Wellington

Kate Stewart

• There are now 7,000+ packages in the Ubuntu main distro and over 19,000 in universe.
• There’s a tremendous rate of change.
• This isn’t a solution - it’s a description of an interesting problem space.
• Many projects feed into Debian, who Ubuntu feed from, and then Ubuntu release multiple images to the world at large.

• Stabilising this is hard, especially since Debian is a time-based release.

• Understanding whether there should be exceptions to the freezes is really challenging, especially when they’re to packages with many dependencies upon them.

• There’s a planning phase, a feature development phase, and a stabilization & QA phase.
• There’s a continual set of imports from Debian, mostly from Unstable. LTS comes from Debian Testing.
• Ubuntu archive contains ALL THE THINGS.
• 6 architectures, with 73 daily images (core Ubuntu + Flavour images).

• Number of packages varies by the edition of Ubuntu, as well as the flavours.

• Lots of facts: how do you convert facts into knowledge?

• Image build summary files, the Package control files provide some core information.
• Germinate examines “seeds”, groups of packages, and examines the dependencies between them; e.g. “desktop” is a seed, “live” is a seed, and so on and so forth.
• Shows forward and reverse depenedencies.
• There are other tools that help understand these dependencies: madison, apt-rdepends, apt-cache, diff-manifest.

It’s an interesting problem, but there’s no good solution for the problem, as yet. But there are some steps towards one: - You can graph out seed-based dependencies; you can use rdepends to extract information for driving regression testing.

Q&A

• Look at the bug trackers alongside the package mapping to see how past bugs/changes have created inter-dependencies. But how do you mine those bug trackers?
• For visualisaion, are you familiar with Gapminder? It shows changes of multidimensional data over time. (http://www.gapminder.org/)[gapminder.org]
• Consider using the Debian popcorn database to understand the numbers of people likely to be affected by changes.

Selena Deckelmann

Slides at: Slideshare.

• “Success Engineering” - that clearly will work, then.
• Plan for the worst. Minimise risk. Fail. Recover, gracefully.
• “You can’t eliminate risk.”
• alt.sysadmin.recovery shoutout.
• Failure is an option. Admit it.
• The open source world has failure and recovery as a core competency, but perhaps not systematically enough.
• Dr. Jerker Denrell publishes fantastic papers on the topic from a business perspective. “Predicting the Next Big Thing: Success as a Signal of Poor Judgement.” Looked at people who had predicted Black Swan events, and found there was a negative correlation with general quality of judgement.
• Try “Everything is Obvious Once You Know The Answer”

• Whatever, science, blah, onto the entertaining anecdotes!

• Rats like fibre optic. And we can use stories about this to help inform our planning.

• Document, Test, Verify is like Stop, Drop and, Roll.

Documentation

• Documentation tools are mostly pretty terrible, and there’s good work that could be here.
• Making time to update documentation when you do stuff.

Testing

• Verify your success criteria. What does success look like, what are you trying to achieve.
• Make sure you actually write tests, however simple, and have a buddy sanity check your work.
• Have a plan: make sure you involve other people with it, too.
• There are no shortage of testing tools, which should be repeatable.
• Do stuff in repeatable shell scripts.
• Have staging environments.

Verify

• What does pg_dump -d actually do. Well, it depends.
• Needed a plan for what to do if things go wrong. Staging environment. And test your rollbacks, not just implementation.
• People are really important. Having a buddy.

Failure to Imagine

• Telling externals they need to tell you when you have a problem is not going to work. Trust no-one.
• Share your stories of failure and talk to a diverse group of people, people who are different to you.
• Sharing lets you head failure off at the pass.
• People who are different to you means outside IT - business, musicians, the construction industry.
• Go and physically look at things you might need to do, don’t just sit in a room.

Reflection

• The post-mortem/debrief.
• Keep a notebook of your work, learn from it.
• Plan to have a post-mortem, even if there’s success.
• Document your plan with a timeline, allocate time, and actually test the plan.
• IRC is great, speaking is better. A headset is great.
• Have a timekeeper and alert people to when you’ve hit your drop-dead point.
• Limit improvements to 1-2 things. And endless list will never be worked upon.

Read the DailyWTF.

When people say, “Avatar had $50 million in subsidies” it conjures up a particular image: Joe Taxpayer ladling $50 million to a bunch of film moguls. It’s understandable, not least because it’s how things like SMPs used to work, but it’s not true. Subsidies are many and varied; I can think of at least three major types off the top of my head:

Exemption

The aforementioned subsidy for film production is a great example of this; film-makers get a break on their taxes in New Zealand. The thing is that when people toss around the idea that these cost something in a direct form they’re being, charitably, economically illiterate. No-one gave anyone anything. We simply neglected to take something. There is, you could argue, an opportunity cost, but in many cases, we can have something we wouldn’t otherwise get (Avatar work), or we can get nothing.

Framed that way, they sound quite tempting, right? Not always. We don’t have a capital gains tax in New Zealand, which means a major sink for our investment dollar in the last 7 or 8 years, even more than in the rest of the Western world, has been property speculation - not even property development, but borrowing money to buy houses that already exist in the hope a greater fool will make us rich. Unlike making movies, it’s not an activity that really spreads the wealth around; a built house requires pretty minimal upkeep. It may even have a perverse result, as rents rise to cover the costs of mortgages taken out with a view to capital gains, more money goes to service debts raised with offshore borrowing, rather than into the New Zealand economy, and we skew the view of what we invest in from activities that generate good first-world employment (special effects, software development, what have you), and into hoarding bricks and mortar.

So if we do have tax exemptions to attract foreign dollars, we ought to be quite careful that they’re an overall benefit.

Paying Your Bills

You’re a high-carbon-output business. Well, in New Zealand, I’m giving you money. Lots and lots of loverly money. That’s because I’m paying for your carbon credits.

(No-one’s paying my employer’s carbon credits, we just hand to spend a lot of money doing things like reducing the footprint of our server rooms with virtualisation and consolidation programs.)

There are no shortage of activities like this. Cities build sports stadiums for profitable sporting businesses to use. We cover the costs of businesses that might otherwise go broke.

Giving You Cash

Our domestic cultural industries are beneficiaries of this one; we ladle money into TV, local movies, ballet, opera, symphony orchestras, you name it. Outrageous Fortune collected almost $50 million during it’s run; whatever my share of that was, was totally worth it to hear the word “nungas” on TV, I might add.

In business these used to be our standard subsidy for farmers and other industries we’d decided we want to cultivate but couldn’t actually, you know, turn a profit. Sometimes that’s a good thing - I’d rather live in a New Zealand with an NZSO than not, for example - but in the case of SMPs, it damn near broke the country.

The Bigger Question

Implicit in the criticism of subsidies for Jackson’s work is that this in some way proves his companies haven’t really added value to the New Zealand economy, that he’s bludging off us in some dirty, underhanded way, slipping dollars out of our pockets.

As I noted above, it’s not really true; tax breaks cost us nothing in and of themselves, and have arguably helped funnel hundreds of millions of dollars into the country for productions involving the likes of Jackson and Rob Tapert (remember him? Been making internationally-known and profitable TV shows in New Zealand for the best part of a couple of decades?).

Moreover, the roots of the idea that there’s something wrong with encouraging particular industries is rooted firmly in the neoliberal religion that markets are perfect, governments are crap, and that governments oughtn’t pick winners. Now, if that argument’s coming from Richard Prebble or Roger Douglass, well, I think it’s a bunch of crap - after all, the East Asian neighbours, such as South Korea, we looked down on as third world countries when I were a lad have shot past us on the back of heavily controlled, directed, managed economies, not neoliberal paradises. Come to that, Britain’s economic rise was butressed by mercantalism, and China’s development is as a result of tight monetary control and a willingness to direct the economy whereever deemed strategic.

But I digress. I think it’s a load of crap. What’s odd, though, is that the flurry of concern about these subsidies so often comes from people who seem so otherwise uninterested in neoliberal dogma. The same line of thinking that argues taxs breaks distort the economy, the government shouldn’t pick winners, is the same line that says privitisation was and is good, that the Employment Courts are unfair and unreasonable, and so on. How many of the people bitching about Jackon’s “subsidies” are actually interested in that line of thinking in any genuine manner? How many are simply looking for another club to belabour someone who appears to have committed the great Kiwi sins of being very successful in his fields, wealthy, and uninterested in issuing grovelling apologies for daring to achieve these things?

You know, when your staff leave a child with second degree burns you’d think you could at least, I don’t know, proffer an apology to the parents and the kid and not, say, delete the negative review from a supposed restaurant review site (which makes me wonder; if Ernesto can get a review about them crippling a customer yanked, what else will DineOut pull? If you aren’t allowed to say bad things, what’s the point of the site?[1])

I don’t think I feel like playing Russian roulette with Ernesto any more; we’ve been there in the past, but when your response to leaving someone burnt enough to need morphine is to... attack the victim for posting a negative review, I don’t have any confidence that they give a shit about the safety of their staff or customers.

Since they’re next to Espressoholic and over the road from Scopa, that’s not too hard.

[1] Hoovering up donations from people under the misapprehension they provide a useful service, I guess.

Crane on the Dock

Autumn has begun to feel like autumn; for the first month or so we enjoyed a late summer of cool (but not cold), crisp, brilliantly clear days; that particular spell of the closing of summer and the opening of the colder season which is perhaps my favourite time of year.

That is rounding to a close now; the weather is closing in with early nights, pitch black by the time I trek home from work; enlivened by the lights of the harbour, perhaps, but during the working week the only glimpses of sun are those of a morning walk or time snatched from the office during the day; not yet, though, closing me in the suffocating period where the day is black when leaving the house as well as when arriving back at it.

Red Leaves of Autumn

There are pronounced pleasures, though; the ducks at the Botanic Gardens have not grown fat, as they do in Spring, on the offerings of the people rushing to offer bread to ducklings; they’re eager to enjoy the ministrations of a small girl and her gifts, and my heart soars as my spring enjoys her autumn.

Foggy Autumn Morning

With autumn comes thick fog, rolling in from the sea over Rongotai and up Hataitai.

Putting the Queen on notice

I assume this is yet another set of buildings the NZTA are deliberately running down so the Wellington City council can pave over more of the central city, which annoys me, but it’s pretty funny seeing The Queen getting a rude letter about the state of her properties, nonetheless.

Peacock Costumes at the Sevens

This was easily the best Sevens costume, originality and cool-wise, I saw from the gaggles of people walking past work and our local coffee spots: Four women dressed as peacocks, with backs covered by a tail of real feathers. Nicely done, and something a bit different.

They’d also spent some time getting the noise of peacocks calling down to a fine art.

Last night I uploaded an old picture of horrid fog over Hataitai. A couple of hours an even worse one descended. Airport closed to 10:00, it’s like it’s 2005 all over again.

My bad.

Please send ark.