Anthony Bailey

All Your Face Are Belong To Us

Tue, 31 Aug 2010 09:23:06 +0100

Well someone had to make the obvious joke re Facebook applying for a trademark on the term “face”, even if I am late as usual.

Surely You're Trolling, Mr Allen!

Sun, 29 Aug 2010 19:11:39 +0100

Reading about Paul Allen’s recent round of “…but on the web” e-commerce patent suits reminds me of one of the most objectionable aspects of the current patent system: the way that when a new technology or medium appears, rights to using it in combination with existing ideas seem to get assigned in a “first post!” manner.

It always makes me think of this excerpt from the “I Want My Dollar!” story Feynman tells in “Surely You’re Joking, Mr Feynman!”:

…during the war, at Los Alamos, there was a very nice fella in charge of the patent office for the government, named Captain Smith. Smith sent around a notice to everybody that said something like, “We in the patent office would like to patent every idea you have for the United States government, for which you are working now.[…]

I say to him, “That note you sent around: That’s kind of crazy to have us come in and tell you every idea. […] There are so many ideas about nuclear energy that are so perfectly obvious, that I’d be here all day telling you stuff.”

“LIKE WHAT?”

“Nothin’ to it!” I say. “Example: nuclear reactor under water; water goes in, steam goes out the other side, pshshshsht - it’s a submarine. Or: nuclear reactor; air comes rushing in the front, heated up by nuclear reaction, out the back it goes, Boom! Through the air - it’s an airplane. Or: nuclear reactor; you have hydrogen go through the thing, Zoom! - it’s a rocket. Or: nuclear reactor; only instead of using ordinary uranium, you use enriched uranium with beryllium oxide at high temperat ure to make it more efficient - it’s an electrical power plant. There’s a million ideas!” I said, as I went out the door.

Nothing happened.

About three months later, Smith calls me in the office and says, “Feynman, the submarine has already been taken. But the other three are yours.”

I am not a third party

Wed, 25 Aug 2010 21:43:35 +0100

You may be aware that because of its unsuitability as a mechanism to give third-party applications suitable access to a user’s account, Twitter are phasing out Basic Auth at the end of this month, supporting only OAuth from then on.

This is annoying for those of us who individually automate some aspect of our Twitter use - Dave Winer’s reaction is illustrative. I recently sighed and reluctantly updated the relevant bit of my homegrown blog admin. (I followed the same kind of approach I used recently for the Facebook graph API.)

The reason it irks is that although OAuth is surely appropriate in a third-party developer API, it is not so for a client API that allows an individual user to script their regular interactions with a web presence. OAuth is overly heavyweight. It makes assumptions that are not true. And it does not reflect the simple client usecase - a way to automate the things I can do by hand without resorting to robot-clicking and screen-scraping.

If some action or resource is available after a user logs in with a username and password, I think ideally the same action or resource should be available for scripting through HTTPS + Basic Auth with the same credentials - surely that is exactly what this authentication mechanism was intended for? I am the first party, not a third.

Headius: My Thoughts on Oracle v Google

Mon, 16 Aug 2010 23:45:57 +0100

Headius: My Thoughts on Oracle v Google:

A great piece by Charles Nutter laying out the context and content of Oracle’s patent claims against Android.

Recommended Listening: CometD and Push Technology Greg Wilkins...

Wed, 11 Aug 2010 13:32:00 +0100

Recommended Listening: CometD and Push Technology

Greg Wilkins talks about Comet (that “long-poll” AJAX approach to approximate two-way communication between client browser and web server) in this episode (48min) of Phil Windley’s Technometria.

As well as introducing the technique he offers opinions on how to use it both clientside and serverside, and how other technologies such as websockets and SPDY fit in.

I felt vaguely ignorant of this stuff before listening to the interview; afterward I feel vaguely informed (albeit by an interested party who maintains the CometD library, and Jetty.)

Re-opening Portal (my favorite custom maps)

Sun, 11 Jul 2010 19:38:00 +0100

A few co-workers recently succumbed to peer pressure and got around to playing Portal. There has been much quoting of GLaDOS and xkcd recently as a result.

I loved playing the game and many user-created maps. So for those who have tackled the Advanced Chambers and still want more, I’ve put together a list of my favorites.

(I won’t link directly; you will probably find them all on both My Aperture Labs and Thinking With Portals but I’m not confident about either site’s permalinks.)

My tastes in brief: I tend to like puzzles and neat uses of physics; I’m really not very skillful and if much dexterity is required then I want reassurance that I’m not wasting my time.

I’ve already previously praised the huge, wonderful conversion Portal: The Flash Version. Others that suit my tastes, in vaguelly descending order:

*UPDATE* Marf Aperture by MasterMarf (may be the best one yet!)
LoveGLaDos by Graxthal (gated right at the start - see my note on crouching, below)
Weighted Companion Cube Map by MrCow
Rainbow Palace by Duffedwaft (has one awkward twitch-turn when falling to the first cube)
Spheres of Influence by Chris Cummings (take the turrets bowling, take them bowling)
Portal HPC by Omega Studios
Infrared by Omnicoder (on-the-edge of needing too skill for me to enjoy it: Omega by the same creator went over that edge)
Tasty Treats by Duque-Rois
Conundrum by AmnDragon

You could also check these out (especially the first) for some unusual features:

Chamber23 by Tefkor17 (features an amazing Rube Goldberg contraption)
Maybe Black Mesa by Sean M (nice Half-Life decor)
Companion Cube Rescue by SpikeX (explores really huge distances and big turret fights)

Please note that there’s one principle that the original game never trains you on but that custom maps sometimes assume you know: crouching as you go through a Portal - so that more of you is closer to its center - increases how much momentum you “conserve”.

(So that you know if you’ll like what I like, I should explain in more detail what I want from a game that requires both thinking and acting. Working out a clever solution is fun. Mastering a difficult gameplay sequence is tolerable. But I really don’t like having to do both at once, especially when I’m not sure whether I’ve got the wrong idea altogether, or am just too crap to execute it. I recently enjoyed playing Braid, but I did find it suffered from this quite a lot.

Also there are plenty of maps that simply defeat my meagre skills and patience. For example, I suspect that the Logic Portals series is excellent: it takes training seriously and it made a determined affort to teach me how to step fling, but I just found it too tricky. Anything involving twitch-reflex portal placement is no good for me. And I found Portal: Prelude way too hard.)

Prince pun

Wed, 07 Jul 2010 10:25:39 +0100

My late joke re the Prince thing: though still mostly downloaded, he’s now the artist formally known through Print

PeepCode: Rethinking Rails 3 Controllers and Routes

Thu, 03 Jun 2010 19:50:43 +0100

PeepCode: Rethinking Rails 3 Controllers and Routes:

The scales fell from my eyes as I read Geoffrey Grosenbach’s post on how Rails’ restful route to controller action translation has become a pointless, wasteful abstraction.

Exporting your Facebook graph

Mon, 10 May 2010 00:03:35 +0100

I’m not a real Facebook user. I have a “ghost” account on there that puts my tumblelog on the wall, and redirects people to my real home outside the Facebook walls. The ghost also acts as my doorway into searching the Facebook network, and helping my partner with the Miss Fitz-Poste’s Modern Mixers fan page.

One of the things that has kept me from spending time there is the lock-in factor. Although I can write my own web apps to publish content, I much prefer to use services provided by others - but only if I have the ability to migrate that content elsewhere. As an experiment, I tried using the new Facebook graph API to export data from their site.

Happily, it’s pretty easy to do this. Although the pseudo-privacy controls means you have to jump through some OAuth hoops, you can do it all from your browser.

Register a Facebook app. OK, this sounds like a big deal, but you just have to provide any old name, and a site URL (later known as the App URL.) The key thing is that this does not need to be a URI that addresses anything real. For my first test I used an endpoint underneath anthonybailey.net, but subsequent trials showed you could use a non-existent domain, or localhost - anything you like so long as you don’t mind that your browser is later going to fire it a single request containing a one-time secret code.
Negotiate a captcha. Facebook registers your app and assigns you an App ID and an App Secret.
Log into your Facebook account so that your cookies are all properly baked.
Visit https://graph.facebook.com/oauth/authorize?client_id=&redirect_uri=/oauth
Facebook will ask you to grant access to your account to your app. Make it so.
You will now be redirected to /oauth?code= The fetch will fail, because the endpoint doesn’t exist, but your browser should still show you the address, containing the code you need.
Visit https://graph.facebook.com/oauth/access_token?client_id=&client_secret=&redirect_uri=/oauth&code=
Facebook will return a body with content definining an access_token and a time in which said token expires. (From what I’ve seen the latter is a duration in seconds between one and two hours. You can get a fresh access token by returning to step 4.)

Now you can go explore your personal graph of JSON objects, and those of any other areas of Facebook your account has access to. You can see what relationships there are to follow by starting at https://graph.facebook.com/me?metadata=1&access_token=<...>

Clearly once you’ve registered the app, a little script ought to be able to perform the brief dance that gets the access token so long as it can supply your Facebook account cookies. I haven’t bothered to automate it myself because I’m still not living my life in Facebook.

(You might ask why, given that I now seem to have an export mechanism? OK, I oversimplified earlier in order to get to the useful bit more quickly. Actually, for an identity/social service like Facebook, my preferences are somewhat more complicated.

On the plus side, I could live with the most transient or socially intertwined contextual Facebook content (quick comments on other’s stuff) being hard to export cleanly and effectively. However, I really want the most core information about me - contact details and bio - to be served directly from a resource addressed under a domain that I own. It would be OK to have a third-party host that information behind the scenes, or to explicitly import it from my primary resource, but I don’t want an address under Facebook’s control to be authoritative for my identity, or even to imply that it was.)

This is not an XKCD strip

Fri, 30 Apr 2010 14:04:02 +0100

Team: *lunchtime discussion about the differences between company "director" and "acting director" roles*
Andy: ...and we would call the other one "director prime". Right, this has got too nerdy, end conversation.
Anthony: But we were only one pun away from Star Trek... do you think there's a Godwin's Law variant specific to nerd talk? Some topic whose probability of being discussed gradually approaches 1?
Andy: This topic.

A fair few people mislike Facebook’s reigniting their...

Wed, 28 Apr 2010 21:04:00 +0100

A fair few people mislike Facebook’s reigniting their beacon and tracking your social activity on other sites via their new “Like” button.

The privacy concerns don’t affect me very much since I’m never logged into Facebook. I do find it a bit annoying that e.g. the Lovefilm site is now wasting space trying to get me to sign up to a social network rather than, say, helping me find stuff to watch, but whatever.

It did occur to me, though, that those who do care a lot could create/install a simple browser add-on that amends the buttons in place so that they do something like demand.openid.net when they are pressed - compile a per-website record of users support so as to allow a collective petitioning of those sites to drop the feature.

Eight bit Doctor Horrible adaptation. Wonderflonius!

Thu, 08 Apr 2010 20:09:28 +0100

Eight bit Doctor Horrible adaptation. Wonderflonius!

Thoughts on SRC 2010… Last week-end was the third Scottish...

Tue, 30 Mar 2010 22:05:52 +0100

Thoughts on SRC 2010…

Last week-end was the third Scottish Ruby Conference (well, technically the previous two were Scotland On Rails Conferences.) It continues to grow - three hundred attendees and three presentation tracks. Purely selfishly I think I preferred it just a little smaller and tighter, but it has scaled very well and I still enjoyed plenty of good talks and chats, so the total value from the event must be way up. Congratulations to organizers Paul, Alan, Graeme et al.

Overly abbreviated highlights: Tim Bray’s keynote on where Ruby isn’t (concurrency, enterprise and mobile) was provocative and very popular. My favorite talk was New Bamboo on how to do speed-run coding; amusingly presented, with a graph-backed economic motivation followed by dozens of juicy details. There was also an enlightening fight over the value or otherwise of mocks, about which I must not forget to blog in more detail. Finally I ditched the Saturday wrap-up party in favor of sleep before a coding retreat on the Sunday - I’ve always loved shorter gregarious kata sessions, and a full day did not disappoint.

Enterprise Java

Wed, 10 Mar 2010 15:22:30 +0000

class $$$
{
    public static $$$ $$$$( $$$ $$$ )
    {
        return $$$;
    }
}

http://www.w3.org/TR/html5-diff/

Fri, 05 Mar 2010 23:35:05 +0000

http://www.w3.org/TR/html5-diff/:

This document makes me happy.

Brian Swan gave a strongly anti-mock talk as part of the most...

Fri, 26 Feb 2010 01:15:24 +0000

Brian Swan gave a strongly anti-mock talk as part of the most recent ScotRUG. It was pretty much point-for-point how I feel on the subject (apart from the bit at the start on “no getter” coding, with which I have had little contact.)

Buzzing noise - irritating but deliberate?

Mon, 15 Feb 2010 19:58:26 +0000

There seems to be a blog consensus that Google screwed up the launch of Buzz and that even the idea there could be privacy concerns caught them by surprise.

Well, maybe. Certainly they could have done better, and lessened the worst kind of damage.

But I speculate they knew the auto-follow would cause disquiet, and did it anyway. Every previous Google attempt at social networks (Orkut, FriendConnect) had mostly failed to catch on due to network effects. To be useful, you have to be large. Auto-follow gets you the critical mass you require very quickly.

(Myself, I’m waiting for Buzz to be enabled for non-Gmail Google accounts. I have a “public by default” attitude and a Google profile linked to a Reader account and associated with a homepage that exposes tons of contacts information - so auto-follow will suit me very well once they offer a Buzz detached from the Gmail interface.)

Continuity is a great little Flash game. It’s an original...

Wed, 03 Feb 2010 20:34:37 +0000

Continuity is a great little Flash game. It’s an original hybrid of platformer and sliding block puzzle. Playing through it cost me an hour or so of sleep last night.

https://panopticlick.eff.org/

Thu, 28 Jan 2010 00:48:47 +0000

https://panopticlick.eff.org/:

The EFF have been looking into how many bits of identifying information your browser provides to the sites you visit. They suspect: very many. They have an experiment called Panopticlick running to test this out for real. It’s fascinating. As an example, my own fingerprint was unique amongst the 50K they’d sampled so far.

Uncle Bob: mocking mocking and testing outcomes

Mon, 25 Jan 2010 10:46:13 +0000

Uncle Bob: mocking mocking and testing outcomes:

Bob Martin wrote a great article on the overuse of mocking frameworks.

His position is that whilst they are sometimes useful, they shouldn’t be the hammer with which you hit every nail - which is how they are often presented.

He advocates hand-rolled mocks for the simple cases, and realizing that when this gets painful, you have a code smell re the coupling of the classes in the system under test.