Hey! Windows Party (tm)

We're all aware by now that Microsoft have pushed the boat out with an instructional video, designed to allow sad-cases like ourselves to engage in small-talk and nibbles.

An undeniably useful resource for any IT professional who would otherwise routinely fall short of the standard achieved and expected by all you normal folks out there.


Best of luck with your efforts, but don't forget to adhere to your EULA:

MICROSOFT SOCIAL EVENT LICENSE TERMS

MICROSOFT WINDOWS 7 LAUNCH PARTY

These license terms are an agreement between Microsoft Corporation (or based on where you live, one of its affiliates) and you. Please read them. They apply to the event named above, which includes any invitation you received and any resultant stains, damage, tearing, traffic cones, makeshift togas or unexpected intercourse. As if any of you sad bastards would get close enough.

By attending this event, you accept these terms. If you do not accept them, do not attend this event. Instead, return to the beginning. You know. When you were 16 and you decided this was interesting. Before the initial thrill of creation was overtaken by a bleak unending round of frustration and social exclusion; when you were young and interested in PEOPLE. When you glanced across at the pretty brown-haired girl with the freckles and she smiled back. Before you started routinely sucking in your gut and trying to hide the soup-stains on your shapeless brown chinos and navy-blue-smart-casual polo shirt. Remember? Oh Jesus Christ. REMEMBER?

You cannot obtain a refund, but feel free to contact Microsoft or the Microsoft affiliate serving your country for information about trying to get one anyway. Our Marketing and Sales Teams can always do with the boost afforded by listening to your sad little problem in your sad little life. Just before they piss off early to that charming beer garden to flirt and laugh in the sunshine.

If you comply with these license terms, you have the "rights" below for each license you acquire.

1. OVERVIEW. These license terms permit organisation and attendance of one event at one location. No events at any other time or location are allowed. Or, let's be honest, expected.

2. INSTALLATION AND USE RIGHTS. Before you attend the event under a license, you must assign that license to one pokey bedsit or parents attic room.

3. ADDITIONAL LICENSING REQUIREMENTS AND/OR USE RIGHTS.

a. Trial and Conversion. The party is licensed on a trial basis. Your rights to attend the party are limited to the trial period. The length of the trial period is set forth during the activation process, or until 9:30 when no-one has turned up or 10:30 when everyone has gone home.

b. Media Elements and Templates. You may copy and use images, clip art, animations, sounds, music, shapes, video clips and templates provided for the event. If you wish to use these media elements or templates for any other purpose, seek help. Seriously. And what's that smell?

4. SCOPE OF EVENT. The event is licensed, not sold. This agreement only gives you some rights to attend the event, in some very limited ways. Microsoft reserves all other rights.

You may not

· work around any social awkwardness or lack of humanity in the event

· schedule any further event without express permission (No, that's not irony. We mean it.)

5. BACKUP PLANS. You may make one backup plan in the event that your event is a non-event. They're running Buffy from the start again, for instance. Or there's that Summer Glau fan-fic you squirreled behind the skirting board, under the bed.

6. DOCUMENTATION. Eh? What did you plan on writing about this exactly? And who would you tell it to? You don't have any friends and you won't be bearing any progeny. You know it. We know it. So go on, knock yourself out. But if you do, we own it, right?

************************************************************************************

LIMITED WARRANTY

A. LIMITED WARRANTY. If you follow the instructions, the event will perform substantially as described in the Microsoft materials that you receive in or with the event. And, believe it or not, that's the best case scenario

B. TERM OF WARRANTY We'd like to take a bit of time to warn you that you are not covered for hangovers, excessive mirth, lovebites, cat-fights or unwanted pregnancy. Safe ground there we think. You are not covered for despair, loneliness, isolation or the feeling that it could all have been so, so different. We're not idiots.

iPlayer, iPhone, iGiveIn

Short and a bit unfocussed this time, still.

For some time I've resisted buying / upgrading to an iPhone.

Previous resistance to an iPhone is a product of: the unbearable smugness of existing owners; my capacity for ruining anything vaguely stylish with oafish behaviour; a general feeling that I want to support something more open.

Anyway. Capitulation. I've got one.

It's still in perfect working order and, yes, I'm getting smugger ("Haven't you got an electronic compass in the old 3G then? No? Oh, ever so handy.")

Lot to report at some point, but mostly I'm very impressed with the mobile iPlayer / iPhone combination, particularly when it comes to the God of all media, Radio.

Access to listen-again material is superb but live streams aren't yet supported (come on, come on, I'll pay).

However, until that glorious day, everyone should purchase a copy of WunderRadio from the App Store. This allows you to listen to a couple of flavours of live BBC streams as well as thousands of others. Well worth it's paltry £4ish purchase price.


Two other small criticisms as far as the iPlayer is concerned:

1) I've signed up for Labs features, but there didn't seem to be a way to easily detect if what I was looking at was Beta-ish or not. This makes it a bit of a chore to decide whether or not to feedback.

2) Please, please, support playlists of some kind. I can already see what I've recently watched / listened to, but to be honest that isn't much use.

What I really want is a way to quickly add any program that takes my fancy to a queue, and then to be able to work my way through that queue in the iPlaeyer itself. So a button somewhere on a program says "Queue This" and it's added to the end. And playing something from the queue removes it.


That way I don't miss anything I could fancy a look at in passing on my way to something else, and I can probably stop watching late night episodes of "Make Me a Supermodel" and "America's Hardest Pets Do the Funniest Things". Thanks.

PSI, One Million Pounds and the Romance of Moths

Not a usual choice of subject, but stimulated by a typically heated discussion in another place. Browse by all means, but be warned that there's a fair bit of ill-tempered, foul mouthed, heated discussion in which few people come out of it well, even if they've convinced themselves they do, myself very much included.

To précis, the discussion revolves around whether or not the James Randi Million Dollar Challenge is a fair way to allow claims of supernatural powers to be tested, and consequently whether we can consider "PSI" to be real.

To simplify, in this context, I use "PSI" as shorthand to mean "Can I communicate over a distance without using the usual complement of 5 senses".

Cards on the table first: I don't believe any human can communicate in this way, or that animals do. I don't believe in telepathy, precognition, telekinesis, clairvoyants, mediums, homoeopathy, dowsing or God.

None of which in any way ruins the unexpected pleasure at a phone call from an old friend I may have been thinking about, or a chance meeting with a stranger sharing an acquaintance or seemingly eclectic set of interests. Just make sure they're recognised for what they are; pure coincidence but no less a source of unalloyed delight.

"You can't tell me there isn't something to it"

For years my brothers large, fierce dog would spring up and bark loudly at strangers approaching and knocking on the front door. But if I approached and knocked, he never did. I couldn't be seen approaching; I don't ape Dicky Rice's whistle**; I don't drive an ice-cream van with a customised chime; the ground doesn't tremble to the unique signature of my footfalls.

What he did, of course, was smell me. Through locked doors and intervening rooms, and down the garden path. He knew I was on my way for that reason before any coarse human sense had a clue, so by the time everyone heard me banging to get in, he knew he could stand and wag his tail and get his ears scratched without any drama.

Oddly, when someone pointed this out to me, far from being disappointed that some mysterious other sense wasn't involved, I find it almost thrillingly miraculous. Achieving this using a sense that humans apparently blunder around with by comparison seems far more unbelievable than pressing some nebulous and unshared dog-sense into service.

(Compare with the Emperor Moth. Able to detect a female's scent from 6-7 miles away - a long and possibly fruitless flap ahead before he has a shot at any action; spare a thought for him and his unused moth knackers. And spare a thought for her by the looks of things. No wonder she keeps a distance.)

So my position is that the sort of PSI decribed here doesn't exist, but what does exist very clearly makes up for it.

Yet I'd still maintain that the existence of PSI is scientific fact.

"Say what now?"

There's a superb exchange I can only half remember and relate (any pointers would be useful). Two protagonists in a film, one a believer one a sceptic:

Believer: "So do you believe in mind-to-mind communication?"

Sceptic: "Absolutely."

Believer: "Really?!"

Sceptic: "Yes. But it doesn't always work. One makes the right sounds and gestures, but of course sometimes the message doesn't get through".


So Question 1 is: "What message?"

Believers in PSI rarely claim very specific powers to communicate detail. No shopping lists or tips on Sudoku. They are usually broader in terms of feelings and intuition, and nothing intrinsically wrong with that.

The barking dog is a good example. His senses told him someone friendly was on the way and he responded accordingly. The very basic message my scent got across was "someone friendly is nearby" and the dog filled in all the blanks.

We shouldn't underestimate the importance of this: however clumsy it may seem to us by comparison to texting "STK ON BUS WIL B L8", it confers an enormous evolutionary advantage, especially if your bugle is sensitive enough and everything around you hums with information.

So the message I'd personally find compelling and useful is that someone's feeling happy / sad / friendly / antagonistic / comfortable / in distress.

"Is that the blower?"

Question 2 is - what equipment do we need to communicate? At it's most basic level we need a transmitter and receiver and a power source for both.

Supporters of PSI say "we already have all three - the brain".

Sceptics say "the brain has nothing that can do the job"

I say they're both right (and wrong).

Transmission of information to and from the brain without any intervening skin, mouth, eyes, ears, nose or tongue is already scientific fact.

Given the right setup, a SQUID is a mind-reader. The important thing about this at the moment is not that you need a huge expensive setup to detect brain activity, but that the brain quite evidently does broadcast what it's up to.

It does this in a way that doesn't involve waving your arms or flapping your gums, but by exquisitely modulating the magnetic field generated by clusters of neurons firing. That this is considered something of a waste product of the brain's activity is of no consequence.

Certainly that broadcast needs significant resource to detect and measure, and interpreting the content is a fledgling discipline, but the basic criterion for transmission is met.

So we can drag a signal out, and it turns out that we can follow roughly analogous means to ensure reception.

Transcranial Magnetic Stimulation (TMS) is actively used to alter the brain's processes, once again without "resorting" to the standard means for getting a point across. There is controversy as to how useful a tool it is, however few people deny that it's a cause, and it has an effect. Determining whether that's currently of any use or we're just randomly poking about in there is the subject of ongoing debate.

So to summarise, we have:

• An expensive, fragile means of reading the activity of the brain in a way that we only begin to think we can interpret
• An expensive, fragile way of affecting the behaviour of the brain where we don't have much of a clue what we're telling it.

Technologically though, I'd maintain that on this basis PSI is real.

We can easily postulate a gedankenexperiment where one person wanders around all day like a cyborg Marge Simpson, and another with a pair of giant, unfashionable and eccentrically fitted spectacles.

Through the miracle of the internet and suitably vast processing power, the mood of one may be communicated directly to the other. None of the standard senses come into play, it's genuine mind-to-mind communication. In the fullness of time, experience and expertise, you could extrapolate to see a charming necklace performing the same function.

For me, that would qualify as PSI with some assistance. But it can hardly be regarded as a victory for believers.

The means of transmission, reception and the medium are all there, but you're faced with crushing problems of interference from other sources, with orders of magnitude more power, and formulation and interpretation of the message itself.

No one could say that there isn't the possibility that one persons brain activity is picked up and in some way influences that of another unaided, but this is vanishingly small.

And never demonstrated in any replicable way.

** Dicky Rice was/is a council electrician with a legendary whistle so tuneless and piercing that he was, in all seriousness, more than once threatened with a good leathering if he didn't shut up. We can only aspire to be like him.

Solving the Gmail "Sender" Problem...

...if you have enough patience and control.

There are some long standing threads on the Gmail Help Discussion group concerning the problem with unwanted "Sender:" headers. A lot of very strong opinion and with fair reason.

Google promises something will be done, but there's no timescale.

For those of you who don't want to wait, and who are willing to flex a bit, there is a possible alternative.

A Brief Recap

You use Google Applications For Your Domain (GAFYD).

You let it manage your domain, www.onemanband.co.uk, and you have the email address mrsean2k@onemanband.co.uk

You work for or on behalf of another company, www.muchbigger.co.uk

You want to "brand" all your email as mrsean2k@muchbigger.co.uk.

Much Bigger manages it's own email with it's own email servers but forwards any incoming email on to mrsean2k@onemanband.co.uk

You setup an email alias in GAFYD to allow your email to appear to come from mrsean2k@muchbigger.co.uk

GMail sets the "Sender:" header with a value of mrsean2k@onemanband.co.uk. Some clients will render this and the illusion is destroyed.

So if this isn't pretty close to your situation, you'll want to find something better to do with your time.

What Do I Need to Know?

First, this is a compromise:

You can't have mrsean2k@muchbigger.co.uk, you have to make do with mrsean2k@m2.muchbigger.co.uk or your subdomain of choice.

Second, you need a friendly admin:

You need access to the DNS records for muchbigger.co.uk, or a friendly face at Much Bigger willing to access them for you.


If both these conditions are acceptable:

Create an A record for the subdomain m2.muchbigger.co.uk. It doesn't matter what this A record resolves to, but you'll need to go through the Google verification process to prove that you control it.

This may leave you with a further subdomain, googlesrandomid.m2.muchbigger.co.uk depending on how you elect to verify.

Follow the instructions from the Gmail dashboard and set up the full complement of MX records for the new subdomain.

Wait for the changes to filter through.

In "Domain Setting" -> "Domain Names" add a domain alias for m2.muchbigger.co.uk

That should be it.

You should now find that incoming mail is delivered directly to your GAFYD account, and if you respond using the mrsean2k@m2.muchbigger.co.uk identity, no Sender: header is sent - the email appears to originate seamlessly from Much Bigger.

Example DNS Entries

In trying this workaround, I implemented these steps using a sacrificial domain I own. These are the DNS entries I ended up with on my registrar of choice, www.register.it with some identities changed to protect the innocent.

Entries for the m2 subdomain below are typical of the changes you'd need Much Bigger to make on your behalf:

Name	Type	Value
muchbigger.co.uk	MX 10	mail.register.it
m2.muchbigger.co.uk	MX 10	ASPMX.L.GOOGLE.COM
m2.muchbigger.co.uk	MX 20	ALT1.ASPMX.L.GOOGLE.COM
m2.muchbigger.co.uk	MX 20	ALT2.ASPMX.L.GOOGLE.COM
m2.muchbigger.co.uk	MX 30	ASPMX4.GOOGLEMAIL.COM
m2.muchbigger.co.uk	MX 30	ASPMX3.GOOGLEMAIL.COM
m2.muchbigger.co.uk	MX 30	ASPMX2.GOOGLEMAIL.COM
m2.muchbigger.co.uk	MX 30	ASPMX5.GOOGLEMAIL.COM
m2.muchbigger.co.uk	A	195.110.124.188
to-windows.muchbigger.co.uk	A	195.110.125.55
muchbigger.co.uk	A	195.110.124.188
to-linux.muchbigger.co.uk	A	195.110.124.133
www.muchbigger.co.uk	CNAME	muchbigger.co.uk
ftp.muchbigger.co.uk	CNAME	muchbigger.co.uk
pop.muchbigger.co.uk	CNAME	mail.register.it
google7cd8065b1aa54390.m2.muchbigger.co.uk	CNAME	google.com

Conclusion

If you can live with the limitations, sending email with these setting eliminates any clues that you're using an alias; the From: header is the only clue, and this is set correctly.

Don't Put Google into Google or You'll Break the Internet

At least according to the IT Crowd.

Frustrated by some of the limitations in search capabilities supported by my otherwise excellent email provider Fastmail, I've signed up for the paid version of Google Applications For Your Domain (GAFYD).

It's too early to call whether this is a permanent state of affairs and whether or not it will last past the free trial period. I really want it to work, but although there are touches of genius in it for your 25 notes a year, there are some bizarre limitations and omissions.

Rather than go into detail at the moment, I'll list these as dispassionately as possible - some of them may be show-stoppers for you, and it makes sense to know about them before investing a lot of time.

Also bear in mind this state of affairs may change, or that I may just have been too stupid or too impatient:

Q Can I transfer my old Google identity across to GAFYD?

A No. Not if you're talking about flicking a switch and having your data, logins etc. transferred lock-stock and barrel from your old free account to your new paid one. You can approximate the exercise, but it isn't the experience you might expect it to be.

Q Ok, can I transfer my calendars across from my personal Google Calendar to the GAFYD version?

A No. You can get close though. You can invite yourself to view your own calendars and give yourself full permission, but the calendar is still "owned" by your old account.

None of the colour choices and other settings you already have will be copied across either, and many of the public calendars and other utilities that are available in the personal version seem to be absent in the GAFYD version.

Q Can I transfer my Google mail across then?

A Yes and No. You can't just appropriate your old inbox etc., but there is a migration tool, and you can forward mail that was destined for your old account pretty much seamlessly, or grab it with POP, optionally leaving it in the original account.

You'll also need to manually export / import your contacts from one to the other.

Q Can I access Google Reader, Google Groups?

A Not from your GAFYD account, no. You still need to maintain a free personal version for that lot. GAFYD is skewed heavily towards office use, so you get email, calendar and document functions. Even then they are a relatively pared down and business like version: fewer themes, less customisation, fewer gadgets.

In short, check that GAFYD supports the bells and whistles you like before committing.

Q Can I seamlessly manage multiple personalities in my email as Fastmail allows?

A No you bloody can't. In this respect GAFYD trails Fastmail by quite a distance. For one thing, Google always sets the "Sender:" header, so your efforts to communicate from mrsean@megatron-industries.com will be for naught when the client at the other end reads the header and reveals the awful truth to your correspondent.

In fairness this can avoid problems with your email being classed as spam, and Google are committed to allowing this behaviour to be optional "soon".


More later but, in summary, check that no-one is complaining that your favourite feature is missing or poorly implemented before putting the effort in.

Filthy Bandwidth Thievery: Avoiding Disaster by Diversity

My broadband supplier of choice, www.adsl24.co.uk via entanet, has dropped me off the map again. I'd have to say they are generally excellent with a no-nonsense approach to bandwidth and port-blocking (there isn't any), and I've been kept well informed.

Unfortunately knowing that they know something's up and that they're trying to fix it earns 8/10 for effort but 1/10 for usefulness. It's great to know a cable's been sliced but not as good as never needing to hear that in the first place.

It's unreasonable to expect them to be able to control some faults, particularly as the evidence so far suggests that there's been a renaissance in the use of blind scythe-wielders for digging up roads in the Sheffield area.

Although I can get a decent connection by borrowing next doors wifi, which remains unaffected, the VPN I use to connect to work requires a known, fixed IP address. No dyndns or similar will do, and no arguments about it.

So I'm thinking of offering to pick up the bill for my neighbours broadband connection as long as I can specify one with a fixed IP and use it on a slightly more formal basis if I'm in this position again.

The question is: which provider?

Unfortunately leaving them with the current incumbent, AOL, isn't an option as a) AOL don't offer fixed IP addresses under any circumstances and b) AOL have been arses in every conversation I've had with them to get this information.

What I'm interested in getting is a reliable provider that has as little in common with my own as possible, i.e. the maximum possible diversity in the equipment, software and physical route.

This would mean I stand a decent chance of dodging the bullet if there's a planned or inadvertent outage in the offing.

I'd anticipate quite a bit of Googling before settling on one, but if anyone has any suggestions how I can determine something like this, I'd be glad to hear them.

Bashcasting Your Playlist

After some reflection, a bit of swearing at the Bash reference, which may be canonical but is relatively unhelpful, an easy solution for Ubuntu users to tweet their tune of choice. None-Linux users are out of luck, sorry but you asked for it.

The prerequisites:

Amarok. I'm using Version 1.4, largely because a) it came installed and b) there's a different scripting / plugin system for 2.0

A command-line twitter client such as Twidge. This allows you to tweet a text fragment from a bash script.

Next a script that can detect which track you're currently playing, store that information to disk, and then tweet the results courtesy of twidge. After a bit of messing around, and bearing in mind I only want to tweet changes once-in-a-while rather than a note-by-note account, I settle on this:

#!/bin/bash

oldnotification=""
notification=""

while [ 1 ]
do

  #
  # Use dcop to ask Amarok what it's up to
  #
  artist=$(dcop amarok player artist)
  title=$(dcop amarok player title)

  notification="$artist - $title"

  #
  # This loop runs continuoulsy, so only leap
  # into action when there's a change to report.
  #
  # If there's no change, sleep for 10 secs. This means you may
  # miss changes to tracks that are < 10 secs, but also that you
  # don't mess around with checking state etc.
  #
  if [ "$notification" = "$oldnotification" ]; then
    sleep 10
    continue
  fi

  #
  # Keep a note to suppress duplicates when playing
  #
  oldnotification="${notification}"

  if [ "$artist" != "" ]; then

    #
    # Store the notification in a file.
    # This would allow more than one app. to use it if necessary
    # In an attempt at consistency, all tweets from Amarok are prefixed
    # with {rok} to allow filtering etc. if needed.
    #
    echo "{rok} $notification" > ~/.tweetsig

    #
    # Tweet it to the world.
    # In my case I don't want to flood with too many messages
    # so I give it 10  minutes before trying again
    #
    twidge update < ~/.tweetsig

    sleep 600
  fi          

done

Pretty self-explanatory. Change the "sleep 600" or remove it if you want to keep your audience more thoroughly up to date.

To turn this into an Amarok script, all that's necessary is to tar it up according to a naming convention:

tar cvf NowPlayingTweet.amarokscript.tar NowPlayingTweet

Then in Amarok, use the Script Manager to browse to the newly tar'ed script, install it, and run it.

Not perfect, but suited to my purpose. With a bit more time, perhaps an environment variable for the interval I tweet what's playing, and I'd prefer it only to tweet those I've finished listening to, not stuff I decide to skip past, but that'll do for now.

Who cares what music I listen to anyway?

Or: How To Stop Spamming Twitter With Your Playlist.

Although the NowPlayingTweet script / Twidge / Amarok combo basically works well enough, there are a couple of issues that mean I'm leaving it off just now, namely the...

Easy Technical Issue

There are a couple of glitches that mean the odd tweet is duplicated, appears blank etc. Scanning the code snippet I pinched, this just looks like the way a change in status (i.e. track changing) is detected fires too often. This manifests itself as duplicate tweets, the occasional contentless tweet etc.

This is readily addressed by brushing up on me Bash scripting and sorting it out and is of no further concern here.

However that still leaves the...


Slightly Harder Social Issue

If I sit and listen to music all day (which I often do) that's around 15 songs an hour, for a total of around 120 tweets per day.

This seems like a lot to me, in fact a potentially very irritating number for anyone following. Hard to know. So I want to strike a balance between spamming twitter with junk and tweeting the odd message that might have someone say "Ah! I could stand a bit of Burt Bacharach myself" or "He'll be cutting his wrists if he's not careful"


The two ways I thought this could be achieved are:

Tweet on change of Album or Artist (or either or both)

I listen to a lot of complete albums, particularly as an aid to concentration when I work. This would mean one tweet every 30 mins or so which seems reasonable.

The downsides to this are that it squashes a bit of the serendipity of seeing a song-title you've long forgotten or are intrigued by and, if you change on Artist, compilation albums still produce a flurry of tweets (a chorus?)

Tweet every N minutes

So take an interval of, say, 10 / 15 minutes and tweet all of the Artist / Album / Track info. This seems like a better option, if a bit more complicated to script. The number of tweets comes down to something passable, but you still get the serendipitous effect that tweeting on change of Album lacks.



The second gets my vote - and my time when it's available - unless any other inspiration strikes.

Tweeting Blog Posts

In the sort of horrible electronic incest deservedly lampooned in Viz some years ago, twitterfeed allows posts blogged here to make an appearance on twitter.

Once you've signed in, supply your twitter credentials, confirm you own a blog, supply the RSS feed for it, and Bobs your platitude at a frequency of anything as high as hourly. More prolific bloggers requiring a higher update frequency should perhaps consider getting a life.

The only issue I can see with this is that it's also possible to tweet directly to a blog. Exactly what sort of internet-crushing feedback loop this sets up remains to be seen.

Twitter and Amarok

Against my better judgement I'm playing around with twitter (mrsean2k, 1 (spam) follower) ETA: some non-spam followers, woohoo, electronic chums!

Although I haven't "got" it yet, it's got me dipping my toe in the water with a few things I've meant to get round to, and this is as a result of wanting to have Amarok post my "Now Playing" information as a tweet.

To cut a long story short, the sequence of events is:

• Install the command line twitter client twidge essentially this will let you tweet from a batch file
• Get a copy of the Now Playing Signature script
• Amend the script so that the generated signature is a single line, and so that after the sig is generated, twidge will tweet it automatically
• Install and run the script in Amarok

I'll write these up in a bit more (or at least some) detail shortly, but the upshot is that my song choices are automatically posted as tweets. Some tidying to do, and I think I'd be tempted to alter it so that it only tweets on a change of artist / album as opposed to every track, but it was surprisingly easy to do.