Hacker News new | past | comments | ask | show | jobs | submit login
Poor Man's Email? (scripting.com)
66 points by brlewis on Mar 4, 2009 | hide | past | web | favorite | 36 comments

His speculation on Google 160 character limit is curious. Twitter made it 140 to allow for usernames and the colon prefix. There is a real reason for it, not a sacred "140 characters is arbitrary and magical" feeling Twitter is holding on to.


Since 160 characters is/was the common limit to SMS -- which I think carriers themselves define, it feels like the author is missing a pretty key element in Twitter's implementation. The reserved 20 characters _is for_ metadata. I don't know the SMS protocol at all, but I suspect if Twitter could have dropped metadata into a header or somesuch, they would have.

Yes, the idea that there is any extra room in the SMS protocol for metadata that Twitter is somehow missing is completely retarded.

(Source: I spent 2 years building SMS-based applications for a mobile content company. I can format my SMS in binary, and trust me, there's no space in there.)

Most people don't use Twitter over SMS any more.

I never got that unless you followed about 4 people.

May I ask what company/product you worked for?

I am launching my own SMS content service very soon, just curious.

They are called Boltblue (Boltblue.com). In terms of product offering they're very unexciting: just your bog-standard ringtone/wallpaper company. But I had to implement payment systems and an SMS gateway in Java, going all the way down to SMS binary format because (at the time) they sold black-and-white "service logos" for Nokias, a crazy hack which fitted 160x30 pixel binary images into multi-part SMS messages. (There are also a couple other mobile products where it's useful to be able to format your own SMS)

I learned a lot and the people who worked there were great, although it's not a company that's going places.

The interesting story is how mobile operators manage to charge more for sending 160bytes of data than they do for a call. And who was the genius that created the $Bn SMS business out of some spare capacity left in the GSM spec as a debugging feature.

Well, SMS is win-win from that perspective really. Sure, the 160 bytes of data costs 10 cents, which is a little crazy, but 160 characters replace a conversation that might last two minutes or more, with greetings and small-talk and such. Text is more efficient than speech for conveying information, and the social conventions around SMS make it even more so.

So consumers save money by sending a 10-cent SMS instead of making a 20-cent phone call.

I don't know where you live, but here in the states, an SMS costs 20 cents, for both the sender and the recipient, netting 40 cents for each message... It's a fucking scam that the clueless people put up with because they just don't care...

Interestingly when roaming in Europe, T-mobile charge me 40p/SMS and 20p/MMS. It is actually cheaper* for me to write down what I want to say by hand, take a photo of it, and send that than it for me to tap out an SMS!

Why is this? Well it suggests to me that they want to encourage people off the old SMS infrastructure, possibly due to capacity or manageability issues, and shift them onto the much newer GPRS/3G/whatever network.

* Yes I know you can send text by MMS and I have an E71 so typing isn't a chore anyway.

No - they know most people are going to use SMS and want to gauge them for a s much money as possible. The SMS infrastructure is fantastic (for them), they use a spare 140bytes in a message the phone is going to send to the tower as part of a 'I'm still here' message anyway - and then charge you 40p for it!

Or they could do some actual work and charge me 20p for it! Why charge less for more data? There must be a reason.

Here (Ireland) it's pretty typical for phone plans to come with all sms free, or at a low price, such as 3c a text. Even on the worst available plans you might pay 9c per sms (and only the sender pays).

I know that twitter started as an SMS app, and it may still be popular as such (I always use it on the web), but as internet-on-your-phone becomes more ubiquitous, sticking to the limitations of SMS becomes unnecessary. I'm not sure how far we are from that SMS to internet swapover, but my guess is not very far.

sticking to the limitations of SMS becomes unnecessary

I think the 140-character constraint is a huge part of Twitter's identity and success. It forces concision. It's one of those constraints that breeds creativity.

Respectfully, I disagree. The "SMS isn't important" argument is why Pownce failed. 50% of twitter's content is sent in via SMS. It's not consumed via SMS so much, but it sure as hell is produced that way.

And remember that the mobile Internet in the US is pretty good. In much of the rest of the world (the places where people have mobile phones but don't have computers) mobile Internet is still slow and poky. The switch-over from SMS to Internet hasn't happened even in the US yet, so it will be a long while before it becomes a good idea for Twitter to drop SMS, on the order of 5-10 years.

50% ? I'd be interested to see some data on that. Do these come up saying "from sms" rather than "Web"? If so, I've never seen one.

This pegs 'txt' as pretty darn low (3%): http://tweetstats.com/twitter_stats

Also months stats from Jan: http://dcortesi.com/2009/02/19/the-real-top-20-twitter-appli...

Wow, I stand corrected. I must have been thinking "50% of twitter is not from the web" and translating that to "50% is from txt". Apologies.

I stand by the value of SMS as a driver of valuable content, and in the developing world.

Possibly, but since it only accounts for 3% of tweets, I think they could pretty safely change the limit if they felt like it.

I know a few people who receive SMS from Twitter on their phone. I imagine there's more people receiving Twitter SMS than sending it.

More people want the option of receiving Twitter by SMS if they have to than actually do by choice day-to-day. If being out of good signal meant not being able to use the service at all, people would be less keen on incorporating it into their lives.

The limit is 140 octets. That can mean 160 7-bit characters, 140 8-bit (1 byte) characters, etc.

70 Unicode16 chars :) I wonder what twitter does about unicode here...

UTF is variable width; it'll use 8 bits/char if it can, and only use 16 bits if it has to.

Funny, I've been calling it a poor man's IRC from the start. I'll still take good mailing lists over Twitter any day, though.

At least a decent mail/news client will track and aggregate all of your unread messages for easy consumption to keep up with the topics. Explicit threading in mail/news is just icing on the cake to make conversations so much easier to stay on top of.

Twitter is the smoke signals of internet communication by comparison.

Twitter also happens to be 95% noise. At least if someone is rambling in a mailing list, I know they're rambling on-topic. The tweet that made me stop opening my Twitter client entirely? "Just saw dog poop in the street."

But one thing I never thought of Twitter as was Poor Man's Email, which is how Google CEO Eric Schmidt described it to analysts yesterday. ...

I couldn't imagine two things being more different, Twitter and email.

1. Twitter is primarily one-to-many, where email is primarily one-to-one.

2. Twitter is by default public, where email is by default private.

So why would Schmidt equate them, if they are so different?

This is the interesting question raised by the post. I'm not sure how to answer it; but my guess is that the answer is unlikely to be found by thinking about the details of the SMS protocol.

Here's my best stab at it, from the perspective of a guy who saw the internet in its infancy (not myself):

Email was the first great asynchronous digital mass communication tool built on the store-and-forward idea. Twitter (or SMS'd tweets, really) is another, much more limited form of the same technology.

Winer's argument rides on adding metadata, ergo why the SMS discussion is important. For Google to play in his argument, they need to solve the SMS problem, or ignore it altogether.

The way I see it, since both are communication tools, this is why are equal. They soon diverge. The public and ad-hoc nature is where it is substantially different.

A gross simplification of primary use cases:

IM: 1 to 1, private, active

Email: 1 to 1, private, active

Twitter: 1 to many, public, passive -> active

The passive/aspect of the conversation becomes more organic in a public space -- changes into an active conversation when someone tweets back.

Why would he equate them? Only because it's a textual communication medium, IMHO. GTalk is in a way, an extension of GMail. If you're not there, it pops in your inbox. I don't see Twitter merging with email until it becomes part of the active conversation, something which should be pointedly and purposefully in your inbox.

I think Google will try to create a Twitter clone but in the same fashion as Youtube, will end up discontinuing it and buying out Twitter.

I believe it's almost like a hazing ritual or proving ground. Twitter has enormous potential to integrate Google's Adsense (or more recently, FriendSense) or become another distributive voice for their products.

The question though is, does Twitter want to be bought? Is Google too big now to let Twitter remain truly innovative (I know the catch to this argument is that Twitter has not been very innovative at all actually)?

I have a feeling some big things will be rolling out the door. The founders have been smart enough to let the user base define the product, and I certainly believe that when the time is right, Twitter will be a killer.

Hopefully they don't wait too long though.

>I think Google will try to create a Twitter clone but in the same fashion as Youtube, will end up discontinuing it and buying out Twitter.

They already did this: Jaiku


What's the point of trying to buy out Twitter when you've already bought out a Twitter-clone.

Jaiku is a great twitter clone and can be improved much more to dethrone it.

But the first thing they need to do is change that stupid name.

See, I agreed with you that Google would try to clone or buy out Twitter. That's what I thought they were trying to do when they bought Jaiku -- but then they sat on it for two years and then killed it completely, throwing the source overboard to save it at the last minute, like Netscape did.

So that leaves a buyout -- in which case, what are they waiting for? Google is not known for hanging around debating acquisitions. It gets in early (with the notable exceptions of YouTube and DoubleClick). So is it waiting for Twitter to prove itself? With CNN constantly plugging Twitter and even members of congress getting in on the act, how much more does it have to prove?

I think the time for Google to get involved has passed. They will partner with Twitter as an independent company, if anything.

IMO I think Google is hoping that twitter becomes a better search platform to compete against it. Google is the top dog and nobody is even close to its dominance in the arena, it needs competition if it wants to avoid the whole anti-trust thing.

I am surprised that Windows 7 doesn't have an inbuilt Twitter competitor (or does it?). With ICQ, lots of companies managed to grab a share, with Twitter nobody even seems to be trying. Weird.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact