacts for four email addresses, you have one contact with four email addresses. But you don't put an email address into a group, you put a contact into a group, and AFAICT there is absolutely no way to say which email address you want for this particular group.
My impression is that it simply always uses the first-listed email address for that contact. But AFAICT there is no way to re-order the email addresses, short of erasing and retyping them! (And of course, in typical Google fashion, they completely ignore the fact that different groups might be different contexts, and call for different email addresses.)
About every six months, some recruiter from Google tries to lure me in. I try to be polite, but this sort of crass incompetence keeps leaving me feeling like I would never want to work for a company that would put up with nonsense like this. They're the anti-Apple: as far as I can tell, they simply don't care about the user experience enough to put the slightest damned effort into it. As far as I can tell, I would find working there to be incredibly demoralizing.
(Or, I suspect more precisely: they don't care enough unless their corporate case of ADD has latched onto this topic right now. In which case it gets huge attention until the company gets bored, and wanders off to pay attention to the latest shiny, dropping all effort to make the existing software function right.)
Folks constantly ask me whether they can trust Querki, which after all is a much smaller company than Google. This is my heartfelt rebuttal: while my resources may be slim, I care passionately about making Querki as good as it can be, and supporting the users. I don't think you can say that about Google for any products except search and advertising. Everything else is just another Technology Preview, to be pushed for a little while and then abandoned.
(Rant mode off)
As to the question at hand, I eventually found this article. The secret turns out to be contained in the comments down there: if you scroll the left-hand bar way down, and open "More", you can abandon the goddamn Google Contacts "Preview" (never mind that it's been the status quo for years), and go back to the old, ugly but actually functional Gmail-style Contacts. That UI actually works -- there is a way (easy to use, although with crappy affordances) to say which email address to use for a given group for a given Contact.
Which I guess just underscores the point. Google got distracted by a New! Shiny! UI!, pushed everyone into it, and then lost interest and never actually finished it. So the old UI is still hanging around, for those of us who care more about a product that works than one that follows the latest visual-design guidelines...
no subject
(Or, I suspect more precisely: they don't care enough unless their corporate case of ADD has latched onto this topic right now. In which case it gets huge attention until the company gets bored, and wanders off to pay attention to the latest shiny, dropping all effort to make the existing software function right.)
...yeah. In this way, I think of Google as a bit like a preschooler - or a programmer addicted to green-fields development over maintenance, improvement and refinement.
I've been looking at PaaS offerings recently, and I realized that while looking at Google's that I was applying the filter, "assume that anything not currently implemented/offered will never be implemented/offered, no matter how much of a good/obvious idea it seems". Sadly, upon rational examination, this filter seems sensible.
So...
Google prides itself on being a "data driven" company. They treat design that way - if you can produce data showing that one design is measurably better than another then you have a chance - maybe even a good chance - of getting it implemented.
This has led to them doing ridiculous things like testing many many shades of blue and picking the one that had an 8% higher response rate.
They tend to roll out features and changes without any testing or without adequate usability testing because usability testing produces squishy non-numerical things like "comments" and "feedback" that are hard to quantify and see above about number-worship. Usually the new UI is just horribly worse and nearly impossible to use. They killed Picasa this way and the latest revamp of G+ has just about killed that, too.
I myself have never tried to use contact groups but I can see how that behavior you describe would be stupidly annoying. Even without groups I can't get Google to forget email addresses that are no longer valid nor ever be certain that it's actually sending to my friend-who-is-also-a-coworker's correct address for any given email.
Re: So...
I mean, architecture, analysis and UX all tend to go hand-in-hand. You have to actually understand the problem before you can get *either* the data or the UX right; if you *do* have the data right, you're often halfway towards getting a decent UX. And if the data model is wrong, no amount of visual design will save you.
The thing that ties this mess together with the nymwars (which was when I really got pissed off at Google) is their persistent failure to grok that people aren't as *simple* as they want. They seem to have this extremely simplified, monolithic concept of "person", and they desperately want to believe that that's sufficient. In reality, that's not only stupid, it's dangerous: in the real world, people have different identities that they use in different contexts, and mixing them willy-nilly like that can have serious real-world consequences.
And yes, getting identity right is hard work. But having put in that hard work for my little startup (indeed, I'm refining the identity systems this week), I don't have a lot of respect for the gigantic monolith of programmers being too collectively lazy to do so.
(I should be clear: I don't blame the rank-and-file engineers for this problem. This is all about failures of management priorities, far as I can tell...)
Re: So...
I was off G+ for about three years after losing the nym wars. I'm now back as the name people know me by. It's true that people are complex, but it's also true that people have a lot of investment in their self-created and -controlled identities.
people have different identities that they use in different contexts, and mixing them willy-nilly like that can have serious real-world consequences
The incident that caused me to drop Google search came when someone was at my house and we were researching a medical condition. Search terms from that history popped up as suggestions in my browser DURING A PRESENTATION AT WORK. oh my gawd
no subject
no subject
And in a way, that's the root of the problem. AFAICT, Google regards all other efforts as being nothing more than alternate ways to feed their maw of Own All The Information. Their money comes from having All The Information (to improve their Search and Advertising results), so that's the corporate priority. But the implication is that they don't actually care about all those other products *as* products -- they only care about making them good enough to provide them with Information, and anything beyond that is expendable...
no subject
no subject
As for multiple emails, I'd settle for a way to tell it not to hang onto every email address for someone that they've ever used to send you mail. Because when someone gets a new address they want to use, that doesn't mean I want to delete every email they've ever sent me from the old one, which seems to be the only way to keep it from getting back into the autocomplete option list when you start a new message to the person.
no subject