How could Labour win the next general election?

A disclaimer: this is an issue I don't have real depth of knowledge on. I'd love to see Nick Robinson, Ian Hislop, Johnathan Dimbleby, Martha Kearney and a non-BBC commentator as a panel facing a public audience to thrash it out, absent any party lines and attention-seeking from politicians.

What could Labour do to win it, then? The public reviles politicians, particularly Labour, and particularly the incumbent leader. Even impressionables like me have finally got tired of Brown and the various expensive schemes like ID cards that have proven foolish.

Reading Ian Hislop's page on wikipedia, he has apparently expressed support for Vince Cable as potential treasurer, which made me wonder. What if the Lib Dems joined up with Labour? They could give Cable the treasurer spot, put in a smoother operator like David Milliband to run the show and set a fixed date 6 months away to have an election. This should give the country time to get used to the idea, push through some high-profile Lib Dem policies (needed to demonstrate that it wasn't just a gimmick) and get some battles fought against the Conservatives in the Commons. They would do well to field Cable nearly as often as Milliband in the Commons if the PMQs rules allow it, because although Cameron might rate higher in public minds than Milliband, Cable must thoroughly trounce Osborne.

They would need, inevitably, to keep a few less well-liked faces around - Mandelson for his astute manoeuvring, perhaps a real surprise like Blair as foreign secretary for his experience and respect beyond the borders.

Policy-wise, having the Lib Dems on board would be a fantastic excuse for rewriting the policy book - they could drop ID cards, 42 days, clarify Iraq and Afghanistan, commit to cutting spending (which was always inevitable, but now acceptable), and so on.

It would be an utterly catastrophic result for the Tories, whose position is currently so defined by New Labour that they would struggle to find a footing, perhaps long enough for some memorable moments like Michael Howard's Paxman interview. Taking a solid devolutionary line would pacify the Old Labour North with the prospect of more local rule, and the lower taxes would be a step squarely into the Tory sweet spot.

The only Conservative defense would be that they are merely rearranging the deckchairs on the Titanic, or the derision of having to resort to joining up with the Lib Dems. But the flaw here is that these are political arguments. The public cares about policy arguments at the moment - what will you do to the NHS? What will you do for the economy? How will I know you are a good politician, not a typical, dodgy one?

But they wouldn't do it, would they? Either they'd rather lose this election and pick up after a Tory term, or they are too shortsighted to try something really radical.

Real artists ship

"Real artists ship"
Attributed to Steve Jobs.

Could Markov chains be useful cryptanalysis?

The Enigma code-breaking relied on the discovery of common pairs of letters. Markov chains are a way of analysing common sequences of symbols. Could Markov chains be useful as a cryptanalysis tool?

Apple, tethering and ARPU

Girtby blogged on Apple allowing carriers to detect and block tethering, affronted that Apple would act so unfairly to its users, because the carriers are all 'Apple's bitches', right?

Unfortunately the carriers are not all Apple's bitches, as you put it. As per the normal business negotiation, each side brings their cards to the table. Apple had the iPhone, for which it wanted moolah. The carriers had moolah, for which they wanted the iPhone and as many restrictions they could get. For each restriction they will be willing to offer more money, so if Apple want to be nice to a small group of users that want to use tethering they have to be willing to give up some revenue for that.

In creating restrictions, carriers push a larger proportion of users into buying addons to their contracts or into using the expensive bit of the tariff (international calls by banning net phone apps, etc), so driving up their Average Revenue Per User. Remember, the better the iPhone, the more Apple can squeeze out of the carrier and the more the carrier has to raise the ARPU to get their profit margin back. So Apple leverage the hype they can build to get a better deal from the carrier. And the carrier know that if you're spending $40/m on a contract, you probably won't mind another $5 for tethering.

Yes, they are all there to take your money. That's Big Business at work. The terms of the deal to prevent tethering, to quietly throttle certain apps, were all written into the contracts at the negotiating table. This was a done deal a long time ago. As time goes on, and more apps and uses are prevented, the terms of that deal will be slowly revealed, clause by clause. Apple has not just turned on its loving fans, it has already sold them. Did you expect anything else?

Exchange

In everything there is an exchange of some problems for others, some freedom for others, some limitations for others.

So as you wait for the new thing - a record contract, a home of your own, a wonder-job - see afresh the freedoms you will be giving up, the joys you will lose to memory. By all means, grasp the new thing if you wish, but expect to grieve the old as you celebrate the new.

With new birth comes the death of the old, and the inevitability of more death. We invite it with every breath. But don't be depressed by it, be inspired! Your death will invite new life.

Life is minor and major, stop wishing for utopia and live it.

[Idea] Fie to bit.ly scammers: A public URL database

I've never understood why URL shortening has not become a part of the browser... surely it's easy enough to shorten a URL by compressing it, so the browser could see abc.ly/q2FFqta8ep and decompress it without a single http request.

Still, the world loves ease of use and hence we get the inevitable abuse of URL shortening to send people to spam/scam/phising sites.

So why not have a public URL database, where a people enter the shortened URL and the database either gives them the URL or tells them it is spam/scam etc? Crowd-source it - let people use browser extensions to report bad links.

Importantly, to avoid it just being yet another redirector, the service would need to provide public database dumps so that others could provide competing services, verify the data, etc. Vitally, ISPs could provide access or even use the list to warn the users about the content. Think of it a bit like the freedb or the imdb - public service databases.

It would allow the browser to look up every link on a page to find out where it ultimately points, and take the URL shortening services out of the loop entirely. It could even communicate via the minimal overheads of http error codes - 404 if the link is not known, 307 to provide the correct redirect (skipping advert pages, further pages of redirection), 303 to indicate a dodgy site and supply the URL of a relavant page (e.g. the Dodgy Scammy Warning page).

Come on people, one link gets read thousands of times, so if a small proportion are willing to mark dodgy links, that could save a lot of people a lot of headaches. And we could always use it to tag links, and vive la web semantique.

Take me back

Take me back
to simple.

Take me
and return
to humble and quiet faith.

To quiet and humble
sweet and waiting
patience and hoping
white.

Let my arrogance and force,
my pride and envy
drop from my shoulders
ease my burden.

Give me
gentle joy of being.
No more
fearful waiting.

Soft white curtains billow
in the sanctuary of my mind. Sought
and seeking
to escape frought
and fleeting.

Take me back
freedom.