Update 2:

Via a Twitter post from Engtech (which I found on FriendFeed), I came across a suggestion from this guy about how to re-enable your Twitter cache. I tried it and it seems to have worked. And now — finally — there is a Twitter blog post about the difficulties (hat tip to Frederic from The Last Podcast).

Update:

There’s a yellow box at the top of the home page for Twitter users that says:

“Due to some cache changes we made Friday, you may not be seeing all updates in your timeline (don’t worry, they’re still there!). Thank you for your patience while we fix this issue.”

As Nav notes, this is the equivalent of flowers after a fight. I think we could use a bit more than that.

Original post:

MG Siegler of ParisLemon and VentureBeat (and my rival for number nine slot on the Techmeme 100) is keeping the pressure on Twitter about the system’s recent problems — intermittent dropping of people’s messages, etc. — and so he should. For a service that is supposed to be all about real-time communication, the Twitter gang have been doing very little of that about their own issues. People continue to complain on the Twitter customer-response forum, and have gotten little response.

The last update from the team on the forum is almost two days old, and says that the problem resulted from a caching fix — which ironically was supposed to make the system more robust and faster, but seems to have opened a black hole into which people’s Twitter posts randomly fall. The last post on the Twitter blog is from five days ago, and it’s about how great Twitter is for helping that guy get out of jail in Egypt. It’s a good thing he’s not there right now, because he’d be out of luck. The last Twitter status message is from two days ago.

As MG notes, one of the big problems with this particular Twitter outage is that it isn’t really an outage at all — it’s just a seemingly random disappearance of posts, which makes it difficult to tell whether there’s even an issue at all, as I mentioned in my earlier post on the weekend. How many people are thinking “Oh well, maybe my friends are just too busy to respond to my posts.” That could push them away from Twitter gradually, I think. If you’re basing your business model on being a viral communications app, you had better communicate.

And that goes for the Twitter team as well — I’m sure they’re busy fixing things, but they should be out there communicating it. And as MG and others have pointed out, this is a particularly bad time for Twitter to be failing: the massive Web 2.0 conference starts tomorrow.

There’s a nice overview of the music industry’s dilemma in the current issue of The Economist. Nothing that surprising, but some worthwhile points — including a telling anecdote to start the piece, in which one of the major record labels has an epiphany about where the CD business is going:

“At the end of the session the EMI bosses thanked them for their comments and told them to help themselves to a big pile of CDs sitting on a table. But none of the teens took any of the CDs, even though they were free. “That was the moment we realised the game was completely up,” says a person who was there.”

There are also some numbers, of the kind that should make many a label stop and think, if they aren’t already:

“The volume of physical albums sold dropped by 19% in 2007 from the year before… For the first half of 2007, sales of music on CD and other physical formats fell by 6% in Britain, by 9% in Japan, France and Spain, by 12% in Italy, 14% in Australia and 21% in Canada.”

And some perceptive points about the slippery slope that the major labels find themselves on at the moment:

1. “Because sales of CDs are tumbling, big retailers such as Wal-Mart are cutting the amount of shelf-space they give to music, which in turn accelerates the decline.”

2. “Artists are receiving far less marketing and promotional support than before, which could prompt them to seek alternatives.”

3. “Record companies face such hostile conditions that their backers, whether private equity or corporations, are loth to spend the sums required to move into the bits of the music industry that are thriving, such as touring and merchandising.”

But perhaps the best point in the piece is the observation that the industry could have saved itself a lot of the pain it has seen over the past several years, and likely become stronger to boot, if it had shown more flexibility when digital music first became a reality, instead of deciding to sue everything that moved and call that a strategy. Now, they could be too far down the slippery slope to irrelevance to recover.

Just a short post to note something that I think every current or prospective Web 2.0 startup should probably read — or actually, several things, all of which are related to the demise of Kiko, an AJAX-driven online calendar that got its start in Paul Graham’s YCombinator summer camp for geeks. Kiko has effectively shut down and has put itself up for sale on eBay. The first thing worth reading is a post at the blog On Startups, which looks for lessons in the failure of the well-regarded calendar app.

The post’s lessons are not exactly rocket surgery, so to speak, but they are worth reading nevertheless — including “Google is the new Microsoft” and “Have a plan B.” Equally interesting and worthwhile, ironically (since he is critical of the On Startups post), is a post by one of the members of the Kiko design team, who posted a comment to the On Startup blog with a link to his own version of the company’s demise. Richard White’s lessons go a little farther than the simplistic “Don’t take on Google” — he notes that the day Google’s calendar launched was actually one of the highest traffic days for Kiko, because all the stories mentioned it.

Among other things, Richard (whose post has a comment from Narendra Rocherolle of competitor 30boxes) notes that Kiko lost its focus at a crucial time and thus its launch was delayed — allowing 30boxes and Google to grab more of the spotlight — and that the team tried to make the app a little too feature-rich. Kiko co-founder Justin Tan also has a post-mortem in which he mentions staying focused, and argues that an online calendar is a worthwhile thing to have and not necessarily doomed to failure (Don Dodge disagrees). All in all, definitely worth reading.

Update:

Paul Graham has his own thoughts on Kiko’s demise, which boil down to “don’t fight Google”, but David at Signal vs. Noise disagrees, and Scott Karp of Publishing 2.0 says that if Google is the next Microsoft then that’s actually a good thing. Umair at Bubblegeneration says not to cry for Kiko.