Previously, I wrote about how to convert SVN-tracked packages to Git. Unfortunately, not all packages can be converted just like that. For instance, we’ve only recently imported python-docutils into the Subversion repository of the Debian Python packaging team, and I am not going to push Git on those people.

But there’s git-svn, so I don’t have to. In addition to being an excellent conversion tool, git-svn also allows you to track and interact with a Subversion repository.

The challenge in this case is that the python-docutils repository only tracks the debian/ directory, not the upstream code. This approach relies on svn-buildpackage to combine the upstream source with the checked out code to build, which works. However, it makes it unnecessarily cumbersome to work on e.g. Debian-specific changes to the upstream code.

Let’s see what we can do to improve the situation without influencing the whole team.

git-svn init --prefix=debian/ --stdlayout svn+ssh://svn.debian.org/svn/python-modules/packages/python-docutils python-docutils

git-svn init -R upstream --prefix=upstream/ -T trunk/docutils svn://svn.berlios.de/docutils python-docutils

git config svn-remote.upstream.tags 'tags/*/docutils:refs/remotes/upstream/tags/*'
git config svn-remote.upstream.branch 'branches/*/docutils:refs/remotes/upstream/*'

git-svn init -T trunk/docutils -t 'tags/*/docutils' -b 'branches/*/docutils' svn://svn.berlios.de/docutils .

git-svn fetch --all

NP: Queen: Queen II

I recently asked for input on document management systems and received a good number of replies from people. Thank you!

In the following, I present my evaluation of each of the systems in the order in which they were suggested. In all cases, I hope to be shown wrong!

Plone

Having been involved in Plone development for several years myself, I was a bit surprised when two people suggested it for my requirements. Plone to me is a content management system, not a document management system. It’s optimised for portals with members publishing content on webpages, along with the occasional non-HTML file. Elaborate and advanced workflow definitions regulate the cooperation between contributors.

Here are the relevant points I found:

• Its default storage (ZoDB) is excellent for content, but for plenty of large files, one needs to set up external storage. Alternatively, a tool like Reflecto can be used, but it requires reindexing of the Plone catalog whenever files are manipulated directly and not through Plone. I have not tried this, but it sounds like it’s the wrong tool for the task.
• Objects can have keywords (i.e. tags), but those cannot be defined on-the-fly, nor can they be edited via external storage or WebDAV. An ontology product as well as a keyword manager exists, however, and something could be implemented using lovely.tagging or similar
• Plone uses AJAX very sparingly, and not at all for file management (although this is in preparation). So moving files between folders is painfully slow, all the more if you have to do it in bulk.
• Bulk upload/download support seems to be available only via Flash or Java.
• Even though Plone 3.0 supports revisions and version control, it uses an exclusive-access-scheme with locks, and lacks any form of merging support (binary versioning). Locks can be broken by others.

All in all, I am confident that a document management system could be implemented on top of Plone, but it would be quite an involved task, and the result still suboptimal.

Thanks to the folks in #plone/irc.freenode.org for their help and pointers.

Twiki

Twiki is advertised as an enterprise wiki, and even though it supports batch uploads, its intended use case is to be a wiki: documents only exist in the form of attachments to “topics” (wikipages)

Nevertheless, it’s worth to mention that attachments are stored on the filesystem and can be directly manipulated, Twiki will automatically detect changes. Also, a WebDAV plugin exists, but it’s not been ported to Apache 2.0 yet.

In conclusion, I don’t see Twiki as suitable for the job. However, I have been referred to someone who is using it as a DMS, and I am waiting to hear back from them.

KnowledgeTree

There are two things I noticed almost immediately when I surfed to the KnowledgeTree website: right at the top, it says “Open Source Document Management Software” and “Try It / Buy It” next to it; well, I can deal with that, but then, on the product overview page at the bottom, I see them advertising “leverages best of breed technology: PHP, Java, … MySQL” and I suddenly feel very uneasy. It’s not news that PHP and MySQL are very often chosen for projects, but to call them “best of breed” is like advertising your car to be built of the “best substances in the world”: wood and spit.

NP: Pulp: This is Hardcore

You are probably using DHCP on the machine currently in front of you. The “Dynamic Host Configuration Protocol” is a way for your computer to obtain an Internet address from a pool of available addresses, and to return it to the pool when you no longer need it. Basically every Internet service provider uses DHCP, or something similar.

As a network operating system, Linux has DHCP support (and had it for ages). In true Unix fashion, Debian sports at least four DHCP clients. Debian’s default is dhcp3-client, also known as dhclient.

The theory is that the client requests an address lease from the server and periodically renews it. This process yields a number of events, to which the operating system can react. For instance, initially, the client issues a PREINIT event to get the interface into a state where it can talk on the network, and a BOUND event as soon as it acquired a lease, or FAIL if it, uh, failed.

After a certain period of time, the client tries to renew the lease. If it succeeds, it issues a RENEW event; if it fails, it yields EXPIRE.

So much for the theory.

It seems that dhclient is rather stupid, which I tried to document in bug #459813 — it does things differently: given a lease, after a certain period of time, it just issues an EXPIRE event, which causes the operating system to deconfigure the interface and take down connectivity. Then, the client spits out a PREINIT event, followed by BOUND or FAIL, as appropriate.

I have not quite investigated what all this means, but this much is for sure: periodically, your machine goes offline, only to come back online a second later. If this were Windows, one would probably knock on wood and be glad that it works at all. But we’re on Linux here, Debian even, so this cannot be.

I’d love to be proven wrong, so if you have a minute, please try to verify. One way of doing so is to insert

echo "$(date) got $reason" >> /tmp/dhclient-script.reasons

towards the top of /sbin/dhclient-script and monitor the output file. Once your client renews, it should read:

Wed Jan  9 11:53:27 CET 2008 got RENEW

but instead you’ll see

Wed Jan  9 11:53:27 CET 2008 got EXPIRE
Wed Jan  9 11:53:28 CET 2008 got PREINIT
Wed Jan  9 11:53:33 CET 2008 got BOUND

and if you look closely enough, your interface will be unconfigured those seconds between EXPIRE and BOUND.

I only spent about 2 hours with dhclient last night, as part of my efforts on netconf, but I already found bug #459813 and bug #459818, two bugs which don’t suggest that dhclient is quality software. If I find more, I’ll have to consider a different client, hoping that there are others which separate the DHCP client from the operating system calls in a similar way as dhclient does.

NP: The Flower Kings: @Live Recording, Uppsala City Theatre, Sweden, 10 February 2003

The theory of reasoned action(TRA) was first proposed by Azjen and Fishbein (1975 & 1980)

So it was first proposed in 1980? Or in 1975?

Endless lists of references, unrelated. Bloat.

Passive voice.

Single-author papers speaking of “we”.

“The qualitative interview is used in qualitative research of all kinds, whether positivist, interpretive or critical. It is used in case studies, in action research, in grounded theory studies, and in ethnographies (Hesse-Biber & Levy, 2006; Klein & Myers, 1999; Myers, 1997, 1999; Northcutt & McCoy, 2004).” — pointless, contentless, waste of paper

This study aims to address these issues and answer the above question through a number of objectives

Studies aim? You answer through objectives?

Conboy’s thesis 07 18:35 < madduck> and if (Smith 2003) already said what you say, quote it in full, leave it the fuck out, but do not do: “It has been shown that foo (Smith 2003) and that bar (Smith 2003). Furthermore, it seems that water is wet (Smith 2003). Finally, some concerns have been raised that moo (Smith 2003) 07 18:35 < madduck> especially not if Smith 2003 is a fucking 700 page book

Klein and Myers (1999) recommend that “the intellectual basis for the research should be as transparent as possible to the reader”.

””“Cooper and Schindler (2003) recommend that, particularly in instances where the study is broad and exploratory and where limited research currently exists, it is vital that the researcher parses the research project into a set of clearly defined steps.”“”

—> and I need references for that???

NP: And all because the Lady Loves: Sister Bridget

I must have generated this list in a hurry, for I surely left out quite a number of bands:

• Rocket Scientists

• Queensrÿche

• Neal Morse

• Barclay James Harvest

• Negua

• Oceansize

Also, several people have responded with recommendations:

• Cave in

• Seachange

• Agalloch

• Green Carnation

• Fates Warning

• Pain of Salvation

• Evergrey

NP: Rocket Scientists / Oblivion Days

I will never fly through or into or out of London Heathrow ever again. Excuse the following rant, I need to vent. And I am not going to do so in well-mannered words.

I had a three hour layover there today. Of those three hours, I didn’t get to sit down at the gate (or in a pub) to wait five minutes because I was busy waiting for the bus to take me from terminal 4 to terminal 1 (a total of 1:10 hours time lost), spent another 70 minutes being scrutinised — or rather treated as a terrorist — by security, and scuttled through the endless corridors and mazes that make up this shithole of an airport.

I kept a tally: not counting repetitions by the immensely annoying PR announcements, my ears had to absord the word “security” 16 times during that time; “regulations” topped that with 23; I had to show my boarding pass 4 times, and my passport 5 times. I had to take off my shoes three times, completely unpack and repack my bag twice, and twice had to explain my intentions for (daring to) travel to Ireland.

But by far the worst were those 70 minutes of scrutiny behind the curtains. It all started in the regular security queue, when, after I had taken off my shoes and belt, emptied my pockets, saw my bag scanned twice, a security guard walked up to me and started to search me. He didn’t say a word, he didn’t excuse himself, suddenly he was just searching me and doing so not exactly in a very respectful way.

I then made the mistake to ask him to give me a break, out of surprise — the guy was giving me a treatment you’d usually expect from small-penis US cops getting their daily adrenaline just before hand-cuffing (or beating) a shoplifter. I was startled for a moment, but a second later I realised I should have kept my mouth shut. It was too late, he had already signal two others and before I knew it, I was escorted to their office where they took my data and announced that the officer had the choice to press charges against me for abusive behaviour. Hello, pot? Yeah, kettle on line two…

I asked what the reason for the charges were and was told that “these honorable people ensuring everyone’s security at the airport have a right to their own security, and thus any form of abuse of violence could not be tolerated.” It was almost like in the movies, and so I wasn’t surprised when noone wanted to hear my story. I inquired about the guy’s name and was refused.

At that moment I went into a trance and just started smiling, kept my eyes closed most of the time and answered their questions. For one, it seemed like the thing to do; in addition, it was probably the only way I could keep calm while exposed to this fucking security theatre, performed by a bunch of monkies in human clothing with a collective IQ below room temperature (Celsius, mind you).

This wasn’t the first time that I had been searched and questioned and treated like a potential terrorist (if they keep doing this, I might just become one with the expressed goal to shove a cricket bat up the rear end of all those responsible for the fucking circus that airports have become these days). However, this time was by far the worst, and it wasn’t even within the US — but then again, the UK isn’t too far behind, I guess.

Someone previously agreed with me that Heathrow is a dump, but that Geneva weren’t any better. I had to disagree, for in Geneva, and every other airport I’ve visited so far, there was at least a minute display of decency, sometimes humour, but in all cases politeness. What happened today at Heathrow was everything but, rather like the opposite of what I have come to believe would be the English Way.

I will not touch ground at Heathrow ever again, and I urge everyone to do the same. Stansted, Luton, London City — they may not be all that great either, but they aren’t Heathrow.

NP: Agalloch / The Mantle

top-aligned documents have the top 3mm cropped in the scanned image

Dear Lazyweb: I am looking for a command-line tool to rip DVDs into .avi or .mpg files, and I want the subtitles to go into a separate file such that the movie itself is not tainted by the subtitles.

I looked at mencoder and transcode, but neither seems to do what I want, or at least it’s not obvious from the manpages.

Does anyone know of a tool, or could tell me how to make mencoder or transcode do what I want? Let me know, please!

NP: Green Carnation / Light of Day, day of Darkness

last 10 questions answered negatively

• migrate subtree from one repo to another
• undo a commit
• permanently remove a file
• have to update to get new log entries

Following the week of Debconf6, I was one of eight leaving Oaxtepec for Mexico City, looking at several days to spend at will in the country. Some of us went by bus, others with a van to the airport to help one of our colleagues, but despite the two different routes, we arrived at the Hostel Moneda almost at the same time, checked in, had dinner, then ventured to the big square close to our hostel for an open-air concert by \Tijuana Nortec Collective\_, probably organised as part of one of the many protests going on in the city. As we could not find any drinks at the venue, we eventually wandered off to end the night over a last beer on the beautiful roof-top terrace of our hostel.

With big plans involving a rental car and driving it to Oaxaca on Monday, our group of four rose (almost) early, found and said goodbye to Michael Banck, who had come to the city with us the night before (he continued to refuse to join us, muttering something about having to catch a flight or a similar petty concern) and were about to head out the door if it hadn’t been for Micah Anderson trying to catch our attention by whispering three times in short succession that “Jeroen is still asleep”. This registered as “we have to go to his room to say goodbye” with Biella, Vagrant, and myself, so it came rather unexpected to hear Michael claim that Jeroen had in fact planned to come with us (and told Micah but none other in the group). Flexible as we are (and quick as he is), this didn’t pose much of a problem, and before too long, five geeks left the hostel with backpacks for the metro for the car rental company at the airport. Next time, though, Jeroen… do tell a little in advance, and maybe to someone else but Micah.

When we were finally quenched in the car — I still didn’t get a new driver’s licence after having \lost mine in Thailand\_, so I had to submit myself to the driving custody of Biella and Micah — we brushed by the local authorities and reached Oaxaca, a “romantic town” about 500km south of Mexico City, long after dark. With some difficulties, mainly due to the yearly teachers’ protest for higher salaries, who were all camping out in the streets, we found \our hostel\_ and ended up in bed soon thereafter.

When on Tuesday the five geeks that we were finally managed to commence roaming the city, we had “fixed” the hostel’s computer by popping in an Ubuntu Live CD, thus made friends with the other guests, eaten breakfast, written plenty of blog entries, and waited for the hotter hours to pass. The afternoon was thus spent ticking off the main tourist sites (such as Santo Domingo) and browsing around the old market. I am not that much of a market lover, but this Mexican market, although similar in spirit to the various Asian ones I’d seen in the last few years, was refreshingly different, mainly in stock and ambience.

Wednesday brought the first complications within our group, but in retrospect it’s difficult to say what exactly went on. The plan was to touch down at a large, traditional market that kept switching locations with days of the week, but due to some moodswings I ended up being the only one listening to the directions (in Spanish) and directing our crowd to the north of the city, where noone knew of a market on Wednesdays, only other days of the week. While I was still trying to cut the losses and find an alternative, my group had decided to return to the hostel, which, given my mood at the time, seemed like a welcome way to get away for a couple of hours to sort out my thoughts.

Even though the five of us got along pretty well, I just do not like travelling in groups. On the one hand, everything just takes way longer, thus leaving less time in the day to do stuff. On the other hand, the dynamics are fragile, especially if the members do not really know each other, which was our case. We had too much fun to warrant bad feelings along the lines of “I wish I had gone alone”, but I feel yet again reassured that I am made for travels in groups of two, at most.

In any case, Biella, Micah, and I later XXX Most of the afternoon I spent doing what I always do in foreign cities if I can: trying to get myself lost, which is near impossible in Oaxaca. Accepting the grid structure of the roads, I found a cute café and enjoyed a cappuccino, then headed for the hill in the northwest of the city, where I strolled between the observatory, planetarium, and amphitheathre (all of which were closed), finally ended under a tree to read Haruki Murakami’s “Norwegian Wood” a second time around, and eventually linked up again with the rest at the hostel towards the evening.

I did not like “Norwegian Wood” the first time around, but in Myanmar, talking to the Portuguese traveller I met there, I could not explain what it was that I disliked. Thus, I promised to reread it and to try again to describe my problems with the text. Great reason to read a book, eh? Let’s see how long I last (in the mean time I have finished and \reviewed it\_).

difference growing bigger.

mark is doing an experiment, if it does not work out, he’ll just return to his money which cannot leave africa. jane silber is in charge of canonical, so she’ll be responsible.

ubuntu currently financed through interests only, no risk.

mark is about business, about marketing. it worked, ubuntu is a hype and he hasn’t spent much on marketing, it was all mouth to mouth propaganda.

mark is hbd, global venture capitalist (is it really global?)