Commonly held notion: “The longer I work on this, the better it will be.” Maybe up to a point. But after a while — and it might be just a short while — you’re being overly fussy.
There’s an optimal release point for anything you make. That’s when you should get it out there. After that, you’re just fiddling for the sake of fiddling. And you might even make it worse. Sometimes what you make will be just fine if it’s released after three months — but add another six months (or longer) and it turns into a jumbled, complex mess.
Case in point: “Chinese Democracy,” the Guns N’ Roses album that Axl Rose worked on for over a decade, going through at least three recording studios and four producers. Everyone knew that it wasn’t getting better with more time. In fact, it became a running joke in the music industry. It was just a sandbox for a control freak who couldn’t let go.
Frank Sinatra, on the other hand, was known as “one take Frank.” He’d walk into the studio, sing a song live with a full band, turn around, and walk out. Quincy Jones produced Sinatra and described recording an album with him:
He came in at 2 p.m., and in less than two hours we had rehearsed, had keys and routines on ten songs…Frank is one take, that’s it. If the band’s not in shape, he leaves them behind…he came in at 7, and at 8:20, baby, we went home. None of that three month stuff.
U2 singer Bono always respected Sinatra for that approach too:
It’s all about the moment, a fresh canvas and never overworking the paint. I wonder what [Sinatra] would have thought of the time it’s taken me and my bandmates to finish albums, he with his famous impatience for directors, producers — anyone, really — fussing about. I’m sure he’s right. Fully inhabiting the moment during that tiny dot of time after you’ve pressed “record” is what makes it eternal.
Sinatra’s one take style produced classics. Axl’s dithering produced a pile of mush. We can all learn something from that. It’s easy to fall into a trap of nitpicking over things that don’t really matter. Instead, focus on the essence of what you’re doing. Press record, get it done, and get it out there. (And that’s even more true if what you’re creating is something you’ll get to improve upon after it’s released.)
Graham
on 06 Feb 09Another classic example is Miles Davis’ “Kind of Blue.” It wasn’t done in one take, like some rumors state, but was damn near close. And its the probably the best selling jazz album of all time.
http://en.wikipedia.org/wiki/Kind_of_blue
Scott
on 06 Feb 09And that’s even more true if what you’re creating is something you’ll get to improve upon after it’s released.
Well stated Matt. We too often forget that software is called “soft” for a reason.
Troy
on 06 Feb 09You know, I kind of disagree. I’m a musician, and I record music. I think Axl was in a very different mindset than Sinatra, and that’s neither a credit, nor a dig, at either.
Frank did what he did, baby. No more and no less. In his day, there was only ONE way to make recordings, and he mastered that paradigm to full effect.
Axl, however, had always had en eye to the future, of the industry, of producing and recording techniques, even to the point of hiring the most avant-garde players in their fields. Buckethead, anyone? This guy was betting his chips on the music industry evolving, and that being the case, why not exploit modern techniques for recording music, ad nauseum?
I'm not making any opinions on Chinese Democracy, but I think it's at least worth it to hear the work that's taken 14 years to make, even if it's to reward the guy for actually crossing the finish line against long odds. And I mean, hey, is there any modern rock singer better than Axl? I'll take a digital Axl over any one take from any given modern rock band.Jimmy Guterman
on 06 Feb 09Nick Lowe, when he was working as a producer of punk records in the late ‘70s, used to have a motto: “bang it out and we’ll tart it up later.” He did the bang it out and never got around to the tart it up. Ideal approach.
Rob H
on 06 Feb 09@Scott @Matt, I agree completely with your last statement of focus on the essence of what your doing. Some of my best video reviews are the ones when I don’t rehearse and just do them on the fly. It comes out so much easier and I don’t over think it .
I’ll have to remember this for now on.
Thanks R
Mauricio Gomes
on 06 Feb 09I think the corollary isn’t necessarily true either, though. More time doesn’t mean a worse product either.
It took Howard Hughes so long to release “Hell’s Angels” that when it was finished sound technology had taken hold and he had to redo the entire film with sound. It turned out to be a huge success.
Joel Hughes
on 06 Feb 09I couldn’t agree more! Over that last 6 months I’ve been building a web app in my spare time. It’s been hard work and a huge learning curve (developing a concept completely on your own terms is very different to those done with/for an employer).
Anyway, I worried over when to release Good Baad and how feature rich it should be. In the end I just got it out there – I’m very pleased with it but it’s probably not the overnight success I (secretly) hoped for. And because of this I’ve seen that while it was a good concept it may not be worth spending too much time on.
So no I’m coming up with something new. Getting concepts completed sooner rather than later allows you to learn from mistakes more quickly – at least that’s the way I see it at the moment!
mattias
on 06 Feb 09Working on something for longer also means that the expectations are going to be higher. The product might not get worse over time, but most likely the added details will not make up for the much faster rising demands.
Zach
on 06 Feb 09Look at how the Beatles progressed, they were certainly never a one take band, but they definatly were relatively speaking compared to how the last couple albums were made…. when they stretched the defination of a band.
Ed.T
on 06 Feb 09I used to work with a gentlemen who after editing a proposal enough times would say we were just changing “glad” to “happy.” There is a bit of genius in knowing when.
ratchetcat
on 06 Feb 09Good post. Set a deadline - even if it’s completely arbitrary - and try to meet it. Less time to complete a project almost guarantees prioritization, focus, and energy.
(I try to do most of my sketches in a single hour, for example. This forces me to optimize compositions and technique for greatest visual impact in the fewest lines.)
Michael
on 06 Feb 09One take, but how many years of practice behind the take?
Steinblock
on 06 Feb 09And less time does not mean better product either rather it often means worse product.
Chinese Democracy is great album. Who cares it took decade to release? When working on the projects its about the trip as much as it is about the destination. If you can afford to take decade to make something you feel is perfect nobody’s got to say anything about that.
Rushing it out is never good policy. Half product is better than half arsed, but that depends on your customers definition of “half”... Half sometimes does not cut it.
John Moody
on 06 Feb 09It’s funny – I had spent months toying around with a deperately needed website overhaul for my company. I had maybe 15 abandoned designs sitting on my hard drive.
Then yesterday, I pulled a Sinatra, said “enough’s enough”, and just did it. At 12:15 last night, I pushed the new site up. It’s not perfect, and I’ll tweak it some in the days to come. But it’s up there, and it’s like a great weight has been lifted from my chest.
Just get ‘er done.
Brett
on 06 Feb 09Your opinion is absolutely subjective… and Chinese Democracy (an album I really liked) is not a good example.
Can you say exactly the same with “Black Ice” from AC/DC, they also waited a decade or so to release it and it is a very good album too.
The whole idea might be ok, but the examples are far from it.
Geoff
on 06 Feb 09There is a saying in the quilting profession: “Finished is better than perfect.”
Raymond Brigleb
on 06 Feb 09I was just thinking about that the other day. Do you realize it took Guns ‘n Roses longer to produce one (crappy) album than the entire recording career of the Beatles?
Kind of puts it in perspective.
I always liked George Harrison’s quote about that. He said they had recorded the first album in an evening. “The second one took even longer.”
Keith
on 06 Feb 09Talent plays a part there as well…
If you suck…you’re going to need more time and if you truly suck…no amount of time is going to make your work decent.
Axl Rose was a product of a time, a place, and GnR had like 3 great albums in them with the original lineup.
Comparing the processes behind U2’s work or Frank Sinatra’s work to Axl Rose’s is like comparing Degas or Vermeer to some dude at the Starving Artist expo at the Airport Marriot.
Nathan L. Walls
on 06 Feb 09@brett I haven’t listened to Chinese Democracy, so I can’t comment on quality or lack thereof. The question is, was it worth 10 years in the studio?
For Axl, it apparently was. For fans, perhaps.
But the point is, there’s practicing and revision, and there’s shipped and live. I know I’ve spent too long fixing just one more thing. It’s common in writing and software, too. Really, any human creative endeavor where doubt can creep in and the inner critic says, “it’s still not good enough.”
The question here is: Is it worth 10 years in the studio for you, for the creative thing you’ve been finding a reason to say, “not yet”?
Todd Austin
on 06 Feb 09Progress is a much better goal than perfection.
Ricky Irvine
on 06 Feb 09Hear, hear!
stevent
on 06 Feb 09This post makes a very valid point, but you need to find a better analogy. “Chinese Democracy” is quite the opposite of “a pile of mush” according to Metacritic.
indi
on 06 Feb 09Brian Wilson and Good Vibrations come to mind here. Lots of takes on the various pieces and lots of postproduction putting it together produced a masterpiece (#6 on Rolling Stones 500 greatest list) I agree with those saying it is an art knowing when enough is enough. Software is more forgiving though, especially web sites.
In my day-to-day work I run into this more with emails. You don’t want to say too much but you don’t want to say too little. Business emails become a permanent record for a project. I’ve seen hastily sent emails that caused a lot of grief, espcially ones where the sender hit “reply-all” instead of “reply”. On the other hand I hate it when I get ready to leave for the day and close a mass of working windows only to find an unsent email sitting there from the morning that I had planned on tweaking but never got around to it. So point well taken.
Nathan
on 06 Feb 09One point that should be made is that Frank Sinatra rarely wrote his own songs. Sure they might have a hand in arranging them, but comparing it to writing, recording and producing one’s own music is unfair.
Also – the album as a whole being taken as a legitimate piece of art was a long way away. Having Sinatra show up, bang out some singles, throw them together on an album with a picture of him on the cover is a far cry from what goes into making a (good) album today.
I’m not saying Chinese Democracy is proof of this, but if you look at most of what would be considered the greatest ‘albums’ of all time they would be much closer to Axl’s process than Sinatra’s.
Dennis Eusebio
on 06 Feb 09Jay-Z is another great example of this. Check out ‘Fade to Black’ on his process. His recordings are often finished in one sitting and never written down.
Anonymous Coward
on 06 Feb 09You know where you are? You’re in the Jungle, baby. You’re gonna die.
Can I get a Sha, na, na?
Gabe da Silveira
on 06 Feb 09Music doesn’t have features or bugs.
Andrew Bennett
on 06 Feb 09stopping before you get overly fussy is key to being a great developer.
But…it should be noted that the time for stopping is really all about ‘how easy is it to iterate this design/product later?’
If it’s a website that you can iterate 27x a day if you don’t like something later, then stop asap and PUBLISH!
If it’s the 3D design for a new plastic mold that’ll take 6 weeks to make out of steel and the company’ll only have 1 chance to go into production this Christmas season, then TAKE A COUPLE MORE DAYS before stopping being fussy!
Have a great w/end everyone.
carlivar
on 06 Feb 09Apples and Oranges.
1. Frank Sinatra doesn’t play any instruments.
2. Use Your Illusion I and II were also fussed about quite a bit, but are awesome.
ML
on 06 Feb 09Carlivar, even within Guns ‘n Roses output, the theory holds true…
Amount of fussiness on G’nR records (from most fussy to least fussy):
Chinese Democracy
Use Your Illusion I and II
Appetite for Destruction
Quality of G’nR records (from best to worst):
Appetite for Destruction
Use Your Illusion I and II
Chinese Democracy
I don’t think it’s a coincidence.
JB
on 06 Feb 09This is another one of those cases where there are great examples on both sides. Sgt. Peppers’ and Pet Sounds were fussed over ad nauseum and many people think they are some of the best rock albums ever made.
The Beatles were notorious for recording take after take of basic tracks for the same song. Even simple songs, like Mother Nature’s Son, you look it up and it turns out it was Paul’s 46th take or something of the song.
Trust me, there are loads and loads of bands that head into the studio, knock out a few tunes in a few hours, and leave with a rough mix of absolute crap.
This actually goes along with the “don’t copy 37signals” post above. Don’t copy these guys either. Dylan recorded some albums where he threw out whole bands and rerecorded whole songs with completely different players, lyrics, keys, arrangements. He recorded other ones off the cuff. Many people have spent years on opus albums and been happy with them, many people have done the same and failed. A few people of incredible ability have quickly banged out something good, but many, many more people trying to do the same have failed utterly. This is more a reflection on how easy it is to crap something out vs. polish it than it is on which process works better.
JB
on 06 Feb 09Good performance is not connected with long time performance, but to the time one takes preparing for the performance…
jb
Nivi
on 06 Feb 09I like this quote from Immad Akhund: “We kept to minimum feature spec. I think that is always very important. It is hard to determine what to do until you launch.”
It is as if he is saying that they built the minimum viable product, so they could launch quickly, so they could figure out the product they should really build.
Fred
on 06 Feb 09The thing is, if you try the quick, spontaneous thing and it doesn’t work – you’re going to have to spend more time. Music is great when it just ‘happens’, I’ve played many solos I wish I could play again, but I can’t make those magic moments just happen. The easy stuff also tends to get used up early in a music career.
Mark
on 07 Feb 09There is also the whole “did it my way” aspect of this. Assuming both Axl and Frank cranked out products that suited their individual means, such as the mantra followed by Apple, 37Signals and others, then who is anyone to judge how long that individual process should take?
Some people find their soulmate right off the bat and are happy from day 1. Others find they need to work at their marriage for years to get to that point. Bottom line, as long as the point of happiness is achieved, who cares how long it took to get there.
GeeIWonder
on 08 Feb 09I think this post does a great disservice to Ole’ blue Eyes. Especially after his down-and-out phase, he knew better than not to fuss, or to have a ‘one-take style’ for his craft. Sure for movies, he was ‘One-Take Charley’ and I’m sure he put all the weight on giving Bono a chance to hatchet his songs that, let’s face it, he deserved. But don’t mistake the exceptions for the rule:
If you want to know Frank’s thoughts, get ‘em from Frank:
“Never ignore an inner voice that tells you something could be better. Even when other people tell you it’s okay.”
Or this, from Bill Zehme’s classic and authoritative “The Way You Wear Your Hat”
“When making record, however, he was the perfectionist obsessed. He stopped and started endlessly, catching himself, correcting infinitessimal clams: “Once more, for me,” he would say, ten times, twenty times per song, halting the music, shifting his phlegm.”
Johan Strandell
on 08 Feb 09If you’re going to consistently do one-take recordings you actually need to spend a lot of time, with practising. (Yes, this is similar to Michael’s sentiment above.)
There are artists that are able to nail recordings in one take time after time, but they all either spend a lot of time practising and obsessing over details, polishing their technique until it’s perfect, or have played the music live a large number of times. You can get lucky once or twice even if you haven’t practised, but counting on luck or momentary divine inspiration isn’t really a sustainable strategy in the long run.
And that’s without considering all the music/art that simply isn’t possible to produce in one take. Loveless was recorded in 19 different studios, and is a classic album.
There is no simple way to produce great results; it takes time regardless of how you’re doing it.
James Higgs
on 09 Feb 09On one level what you say is blindingly obvious: things don’t necessarily get better the more time you spend on them. Well, yeah.
But there are so many examples of the opposite that it’s hard to know where to start. But let’s pick one: James Joyce and Ulysses – took 14 years to write, and debate continues today about a definitive version of the text, and yet is an absolute masterpiece. Had Joyce been less of a perfectionist or less of a control freak, his great novel would have been much the poorer.
Proust – almost an identical story to Joyce’s, except that it took him even longer to complete.
Wagner – the Ring Cycle took over 25 years to complete (with a long break) and yet gets better the more time he spent on it.
Musil – The Man Without Qualities – a gargantuan, unfinished marathon of a book. As with Joyce and Proust, it took years. At the time of Musil’s death, more than half of the material had not been published, yet it is one of the most fascinating and brilliant novels of the 20th century.
Herbert von Karajan’s recording of Mahler’s Fifth Symphony – he scheduled over 100 3 hour rehearsals for a piece that lasts between 65 and 75 minutes depending on the performance. It’s still one of the most highly regarded recordings of the piece after more than 30 years.
Enough examples. I believe that your problem is that you use artistic examples to back up ideas you have about non-artistic activities (like building websites). There is no functional requirement for a piece of art. Whatever process is used to create a piece of art is the correct process. By definition, the art could not have been made another way. There is little or no virtue in getting a provisional version of a piece of art in the hands of the public early (although Joyce did publish sections of Finnegans Wake in advance of its completion, but largely because he was afraid he wouldn’t live to complete it, and Musil allowed the first part of TMWQ to be published because he needed the money).
I have a great deal of sympathy for the principles of Getting Real, and I use and like several 37s products, and I find much of your advice about software development inspiring. But I really think you should lay off the comparisons with art. It makes it look like you’re philistines.
john
on 09 Feb 09James somehow beat me to it. Didn’t really think there’d be a Joyce reference, but yes, Joyce took (12) years for Ulysses and even more for Finnegan’s Wake.
How about MBV - Loveless - only a top 50 albulm all-time.
You can go either way w/ this sort of thing. Personally my skill is in “letting it fly”.
Also, Sinatra is a … lounge singer. Ok?
john
on 09 Feb 09omg, and a loveless ref above that? how lame am i (so lame)
john
on 09 Feb 09or if you are programming heart valves or space shuttle software
Jayme
on 10 Feb 09How true, how true. I’m a notorious fiddler (or as I call it, Noodler). I love to noodle, noodle and noodle some more. But then, nothing ever gets done. So thanks for the reminder. :)
Denis
on 11 Feb 09Jayme,
I have to say, posts such as this are one the reasons I love this blog.
Having studied and performed with a number of professional musicians (all the Marsalises, BB King, and a number of famous others) I think it’s feasible to point out examples of both keeping it simple and real as well as going overboard with detail-finickyness and not arrive at a conclusion. It has to be a smart compromise between both realizing time and money constraints a not losing the soul or essence of something. Good music, most especially as it is somewhat subjective, can be hard to articulate and capture the vision of what the artists are looking for.
There is also something to be said for quality artists and musicians who can walk into a session and lock it down in under 2 hours versus others who struggle to make the music come together the way they see it.
To prove your point, Jimi Hendrix played bass in All Along The Watchtower after the bassist couldn’t nail it and Jimi thought he could do it himself while the bassist when out for a cigarette, Michael Jackson recorded Thriller in one take, and most (if not all) jazz musicians never record more than 3 takes. It is possible to capture something the first time round and nail it, but how many performances can you think of where the performer wished they hit that big note or handled their music differently.
On the other hand, the Beatles, Santana, and most other musicians aren’t strangers to recording takes multiple times, early G-N-R included (though mostly due to intoxication issues).
Once you commit the music to “wax” or to CD, etc., it’s not so easy to go back and change it once the discs are burned. Conversely, it’s A LOT easier for 37sig to cascade a change to every user within minutes, thus some metaphors are appropriate others are not.
Keep rockin…
john
on 12 Feb 09I was in a band that recorded one take, if you fucked up, everyone who heard it knew.
This discussion is closed.