Given my own experiences with AT&T, I find it difficult to find any sympathy for them.
I do have some sympathy for increasingly (re)balkanized electronic populations. The Internet, for a time, opened an avenue past local, restrictive controls.
Unfortunately, although perhaps inevitably, the U.S. and related players felt compelled to replace them with a global overlord.
At which point, balkanization starts to look at least partially useful, again.
Hopefully, though, we will not regress simply to state-controlled balkanization. Grow orthogonally, and develop independent networks. Decreasing costs may favor this.
"Unfortunately, although perhaps inevitably, the U.S. and related players felt compelled to replace them with a global overlord."
I am starting to suspect that creating a fully-controlled global network might have been a conscious effort since the early stages of the internet, especially given its history with DARPA. Perhaps this effort didn't begin that far back, but certainly since the mid 80s or early 90s, at a point where internet started crossing the critical threshold from hobbyist or military interests, to everyday/commerce interests. Certainly, nobody could have predicted how popular the internet would become, but once its potential became clear I can only imagine how irresistible the opportunity for its control would have been.
The more I think about the current state of affairs, the more I understand now why the Chinese government has resisted efforts by foreign data providers to become primary providers, and instead creating their own versions - it seems they had the foresight that none of us (engineers) had. They smartly created their own search engines and social networks to stop/limit the funneling of their data to foreign companies with the side-benefit of having near-full control over domestic data.
This is hardly the only circumstance, but as a concise and personal (as opposed to commercial) example:
Something over 10 years ago, I ordered DSL from them. It was my only high speed option at that time. Actually, to be fair, it was I think at that time Ameritech, which subsequently fairly quickly got sucked into SBC, which is what AT&T really is, now. (SBC bought the "shell" of AT&T some years back, that included perhaps principally the name and branding. All of the old AT&T "substance" had long since been hollowed out.)
Placing the order took less than 10 minutes. The call was answered, by a live customer service rep, within a minute.
Accomplishing the installation took three half-day appointments (they wouldn't provide a narrower window). They simply blew off the first two.
The third appointment resulted in bare connections hanging on the side of my building. The third party contractor they sent reused some old, old wires someone had left hanging there and didn't even bother to wrap the splices.
(Fortunately, my downstairs neighbor was friends with a real, honest to goodness union lineman for the voice side of their operations, who was over visiting subsequently and cleaned things up a bit as a personal favor to my neighbor.)
Their network was for shit. tracert would show requesting bouncing through sometimes over a dozen of their servers before even making it out onto the wider Net. And it got to the point of going down at least once a week.
One time, I spent a half hour or so on the phone just talking with and calming down one of their contracted "service" reps, who was so frustrated that she was quitting her job the following week and going back to Maine. She was nice, and straightforward, and with a good sense of humor, so I didn't mind taking the time.
She told me that, in her contract position, all she could do was file a ticket. She couldn't even examine the status of an already-filed ticket. Even for them, the subsequent repair process was a black box.
I guess I should add that a call for service (e.g. those weekly network dropouts, for example) took, at a guestimate, on average about 30 minutes to get out of queue and reach a person. During this time, sales calls (calling the number to order service) continued to be answered almost immediately. Priorities could hardly be clearer.
Add to this the 700+ million in tax breaks and subsidies that SBC (now AT&T) received in return for committing to universal high speed Internet access in this state. Whereupon, they turned around and immediately deployed and hired lawyers and lobbyists to get them out of their half of the deal while keeping the incentives.
Their lawsuits and lobbying against any and all attempts of municipalities to deploy their own networks, most often only after AT&T has refused to provide said service themselves.
On and on...
Anything that keeps them from growing further? I'm all for it.
Balkanisation, is a geopolitical term, originally used to describe the process of fragmentation or division of a region or state into smaller regions or states that are often hostile or non-cooperative with one another.
It was more for the benefit of others who may not have read it on Wikipedia.
Making a reply in a thread does not always mean you're directly and only replying to the person you're replying to -- it can be more than that, it can be a reply that moves the discussion along, and gives off information and insights to anyone who's reading the comment. Seeing as you're a new HN user, I encourage you to give the 'comments guidelines' page a read: http://ycombinator.com/newsguidelines.html
Indeed. I was just about to look it up before reading an article linked also in the comments. Now thanks to mjallday's note I've got the context I was looking for and don't need to drain thirty minutes doubly off-topic reading about Balkanization. (Not that I'd mind, but Wikipedia can be dangerously engrossing like TvTropes...)
I do have some sympathy for increasingly (re)balkanized electronic populations. The Internet, for a time, opened an avenue past local, restrictive controls.
Unfortunately, although perhaps inevitably, the U.S. and related players felt compelled to replace them with a global overlord.
At which point, balkanization starts to look at least partially useful, again.
Hopefully, though, we will not regress simply to state-controlled balkanization. Grow orthogonally, and develop independent networks. Decreasing costs may favor this.