Yes, sometimes.

Or because the stories of scaling have potential lessons?

Yes.

Let me add a few observations of my own:

Are people upvoting because Twitter is a genuinely new and popular service, and hence very fashionable at this moment? Yes.

Are people upvoting because Twitter's design has just been sitting there, right in front of our faces, for years, and yet we plugged away on email and IM and static HTML sites and blogs and Facebook and never even thought of trying something like this, so we're fascinated to discover it in the same way that biologists were fascinated to discover the platypus? Yes.

Are people upvoting because they're still trying to understand what Twitter is, and how to build something like it themselves, but they're confused because it isn't really email, or IM, or a microblog, or Facebook? Yes.

Do half the folks on news.yc look at Twitter and see something that they simultaneously crave (a startup with a viral app) and fear (a startup with a viral app that is pissing off its customers) and envy (because the customers moan, and they complain, but they don't abandon Twitter for the alternatives, at least not yet)? So, try as they might, the entrepreneurs can't possibly look away?

Yes.

Overall, I think Twitter resembles the conference circuit. A small circle of people dominate the crowd's attention, largely by talking voluminously among themselves. You might be able to learn something new, but you'll have to filter out a lot of noise. And you're probably better served by finding a person or two and having an actual conversation, or writing an essay, or actually doing some work instead of being interrupted every few minutes.

Amazing fact: Twitter works perfectly well in asynchronous mode. Turn off the alerts, visit Twitter once or twice a day and catch up. Catching up on Twitter is pretty easy because everything is very very short by necessity.

Web usability guys like Jacob Nielsen have been saying for a decade that the essence of writing for the web is microcontent -- we should all stop writing paragraphs and concentrate all our energy on writing decent titles and headlines. Well, Twitter is a machine for forcing people to write good headlines and nothing else. As a result, I find it wastes far less of the reader's time than any other web technology to date.

;)

1) Twitter is like any other social network-- its value to you is dependent on which of your friends is on it. For most of us who have lots of friends that aren't 100% plugged-in to the hottest web trends, this severely limits twitter's value. However, the technorati get a lot of value from Twitter in that it's an elite social network (aka mutual backrubbing), and a way of getting to each other's inboxes (via @ and d messages). The electronic equivalent of TED and Davos.

2) Twitter is basically a blog platform in disguise-- where the hordes of wannabes who follow calcanis and ev and scoble really what they're doing is a fancy equivalent of subscribing to their rss feed leaving comments on their blog entries, with the added social bonus of making it seem like you're with the 'in' crowd since you're on twitter leaving @messages for loiclemeur like the big boys do (see 1 above).

I think this might by one reason why twitter is so popular - it can be used for many different purposes.

1) The asynchronous nature of Twitter networks models real life quite well - you might be interested in everything a person has to say, but they don't even know/care about you. This works with Twitter, because following is a one-way concept.

2) This makes it hard to get spammers onto Twitter, because the only way for spammers to get their posts listened to is by having people follow them. Also, services like the Twitter Black list exist to help people sort the wheat from the chaff.

3) The best "feature" of Twitter is its tabula rasa attitude to what you post - as long as it fits in 140 characters, Twitter doesn't care. This has allowed people to build fun minigames (@lyricbot), query/response info bots (@weatherbot), and plug Twitter into their external apps (Remember the Milk, Buxfer, etc). This sets Twitter apart from something like Pownce, which had a similar idea (complete with the asynchronous fan/friend model). The difference (IMO) was that Pownce tried to impose some sort of order on its users, which limited the way it could be used.

Unlike a social network, I don't need to be friends with someone to see what they post. I can stop following someone whenever and follow them again later. In short, it has the useful content of a social network without the spammy apps, useless profile information and dubious "friendships". It's a better version of irc.

Twitter feels a bit like a 24/7 cocktail party full of intelligent and interesting people, where you can chose who to listen to. I think they've also evolved (or perhaps resurrected) a cultural norm of openness where you can listen to and engage in conversations with anyone. I think that, along with the asynchrony and discoverability, is what makes it slightly magical.

That being said, it has problems: - It can definitely be a waste of time.

- There aren't good tools for filtering, prioritizing, and contextualizing the feed.

- There's an excessive density of 'professional' social-media types who seem to spend all day twitterring about nothing but Twitter.

- I don't understand the appeal of the IM-type clients at all. They look pretty, but I can't imagine anything worse than being constantly interrupted by windows popping up all over my screen with random, disconnected tweets.

Twitter, by definition, appeals most to the people who are somewhat extrovert, like to talk/write a lot about themselves (see Michael Arrington complaining recently that Twitter's downtime denied him from broadcasting, live, his food poisoning. I Am Not Making This Up)

Therefore, these people tend to talk a lot about what they are doing - and thus talk about Twitter, which makes Twitter more well known in our circles than much bigger websites.

A classic example is Apple's products, in the pre iPod/OS X days. They had a tiny market share, but a very high mindshare. The reason, in my opinion, was that Apple's user base consisted of many creatives - designers, writers, musicians etc. If Apple computers' users were mostly biologists, civil engineers and plumbers, you wouldn't hear about it nearly as much - even if they had the same market share and the same customer loyalty.

Hype seems to be much cheaper now, though. Many Apple products truly had groundbreaking engineering and design. Many Apple users were well known because they were talented and creative. Compare this to to Twitter's technical "challenges" (I mean, seriously - it's very minimal in UI/functionality side, as it should be. They got plenty of money and can attract top talent. If not scaling, what have they been doing for the past year?), and its most well known users...

I've thought about how I would have made Twitter, but the truth is that the Twitter problem is probably not something that is easily fixed. The people at Twitter have had a lot of real world experience and have more data to base their development on.

That, and also how the spectators and critics are investing in themselves by how much they complain and speculate about how each little thing is being mismanaged. It is Twitter, after all.

It's a social thing and I love it.