I'm skeptical of a lot of this stuff, but I'm glad that people are working on it.

Technologies of the Decentralized Web Summit

Expose localhost to the Internet securely

Maybe useful someday. Demo, test, run lightweight services from a Mac mini under the end table.

"in China ... WeChat is the web"

Okay divide by two given this guy's motivation to pump Kik's valuation but still.

Less breathless very interesting look at Chinese mobile UI patterns.

Motherfucking Website. Write once, run everywhere.

Wonder what actual commercial sites try to hew close to this. A lot of the world is going the other way.

The graybeards are gazing into their palantirs, warning of doom to come. The architects are looking at how their plans for the agora are being subverted by the walls the wealthy and powerful are building. Protocols are being designed, spells are being cast, all to keep the web ungovernable. The efficacy of magic, however, requires collective belief. Do users want to live on a perpetual frontier, or will they prefer the safety of the walled gardens of mobile apps, Facebook, WeChat, and Line?

Brewster Kahle on creating a more secure, distributed web. Calling for a web that's distributed (storage, bandwidth, authentication), private and (this is new): versioned.

Maciej Ceglowski on making the Internet more human. The diagnosis is more precise than the prescription, but the diagnosis is valuable all by itself: clearly voicing our anxieties, showing how they arise from concrete developments pulling the net away from our values and aspirations.

It's not clear to me that the next wave of change is going to come from the last generation (which I count myself a part of.) But the web is going to stick around, and since the web is such a thick layer in the stack (it can be used to carry pure semantic data, or that plus presentation, or add application logic ... ) and is still the connector, we will probably never stop working to adapt it to changing capabilities, changing desires for what technologies have a home there and what we want to do there.

Camlistore is a set of open source formats, protocols, and software for modeling, storing, searching, sharing and synchronizing data in the post-PC era.

Basically, this is designed to be How Individuals Store Their Stuff. Forever.

Ambitious, but it's from Brad Fitzpatrick, so it might actually work.

Sandstorm.io

For Linux (dedicated box at home, EC2 instance, etc.) Currently mostly CMS, email and collaboration apps.

Running my own Flickr looks interesting.

Edit: see also Portal, which is also looking at interesting stuff like having apps use Camlistore for storage.

There's a decent definition of federation here, in the discuss of Gmail and AIM. Basically: people using service provider X can interact with people on service provider Y.

• email
• XMPP, eventually - although that didn't save it

If that definition is right, it only really applies to interactions that peer-to-peer, mediated by servers. But identity/ single sign-on is called federated as well, and is really just a peer-server-server interaction, so ...

I have some .htaccess rules that are supposed to prevent hotlinking images on the site. Unfortunately, they also (I think) broke images when my RSS feed was viewed in Google Reader. If I'm right, this should now be fixed. Let's see:

"a format for syndicating social activities around the web"

I had no idea this existed. Or maybe I just didn't care when I heard about it the first time. Supported by Facebook, Google, MySpace; not used by LinkedIn.

From the rationale:

And if I haven’t made it clear what I’m talking about, well, we’re starting with an assumption that activities (like the ones in Facebook’s newsfeed and that make up the bulk of FriendFeed’s content) are kind of like the synaptic electrical impulses that make social networking work. Consider that people probably read more Twitter content these days than they do conventional blog posts — if only because, with so much more content out there, we need more smaller bite-sized chunks of information in order to cope.

So starting there, we need to look at what it would take to recreate efficient and compelling interfaces for activity streams like we’re used to on FriendFeed and Facebook, but without the benefit of having ever seen any of the services before.

In other words, how can we make the user the center of gravity again, instead of a bunch of applications that want to own the user.

Exciting!

Since moving to San Francisco and talking to people about joining the tech industry, I've been asked several times "so, what would you find it exciting to work on?" When I answer "feeds - you know, RSS, Atom that kind of thing" I invariably get a blank look. Maybe if I develop my thoughts I'll do a better job communicating my enthusiasm.

I use a feed reader daily. I use it to keep up with blogs that I'm interested in, of course. I've written before about how it changes the definition of a high-value blog, from high-volume to good signal to noise ratio.

That's the canonical geek use case for feeds. It's not going to take the world by storm, even with simpler tools.

But there are other things I use feeds for. I use them for anything that I want to check regularly that has a feed available:

• My friends' Flickr photos (based on my Flickr contacts)
• Comments on my Flickr photos
• Local events (based on my last.fm profile)
• Comments on MetaFilter threads that I care about
• Trips my friends are taking (through Dopplr)

This is starting to look a lot like Facebook, right? And one way to look at Facebook is a feed reader in a social context. The design is nicer than MySpace's, but what really distinguishes Facebook is that central news feed. "Like" is a really simple method of subscribing to a feed, and becoming friends with someone subscribes you to multiple feeds about their activity.

Facebook's popularity means that hundreds of millions of people are looking at a feed reader every day. Many of the applications built on the platform are only possible because it's such a good interface for consuming feeds. Suffice to say I don't think Facebook is a fad that's going to die out. People's activity online is increasingly going to revolve around subscribing to, consuming and interacting with feeds.

But short of building a Facebook competitor, what are the opportunities for cool new applications? Watch this space.

Currently using Simplenote+Notational Velocity+Dropbox, thinking about switching to PlainText+Notational Velocity+Dropbox.

• Do people know what they're talking about when they say agile development?
• Academic papers on Hadoop and MapReduce
• What happens when you get drunk?

Sync THL between machines using Dropbox
It's been so long since I've had a reliable internet connection that I have no idea what they're talking about here. Other things that I have never used which confuse me: Spotify, Netflix VOD, Foursquare, etc. A lot has happened in the last 18 months! Or so I'm told.

Good tool for checking domain availability.

No need to install software for the one I'm using; just plug in a URL and go. Useful if you, say, want to watch I'm On A Boat feat. T-Pain in a country where it is "unavailable due to copyright restrictions." (I'm surprised Youtube video works, actually, since I'm guessing the Flash player isn't downloading video through the proxy.)

"We’re really excited to bring you forms! Create a form in a Google Docs spreadsheet and send it out to anyone with an email address. They won’t need to sign in, and they can respond directly from the email message or from an automatically generated web page. Creating the form is easy: start with a spreadsheet to get the form, or start by creating the form and you’ll get the spreadsheet automatically."

Offhand I can think of, oh, a zillion ways to use this.

boo - 5,000 row limit?

Guy from Elephant Disk is here, they're one of EC2+S3's biggest users. They only use it for non-real time stuff right now. He's terrified of stories about companies that use EC2 as their only web servers.

Issues

• EC2 instances have no persistent storage - need to code your own persistence solution.
• No static IPs - need some sort of dynamic DNS client.
• What happens if Amazon pulls the plug or changes something crucial? No SLA.
  • AWS is actually showing up in Amazon's financials. Helps assuage fears of it disappearing.

Companies built on EC2

• RenderRocket
• The AWS forums have lots of good examples and feedback
• Many more companies using just S3

Some companies have started to wrap services around AWS (e.g. RightScale) - check the partners page of AWS.

But: if you have something that parallelizes well, need scalability on demand, and can deal with the constraints, AWS is the shit. "AWS is the perfect startup environment; my credit card loves AWS."

Useful now that I'm setting up automated backups of this site.

"Ning is cool, but can I ditch the social networking crap?"

So as part of the same gig I've had since I moved to LA, I am now doing some work for a webcasting provider called Interactive Video Technologies. They recently rolled out a program called IVT Studio that's used to create pre-recorded video for on-demand streaming, usually incorporating a Powerpoint deck.

The software is polished and IVT clearly put a lot of effort into making it frictionless to use, which - yeah, I'm pushing product here, but I'm serious too - is pleasing to me. Makes it easier to get out of bed in the morning.

I don't think BlogFS actually works right now, but used with MacFUSE the possibilities are sort of amazing. Start with finding stuff from your blog posts (even private posts) via Spotlight.

e.g...what if BlogFS was really smart about OS X integration?

• What if copying a file to a BlogFS-mounted folder uploaded the file?
• What if creating a shortcut to a file in a BlogFS folder (depending on where you're accessing the blog from) ...
  • gave you the local file
  • ftp'd the file from its source

Update: Detailed look at MacFUSE + sshfs

Yeah really. Based on rural telco kickbacks.

Search with a live guide. This is my personal high water mark for Bubble II, VCs will give money to two guys and a pet rock. How can this ever scale?

Tons of possibilities: "text ALLNIGHT to 414141 to get directions to the party", or scavenger hunts ... geekfun.

The example is for budgeting, but I can imagine a lot of uses.

Finally someone tackles the central issue: where's the bottleneck? For what it's worth, I think there's too much money in getting bits to and from the user for this bottleneck to last. In the long run we'll all have multiple competing sources of connectivity.

However there's another point: if the bottleneck is in part an artifact of government-erected barriers to entry (rules on spectrum use, rights to lay fiber) than the government has a positive responsibility to ensure that the oligopoly it created is working for the public good. This pattern crops up all the time - eliminating rent control should increase supply, but not if cities prohibit developers from building low-income housing.

I'm never, ever going to use this, but that's some crazy shit.

"Today there is rough price parity between (1) one database access, (2) ten bytes of network traffic, (3) 100,000 instructions, (4) 10 bytes of disk storage, and (5) a megabyte of disk bandwidth. This has implications for how one structures Internet-scale distributed computing: one puts computing as close to the data as possible in order to avoid expensive network traffic."

Paper claims this situation is unlikely to change: historically, telecoms prices have not declined the same way chip prices have. (But I thought we had all this dark fiber, and it's the last mile that's the problem?)