Ansuz

IDSgrep 0.1

I've just released the first packaged version of IDSgrep, which is an implementation of the ideas I posted last month about Ideographic Description Sequences. It's meant to bring the user-friendliness of grep to kanji dictionaries. Compiling it will require the usual Unix tools, and using it effectively will require a copy of KanjiVG, but you can look at the screenshot of it in action on the SF.JP site.

It'd be really nice if I could publish a paper about this. I'm looking at some academic-type computer science conferences, but it might actually be more on-topic for the more industrial or open-source type of meeting. If any readers have suggestions on what might be a good venue, I'd like to hear them.

SOPA/PIPA protest disappointments

As you probably know by now if you live under a rock and get all your news through the Net, several popular sites are protesting current US proposed Net censorship laws. I'm glad to see that happen, and I'm glad that a lot of people are paying attention, and I don't want to understate how glad I am of those things. But I'm also disappointed by a lot of what I'm seeing, too.

Distributed version control is not my favourite technology

Not too long ago a free software project I'm peripherally involved in decided it was time to replace its old and not broken version control system with something new and broken, and the lead maintainer conducted a straw poll of what the new system should be. My suggestion of "anything, as long as it's not distributed" was shouted down by the chorus of "anything, as long as it's distributed." Having lost the argument in that forum, I'm going to post my thoughts on why distributed version control sucks here in my own space where it's harder for me to be shouted down.

2011

It's the end of 2011, and I'm writing this from my parents' home in Nanaimo, where I'm visiting over the year-end holidays. If you ask me how this past year has gone, I'd have to say it's been mixed. Some good things have happened; some not so good; and my current situation is what I'd call metastable.

Ideographic Description Sequences: some thoughts

I went through a bit of a crunch to get Tsukurimashou 0.5 out the door before my year-end vacation. With that done, and at least 99 kanji to do before the next planned release, I have a chance to sit back and think about some longer-term and spin-off projects. Here are some ideas on kanji searching.

UPDATE: A prototype implementation of the system described here now exists as part of the Tsukurimashou project, and you can check it out via SVN from there. Packaged releases will be available eventually.

Tsukurimashou 0.5

I've released a new version of the Tsukurimashou fonts (project home page). This one contains 776 kanji, including all those taught in the Japanese school system through Grade Two and half of Grade Three. The bigger news, however, is that I've also added a set of fonts for the Korean hangul writing system. Those should now be beta quality - you should now be able to write the standard modern Korean language in its entirety. Downloads: source code; precompiled fonts; demo PDF files.

These fonts are far enough along now that I'd really like to create a bit of "buzz" around them; that's part of the sneaky plan behind my recent technical postings about my experiences building the fonts. I'm hoping that a lot of people will read those, and, especially, share them on systems like Twitter and the other one. In the new year, after I've posted a couple more (I'm aiming for weekly technical postings), I'll evaluate whether they are attracting third-party traffic and whether I want to continue them. They take up time I could be spending on writing code, but having people use the software is important too.

Building a build for something weird

Here are some thoughts on the Tsukurimashou build system. You can find the code, and some documentation of how to use the build system, in the package, but this posting is meant to look more generally at some of the issues I encountered while building a build for something weird.

The thing is, Tsukurimashou isn't a piece of software in the normal sense, but a package of fonts. It's written sort of like software, using programming languages, but the data flow during build doesn't look much like the data flow during build of the usual kind of software package. As a result, although it seemed like using Make was the thing I wanted to do, the way I've written my Makefile doesn't look much like what we might expect on a more typical software project. Working on it has forced me to see the structure of the project quite differently from the way I'd usually look at software, and maybe some of the ideas from that can be applied to other things.

On begging

December 2011 on Earth, but it is eternal midmorning on the third layer of the Astral Plane. THOMAS OF AQUINO, NICHOLAS FLAMEL, and K'UNG FU-TZU sit at a card table, in that order clockwise around the table. At the fourth, otherwise unoccupied, spot sits an ominous blue-painted Chinese porcelain ginger jar.

Code refactoring by combinatorial optimization

I encountered an interesting problem on the Tsukurimashou project recently, and some inquiries with friends confirmed my suspicion that if anyone has solved it, they've done it in a language-specific way for ridiculous languages. It appears I have to solve it again myself. Here are some notes.

Manitoba Writers' Guild opposes fair dealing

I just got a message on the Manitoba Writers' Guild mailing list that disappointed me; it was forwarded from the Association of Manitoba Book Publishers, and opposed Bill C-11's expansions to fair dealing. Here's the original message (I think that's a publicly available posting, but if not, let me know and I'll post a mirror); my response is below. I don't know that anyone over there will read it - I'm not sure who reads the mailbox that is the "From:" on these mailing list messages - but maybe readers here would like to see it.

Solved: Compose key doesn't work with SCIM, anthy, and Xorg

Posting this for the benefit of people who may have the same problem, and for my own future reference if it happens to me again.

The problem: Xorg (new name for XFree86) installation, configured with SCIM-anthy for Japanese-language input, and I want to also use the Windows keys as Compose keys when not in Japanese input mode. Compose and Japanese input both work in GTK applications, but in QT applications (notably, Konsole), Japanese input works and the Compose key doesn't. The Compose key is properly configured in KDE, xorg.conf (XkbOptions setting and UTF-8 locale), and xmodmap. If I disable SCIM entirely, then Compose works everywhere, but I lose Japanese input everywhere also.

Air Canada's bug letter

I got the bug letter from Air Canada in response to my complaint about the Typhoon Roke aftermath. They say they "regret" what happened, but they very noticeably do not acknowledge any wrongdoing at all on the part of any of their employees; they claim what was done was in accordance with their policy. They spend a few sentences chiding me for not giving them a cellular phone number - as if it were somehow my fault that they didn't keep me informed, and as if that had any relevance to the lack of announcements in the actual airport terminal - in the same paragraph where they explain that they would only have phoned me anyway if the cancellation were announced at least three hours in advance of the scheduled departure, which was not the case here. They conclude by offering me a discount code good for 15% off on a future booking. This isn't an apology because it lacks the critical defining features of an apology: it doesn't acknowledge that what was done was wrong, and it leaves open the possibility of doing it again.

Chessudoku (much) cheaper

Without my permission, Lulu.com changed the downloadable version of my book Chessudoku from a "download" to an "eBook." It'd still be a PDF-format file; the only practical difference is that they'd pay me a lower royalty. So in other words, they just told me (except they didn't really tell me, I had to find out when I logged in just now) "Hello, we are going to pay you less than we promised, thanks!" I think that's a violation of the spirit of their contract with me, even if they might technically have been within the letter of fine-print terms reserving the right to change their pricing structure.

I've removed the "eBook" version from their system. I've also lowered the price on the hardcopy version a little bit, to US$12.50. I'm considering removing that from their system too and terminating all business dealings with Lulu, but at the moment I don't have a suitable replacement and it'd be a shame to take the hardcopy version out of print entirely.

I'm posting the PDF version here on Ansuz for free. It is Creative Commons licensed. If you like it, please share it.

Tsukurimashou 0.4

I'm very happy to announce a new release of the Tsukurimashou font project, version 0.4. See files from this release on the Sourceforge releases page, and the summary for the project. The current glyph counts are 2021 overall, 573 kanji; this version now includes all the grade-school kanji through Grade Two. あなたもは、７歳みたく書きことができますよ！ (You, too, can write like a seven-year-old!)

Now, somebody restrain me before I add the 11172 Korean hangul syllables.

Security issue

Folks, we had a break-in on this Web server. I think it's cleaned up now, but if you have created a "visitor" account (I think very few people have) it would be wise to change your password just to be on the safe side. Discussion below.