Danny Angus

Looking through a glass onion

2009-11-03T09:35:00.002Z

In recent days it seems as if the BBC have hired someone from the news desk of the Onion.

In the words of the famous saying I rolled on the floor and laughed my ass off when I read this gem:

Paperboy offered £6.93 redundancy

I got this piece of paper and I didn't really understand all the words.

Exposed, the GPL's fatal flaw

2009-10-30T11:13:00.003Z

Thanks to a tweet from @wrowe I read this post and this open letter fromRichard Stallman to the European Commission opposing Oracle's acquisition of MySQL (via their acquisition of SUN).

The thrust of the argument is that MySQL is currently licensed under the GPL.
The GPL denies anyone apart from the rights holder the right to sell the product or a derivative of it.
Therefore if Oracle become the rights holder they have an exclusive right to sell the product.
They are, for the same reason, under no obligation to feed their changes back into the open source code base, and have no commercial incentive to do so.
Likewise the opposite is true, Oracle's competitors will gain no commercial advantage from taking a fork of the GLP'ed code and developing it, because the GPL forbids anyone from selling that derivative or from distributing closed-source enhancements.

On the other hand the Apache License [sic], does allow companies and individuals to redistribute for profit, and to redistribute closed source product based on the open source code.

It would seem, then, that had MySQL been licensed under the Apache License Oracle's competitors (whoever they may turn out to be) could have competed from an equal position, by taking a fork of MySQL and developing it, or developing new products derived from it.

This is an approach taken by may, and these days, most of us will benefit in some way or other from products which have benignly licensed components within them.

Could this, then, be the long awaited moment when the aggressive social evangelism encapsulated in the GPL tests capitalism in the ultimate showdown?

I think that a lot of people have been waiting for that, but sadly for them it looks like their champion has taken a fatal hit in its Achilles heel.

Wear it pink day

2009-10-30T09:49:00.000Z

Today, Oct 30th is Wear it pink day, and to show our support we're all wearing it pink, as these photos show.
Thank god the photo's aren't any clearer!

OMGWTF US MPG != IMP MPG

2009-10-29T16:56:00.002Z

Kevin pointed out that my tweet which said:

According to national geographic this mth, "average american car gets 21 mpg". Omg, my 2L clio sport gets 31, and I drive like an idiot.".

is misleading.

According to this authority (picked at random from the net of nonsense and with no check by me) what we're saying is the US cars average 25.219946996999997 UK MPG. [1]

And my clio only gets 25.812901195923956 US MPG. Not so smug now, am I?

[1] How precise! But don't forget kids, as Uncle Danny never tires of saying, precision is not the same as accuracy.

Android's Elephant in the room, and the its-not-an-iphone award

2009-10-29T09:45:00.004Z

The elephant in the room for Android, from my point of view as a G1 user, is the woeful bluetooth support.
My previous phone, a w810i, supported a bewildering array of bluetooth uses, but the ones I found most useful were push/pull of my contacts and using the phone as a presentation tool.

Well hallelujah, android 2 platform highlights are saying that the new major version will contain

New BT profiles: Object Push Profile (OPP) and Phone Book Access Profile (PBAP)

OPP is the one I want for managing and exchanging contacts, PBAP is for integrating with my car. Although I have to say it does this already.

I'm also pleased to see that there are a number of usability improvements, in particular for the soft keyboard, which I can't really use very well because of my phat phingers, contacts, email, and a digital zoom on the camera.

Oh and in case you were wondering, I think the ring volume is too quiet, and the battery life sucks. But since those are hardware issues I guess that after a couple of months of living with it Android gets a big thumbs up from me, and so is the recipient of the its-not-an-iphone award.

qr code

2009-10-10T18:02:00.002+01:00

At work we already make good use of onscreen barcodes for use in fulfillment processes, but I hadn't really thought much about putting anything like it on here.

However inspired by google's logo the other day I realised that it might actually be useful to put a link code on my blog for phone users who may want to subscribe.

So thanks to this great site you can now subscribe to this blog by scanning the QR code below (also in the left margin).

Sadly for me it opens in the browser and doesn't get anywhere near a feed reader, but you may have more luck!

I guess I should do something work-related with this idea next..

Lucky Firefox!

2009-09-28T16:30:00.004+01:00

Some folks were discussing the application of the "ribbon" to firefox the other day.
More on that here.

One thing that came out was the notion that the ribbon is a tool best suited to document production.

If that's the case, and I am in no position to doubt it, please can UI designers allow us to move it to the side.

The screenshot graphically demonstrates the problem I have on my 11" widescreen laptop. The ribbon eats valuable height for lunch, while two whole football pitches of space exist at the side of my document.

I could probably get more buttons into that space than I would ever need, and I'd get the height back.

Stepping down

2009-09-24T21:33:00.004+01:00

Today I indicated to the Apache James project management committee (PMC) my intention to step down after managing the process of electing a successor.

I'm very proud to be associated with The ASF and the James project, and even more proud still to have served as the project's second chairperson, and wish my successor all the luck in the world.

I also took the opportunity to stand down from the Jakarta PMC, another project I am immeasurably proud to be associated with, which I haven't contributed a meaningful iota to for a couple of years other than to clog up their decision making process a tiny bit.

So it seems like an appropriate moment to pause and reflect nostalgic-ly on the times shared, the problems solved, the achievements and the failures, friends made and lost, arguments lost and won, flame-wars, troll-baiting and so on.

But I can't be bothered with all that! And to be honest I'm looking forward to a refreshing lack of commitment, no more nagging guilt at not making enough time, and above all the chance to pick and choose how and when I contribute in an arbitrary and selfish way once again.

I hope that this means that I can spend what little time I do have poking about in code, (I have long envied those who have time to experiment in the labs), arguing with my peers and not trying too hard to keep up with mailing lists.

Load Balancing Apache httpd 2

2009-09-15T23:18:00.002+01:00

Yes.. hear it here last!
I've been telling everyone I meet about JimJag's stupendous presentation at Apachecon US 08, in which he showed us, in glorious detail, how to use mod_proxy properly and mod_proxy_balancer.

I was impressed, and so have those who've also seen the slides.

Now its your chance, I recommend reading Jim's slides here.

If you have real life experience of using mod_proxy_balancer, let us know if he oversold it, or if it rocks your world. I *still* haven't had time to use it in for real yet.

Father Spodo Kommodo

2009-09-15T19:24:00.002+01:00

Eternal Thanks are due to this site for reproducing the list of names that Mrs Doyle guesses in Father Ted (possibly the best comedy every to have been shown anywhere in the multiverse) , in the episode I have come to know as the "Father Spodo Kommodo" episode!

That List in full then..


Fr. Andy Riley,
Fr. Desmond Coyle,
Fr. George Byrne,
Fr. David Nicholson,
Fr. Declan Lynch,
Fr. Ken Sweeney,
Fr. Neil Hannon,
Fr. Keith Cullen,
Fr. Ciaran Donnelly,
Fr. Mick McEvoy,
Fr. Jack White,
Fr. Henry Bigbigging,
Fr. Hank Tree,
Fr. Hiroshima Twinkie,
Fr. Stick Bubblecart,
Fr. Johnny Hellzapoppin' ,
Fr. Luke Duke,
Fr. Billy Ferry,
Fr. Chewy Louie,
Fr. John Hoop,
Fr. Hairy Cakelinum,
Fr. Ebula Conundrum,
Fr. Peewee Stairmaster,
Fr. Tight Head Lips,
Fr. Jemima Racktouey,
Fr. Jerry Twig,
Fr. Spodo Komodo,
and of course Fr. Canabrana Lammer.

Get this thing to market.. NOW please, I'll buy one!

2009-09-04T13:28:00.004+01:00

In the many years between the Mongolfier brothers and the Wright brothers people speculated about a world with aeroplanes, inventors fell over each other in an effort to make this world a reality. People waited for an invention that, although seemingly impossible and often dismissed as fantasy, they felt was inevitable.

The inventors succeeded, and in the century that has passed their work has indeed transformed the world.

So let me introduce you to the invention I'm waiting for, and I'm sure you are too.

Wireless power transmission.

First I want to clear my home of power cables, but why don't you take some time to imagine what else will be possible.

Its easy to dismiss this idea as impossible, reasoning that anyone who stood between power source and destination would be zapped by all that power, but seemingly not so.

Here are two articles from the BBC, the first[1], from 2006, talks about research into a possible means of transmitting power without wires, the second[2], from earlier this year, demonstrates a working prototype.

Hurry up Witricity the world is a-waiting, and when I get one there'll be an award in it for you.

[1]Physics promises wireless power
[2]Wireless power system shown off

Best error message of the [specify timeperiod]

2009-09-04T09:22:00.005+01:00

I suppose the svn logs would allow us to discover exactly who should get the credit for this, but this morning I was privileged to witness the first application error message ever to offer me any signs of shame or contrition.

Vista had summarily decided to die during the night, which it does now and again and I think is just an excuse for installing the updates that I've *chosen* to not bother with for now, and when I restarted firefox it showed me the error captured below.

By contrast see the second image, which represents the way we are more usually treated.

Mozilla Dude, whoever you are, have an award for expressing your human emotion, and making me feel valued in the teeth of failure. I hope you start a trend.

(this isn't my screenshot I found it here, but its the same screen)

And now the shamless norm..

Don't invert your security!

2009-08-28T13:51:00.002+01:00

The most striking lesson I think that anyone can take from the recent apache compromise is this:

The more secure zone should have credentials for the less secure one, not the other way round, and the more secure zone should be responsible for controlling the processes that it is involved in.

This way the less secure zone doesn't have any influence over your more secure stuff.

If you, like me, spend your days making systems interact with one another this is reasonably fundamental stuff. But for those who aren't so paranoid its a lesson well worth heeding.

Danny strengthens in the Atlantic

2009-08-27T10:59:00.001+01:00

Danny has maximum sustained winds of 60 mph (95 kph) early Thursday with slow strengthening expected over the next few days.

Its weird seeing your name used this way, but a great opportunity for toilet humour.

Use http AUTH for control of read and write access to multiple svn repos

2009-08-21T14:06:00.003+01:00

I figured this out, and thought I share it, as I couldn't find anything remotely similar on the interweb.
OTOH I'm sure you all know all about this, and there are articles everywhere and I'm just being dumb.

I wanted a password protected svn, with some people read only and others read-write, across multiple repositories.



## your svn url is .../svn/repository/reponame
## first set up the /svn part
## enable DAV svn and require any user from the password file for any action
## this will allow everyone who logs in to do anything

<Location /svn>
   DAV svn
   SVNParentPath /wherever/svn/repository
  AuthType Basic
  AuthName "Subversion Repository"
  AuthUserFile /etc/httpd/dav_svn.passwd
  Require valid-user
</Location>


## now for each repo appy a stricter constraint

<Location /svn/infrastructure>
# only members of sysadmin group can access infra repo
  AuthType Basic
  AuthName "Subversion Infrastructure Repository"
  AuthUserFile /etc/httpd/dav_svn.passwd

## create a groups file
  AuthGroupFile /etc/httpd/dav_svn.groups
## the only people who can acces this one are the group members
  Require group sysadmin
</Location>

<Location /svn/projects>
  AuthType Basic
  AuthName "Subversion Projects Repository"
  AuthUserFile /etc/httpd/dav_svn.passwd
  AuthGroupFile /etc/httpd/dav_svn.groups

## only members of commiters group can commit
## but any valid user can checkout and browse
## because we're using limit except to narrow
## the restrictions

  <LimitExcept GET PROPFIND OPTIONS REPORT>
  Require group committers
  </LimitExcept>
</Location>

Neat?
Or weel kent?

at the third stroke it will be 123456789 o'clock

2009-08-07T16:18:00.002+01:00

Yes folks, earlier today (I meant to post this then, but you know how it is) it was
12:34:56 7/8/9

this will happen again in 1000 years,

but we can look forward to
6:5:4 3/2/10
and
5:6:7 8/9/10 next year.

Failed to open the relay log

2009-07-29T10:06:00.002+01:00

We moved some hardware about last night.
Well actually Jim and Marcin did.

This AM I got to the office to discover everything apparently working (its early days yet!) apart from one mysql replication.

Now; we've set up this replication to replicate one db using replicate-do-db which has very many of its own crazy gotcha's that I'm not going to get into here.

We also set the slave params by executing a command, rather than in my.cnf.
This allows us to meddle without having to restart mysqld, and protects us from accidentally reverting the settings if mysqld should require a restart.

my.cnf only contains the id of the slave, skip-slave-start, and replicate-do-db that affect replication.

This morning, h'wever, the slave was compaining that it couldn't open the relay log, which was annoying. I found a post here which suggested flush logs that didn't work. The next suggestion is reset slave.

Now I'm afraid of reset slave, but I didn;t think I had many options, so I forged ahead and here's how I got the job done.

NOTE-- I take no responsibility for anything misguded and foolish that you may choose to do in the light of reading this, or any other, bit of misleading nonsense on the internet.

If you have any doubts about your ability to resist doing this without putting a roll back strategy in place before hand then read no further. You have been warned!

Working completely on the slave:
1/ stop the slave .
2/ show slave status
3/ make a note of Read_Master_Log_Pos, which should equal Exec_Master_Log_Pos, and of Master_Log_File
4/ now execute reset slave
5/ now set the params you noted..

CHANGE MASTER TO MASTER_HOST='xxxx.killerbees.co.uk',
MASTER_USER='xxxx',
MASTER_PASSWORD='xxxx',
MASTER_LOG_FILE='Master_Log_File',
MASTER_LOG_POS= Read_Master_Log_Pos

6/ start slave

Job Done.

You can also use this approach to start a second slave on the same master without having to affect the master's operation:
stop slave one,
note the params,
set slave two params,
export-import the data from slave one to slave two,
start slave two.

In complete contrast...

2009-07-28T15:40:00.003+01:00

A colleague pointed this out to me.
Its an online tool to check contrast for accessibility, and you'll be pleased to know that the old-geek retro colour scheme of this blog is A-OK scoring between 6.5:1 and 7:1 contrast for the various text colours.
It is also (apparently) ok for folks with Deuteranopia, Protanopia and Tritanopia.

Go on, check out your own website, you know you want to.

But if you can't read it you can send always an email to danny@killerbees.co.uk

Why we don't need a definition of spam

2009-06-30T13:28:00.003+01:00

In my last, rather facetious, post I poked fun at John Levine for trying to prevent the ASRG from falling out, once again, over the definition of spam.

The catalyst for the current attempt was a post on the list that took issue with a paper I drew up (draft-irtf-asrg-criteria-00.html) a couple of years ago. At the time I attempted to define spam, but the whole review process became quickly tar-pitted in a debate in which despite the general agreement of the group the detailed differences were irreconcilable.
As one correspondent put it yesterday:

attempting to define "spam" is the very best way to ensure that a document is never finished.

So I plumped for this:

Any Message or Messages of the class of Messages which the Recipient wishes to prevent from ever being presented with. It is implicit in this definition that it is unnecessary to ever transport Spam. Spam in this context can also be defined as Messages which it is never necessary to transport. It is not in the scope of this document to attempt to distinguish or justify any more detailed definition of this term. Nor is it in the scope of this document to analyse the reasons why the Recipient wishes not to be presented with the Message or Messages.

My intention was to encapsulate some of the critieria which the ASRG applies to the ideas with which it is presented, some well reasoned but flawed, many bordering on the insane, a very few containing ideas of real merit. I set out to highlight common pitfalls, and ensure that proposals have a net benefit. Some don't. Go figure! Some would be more expensive to operate than transporting and filtering the spam would be, others appear to benefit someone, but only by passing on the real work to an innocent 3rd party.

So it occurred to me yesterday that the document is also addressing the problem of defining spam.The problem being not that a definition cannot be drafted, rather that no definition is universally agreed, and unfortunately each reason for disagreeing with any definition that I've heard has some genuine merit.

So my approach is this, if we cannot agree an academic definition of the problem, but we agree that the problem exists because we can recognise it when we see it, then perhaps we should apply the same standard to any proposed solution.

If we can agree that it smells like a solution, we don't need to agree about what the problem actually is.

Of course the risk with that approach is that by avoiding defining the detail of the problem we're never going to arrive at a solution that successfully addresses the detail , and not just the big picture, because we don't agree what the detail is.

Then again, anything which improves the big picture is beneficial, hence the success of DKIM and SPF, so this may not be a real concern.

Quote of the [specify period]

2009-06-30T09:17:00.004+01:00

This [specify period] sees two quotes, both of them from John R. Levine erstwhile chairman of the ASRG of the IRTF.

The first is redolent of Lewis Caroll:

I think that as soon as you start quoting the dictionary, you've lost
the argument.

The second is priceless given the context:

No, we're not going to define spam

Well done John, you do a great job, keep it up.
Have *two* awards!!

Blog Design

2009-06-27T17:43:00.003+01:00

Well I've finally got round to creating my own skin for this blog. Its a blogger blog by the way.

It was the new head of creative at work, Dino, who showed us www.gridsystemgenerator.com which I used to lay the whole think out by doing little more than setting the class of the blogger "sections" to one of the grid's generated classes.

Marvin the martian

2009-06-02T16:40:00.003+01:00

I was accused at Apachecon EU a year or two ago by Steve of relying on "architect tools" i.e. visio, powerpoint etc., instead of programmer tools, (like what? emacs!) For which he proposed a test framework here.

Well today I was discussing some of the finer points of separation of concerns with the guys and had cause to exhume (and re-label for php) a diagram I drew a couple of years ago for a former employer to illustrate the layering in our java systems to an interested, but unenlightened group of Oracle developers.

It would be a good test case for Steve's framework, because it has to be a particular shape, no matter what information it has to convey.

I could plead that I drew it first then recognised the shape, but I won't bother because life's too short.

Colleague Ed suggested it should be recorded for posterity so here is the original Java version...

"write to: Danny Angus, c/o Google, "

2009-05-27T13:54:00.002+01:00

Blimey folks,

I use the incomparable feedburner for this blog, and if you subscribe by email you get the posts in your inbox.
Imagine my surprise when I read this at the foot of the mail..

If you prefer to unsubscribe via postal mail, write to: Danny Angus, c/o Google, 20 W Kinzie, Chicago IL USA 60610

I wonder what would happen if I did...

Dunkirk spirit

2009-05-27T07:51:00.002+01:00

According to this we brits are going to fight the credit crunch by watching football and going to the pub a little more.

The British economy is also expected to get a massive injection tonight from Brits going out to local pubs and clubs to watch the match on TV.
"Some people may choose not to travel to Rome, but they will generate economic activity at home," Chadwick said."

xml co-processor

2009-05-03T20:02:00.002+01:00

so much of what we do nowadays involves parsing XML isn't it about time we delegated it to the hardware?
Or has someone already done it, or more likely it this a dumb idea?