Stuff Michael Meeks is doing

Up early; snow - and power cut knocked back the heating; cold, adjusted that; packed babes off to school with J. in the car.
Dug backwards through mail; lots more patches merged over the weekend, encouragingly; got some Windows credentials for rdesktop setup, and prodded at NSIS on a machine in Provo. Repeatedly appalled by the NSIs scripting language; any language that requires this macros vs. functions page, is unbelievable. NSIS: combining the readability of assembler, with the linguistic elegance of Visual Basic. Since there are virtually no useful built in functions - you get to copy/paste big chunks of impenetrable (and unreliable) assembler-like code into your 'script'. Words fail me to describe the incredible life-sink that is NSIS, another day robbed from me debugging this total, unsustainable junk.
Dinner, put babes to bed, back to the work. Back to battling the evil NSIS, added ever more debugging output for good measure. Did lots of admin, Clarity, status reports etc. amid the hacking, pondered an LXF column.

2010-11-28: Sunday.

Up early, to let J. sleep. Dealt with babes. Out to NCC, Claire speaking in / out to the creche, and ended up with a sleeping baby on me somehow.
Back for a quick lunch; J. out to visit Bruce (recovering from a minor stroke); played with the babes, cooked dinner, failed to kill all consuming it (somehow).
Bathed babes, put them to bed. Admired the Prusa Mendel which seems incredibly superior, in terms of removing redundant complexity, to the stock version; and looks great in action. Pleased to note that my wooden repstrap y-axis fixings are very similar to his plastic ones. It is awesome to remove a vast number ~all of the bearings (and even use plastic printed bearings in their place), along with a huge number of different screw-fixings of different sizes and lengths, and several lengths of studding from the design while keeping it functional. I love the idea of stacking eg. vertex components vertically on a single bed too - to break them apart before assembly; neat.
Listened to a rather interesting sermon on Gideon's fleece; sleep.

2010-11-27: Saturday.

Lie in; breakfast, pottered around poking at the reprap. Discovered that my grounding issue was down to getting the fixings into the z-leadscrew-base the wrong way up, such that they projected below the machine, and thus horribly distorted it when resting; nice. Lunch, Sandy popped over to take H. and M. out for cake making, and various treats at her house.
Re-assembled machinery variously, with the help (or otherwise) of E. and M. ending up with a Mendel that at least moves smoothly and reliably in each axis (at last). Still wondering when to make the transition and break the working repstrap to make the reprap work.
Helped tidy up, clean etc. Hannah, Nick, Joni (& Sandy) over for dinner - really lovely to see them again; with an added power-cut for amusement value (luckily the meal was cooked by then). Up late talking & enjoying their company.

2010-11-26: Friday.

Poked at mail; purchased a heated bed and some PLA for my repstrap - perhaps, together these will solve my warping problems, and allow me to finish my reprap (who knows); Christmas cometh but once per year. Reviewed and merged a number of nice patches and cleanups from various hackers.
Discovered that upgrading my glib/gio broke default associations, by which I mean getting the wrong browser for links, and Evolution not spawning mailto:'s correctly. Dug /usr/share/applications/defaults.list out of the strace, and fixed it manually.
Posted the minutes from the LibreOffice tech steering call yesterday. Amused by the Red Molotov T-shirt with the legend: Fat people are harder to kidnap; amusing. Elizabeth on her own initiative decided it was time to get Daddy for lunch, only for me to discover (after dis-engaging) it was an non-maternally-authorised interruption, fun.
Out to Derek Llewelyn's funeral, at All Saints; a man of great faith & an inspiration to many it seems.
Chat with Thorsten, quick dinner, out to the School Faire - with the three smallest babes, met a 'Father Christmas' beard-alike, returned fiscally milked. Reviewed some patches - I love to remove these chunks of old, commented code beginning // why do we need the code below ?.
Dusted off, and setup my Win32 machine again - it has been quite some moons since it was needed. Dug at Win32 / NSIS scripts, bitmap branding, and so on: what fun, the things we do to make Brazilians happy...

2010-11-25: Thursday.

To work; dug through the mail, wrote up an interview, prodded at patches we had overlooked here & there. Dug at UK charity law issues a little. Reviewed and merged a few more patches, dug at some icon work.
Poked at auto-migration from older versions, to disable it happening multiple times; if migration fails it leaves a 'MIGRATED' stamp file around to stop it happening again. Ever wondered why OO.o doesn't launch when you start it - that would be the migration, then silent exit: you get to re-start it manually; shame.
Nice interview with new LibreOffice hacker Norbert (shm_get) got published; Cor Nouws is working hard at the next ones already. Hopefully getting to know people helps to build those invaluable human linkages of which a community is composed.
Reviewed the financial situation in the evening, and closed out Towry Law investments, since they are impossibly lamer than Edward Jones. Now need some good, cheap way of buying exchange traded funds to avoid paying 3% on purchase, to those clowns, and a %age per year - rather a large chunk of the return.

2010-11-24: Wednesday.

Up early; helped with babies. Quick mail check, Mike kindly took me to Cambridge. Lots of study, good talk on the Old Testament Law, then double session on Revelation - much more interesting, and less odd than it seems to many (so far). Gave my talk. Amused, on comparing notes that an identical discussion over missing pieces in my 'Big Idea' occured with Mike, but with a totally different thrust; felt vindicated in my defence. Home, dinner, put babes to bed while J. counselled, filed time off.
Tried to catch up with admin etc. Interested by the mooting of rolling releases; I look forward to seeing how that pans out. Pleased to see Alex's gtk+3 / HTML5 demo - interesting. Got my eleventh anniversary of joining Novell E-mail; that's quite a while. Worked late, interspersed by comforting a sad Elizabeth coughing & crying in her sleep.

2010-11-23: Tuesday.

Lie in, feeling poorly, rumbling chest infection on the up. Slogged through mail, Lunch. JP's staff, misc. conference calls of varying interestingness, more mail, boggled at my mobile phone bill, conf-calls from America are worse than possibly imagined.
Up lateish, prepping a talk on Genesis 2:18-25, lots of careful reading, and re-reading of the text in misc translations, the context, digging out the structure etc. fun.

2010-11-22: Monday.

Up too early; packed babes off; to work. Plugged through the un-merged patch backlog, lots of interesting things, and starting to get some good questions on the list as well. Really great to see the Beta 3 announcement with nearly one-hundred-and-twenty individuals contributing code since Beta 2.
Novell announced their merger agreement with Attachmate - a company with one quarter of our employees, and over a third of our revenue; interesting, if the shareholders approve it, we will go private. People, it seems, like to E-mail me asking me what this means: and since I have no idea this just fills my inbox for no good purpose. What I do know is that Software Freedom is the future. If I was a believer in efficient markets, I might speculate that the gap between our stock price now, and the offer price might reflect any vestigal uncertainty as to whether the shareholders, regulators etc. will approve the deal. The SEC filing is quite interesting too.
Saddened to read the latest hand chopping story out there (apparently with diagram: where to chop), quite a contrast to Proverbs 6:
People do not despise a thief if he steals to satisfy his hunger when he is starving.
and the penalty to pay back some low multiple of the financial pain in restitution. Quite how depriving someone of their livelihood is a proportionate response I have no idea, and the very idea of deterrence: by acting un-justly in this case we will scare others away sounds like the cure is far worse than the illness.
Filled out Clarity timesheets; hopefully Attachmate don't require extensive time and motion reporting. Conf call with Frank in the afternoon.
Pushed out a new bootchart2 release with lots of fixes; apparently it is in OpenSUSE, Debian, RedHat, Maegia, Exherbo and presumably elsewhere now download, and NEWS.
J. off counselling, and plugged away at the mail / task backlog, prodded at a series of questions for Norbert.

2010-11-21: Sunday.

Lie in, off to St Lukes in the morning - Baptising Charlie's new daughter, good to meet up with old friends.
Back via nibbles purchasing for a monster family lunch for fourteen, enjoyed getting to know Joe a writer and journalist.
Packed, drove the babes back home; put them to bed, unpacked, sleep.

2010-11-20: Saturday.

Up early, entertained babes while J. went Christmas shopping. Attended the mele myself, got one gizmo for a daughter, missing another two.
Back via the playground, a fine lunch with the family though Ilona off at a baby shower. M&D set off to help set-out Barbara's sixtieth birthday bash. Watched Cinderella with the babes, feeling simply exhausted somehow.
Out in the evening to the party, good to meet up with Andy & meet his interesting fiance Olivia. Fine food, folk-dancing, and family - drove back with Rob, Ilona & a boot-full of balloons for little girls. Bed late.

2010-11-19: Friday.

Up early, dug through the mail mountain. Lots of exciting things happening - quite apart from the deluge of code still going into LibreOffice. Fridrich merged the newest libwpd code changes to improve our WordPerfect filters, requiring some manual fixes to the build (we should improve our dependencies there). Will Lachance's nice announcement of libwpd-0.9.0 has a great message, as well as screenshots of the new features:
it's quite unlikely that OpenOffice.org WordPerfect support will advance unless (1) someone volunteers to do it and (2) Oracle drops their copyright assignment policy. The chances of these things happening seem rather low to me. My personal recommendation would be to switch to LibreOffice as soon as the first production version is released. I expect it to rapidly overtake OpenOffice.org in functionality due to its more open participation model.
Personally, I'm more hopeful that Oracle will see the light, but its great to have Will's support. These improvements (the majority of which are Fridrich's work incidentally) should be included in our 3.3 release shortly.
Got a nice bootchart patch from Harald, and then a set of four more adding standardized (isn't that a great word!) systemd service files; made him a co-maintainer. Roll on the systemd future where we will be able to maintain and update our hooks in one place, rather than trying to tweak umpteen per-system init scripts. Hopefully in a dracut'ized world we could even do that for initrds too (which would be great). Great to see Harald's enthusiasm, and ideas for graphing (eg.) I/O latency per-app, which should be much more useful than CPU for disk based machines. Harald created a new #bootchart2 channel on irc.freenode.net for the cool go-faster-kids to hang out in.
Went through the business cards from recent conferences. Added some more Easy Hacks: python gdb helpers, and a simple macro cleanup. It would be great to have some enterprising hacker, and not-yet-hacker to take on those two respectively.
Got bogged down in more mail, legal pieces etc. Hannah's church friends over for a club in the evening, quick dinner, and drove south to my parents'. Got the sleeping babes unloaded, and re-loaded into their beds. Met new relations Dave & Maureen.

2010-11-18: Thursday.

Up early; taxi with an interesting Nokian to the airport. Wandered questing for Guy, met another Nokian while finding breakfast. Plane back, met Rob + other Collaborans, taxi back with them to admire the new office, lunch, train home.
Checked mail, caught up with family, bed early.

2010-11-17: Wednesday.

Up early; to the conference - signed Christian up for a Wayland un-conference talk in his absence. Disappeared into the free-Lenovo-laptop melee for a while. Back for the (interesting) end of Christian's talk.
Gave a LibreOffice talk, sparse attendance but some good questions and positive feedback, kicked out, backed to the hotel and out for an abhortive wander, then dinner with Mark, Peter, Guy, Aaron & Gary.
On to the match; my first football match ! it turned out that (perhaps) it is a game that has a useful role in building the ability to accept delayed gratification in the masses: or put another way - almost nothing seems to happen most of the time. Managed to get excited at the end, when Ireland were playing a really offensive game to try to equalize.
Wandered off, bid 'bye to the lads, and via a pizza place, to bed.

2010-11-16: Tuesday.

Up early, arrived at the conference, grabbed some food in passing on the way to a pair of Telepathy talks.
Pleased to see Prince William getting engaged, God Bless the future King (unless a better King comes first). Interested to see that apparently confessing: "I am a sinner" - is supporting evidence for stoning you to death (for adultery), but I'm a inveterate sinner and ingrate too. Knowing nothing about the case - I have to wonder: where is the man ? surely it takes two to commit adultery ? not a new blind-ness I suppose cf.: John 8, who casts the first stone ?
Enjoyed a fine lunch, and some great talks - on KOffice, zypper etc. Tried to catch up with mail, and appointments.
Off to the Guinness storehouse in the evening - lots of interesting displays, people, food and music; excellent. Back to the hotel late, up rather talking to Guy, Rob and Christian.

2010-11-15: Monday.

Up shockingly early, off to Stanstead, met a couple of MeeGoCon attendees on the way. Enjoyed some of the keynote demos. The technical steering duo were interesting - Chris Schlaeger announcing AMD getting involved in MeeGo (via up-stream apparently). The second Question was: where is LibreOffice in MeeGo ? which is encouraging.
Pleased to see Mageia supporting LibreOffice.
Somehow ended up topping and tailing talks in the Vavasour suite; good stuff. Clarity, unpaid-leave accounting etc. Out for dinner with some great Intel guys in the evening, understood Wayland a little more, and pleaded with Christian (semi successfully) for restraint on texture memory chew. Bed late.

2010-11-14: Sunday.

Up early; J. drove us all to near Reading for Isabelle's party - a great family bash. On to see David Mansergh at his home, enjoyed a too brief visit; and back home late. Listened to a rather interesting talk on 'The God Delusion'.

2010-11-13: Saturday.

Up early, Jill baby-sat all four; out to training - lots of interest during the day, some extraordinary role-plays. Back in the afternoon.
Played with babes, Fish & Chip dinner; packed them to bed; back to work - wrote LXF column, on-line checkin, etc.

2010-11-12: Friday.

Prodded mail; really, really pleased to see Samsung doing the right thing - using ARM's Mali 3D core for their phones. Anything must be better than Imagination which appear to have an utter contempt for Free Software drivers that goes beyond ridiculous. Imagination FWIW are the root disaster behind many mobile phones 3D hardware as well as Pulsbo and other miracles such as Intel CPUs that come without full specifications, and missing programming documentation for great chunks of the silicon. Hopefully ARM will show the right way to do this with Mali, and win a lot of traction. Building key parts of your product on top of a complicated, closed, binary blob is just a disaster - how can companies ask their customers to do that ? Update: seems I jumped the gun, and Mali still has a big closed user-space blob; that sucks - fix it !
Pleased to see Norbert reply to Roberto's curious blog; and despite being a key LibreOffice hacker who rather knows what he is on about, get told:
"In fact these non functional changes are not a way to get people acquainted with the code - that is something that require time and dedication - but maybe a way to make more complex integrating upstream contributions."
Interesting the certainty (from a non-developer) that doing small changes is not a way to get acquainted with the code (though in fact this is an excellent way to even learn to program: by reading first. I recall learning the syntax, and love of BBC BASIC by typing in games from magazines). The speculation of some malign intention behind trying to get more people involved which is (maybe) to make our lives more difficult in future is an odd thought indeed; why would we want to make merging changes from Oracle more difficult than necessary ? As for the sensitivity of "cutting bridges with up-stream" - from the LibreOffice side, there has always been an open invitation to Oracle to join, and the bridge chain-sawing seems to be in full swing on the other side. As for the idea that LibreOffice programming changes must be discussed and agreed (with a non-technical audience) before implementation - I'm not convinced that that is a winning strategy for attracting independent contributors to balance Oracle, Novell, RedHat etc. so we get a truly diverse and vendor neutral project. Indeed - the very idea that people who did not contribute to the code, should be controlling its ultimate ownership and licensing is fairly offensive to most developers. Having said all that Roberto makes some good points about the social dynamics; It is my hope that we would have a governance represented by a Community of Practise - ie. a meritocracy of those developing, marketing, and actively working to improve the product and project. With regard to code hackers being "a de-facto minority" - I believe we can change that, and we've made a great start thus far, if only by providing easy tasks (I dispute that these have 'very little value') so people can build confidence in having their work accepted. I completely disagree on his copyright assignment point, and I too talk to very many corporate employees at different levels; let us see who is right in the end.
Why is it that while you have a FaceBook friend request open you cannot see more of the details of their page. It is ridiculous to have a person about whom you can see almost nothing, asking to be friends with you - surely ? How are you supposed to know if you know them ? (or if you ask that, is it too late anyway). For some reason I loathe ignoring friend requests on principle, and thus agonise over the regular friend/spam traffic.
Managed to get my counselling coursework done finally; it is frightening to try to answer questions like "Identify areas for possible future self-development / growth" - I mean - with me, where do you start ? it looks like a cliff-face; very useful to understand such erroneous beliefs as "People ought to follow the advice I give" and the impact that believing that can have; anyhow my attempt finished with:
It appears certain that my prior prejudice about those involved in counselling others: that they are almost uniformly emotional and relational basket-cases, is perhaps simply a reflection of their greater ability to face and talk about their failings, rather than any incipient problem in those drawn to this field
Still developing that thesis, lets see how the marker takes it.
Ellis arrived to stay for the counselling corse; Tally baby-sat, set off, A14 closed due to fatality ... teacher absent; drew fun pictures on the flip-chart for a while and returned. Up late talking to Ellis.

2010-11-11: Thursday.

Up early, dropped babes to school (pouring with rain). Back to prod mail, out to collect my missing build machine from the local PC doctor's - fixed a LibreOffice build bug.
Lunch; tried to clear out the backlog of interesting looking patches. Out to a dental appointment - apparently epoxy filling of fissures is the future: certainly quick and cheapish. Sync call with JP.
Tried to get to the Counselling coursework; but failed, bed instead.

2010-11-10: Wednesday.

Slept fitfully during landing, worked on the trains back to Cambridge. Bus to TEAM, and got on with studying the Bible through a haze of tiredness.
J. picked me up, home at last - re-united with wonderful wife, and flurry of small babies; set a master build going, dinner, and put them to bed.

2010-11-09: Tuesday.

Up early, bid 'bye to a much loved Brother, and kind host. Trokked off to the office, prodded mail, out for lunch with Miguel & co.
Rob and sub-team popped in for back-massage and back-flipping machinerisation; got to the mail and blog backlog eventually. Off to the airport, watched Prince of Persia (surprisingly good) on the plane & slept.

2010-11-08: Monday.

Off to the GNOME Summit, caught up with JRB and Christopher. Had a fine tour of the media lab by our friendly GNOME Bengali translator, pleasing to see GNOME running on the "print your head in chocolate" controller demo - if not elsewhere. Lots of interesting ideas on show.
Out in the evening with Federico & Aaron to see Laura, Lucia and Miguel, and out for dinner with the same later; back to watch a silly movie with Thomas for a bit, slept.

2010-11-07: Sunday.

Up in time to discover the hour has changed, and I'm an hour early for Church, prodded the web instead. Off to Missions Sunday there, really great to see Nancy from Wycliffe in Cameroon (where my wife served, short term), and several others.
Chatted with the babes back home; lunch with Thomas, wandered the electronics shops - marvelling at the frightening realism of the immersive first-person-shooters that we have today.
Back, via the summit to Thomas', admired some Ted talk on bonobos; and out to an Ethiopian restaurant with T, B. and Federico. On to the 'Hendrix Experience' - not packed, and with a rather middle-aged bent, but interesting music. Bed late.

2010-11-06: Saturday.

Up lateish, wandered into town, met up with Dave & Srini for lunch, and off to MIT to chat to the mob; great to see Ted again. Back to Thomas', and on to Becky (& Clay & John's) for a lovely dinner, great to see her place & relax.
Back for a little mail & hackery in the evening, while T&B enjoyed a Divali party; sleep.

2010-11-05: Friday.

Up early, to the venue; enjoyed a useful discussion between Lennart and Arjan, chewed over bootchart2 for dracut with Harald. Disappeared to finish slide-ware over lunch.
Gave my talk on LibreOffice (slides). Out to the final party - lots of fun, met interesting new people - dinner with Allison from Canonical, and Chris from TI - enjoyed the hospitality,
Met up with JRB, and off with James to Finale, via lots of tramping through streets waiting for a table, for a fine port and pleasant desert nibbles dish. Back to bed late, tired.

2010-11-04: Thursday.

Up lateish; breakfast, off to the venue; enjoyed the hall-way track. Prodded at my talk, conf call in afternoon.
Caught up with Federico and Rob over dinner at CBC in the evening, then back to Porter for late-night slide creation frenzy.

2010-11-03: Wednesday.

Up at 6:30, clearly sleepig on the plan is an excellent plan. Worked through some of the mail backlog while T. slept. Off to the conference via getting the bank to enable my card, an ATM, and a taxi (arriving somewhat late for Jonathan's talk sadly).
On to the init-track-of-awesomeness - which went rather well, gave my bootchart2 talk (slides) and showed off the all new (on the flight) kernel startup tracing.
Really good to see lots of old friends in the audience, great to see systemd presented well; lunch with more friends. Nice.
To the desktop track with Federico, then long conf-call with the LibreOffice guys; met BDale in the lift and out for dinner & ice-cream with Keith and himself.
On to the Aquarium, for some drinks, nibbles, and to admire Linus' scuba-diving skills in the main fish-tank. Thomas and Becky arrived later; home with them, bed.

2010-11-02: Tuesday.

Up, breakfast, piano & violin practise with the babes; prayed - off to the train. Read and sent mail there via the miracle of 3G goodness. Met an interesting girl - Becky at security - a trancendental therapist trainee off to India for the first time; discussed life, theology etc. with her and a passing Ben, also off to the plumbers conf.
Really sad to see Stormy go from the foundation, she will leave a big hole; I wish her all the best at the Mozilla Foundation.
Interested to read Roberto Galoppini's blog offering buckets of technical advice. Personally, I like Roberto, who has many skills, and much to offer in many areas; however I don't believe that details of coding and growing a hacker community are core competances; lets look at some of the detail:
- Concerns about code cleanup: - He is concerned about removing the fourth(!) boolean type (FASTBOOL) a change that is already included in LibreOffice, kindly done by a contributor (Norbert) - who has since proved to be a super-star: removing the obsolete VOS library, and digging 'Solar' mutex threading nasties out of the code. Is it really -so- dangerous to remove a single obsolete boolean type ? is it bad and silly to just clean up old code ?
- Problems merging up-stream changes: - Roberto makes a good point, that the more changes we do, the harder it is to merge Oracle's contributions purely automatically. Having said that - this is true of any change: and being restricted to only small localised changes was a major failure of the past we don't wish to repeat. Better to have more contributors, than restrict and exclude useful changes in my view. Finally, automatic merging should not be the norm no matter who contributes the code.
- technical debt: - this is a great concept and a good point to raise: Roberto says: coding it is good only if it doesn't increment project's "technical debt". Completely agreed ! However there is an obvious conflict between demanding reduction in technical debt, and criticising cleanups of dirty code. The blog on technical debt, has this good advice: "Refactoring is like paying down the principal" - agreed, but if even the most basic re-factoring: operations such as cleaning up legacy cruft (like FASTBOOL) are a cause for concern, what then for anything bigger ? The argument is self-stultifying here. More good advice might be: "Maintaining an application without any unit tests is like borrowing money each time you add or change a line of code." Problem is OpenOffice has few-to-zero working unit tests. LibreOffice is making adding them a priority, despite the structural problems, and we already have some working properly. Personally, I believe LibreOffice has inherited a vast technical debt from its parent, that we are finally, starting to pay down. A good blog though, worth reading.
- Copyright assignments: - Roberto says: "While individuals may prefer to avoid the burden of copyright agreements, corporations and companies tend to like them more." - to which I can only respectfully disagree. I am sure that some individuals can cope with assigning rights - since they don't (yet) see why that is dangerous. Corporations on the other hand are really waking up to the problems here. Of course, in general, they desire other people to assign all of their rights to -them- - that goes without saying; but assigning their rights to others: that is a much, much harder sell.
- Corporate domination: - again there are real risks here; it is fairly clear that Oracle dominates OpenOffice.org - to an extraordinary extent. By having many more companies and (ideally) encouraging and empowering masses of volunteer contributors to dilute them - I hope we will avoid this risk for LibreOffice. Having said that - creating a structure where technical decisions are driven by FUD from non-developers is not a desirable outcome.
Clearly Roberto makes some good points about self selection of the LibreOffice steering committee - bootstrapping is hard, as is working out whom to include so that things can be moved quickly (StarDivision are past masters at endless rounds of stalling - eg. the 'Corporate Advisory (Delaying) Board'). Thus far we have made a good start in my view. We will have elections and a membership in due course. This however is a fairly normal process - many projects start out as an informal coalition of friends, and like minded people, and eventually incorporate into something formal. In summary some reasonable points amidst some less reasonable ones.
Eventually arrived in the US, after hacking up a nice new kernel boot-graphing feature for bootchart2, and prodding at my slides for tomorrow. Chatted with Thomas, and got somewhere to sleep setup; bed.

2010-11-01: Monday.

Up late; prodded mail, Caolan been doing an amazing job reviewing and merging patches over the weekend; read and commented on various fun packed discussions.
Lunch; annual review with JP - interesting opportunity for introspection. Clarity, fdo account admin, backed up the laptop in preparation for Plumbers / GNOME Summit travel tomorrow - hopefully flight restrictions are not too awful.
Heard rumours that PLA warps less than ABS when extruded; perhaps this was my mistake - thinking that it was harder than that. Printed out expense reports, and collated for posting.
Worked late, and packed for tomorrow.

My content in this blog and associated images / data under images/ and data/ directories are (usually) created by me and (unless obviously labelled otherwise) are licensed under the public domain, and/or if that doesn't float your boat a CC0 license. I encourage linking back (of course) to help people decide for themselves, in context, in the battle for ideas, and I love fixes / improvements / corrections by private mail.

In case it's not painfully obvious: the reflections reflected here are my own; mine, all mine ! and don't reflect the views of Collabora, SUSE, Novell, The Document Foundation, Spaghetti Hurlers (International), or anyone else. It's also important to realise that I'm not in on the Swedish Conspiracy. Occasionally people ask for formal photos for conferences or fun.