Amateur Hour Over At Twitter?

hoping to see better uptime for twitter with new hires.

It is a bit baffling that it takes them so long to establish an infrastructure that scales properly given that they know of their success for a while now and that sufficient funding is available. However, you never know how flaky the idea was conceived in the first place and how much the system suffers from it’s experimental stage heritage.

I hope Twitter gets on top of its reliability problems. Too many things are counting on it now. Wait until the next earthquake when traffic will spike 1000x above where it is right now.

Really hoping this fixes twitter’s issues, as you said, Mike, we’ve become dependent.

Got a dumb question.
How does Twitter make money?

It should be said, thanks to Blaine and team for building something we’re all addicted to. I do tap my hat to you and hope you do something interesting in your career path again. We’ll be watching.

DumbQ - everybody asks that, and nobody knows. No doubt their biz model will become clear before much longer. (though I remember saying that about Napster a few years back)

And given their time in operation, with no money making in sight, why exactly do we expect Twitter to be able to scale to support everything we want?

Is Twitter anything more than a grand experiment in chitchat? I think not.

Wow, someone certainly has an axe to grind.

I echo Scoble’s sentiment re: Blaine and team. Many of us have become addicted to Twitter thanks to their hard work.

I hope that with the upscaling, they retain the functionality that Twitter already has, much less adding new function. It would be a shame to see things taken away because they can’t handle the load.

@dumbquestion - Twitter doesnt make money. Except for Twitter Japan which has ads. Which probably wont earn them much seeing as the majority of twitter activity happens over their API.

I think Twitter should scheduled downtimes so you addicted folks will stop spamming stuff no one cares about.

Is Ruby on Rails the problem and not the guy?

http://antoniocangiano.com/200.....criticism/

Have to agree with Patrick - you guys sound venomous.

Personally, I think Blaine did a terrific job scaling something they never expected would grow like it did. He also built an amazing platform that millions love, hundreds have built apps off of and that works 95% of the time. It was Blaine’s idea to make the messages available via IM and by other means through the API. I don’t know about you, but I certainly couldn’t handle following 800+ people through my SMS.

I think smudging Blaine’s skills is an unfortunate misrepresentation of the hard work, long hours and multiple weekends he spent trying to keep us all happy.

FWIW…he left Twitter last week. This last weekend? Let me tell you, it would have been fixed in hours if he and @meangrape had been around (Jay was out of town).

So what’s up? Burying the twitter disaster thread?
Anyway it could be the rails curse…

So, what you’re saying is:

“I have no new information. I do have a scapegoat, though I have no concept of the technology I’m railing against. So let’s jump on the bandwagon and complain about someone, because it’s convenient.”

Blaine is an excellent engineer who’s put out some really innovative stuff, and _you_ are the reason why many people still don’t consider bloggers journalists. You’re a gossip rag, nothing more.

You need different people at different stages of the product maturity. The article is a bit harsh about Blaine. Don’t forget he helped to get twitter all the way to this level. And I’m sure that they had to be rather innovative to keep up with the quick growth.

It might take some time to see the results of the new people. I’m sure it’s not just “Buy X more machines…”

“Apparently Mike Arrington’s deleting comments from the post about my departure. Classy.” really ? http://twitter.com/blaine/statuses/795211971

Blaine in April 2007, on the ease of scaling twitter:

Amateur Hour Over At TechCrunch? Yes!

Blaine is the main reasons Twitter’s scaling problems were actually contained and under control for so long. To try and link his comments on Rails scaling with what is happening at Twitter now is a true demonstration of ignorance and bad reporting. The Twitter engineering team, which up until very recently was just 3 (!) people, deserves a lot of credit for being able to accomplish so much with so little. Pointing fingers at the developers demonstrates your lack of understanding and the low quality of your reporting. And if only you got some of your other facts right…

I can vouch for that deleting comment.
Comments against Mike et al. are mysteriously disappearing into the ether.

Wow, this is the most unnecessarily vitriolic articles I’ve ever seen. Twitter and Blaine took Rails to places it hadn’t been before in terms of scale and traffic (just as other sites did with PHP, Java, ColdFusion, and so on), and so acting as if he’s some unskilled hippie that didn’t know what he was doing just makes you look like an idiot. Sure, they had some bumps along the way, but no one else had even gotten on the same road as them to encounter the bumps. Being forced to figure out things along the way because no one has done them before doesn’t make you an “amateur.”

Even further, what the heck qualifies you to criticize someone’s ability to scale a website or doing anything remotely technical for that matter? You’re out of your league. Please stick to talking about business or whining about someone not giving you ad money.

i met all kinds of new people due the absolute randomness of direct message distribution over the past few days, at least here in europe.
so it did help the social aspect, but networking really needs work, lots of it.
look what happened to pownce …people just left.
everyone and their brother went on twitter, it’s the default now.

Eran - right. he’s awesome. that’s why he left and twitter hired all these other guys to scale the application.

@scoble sure I hope he lands on his feet, lessons learned and all that.

But I think the bigger lesson is for the industry as a whole - the days of the throwing out a POC and figuring out scale issues later are probably long gone.

I posted on this yesterday.

@Arrington: I don’t know. Aren’t you supposed to be reporting here? Shouldn’t you know if you’re going to post about it?

Love this:

This last weekend? Let me tell you, it would have been fixed in
hours if he and @meangrape had been around (Jay was out of town).

HOURS?!

I’m not sure if that’s a compliment or criticism.

I have to second Tara’s comments. We were lucky enough to have Blaine speak at FOWA Miami back in February and he’s by far one of the nicest and most intelligent guys that I’ve had the pleasure of meeting.

Blaine and the team have done an amazing job scaling Twitter to it’s present form and let’s face it, whilst we might be quick to criticise any down time, can you imagine life without Twitter? No, thought not…

Good luck Blaine - hope rumours of you moving to the UK are true

The most important question is how they’re planning on making money. Doesn’t matter how much we enjoy the product/service if it’s unsustainable. Maybe they just get bought out buy someone else who wants access to the users?

I know form scaling several MVC frameworks including Rails that its NOT DAMN EASY IN RAILS!

Michael, from the SAI:

“I am moving to the UK so that my partner can pursue her career. We’re Canadian and her visa makes it impossible for her to work in the US,”

How hard is that to understand?

Geez Arrington this post seems kind of personal. What’d Blaine hit on your boyfriend?

Blaine is a very smart guy, a fine engineer, and a professional. Twitter would not be many of the good things it is today without Blaine’s excellent work and passion. I’m toasting Blaine’s awesomeness and I know he’ll land someplace where he’ll be able to make another significant impact.

That’s a bit harsh. You’ve taken co-incidence and got causality without any evidence that I can see. Blaine’s leaving? Best of luck to him. He’s a super-sharp guy; I’m sure he’ll have companies beating a path to his door. As a service, Twitter is growing at a phenomenal rate with finite human resources. Something has to give - them’s the breaks. Twitter will be fine. Blaine will be fine. Techcrunch will be fine. Let’s all hug.

Twitter down time is good for the world economy; it’s the only time any work gets done. Don’t knock Twitter down time. It’s not like it’s an essential service anyway. I feel Arrington is being a little harsh and graceless here; Scoble’s comments are more appropriate.

Time for an acquisition. Your product is hot right now and who knows if the scaling issues will ever be fixed.

Wow, what’s the deal with the venomous personal attack? Leave the character assassination to Valleywag, Arrington.

-Charlie

C’mon folks, this is all in “fun.”

Vitriolic? Sure. We’re human. Who ever said we’re supposed to be nice.

If you want to put yourself out in the public, and to put it mildly, be overconfident (”It’s easy, really.”), then you better have a thick skin.

Welcome to Hollywood.

Hmmm… I wonder if the “Cache the hell of everything” makes it harder sometimes to delete posts.

Mike - if you agree with me that twitter has had serious uptime issues, it’s kind of hard not to follow the dotted lines to the chief architect.

Fred - no, see comment 20. scaling rails is quite easy.

How is their infrastructure hosted these days? Are they still gambling on bleeding edge technology or have they moved to a more tried and true dedicated server solution?

I’m not doubting that they have general architecture issues that would persist regardless of virtual or physical infrastructure, but I think they had originally placed too much faith in virtualization hype and attention to architecture suffered as a result (this is pure speculation, take with a grain of salt).

@Arrington - the problem is, you have no clue about this story. You read it elsewhere, copied it, and made it into a personal attack. Your post is the same as blaming the TechCrunch graphic designer for your poor reporting.

The tone of this post is why I’ve grown to dislike Techcrunch over the past months, and why it will never grow beyond a certain point.

This post is bitchier than most found on Vallywag, and contains more opinion that news. Techcrunch seems to have replaced crunchnotes lately for Michael’s bitching and sideswiping.

For a site that’s so hard on traditional media, there’s certainly a lot of lessons that could be learned about what constitutes journalism.

TC has also technical problems : I read in the comments, 10 mn ago, a reference to Blaine’s explanations about his leaving “I am moving to the UK so that my partner can pursue her career…” as in 32. , but this comment disappeared.

Will the person in charge of the service at TC be fired too then?

Software architects are second guessed as much as NFL coaches.

I want to add my name to the list of people thanking Blaine. Getting a site like twitter to scale to the volume it now has on a shoestring budget is non trivial. I would feel lucky to work with Blaine in the future.

Michael Arrington: You’re successful enough of a blogger to not have to resort to this kind of public humiliation of folks to gain traffic. If you have a beef with Twitter’s architecture, I would prefer you leave names out of it. You don’t know what went on behind the scenes, so you can’t judge who the problems should be attributed to. Let’s all make the internet a more hospitable place.

Am i the only one who COULD live without twitter?

Like, happily.

And…easily.

@jimmy, Michael: I’d be fine with the criticism if it wasn’t coming form an uninformed dunce who knows about as much about scaling a high-traffic website as writing informative journalism. Which, by the way, isn’t much.

Michael, I don’t think you know what you’re talking about, and I certainly I don’t think you’re familiar with the ins and outs of Twitter’s technical details. Personally, I’ve had all three of Twitter’s engineers (at the time) explain their architecture to me over dinner at Naan N’ Chutney, and I can tell you that Blaine Cook has *not* been their problem (nor has Ruby or Rails or any of the other technical issues favored by armchair analysts). This baseless scapegoating is unprofessional and your readers deserve better, even if you are unable to give them that.

Crazy! Dont kill him for his slides. Those are like bible for some of us. Any engineer will think multiple times for taking a responsibility like this. And the kind of stuff he had done with limited infra, time is amazing. I think mike, you gotta learn to respect talent.

Maybe new twitter team will remove twitter bugs

Jeremy, I’ve got two rails developers sitting ten feet away from me who seem to have their finger on the pulse of the rails community. And trust me, I toned the post down significantly. but you’re right, i don’t know much about journalism and even less about rails. but i do know that a for profit company that can’t get their shit together two years after launch deserves some criticism.

Wow, TC has got over its love of Facebook. Can anyone say TC loves Twitter ?

How many stories in a day do we need about Twitter ?

Lets all sing , Techcrunch and Twitter sitting in a tree , K—i—S–S I–N–G.

Dude enough of the Twitter love ,

only us geeks have ever heard of twitter anyway

coda - ok, what exactly in the post do you disagree with? the fact is cook said all their problems were solved a year ago. they’ weren’t. he looks a little dumb now. doesn’t he?

@Michael

You _could_ follow that dotted line… but he might just be tired of working nights and weekends. Or maybe he’s got a startup idea of his own to pursue. Or perhaps he’s eloping with the missus and then joining the circus. Who knows? It’s speculation, not news.

Re: 41 - So stability has nothing to do with management and operations? I don’t have the full story of why Twitter has some stability issue and can’t even tell if they are issues of scale or something else. The way you connect the dots is nothing more than wishful thinking. It might be true, but you *are* just guessing. Again, quality reporting!

Someone can’t scale Rails, and that makes them an amateur? Wow, talk about uninformed writing. Some might even say “amateurish”.

Let’s face it - most sites never become popular enough to have this kind of scaling problems. And few have pushed RoR this far. I think we should thank Blaine for giving us Twitter and be grateful for the chances he took opening up the API, and wish the next guy well.

what else you guys got?

“Mike - if you agree with me that twitter has had serious uptime issues, it’s kind of hard not to follow the dotted lines to the chief architect.”

That’s a surprisingly ignorant statement.

You can be the best architect, or anything, in the world but factors outside your control can change the end result in many ways.

Maybe Blaine just became fed up at to the level of staff/software/hardware support he received while trying to scale and support such a massive platform as twitter?

I don’t know if this is true, but I do know your statements are needlessly bitchy and unprofessional.

As for the slides - pfft. Talk about taking things out of context.

@Michael

If Twitter is such a hack-y piece of software, perhaps you should stop using it. No sense in supporting the work of an ‘amateur’ developer like Blaine.

I know it might be hard… you know… not hearing you complain about stuff every 10 minutes on Twitter, but the rest of us (non-250 folks) will survive. Don’t you worry.

I think Blaine’s done an excellent job with twitter. Getting RoR scaling this much is a mean feat. I don’t envy his successor.

I believe Blaine could be right on horizontal scaling. Adding more servers can solve the problem. Perhaps the folks at procurement may have had cash issues? Perhaps they couldn’t find the cash to improve their infrastructure?

One can only guess. But I don’t fault his choice of platform. (even though I’m a PHP guy).

Yea - Twitter in it’s entirety is a masterful bit of communicaion though - it’s like a tamagotchi crossed with a web-app. It’s small and cute and anthropomorhisable - and the fact that it wobbles sometimes doesn’t stop us loving it. Tamagotchi do wobble. They’re kindof supposed to.

And the Blaine Cook? Never met him - but he seems like a tinkerer, and a project-doer and an all-round geek’s geek… whereas the writer of this blog most certainly does not.

Which is my 2C why this article is such an own-foot-shoot.

From No. 61:
> what else you guys got?

Now THAT’S the kind of reply Blaine should have posted here.

Instead, we get the righteous defenders, “We love you Blaine,” which reminds me of the moms in the parks with their kids doing the “Good job!” Sickening.

Wow, this what happens when you annoy the Ruby/Rail’s community in any shape or form (but that is a different rant) - this is not about Ruby Scaling in anyway, although
they have certainly suffered from not rebuilding from scratch when they should have at the outset. This is about the many-to-many problem in databases, and frankly given the timescales they have had the scaling/reliability issues should not have been an issue even if they had programmed the bloody thing in BASIC. So get over the ruby thing - this is just about not understanding scaling in general - as rule 1) performance DOES NOT equal scalability - so should he have gone? yes - about 6 months ago.

sent from: fav.or.it [FID327972]

@Eran: You consider TechCruch real reporting? Please!! This is juicy tech gossip at best. Just read and enjoy it!

Mike, this isn’t valleywag. And you’re not very good at trolling.

Every fifteen years or so, there’s a three foot ‘Nor’easter’ blizzard in the Washington DC region. Happened in ‘79 and ‘96. Shuts everything down for a week: Gov’t, schools, many businesses. You can’t drive until the plow truck comes, which may be five days for secondary streets. A lot of folks freak out at first, then acceptance sets in. Cross country skis and Flexible Flyers and Snow Shoes break out, and parents play with their kids again (rare opportunity to build world class snow forts and igloos). You re-meet the neighbors, shovel drives and sidewalks together, borrow supplies if needed, and pop Advil like candy because you worked long-forgotten muscle groups moving several thousand pounds of snow. And you take a zillion pictures of the most amazing winter landscape you’ve ever seen.

Big snow in Twitterville. Enjoy life for a while.

@Prokofy Neva:
Aww, c’mon, Prok. Can’t you keep your uninformed and deliberately inflammatory grandstanding/griefing to the Second Life world?

@Coda Hale You said “I can tell you that Blaine Cook has *not* been their problem (nor has Ruby or Rails or any of the other technical issues favored by armchair analysts)”

If you know this is the case, then let Mike and the rest of us know what or who is responsible so he can update the post. Otherwise, your comments seem more baseless than Mike’s.

Speaking of amateur hour…

Put it on Google Appengine. Problem Solved

ROR Sucks 

mike I think u should brush up your math. 2!=1
twitted had to hire 2 really good engineers to replace Blaine so what does that say?

Back to seesmic : where’s the disclosure? Take off your lame seesmic widget

Many people who know how to build a single-server web site underestimate the how hard it is to build a massively scaled high-traffic web application. It’s like the difference between building a hang glider and building a jet liner — sure many of the basic principles are the same but the engineering problems are incomparable.

Most of the current high-traffic web sites have had the advantage of many years to build up to their current level of reliability. They have had time to build up their engineering team and to accumulate the hard-won experience learned from mistakes in architecture and operations. Asking for the same level of maturity from Twitter is unrealistic at this point.

Good luck to Blaine in his future career.

So what if this post calls someone out? Blaine gave talks about scaling and worked at a company that has huge scaling problems. That’s opening yourself up to get beat up.

Whenever people ask me for help in scaling I don’t just say it’s hard so they don’t feel bad. It is hard. People who think scaling is easy haven’t ever had to scale.

Just because it’s hard doesn’t get someone a free pass or even a commendation for their efforts. The guy who fixes my car doesn’t get to tell me that he couldn’t fix it because it was too hard. Neither does a company doing business on the web.

It might sound harsh, or crass, but it’s the truth.

I wanna be more like Michael Arrington. Pissing everyone off and yet they don’t go away. Rock on!

Michael, the fact of the matter is that Blaine is responsible for scaling RoR to it’s largest installation and rollout that the framework has ever seen and will possibly see for a long time.

Blaine and the crew did something that **no one has ever done given this framework** and that not many people can attest to having done with RoR.

The two developers you have sitting next to you haven’t done that, and I doubt they could achieve it without a lot of pitfalls and A LOT of long nights. In fact, they’d would probably need to consult with Blaine himself to get where Twitter has gone.

If Twitter was a lesser service, we would migrate away and it’s continued scalability issues wouldn’t warrant a second glance by Mike and co. As it happens, many of us have adopted it as something downright vital and it is only because of Twitters success and utility that it has opened itself up to criticisms.

It is only because Twitter has been adopted into the fold as a useful tool that these issues are worth talking about, so Congratulations Twitter.

john - wait a minute…i mentioned that we have rails guys here in response to the comment that we don’t know what the hell we’re talking about, or what the rails community really thinks about cook. To extend that to “you guys couldn’t do better” is absurd.

re cook, if he did that great a job, they wouldn’t let him leave. he might be the nicest guy in the world, but he’s held twitter back.

Twitter is a simple solution made complicated internally.

I supposed they should have gotten better help since the first day.

Firing now is not the solution, but they need a better framework of doing things and better strategy to win back the users’ confidence. I would suggest a “relaunch” will give them a better image.

I second, third, fourth and fifth everyone here who’s paying Blaine complements.

I’m dealing with scaling the platform that my companies latest venture is developing and I had the pleasure of getting a lot of time to talk to Blaine at FOWA this year. Talking with Blaine ( and Joe Stump, also presenting ), helped me understand the nature of some of the problems that we’ve yet to face, and ultimate extended our plans/design to allow for them when they come.

I can’t credit Blaine’s knowledge and abilities higher, I’ve spoken to hundred of developers over the years, and can tell who’s talking from experience and who’s talking rubbish. Blain is definitly in the former category. He would be a credit to any team.

I wish him the best of luck for the future, and extend the UKs welcome to him should he indeed be coming here.

Michael: Twitter is definitely due some criticism, but Twitter != Blaine Cook. If you knew anything about the internal politics of Twitter, you’d know that there were a *lot* of elements out of his control — like *all* of their operations, which until very recently was under Joyent’s control. Joyent of the where-did-our-ZFS-go debacle. Joyent of the 10-days-of-downtime problems. Joyent of the no-longer-having-the-Twitter-account-because-they-suck issue. Twitter’s architecture is almost totally horizontally scalable, but that doesn’t mean a goddamn thing if the systems crap out under high load, or go down randomly, or perform poorly in general.

But you’re neither an engineer or a journalist, which apparently means you don’t know and aren’t willing to ask anyone.

Wouldn’t let him leave? What is it, Scientology?

Simmer down, cowboy.

It’s another nail in the coffin for Rails. Remember when Web 2.0 companies used to put “we’re built on Rails in their slide decks”? The question is when does someone build a popular Twitter clone on AppEngine, and then Google has to either make sure AppEngine scales to prove it can handle it or pull it down to prove AppEngine isn’t all about cloning all the popular Web 2.0 sites? All things considered, Twitter is a pretty simple application, a relatively small feature set with a huge amount of storage, sounds like the perfect AppEngine-based webapp. Why hasn’t Dave Winer written it yet?

Wait, Blaine Cook killed Michael Arrington’s puppy?

Anyway, ask the two Rails developers to the side of you: a year ago a lot of people thought scaling Rails could be chalked up to that slide.

Anyway, nice rant. Enjoy an extra scoop in your coffee more often.

@LorenFeldman

Douche!

’scaling Rails apps…“It’s Easy. Really.”’

hahahahahahah

Fail

@Lauren Feldman and Arrington: Funny to see you guys (and everyone else) getting so damn worked up about a FREE APPLICATION.

Bunch of children - its a free app. It’s not Blaine or anyone else’s fault that you “depend on it”. Screw you - pay up some money.

This getting shit for free and then bitching about it meme is really starting to piss me off. Imagine if you went to Sams Club or Costco and started screaming at the old lady passing out free cheezits because she didn’t have any milk. That’s what you are doing now.

Don’t like Twitter and the way they scale? Fine. Leave. Get the hell out. If you depend on it so much, then its a mission critical app and maybe you should go build your own damn software and then try to scale that yourself.

Ridiculous.

Re: “if he did that great a job, they wouldn’t let him leave.”

Wait, so on the one hand Twitter is dysfunctional because they’re continuously failing to provide the uptime their users require, but on the other hand Twitter is functional enough to identify and solve their own problems? They’re witnesses in one breath, and in another they’re a foil?

@ Ben. Funny you should mentioned them, they were out in force in London again today.

Anyhow, I actually have no idea what scaling is so I’ll just sit back and watch everyone fighting over what seems a non-story to me

Criticizing Twitter and crucifying a departing Twitter engineer are two entirely different things. Taking one of his presentation slides out of context, conjecture, supposition, lack of any credible witnesses or evidence… This isn’t news, it’s a snipe at a guy who knows a hell of a lot more about his job than you and your 2 Rails buddies. What do you have to gain from attempting to soil Blain’s reputation?

“Sitting next to” Rails developers does not equate to knowledge about scaling the highest-traffic Rails app there has ever been, any more than my having John Resig’s JavaScript book sitting on my shelf makes me a JS ninja.

This story is a blatant smear job, nothing more. We’ll see who looks a little dumb when Twitter’s problems – which you don’t understand – continue long after Blain’s departure.

Being the Chief Architect of any company is more a political position than a technical one. Not to say that there is not a significant emphasis on technical matter but political issues have more weight in some cases.

Case in point: The Silicon Valley Ruby Conference - Your words have to be carefully crafted or you will catch a fireball in the shorts.

stephen - not really. i have no idea if the new guys can fix twitter. it’s just really clear that the old guy wasn’t able to.

WHAAAAAAAAAAMBULANCE!

I’m pissy.

Mike
!!!!!!!!!!!!!!!
!!!!!
!!!!!!!
!!!!!!!!!!!
!!!!!!!!!!!!
where’s the disclosure on seesmic ?
Its so lame so one wants to use it
bs video technology + hype

Mike: I think you are being a little harsh. The architect did a great job to get Twitter to this stage and actually still have a client base. He just could scale any larger, thats all

RoR is nothing more than a prototyping platform and a nice web app infrastructure for smaller sites. They need to migrate their platform to something that scales. I hear J2EE is kind of nice.

@Michael: If you think your Rails guys or gals have their “fingers on the pulse of the Rails community” and they think that Blaine is Twitter’s problem, then they’re lying to you. People who, like, actually know Blaine and other Twitterers can offer you much better information rather than baseless, ad money driven drivel. I can put you in contact with them if you drop me an e-mail (or I’m sure that Coda would be very happy to offer you any details you want).

Of course, investigation probably isn’t your thing since you don’t know much about journalism.

@Coda Hale You little douche bag, don’t go blaming the twitter mess on Joyent. Twitter played that hand and it was a bluff. And now they’re trying to pull the wool over our eyes again by shifting the problem so it’s less visible (”Let’s just always return 200’s. So what if the queued message gets dumped.”), instead of just hiring professionals to solve this.

If there was $ in twitter, someone else would have done it already.

What makes me laugh is that everybody considers Twitter a simple application - I guess that everybody (oh, I mean *3* developers) already integrated web (including REST services), jabber and sms, and with something near twitter’s userbase, also given that most of their problems were from their provider.
Armchair analysts.

Picking some of the funnier moments out of these posts so far:

#35 Twitter will be fine. Blaine will be fine. Techcrunch will be fine. Let’s all hug

#54 Let’s all sing, Techcrunch and Twitter sitting in a tree, K-I-S-S-I-N-G

#61 what else you guys got?

#65 … this article is such an “own-foot-shoot”.