The LilyPond Report #8

Welcome to this eighth issue of the LilyPond Report!

This week’s issue tries to deal with several questions raised last week. Trevor Daniels, who’s our guest again, will introduce us to tap-dancing and to a 6-millions-euros beer bottle. We will also discuss bagpipes, RSS feeds, feature-freezes, Slashdot®-justice, and finish our journey through commercial music softwares’ websites. Finally, be warned — this issue includes a screenshot from Windows Vista. As always, you can post your comments at the bottom of the page, or even register and contribute to the LilyPond Report’s next issues.

 This Week’s Desultory Editorial

Greetings,
good news for the LilyPond Report: the number of readers has been multiplied by almost three in the past few days! Perhaps it was because we talked about beer... But there’s another explanation: as you may have noticed, the LilyPond Report is now announced on LilyPond’s home page. Thanks to Graham for having asked me to add this link!

A LilyPond RSS feed? Here you are!

While I was working on the LilyPond website, someone posted a clever comment to last week’s discussion about RSS feeds. There is actually a much more convenient way to keep up with LilyPond’s development; it is the official LilyPond announcement mailing list. So, I immediately added its RSS feed to the LilyPond webpage; it works well, but it’s just a beginning: John is thinking about cooking us a native RSS feed generator!

Anyway, many thanks to the reader who pointed me to Nabble’s feed. Please remember to sign your comments, otherwise I won’t know who I have to be grateful to!

Now for another subject.

You may think that some discussions on the LilyPond mailing list tend to get too long. Well, let me show you something.

On the Linux kernel mailing list, a 205 mails (and counting) long discussion has been taking place this week. Interestingly, it’s topic was something the LilyPond Report already dealt with several weeks ago,when Laura Conrad stated that “there should be a feature freeze some weeks or months before a stable release”.

Linus Torvalds

We already explained that given the lack of resources LilyPond development couldn’t afford such a thing. However, there’s more to it. While the Linux kernel project has much more resources and many more potential reasons to do so-called “feature freezes”, Linus Torvalds himself explained why he couldn’t agree with the idea of slowing down the development:

The notion that we should even _try_ to aim to slow things down, that one I find unlikely to be true, and I don’t even understand why anybody would find it a logical goal?

Of course, you will have fewer new bugs if you have fewer changes. But that’s not a goal, that’s a tautology and totally uninteresting.

A small program is likely to have fewer bugs, but that doesn’t make something small ’better’ than something large that does more. Similarly, a stagnant development community will introduce new bugs more seldom. But does that make a stagnant one better than a vibrant one? Hell no.

 [1]

You may already know that I am some sort of a Linus Torvald’s groupie, so I may not be in a good position to comment this. But I do think the guy knows what he’s talking about — and I also believe that, in spite of controversies, Torvalds is a true, faithful and legitimate advocate of Free Software.

 News from the Free World

These things happen: this week I was planning to discuss open-source hardware, but a piece of news came up that I needed to talk about. I’m referring to the end of Hans Reiser’s trial.

Hans Reiser and his attorneys

Who is Hans Reiser? Difficult to tell. Many of us know him as a talented Linux developer (you may have been using his filesystem, named ReiserFS). However, the whole community was is shock when, in 2006, his ex-wife went missing and he became the main suspect of her murder.

Nina Reiser’s relationship with Hans was complex and stormy. There were strong presumptions against him, and his behavior after her disappearance was at least questionable. Reiser’s arrogance and bad temper struck everyone, while some commentators noted he might have Asperger disorder. However, there were also unresolved questions concerning Nina Reiser: first of all, her body was never found — and it was suggested that she might be alive and back to Russia where she had a family.

Is Reiser guilty? Was he framed? I obviously wouldn’t try to answer this question, and neither would you. However, I’d like to show you something I got shocked on when the Reiser case was revealed. In early October 2006, this was on Slashdot’s front page :

This almost made me feel sick. (Granted, Slashdot is notoriously full of crap, and I’ve been completly boycotting it ever since.) What’s this, I don’t know this Reiser guy or his wife at all, neither do you, and yet you’re demanding me to pronounce him guilty or not? Like if justice was a game, some sort of a stupid slashdot “poll” no one gives a shit about?

...

I then followed the trial with great attention. And a question came to me: how different were exactly these jurors from slashdot’s readers?

Answer: there’s actually no fundamental difference. In the end, it all comes down to two boxes: “did it”/“didn’t do it”. How can you choose which box to check? “Hmm, he’s a Linux developer, he seems to be a nice guy...”? Sure, you could argue that the jurors have been studying, investigating the case; they got to know the defendant — as much as one can know anyone.

Guilty or not, there was no strong evidence against Reiser, as David Kravets noticed: “no body, no crime scene, no reliable eyewitness and virtually no physical evidence”. In a democracy, there should be no way a man, even a guilty man, could be incriminated based only on circumstantial evidence — or else we could just get rid of trials).

But... Reiser shot himself in the foot. In a 11-days-long testimony, he managed to ruin his whole defence, by being more rude and despising than ever. There was nothing his attorney could do, as Reiser kept on interrupting him and quarelling with him.

How, in the end, did the jurors choose which box they would check? Exactly the same way any random slashdot geek would. “This man is obviously an arrogant jerk; we don’t care that there’s no evidence, let’s just convince him, that’ll teach him”.

Eventually, I’d like to quote a reader’s comment found on the Memeverse blog:

Arrogance usually isn’t penalized with a life sentence in prison. If you’re rich and/or connected you can get away with it. Sometimes, you don’t. At the minimum, Hans did something really, really stupid.

 What’s up with LilyPond?

This week, our visit to Finale’s and Sibelius’ websites is coming to an end — and for the occasion, I’d like to extent this overview to other music software websites such as Encore’s or Igor’s.

All these sites have something in common: the testimonials page.

  We already mentioned (and laughed at) the “I made the switch” testimony on Sibelius’ site. But that probably wasn’t enough: in addition to that, there’s also an awards page, a magazine quotes page, a famous composers page — gee, they even have Steve Reich, for frak’s sake — and two pages dedicated to former finale users. http://www.sibeliusmusic.com/

  On Finale’s site, things are done in a more discrete way. Granted, they couldn’t help claiming that “Finale is preferred by professional musicians and composers”, but no official testimonials page. They do host, however, an American composers forum I’m most suspicious about, a showcase service that is unarguably a good idea (in a trendy web-2.0 superficial way [2]), and promote their product through several conventions — I guess they can afford it, knowing how much Finale licences cost.

 However, Finale did have a testimonial page on the first version of its website, back in 2000; and I honestly can’t blame them for having removed it.

(That’s what puzzles me, with all these aggressive advertising techniques. The first year they tell you “hey, we’ve made this product, it’s simply perfect”; the year after they tell you “hey, we’ve completely redesigned our product: now it’s much, much better than before”, and so on.)

Closing Sibelius’ and Finale’s websites in my browser is actually a relief. Let’s now have a look at two other programs, less famous but more modest.

 GVOX Encore has this simple, clean website with a testimonials page entitled “From Our Dedicated Customers.”. It’s not more ridiculous than the previous websites. Notice, however, a funny detail:

gvox.com favicon
debian.org favicon

  Finally, Igor Engraver is “the next generation”, “the world’s most powerful tool for music notation”. Ehem... Although its website hasn’t been updated since the early 2000’s. They have, of course, a bunch of comments on their website. In one of these, we can learn that Igor was “designed by a distinguished composer, Peter Bengtson” — a funny thing, as one of our most prominent LilyPond gurus is named... Mats Bengtsson!

Igor was coded in Common Lisp for MacOS 8 and Windows NT. It actually reminds me of some design concepts behind LilyPond. They emphasize the quality of its spacing engine, the ability to customize every detail, the support for both contemporary and ancient notation... Please have a look at some of their demonstration videos, and let me know if that looks anyhow familiar.

I hope that this software is not dead yet; and if it is, may the developers have the good idea to open their source code... There seems to be some funny guys among them: for instance, have a look at their “page not found” error message (click on the robot to read it).

Last but not least, LilyPond has its own publications page, as well as some testimonials by many well-known contributors/developers:
 Orm Finnendahl (with a nice picture),
 Kieren MacMillan,
 Graham Percival (though there’s no “bloody” or “mao” in his sentence, it’s definitely his style),
 Nicolas Sceaux,
 Werner Lemberg,
 and many others...

 The LilyPond Feature of the Week

OK. For the past few weeks we’ve talked about features that commercial software have and that we also have (any questions?); however, most of us never regarded LilyPond as a clone of commercial software (while the opposite can be true, as we demonstrated last week). Being Free Software, LilyPond can include some features that nobody would expect from a commercial product.

See how much LilyPond takes care of exotic instruments, for instance (by “exotic” I mean “instruments that are not part of standard classical or pop formations”). Today: Bagpipes.

I cannot find exactly when support for bagpipes was implemented, but it was quite early — in 2000, 2001 maybe? While it initially had been contributed by Andrew McNabb, it was then greatly improved by Sven Axelsson — which makes me discover there are bagpipers in Sweden too (yes I should have known; but hey, I’m a pianist, see what I mean?).

“Amazing”, isn’t it?

Translating the LilyPond Documentation made me completely discover the bagpipes-specific music language, and also gave us a pretty hard time: for example, how would you translate such words as piobaireachd?

 The Snippet of the week

This week’s snippet was (not) added by Jay Anderson, after a long quest, and is at least as cool as last week’s snippet.

(By the way, if you happen to ever write such a cool function, pleeease remember to post it on the LSR — I just despair when I imagine how much nice pieces of code were buried over the years in the mailing list archives...)

Here’s what Jay’s code does: feed it with a simple melody, and it will automatically double it with octaves...

(I don’t know what Jay’s instrument is, but I guess it would have been quite masochistic of him to invent such a macro if he was a violin player )

Well done!

 The Interview of the Week

Last week’s issue was about beer and particle accelerators; I blame myself for having forgotten the most interesting part: you have to know that without beer, there would actually be no particle physics! Indeed, the Bubble chamber that is the most amazing part of accelerators was invented by Donald Glaser after having popped the cap... of a bottle of beer.

There’s more to it. I remember that a decade ago, two beer bottles had caused several experiences to fail in the largest particle collider in the world, and a loss of more than 6 millions euros!

I had to ask Trevor, our man in Geneva, about the story.

Here’s the story from a colleague of mine:

these were found after a routine shutdown of LEP [...] Clearly they had been put there deliberately and as far as I remember it was thought to be an employee or contract worker with a grudge but I don’t think that the culprit was identified.

LilyPond Report — Okay, enough beer! Let’s now talk about your musical life.

Trevor Daniels — When I was a child, my mother had a piano, and I taught myself to play to an elementary standard from her sheet music of 30’s and 40’s popular songs. I joined the parish choir in our home town as a treble, and took part in several school productions of Gilbert and Sullivan comic operettas. At university I joined the G&S society, and shortly after the local choral society. I’ve sung in a choral society, often more than one, all my adult life, so I’m fairly familiar with the standard choral repertoire.

Penny Shaw and Trevor Daniels

Later, Sue and I joined Opera Exchange (now Oxford Touring Opera), an amateur society based in Oxford which used to put on fully staged productions of grand opera. I used to maintain its website; if you hunt around there you’ll find my wife, then called Sue Midwinter, as Martha in Faust. I sang with them too, mainly in the chorus, but once, memorably for me, as Baron Duphol in La Traviata — here’s a picture of me and Penny Shaw as Violetta in the changing rooms (she’s now a professional opera singer).

L.R. — Sue’s description on the website is really funny, I invite our readers to have a look at it!

Trevor — We’re no longer involved in Opera Exchange as it became a bit too far to travel when we moved to Middle Barton; however Sue and I still sing with the village choir, with the Bicester Choral and Operatic Society (BCOS), and with the Stour Singers, a conventional choral society. Currently we’re singing Haydn’s Missa Cellensis (XXII:8) which actually is new to me, as well as the old favourite, Mozart’s Requiem. With BCOS we are currently preparing a concert version of Carmen, and in June we start on a stage production of Cole Porter’s “Anything Goes”, so I have quite a wide-ranging involvement in singing.

L.R. — Two weeks ago we were talking pedagogy with Francisco; what about you? Do you teach music in any way?

It is very important to encourage children to get involved in music. In BCOS we have deliberately chosen recent productions to involve children, and I’m also involved in helping two local schools with their productions of musicals. Most children don’t read music, nor do half the people in BCOS. They have to learn everything; this makes rehearsal times much longer, and it seems to be worse among the younger generations. I already teach tap-dancing in BCOS, perhaps I should instigate some music-reading lessons as well...

L.R. — Do you write and edit music scores?

Trevor — Some years ago, I produced transcriptions of various pieces of music for our local village church choir, using Finale PrintMusic. It was adequate for that, but when I tried to help a composer friend by type-setting his hand-written compositions, I found it was quite incapable of doing the job. He eventually got a grant which provided funding for Sibelius, but under the licencing terms I was not allowed to use it, and in looking for an alternative I found LilyPond.

L.R. — Mmm, I see where this is going Francisco told us that he found LilyPond “after a long search”; obviously you didn’t...

Trevor — I can’t remember any difficulty in finding LilyPond. AFAIR I just searched for something like “music & typesetting” or maybe “engraving” and it popped up. As my background was in computing I was immediately interested; I downloaded version 2.6.5 on 14 Mar 06 and I had my first pdf printed later that evening. It has remained part of my life ever since.

L.R. — I can’t believe you even remember the day where you discovered it!

Trevor — Like everyone else I’d used much of the software which is now freely available on the Internet, but this is the first open-source project I’ve been involved with, and I am quite pleased I’ve been able to pay back a little by helping with LilyPond’s documentation. It’s been fascinating getting to know people through exchanges of emails. It’s surprising how comprehensive a picture of people one can develop this way!

L.R. — Do you think that as a former programmer, you’ve been more easily seduced by LilyPond?

Trevor — Difficult to tell. Certainly I had no difficulty with line-mode input and the concept of compiling, but probably younger people do. I’ve tried to take some of these terms out of the LilyPond documentation. But it is quicker and easier to enter notes by typing than by fiddling with the mouse. When I used Finale in the days BLP I used the typing input mode.

L.R. — Which place does LilyPond take, today, in your everyday life?

Trevor — Well, my wife is convinced I’m a computer geek, as I spend so much of my time at the computer screen. I usually spend a couple of hours in the morning dealing with email, reading the postings on the various lists, and working on LP documentation. Afternoons I try to tear myself away and do something else (this is usually the time Graham seems to sleep, roughly during the morning in W Canada, so that’s quite convenient.)

L.R. — I know this rhythm too! So, you’ve been very much involved in the new Learning Manual and most sections of the Notation Reference; and at the same time I’ve been much impressed by the spectacular evolution in your knowledge of LilyPond; I guess there’s no need to ask whether these two things are linked...

Trevor — Absolutely! Writing the LM as I gained the understanding myself was a great way of capturing one’s own knowledge and passing it on to others at the same time. I don’t actually use LilyPond myself much any more, but I get great enjoyment from learning how it works and helping others.

L.R. — Finally, one embarrassing question (embarrassing for us, not for you obviously): are you really running Microsoft Windows Vista®?

Vista, msysGit and ConText

Trevor — Yes I am; as much of my work is exchanging documents and spreadsheets with non-computer-geeks using MS software, an MS-based system is the best tool for me. (Incidentally, MS Excel is one of the most surprising and best-written programs I’ve come across: did you know it will do regression analysis, t-tests, Fourier analysis, complex numbers, matrices, etc all out of the box?) LilyPond and my ConTEXT editor run fine; it’s ideal for testing the small snippets which are needed for documentation.

L.R. — You even once called yourself a “pioneer”, when you began editing source files under Vista...

Trevor — No one else in the developer/documenter team seemed to use MS software, hence the pioneering; the only tricky part was when I needed write access to the GIT repository; the most time-consuming was finding a means of generating an acceptable ssh key-pair. PuTTYgen generated a key-pair acceptable to Savannah, but I was unable to persuade ssh to use the private key it generated when pushing to git. Eventually I found SetupSSH381 would generate a key-pair acceptable to Savannah and useable by ssh.

L.R. — Well, thank you for having taken the time to answer our questions!

Trevor — Many thanks for an enjoyable chat.

Thanks to Trevor Daniels for this interview.

 The Quote of the Week

This week’s quote is not from Graham Percival (not that he hasn’t tried to).

Nicolas Sceaux introduced us to his own version of Graham’s mao:

Please, this LilyPond, not LaTeX, why in the name of Mandos would LaTeX commands work?

Somewhat refreshing, isn’t it?

And this concludes the eighth issue of The LilyPond Report.

Cheers,
Valentin Villenave