Slashdot Mirror
Rosegarden Developers Interviewed by O'Reilly
rayk_sland writes "Users of the Rosegarden Sequencer project will be gratified to see it featured in O'Reilly's Linux DevCenter web magazine. I am a devoted fan of this program, which allows the user to sequence music using classical music notation, and has many other sequencer features I haven't even properly fathomed (read the article.) The Rosegarden project has recently released a 'pre-1' beta. Almost time for those party streamers..."
Rosegarden looks absolutely fantastic. Unfortunately, its dependancy on KDElibs and QT make it impractical for me to build it in a reasonable amount of time (I'm on gentoo, so binaries for everything isn't really an option) and its layout makes it impossible for me to use it in ratpoison (my WM of choice) anyway.
Looks nice, but definitely could use some work in the dependancy and UI area.
Disconnect and self-destruct, one bullet at a time.
Thats is what i love about oreilly, they have so much good info.
Freedom or George Bush
...but the story submitter said so.
From the page...
Rosegarden is a professional audio and MIDI sequencer, score editor, and general-purpose music composition and editing environment.
What I just want to get are specs... How many tracks at once does it promise ?
Logic Pro does more than enough and Cubase was much slower.
Trolling using another account since 2005.
Harmony Navigator
I've tried Rosegarden a couple of times - looks to be a very promising package - i've even contributed some stuff to the project. Last time i tried it though (probably 6-12 months ago) it had a tendancy to die unexpectedly rendering it fairly useless for anything serious. Its good to see its now approaching a final release. I will be trying this out to see how its been coming along - and maybe ... just maybe I can actually write some tunes under linux...
Electronic Music Made Using Linux http://soundcloud.com/polyp
Wow! I haven't looked at the rosegarden project in a long time, but from looking at those screenshots, they have come a LONGGGG way. I remember when it used to just be a very simple notation program.
Open Source is getting there. Wherever there may be.
...for all those who've been told they were never promised Rosegarden.
I only post comments when someone on the internet is wrong.
Looks like Rosegarden can export to Lilypond, which is by far the best music notation program AFAIAC. For years in our choir there were sheets made using Finale, and when I remade one using Lilypond people were asking me where is the book that this came from, it just looks so professional. They have a great paper on this .
Brainless Manager: Where's my deliverables! We're past our due date!
Rosegarden Developer: It'll be done when it's done. I never promised you a--DOH! Uh, I'll get those deliverables to you by noon, sir. <grumbles>stupid project names</grumbles>
You see? You see? Your stupid minds! Stupid! Stupid!
I think the authors make the point that they want to hold themselves to the standards of hand-engraved scores, NOT computer generated scores. They specifically point out that they don't like most computer-generated prints.
NoteEdit is also decent for sheet music to Lilypond if that's what you're looking for.
Unfortunately the project is half-dead due to percieved disinterest. Doesn't make it less functional though.
Seems like a wise decision would be to follow Audacity's example and to release an OSX port...Windows would be nice as well, but I guess OSX is more likely. In all honesty, many musicians aren't going to want to take the time to futz with compiling relatively unproven software on Linux.
ZuluPad, the wiki notepad on crack
Mod me troll if you want (uh but flamebait preferred :) ), but i'm getting tired of Linux-only releases in a cross-platform world.
And considering that the majority of music users use, well, WINDOWS... (my dad always bought windows music products. I think Rosengarden is missing a huge potential market: Windows users.
I tried freeware Windows MIDI sequencers a couple of years ago, all sucked. So I'd gladly appreciate if the Rosengarden devs made their software cross-platform, now that it's STILL IN EARLY development. This is when major structural changes can be done with nobody noticing.
And finally... for those of you who want to flame me with "this is NOT a Windows thread", I respond: If I don't come in the Linux zone and say that window users needs cross-platform versions of already existing linux apps, then who's going to tell them?
My 2 cents.
Yes, the temperature of Hell has been at zero degrees celsius for the past 5 years or so. That's how long I've been posting here anyway.
Knowledge is power. Knowledge shared is power multiplied.
Even one that uses Cygwin? I'd hate to have to try to run this thing in VMware (Unless somebody can point out a faster way to run Linux in Windows)
like O'Reilly's last interview!
The trouble I have with Linux music notation is that to use it one practically needs to be a skilled programmer not a composer. Lilypond is on the right track, being a very powerfull parser of an input script, it has the potential to become the very best notation engine around. Rosegarden is a valient attempt to do everything for everybody. So it has the unenviable task of trying to code for midi and notation.
Human performance, by nature is not perfect, the task of cleaning up notation created with midi can be very daunting indeed. The skills of a composer are needed, unfortunately these are skills that require study well beyond what the average musician of today has.
To get to the point, having midi input totally separate from the notation gui, and reducing the program overhead to facilitate notation is a goal that cannot be ignored. Perhaps a non midi dependant gui is the only real answer. Programs like Score, (San Andreas), Musedit, Speedscore, Finale and a few others already do this. However few if any have the scoring elegance of Lilypond. It would be great if a Linux gui like KDE or Gnome included an effective music notation gui.
I always think "Shut up! Shut up! Cut his Mike."
BO: Tell me comrades, what makes you want to give away these products and go open source? Communism? Beatles music? Liberal guilt!?!?!
Guillaume Laurent: Well, first off..
BO: Wait, are you french?
Guillaume Laurent: Yes I am from France and I don't see why that should be held against me, in fact the relationaship between the US and France has always been one of [interuption]
BO: Shut up! Shut up! Shut up!
NRPN's to CC's and back again.
Program a graphical interface to MIDI gear.
Track automation layers.
MIDI plugins.
Even things as simple as switching programs don't work flawlessly.
If the Rosegarden developers are reading this, it would be worth their while to spend some serious time with Logic and Cubase and some moderately complex MIDI keyboards and rackmount units.
Start by copying the features from the oldest versions of these programs (I'm sure an Atari + Cubase can be had for very little money), and work up, feature-by-feature. I think some perspective is needed in the development process.
Er...maybe if they trained up some help, they'd have more time...
:-)
Gee, how come we never thought of this before.
Honestly, if it were that simple do you really believe we wouldn't do it ? "Training" people through email just takes too long, and many contributors have just vanished after a couple of patches. We just can't afford to invest this kind of time into some guy whom we have no way to know if he'll stick around. All projects of this scale have the same kind of problem, there's no simple solution to it.
the point that reprints are usually better than the 2004 material.
The quality of score layout has been steadily declining over the past 20 years. Show me a 2004 Baerenreiter score that can compare favorably with their 1950 prints. I haven't seen any and I do get to see a lot of new music.
The unfortunate reality is that revenue in the serious music publishing business is in decline. This makes good-quality engraving (which requires a lot of work and skill) unaffordable except for the most famous composers of the most prestigious publishers. LilyPond is our try to counter that trend, by making software that produces good output without requiring lots of work or lots of knowledge.
Han-Wen
Han-Wen Nienhuys -- LilyPond
NRPN's are expressed as 4 CC's. They're used if the synth sends and receives NRPN data. A fair number of synths use them to get up to 14-bit resolution on controls -- the resolution of a 7-bit CC is often not enough. Google groups will tell you more about which synths use them.
Logic lets you create a knob-and-slider graphical interface to MIDI gear that doesn't have an interface, and it can be fully automated.
By track automation layers I kind of meant two things. First, you want to be able to record a song and overdub knob and slider moves in the sequencer environment, just like real console automation. These moves then appear as separate sub-tracks that can be seen side-by-side. I can't remember if RG supports mixer automation or not (I doubt it, but it needs to). In Logic: http://logicfaq.omega-art.com/html/faq10.htm#10/
Also, it's important to have separate tracks for MIDI automation with a graphical view of what's going on. A typical synth track might have velocity, pitch-bend, mod wheel data, and you want to be able to see these events in parallel in separate sub-tracks. You also want to be able to do all the normal copying, pasting, and looping on these tracks.
As for MIDI plugins, I see none in the list at the DSSI page. I don't mean a MIDI wrapper around a LADSPA plugin (that's what DSSI claims to provide). I mean a plugin that operates exclusively in the MIDI domain -- it accepts MIDI as input and output. Even something simple like a transpose plugin would be useful.
Instead of treating me like a blowhard, offering fairly valid criticism on one of Linux's two flagship sequencer projects (I tried muse-seq, it was even more frustrating), go and play with Cubase and Logic and some real studio gear, and copy and improve on everything that you like. Or do this and then ask the developers to. If this is beneath you, then you'll never compete -- it's not like Steinberg wasn't up to date with each Emagic offering over the years and vice versa.
Its open source. By definition that means the source code is there and waiting for you to do the work. The Rosegarden developers don't care about MS Windows, because that have linux. If you care, either do it yourself, or hire someone to do it for you. For $250,000 I'll port it for you. (It should be easy for you to find someone willing to do it for less, I hate Windows development so I'm charging a lot extra for the pain of doing something I don't like to do)
I did have Rosegarden running without alsa and jack but found that it crashed big time when I used the notation window. There was no indication of why, doing a demesg didn't help and KDE gave me the usual equivelent of a windows illegal operation message after it crashed.
Yes, BTW, I am dyslexic and find writing complex polyphony easier than the English language.
My original post was not a flame. All I was trying to say was that to write a great music notation GUI the kiss principle somehow gets missed by just about everyone. Look into (the late) Frederick Noads Mac program Speedscore, http://www.noad.com/Nsoft.htm he resisted midi for years...for good reason, too bad the guy who now owns the code won't let it out!
I will give Rosegarden another try though, as I do really enjoy causing software crashes. Gives me a good excuse to get out the pen and paper, and do real music notation!
The league for abused ESL bots
Maybe -- but it's the same sort of nit-picking as complaining about books printed using monospaced fonts, or dot-matrix quality print. Other computer-generated scores may have the same information, and may even look superficially similar, but they do tend to be a bit robotic and inelegant. LilyPond's output is clearer, more natural, and easier to read. Which is exactly the point.
LilyPond isn't perfect; there are some things it's still not easy to do (e.g. vocal scores where lyrics are shared or voices switch between staves, or pieces in free time), and placing of marks like dynamics isn't always ideal. (Though I'm stuck with an older version, so it's probably improved since.) But it has far far better instincts about layout than anything else I've used (Cubase Score and SX, Finale, Harmony Assistant). Although the initial entry may take a little longer than other packages, it needs much much less tweaking afterwards, so it works out quicker overall, as well as producing much more professional output.
Result: even though I've shelled out lots of the folding stuff for Cubase SX, I haven't touched its score editor since installing LilyPond. Those nits can turn out to be more important than you'd think.
Ceterum censeo subscriptionem esse delendam.
Sounds like a Training Guide would be useful; that way your marginal cost per interested human would go way down.
I feel fantastic, and I'm still alive.
If people insist on spending money for Windows, they can also spend money on programs for Windows - especially when relevant APIs are very different (MIDI and audio stuff). I've noticed a number of open source projects that provide the source for free (as in beer and freedom) but sell the (supported) Windows build. Sounds reasonable to me.
What is it? I guess it's called other things, but basically it's chord progressions jotted down as numbers. Instead of
C...F...G^7...C
Am..Dm..G^7...C
you would write
1...4...5^7...1
6-..2-..5^7...1
Also, instead of putting chords directly above the words, like most guitar charts do, the words are separate from the chords, with the chords commonly in groups of 4 bars per each line, as above.
Need a Linux consultant in New Orleans?
There are only 3 developpers working on RG, these days more like 2 (Rich became a father recently). So while we do occasionnally use branches, having a 1.0/1.1 tree split is just not feasible.
:-).
As for SVN, yes, we'd like to use it, but sforge doesn't provide it yet (they announced it was scheduled for 2005).
Regarding outside patches being applied : we get less than a dozen per year, most of which are very simple and can't represent stability issues. We've never had a stability problem due to an external patch.
I actually wish your criticisms were true, it would mean much more activity on RG's code
That RG crashes with a relatively high frequency indicates that strictness wrt what gets committed is far too lax.
I'm sorry but you have no clue on what you're talking about here. 90% of the crash reports we get are actually due to the sound libs we depend on (generally people using old version of Alsa). Lately we've received only 1 in several months. Please don't just blindly give stereotypical arguments which simply don't apply to our case.
In general, ad hominem attacks from somebody who is apparently a core member of the RG team posting as AC don't speak very well for the project.
:
Uh, hello, none of us is clueless or immature enough to ever do that, nor would we even accept to work with someone who would.
Seriously: copy down what we say, make a list of the points, and present them to the dev team, if you're not on it.
Is it really that hard to log a feature request on sforge ? Anyway, we're reading you. No guarantee on when what you ask will be implemented, though.
A detailed table-based comparison between RG, Muse, Cubase, Logic
Why not, if we'd ever have the time for it.
Some full songs recorded with RG, with details of how everything was done.
We do link to some composers who use RG
http://rosegardenmusic.com/resources/links/
though they don't give any detail to how they achieved what they did.
Anyway, about feature requests: just make it your goal to have RG assume ALL of the features of the various other sequencer programs out there
:-).
Get real, these products have resources we'll simply never match. This is a relatively small open source project with all the usual problems applying : lack of time, lack of manpower, slowness in communications because it's all done through email, etc... We have to keep realistic about what we can do. So we prefer to have less features working correctly (and even that is not easy) than a lot of half-broken features.
That said, yes, of course we look at Cubase, Logic, etc... for inspiration, and we listen to users. Purchasing old Ataris is not an option though
The fact is, that would be a very poor goal. It's something we actively don't want to do, unless someone else is providing the resources to do it.
We have to balance the needs of quite a lot of different sorts of users. Matching all of the MIDI functionality found in Cubase may make an ideal program for some users, but if we spent all our time doing that, we'd be so behind on other features as to have a program that was useless for most of the people who want to use it now, including all of the developers who caused the project to happen in the first place.
This thread has so far followed the pattern of having one poster or another pronounce that Rosegarden "must" support feature X before it can even be considered 1.0, followed by Guillaume replying half-apologetically to say that we'd love to do that if we could find the time. The plain fact is that we've taken the decision that most of these things are not essential 1.0 features. And there's no reason we should apologise for that. You can disagree with particular instances -- clearly you do, and we'd encourage you to make feature requests with use cases so that we, and other developers, can take them into account when prioritising work in the future. But it's absurd to imply that a program isn't a worthwhile program at all because it doesn't do a particular feature that you personally want. That way lies madness, and/or Mozilla. Far better to accept that there is room for more than one program out there.
We know that for some people, RG is unstable. For others, it's quite reliable. No, we're not using experimental features. Just like implementing features, we don't have the means to maintain a test suite of dozens of distrib/hw config combinations, and sound is still a bitch to configure properly under Linux. Again, we have a clue about software development, we don't release something which easily crashes on our respective setups. The crashes we can reproduce are generally fixed within 24h. But in many cases, the problem just isn't with our code, and we can't do anything about it.
>...we neither have the time to fully document the code...
This is so totally untrue that it is sad the people still say this. Documentation is essential for both bug fixing and new features even if it is a one-person project that is going to last more than a few months. The time saved is enormous when you have to revisit code you wrote some time back. Not only in understanding the code but also to help not writing additional code that breaks previous code. And if there is more than one person on a project, lack of good documentation causes time and quality losses to go through the roof.
It does take a bit of discipline. A couple rules of thumb help tremendously: 1) code must be commented the same day that it is written, and 2) never check in code that's not documented.
How about collaborating with the MuSe guys
:-).
I'm afraid both programs are way too evolved for this to achieve any real results. One program would pretty much have to be thrown away and rewritten to the other's fashion.
what specific things I think would bring it closer
No problem with that, as I said a well-detailed feature request on sforge is the best thing to do. Keep it short though, A laundry list of 100+ items won't help
I hope you don't dismiss everything I've said
Not at all.
Bugs may be reported to bug-lilypond@gnu.org ; as for the Chopin piece, pay some attention to the error messages lily is spitting out. You're shoving a 2.2 file down 2.4 ; you need to run convert-ly first.
Han-Wen Nienhuys -- LilyPond
Probably the first one, saying "error: Incorrect lilypond version".
Why don't you make lilypond 2.4 recognize 2.2 files and automatically run convert.ly on them? That would seem much more user-friendly.
because 1. some conversions need minor hand-editing 2. because we want to encourage users to upgrade their files.
Han-Wen Nienhuys -- LilyPond