When you’re writing code the devil is in the details. You need to be fast, agile, accurate, speak multiple programming languages fluently and know what your customer and their audience need. More or less it’s the antithesis of art, where you can be flexible, exploitative, take risks, experimental, emotional and customer or audience be damned. But then there is TJ Hertz aka Objekt, programming engineer turned one of electronic music’s most meticulous and progressive DJs and producers. Listen to any of Objekt’s tracks and one thing is immediate: they’re immaculate. From, the synthesized sounds that made up his early work to the organic fluidity of his newest LP, Cocoon Crush not one bit sounds out of place.
Then there’s the man’s DJ sets which are world famously adventurous, technically pristine and continuously redefining the notion of what’s possible as a dance-floor DJ. Genre bending, tempo-spastic, groovy, heavy, meditative and heady, Objekt can play mainstage Dekmantel to the depths of Unsound and be at home.
Having the ability to be so naturally flexible but expertly precise comes with a lot of preparation and structuring. In a recent Red Bull Music Academy “Couch Wisdom” session, Objekt touched upon how detailed oriented he is behind the scenes, structuring and labeling his collection. In his own musical house he’s the engineer, architect and designer, making sure the building is functional, secure and aesthetically pristine.
Like any good archivist knows, it’s this attention detail; consistent metadata, defined taxonomies, structure and thoroughness that makes an archive different than a glorified home collection. Inspired by his “anal retentive” (his words) attention to detail, we decide to ask Objekt, “Do you even archive?” Below he divulges how he manages his DAW and sample libraries and his prized Rekordbox. Come to find out, it’s not as militant as one might but expect but also no surprise just like his music, his responses are thorough, expressive and illuminating.
How do you go about defining certain presets, tracks or versions? Do you stick solely to numerical titles or do you get creative with adjectives i.e. reverb good, roomy, dry v1.2? Is there a rhyme or reason to the terminology you use for these definitions?
My naming conventions are pretty dry, to be honest. For tracks, I just number all my works in progress – e.g. WIP 104.31, where 104 is the song and 31 is the version. The versions are typically just “Save As” increments; often the changes are very minor so I often hit 100 versions or more. Sometimes I’ll go back to previous versions and start again from there, in which case I might save a new version as something like “WIP 104.32 (branched from v15).aiff”. I typically give tracks proper names only when they’re finished.
Rekordbox playlists I tend to name descriptively – e.g. I have a “Rollers” folder and in that folder I have playlists called “Loopy AF”, “Less loopy still rolly”, “Broken riddim floaty roll”, “No-kick rollers” etc. Obviously these descriptions only have to make sense to me, so in theory I could call them whatever I like, but I find I need to keep them fairly descriptive otherwise I’ll just forget what I meant in the first place. (To that end I actually made a spreadsheet with all my Rekordbox playlists so that I could remind myself what I meant by “Broken riddim floaty roll”, though I still find myself scratching my head quite often.)
With synth/fx presets, I make them so infrequently that I don’t really have any conventions here.
How do you manage your record collection? How is it cataloged on the shelves and what steps do you take to take care of your vinyl (temp, water, humidity etc.)?
It’s split rather broadly into “Techno”, “House”, “Electro”, “UK”, “Non-dance music” and “Stuff I wanna get rid of” and then organised chronologically according to “last time I took it off the shelf”. So it’s pretty unscientific, and a lot of records just end up going into the “Techno” section despite definitely not being techno. (Also anything below about 120bpm ends up in “House” despite very clearly not being House.) Depending on my mood, this can be either very frustrating or quite refreshing.
No special precautions, apart from making sure everything goes back into the right sleeve etc.
What kind of tiering system do you use for track construction? Will one folder house all samples, versions, VST presets, etc or are things located in different folders?
Given how anal retentive I am about a lot of things in life, my sample libraries are actually pretty haphazardly organised – I have multiple folders across multiple hard drives that I’ve been meaning to tidy up for years now, although you can see traces of my attempts to keep things at least somewhat under control (e.g. certain libraries being organised by company). While working on a track I’ll just drag and drop samples from wherever I like, but generally I’ll make sure to “Collect all and save” all the content in Ableton when I finish the final version, so that if I need to open it up later then at least the samples loaded directly into Ableton will have been consolidated into the project folder (though not necessarily stuff like Reaktor ensembles or 3rd party synth libraries).
The project itself is usually pretty well organised – I’ll group and colour-code channels within Ableton, and within the project folder all the file versions will be quite tidily named.
It was 4 years between Flatland and Cocoon with not many releases in between. You must have an extensive archive of tracks that never will or have yet to see the light of day. What steps, if any, do you take to back up your harddrives? Are they also meticulously structured? If so, how?
I’ve got less unreleased material than you’d think – I have a very slow work rate so most of my WIPs either get finished or aren’t very good to begin with and get abandoned.
I back up all my drives to a bigger local drive with Carbon Copy Cloner and Time Machine, but should really get into the habit of backing up remotely too – I just can’t face trying to upload terabytes of data to the cloud on an underpowered DSL connection. One of these days.
Can you speak a in a little more detail about how you structure stuff in Rekordbox? Are the playlists structured? Is it only tempo and your “banger” scale? How consistent do you try to be with other more whimsical tags and structures?
Well I actually only use the “banger” scale for functional techno, which I’m playing less and less of these days. Everything else is sorted into themed playlists, which are (almost) all sorted by BPM. That way I can pick a playlist relevant to whatever sound I’m feeling at a particular point in the set, and scroll in the broad region of the BPM at which I’m currently playing.
My playlist library is constantly evolving, but in general I have my monthly playlists and my categorised playlists. Monthly playlists are for tracks I’ve acquired that month – each month (sometimes twice a month) I’ll make a new folder with playlists called Club (everything DJable), Slow (60-120bpm), Fast (>140), Other and DnB playlists, containing whatever music (if any) I’ve acquired that month or fortnight, typically in the region of 100-200 tracks. Sometimes if I pick up a particularly large number of tracks in a certain vein (e.g. polyrhythmic stuff, or a certain genre) I’ll make an extra playlist for it that month, just to filter it out of the more generic Club list. So for example my Jan 2019 folder might contain the playlists Jan Club, Jan Slow, Jan Fast, Jan New Beat and Jan Other. And then a level up from that, my Monthly Playlists folder will have folders called Jan 2019, Dec 2018, Nov 2017… etc, each with their own broadly sorted Club/Fast/Slow/etc lists.
Every few months I’ll go through and properly tag all the tracks from the recent monthly lists that I can see myself returning to in the future, and these tracks go into the archive playlists – these tags are much more detailed, and are either genre-based (house, various kinds of techno, electro, disco, various kinds of DnB, slow clubby, slow floaty, fast and so on), mood-based (emotional bangers, curious, hard, floaty etc) or setting-based (peak time, opener, set closer, etc). Out of the tags I make intelligent lists, most of which are just a single playlist corresponding to a single tag (e.g. “Electro”), but in some cases I combine more tags into useful logic, e.g. “Techno [Emotional bangers]”. (I also have a “DECOMMISSIONED” tag, which prevents a track from being included in any playlist should I decide I don’t wanna play it anymore, while preserving the tags I may have attributed to it before, in case I should want to recommission it at some point.) These lists then go into my “Categorised” playlists, which is my organised archive of music that I’ve collected over the past 10 years or more.
Unfortunately, due to various limitations of Rekordbox and Pioneer CDJs, there’s also half a dozen intermediate steps that need to be taken in all of this – e.g. converting the intelligent lists to normal playlists, backing up old playlists, manually sorting playlists by BPM – all of which is pretty tedious. I actually wrote a Python script to read the Rekordbox xml and do some of the donkey work for me, though it’s been pointed out to me (thanks Avalon) that some of the new functionality in the CDJ2000NXS2 players might finally rescue me from all the extra work anyway. I’ve yet to dig into this.
To be honest, my system is a massive micromanaged pain in the arse and I’m not sure I’d recommend it to anyone else. I’m starting to wonder if I should just scrap the lot and do things more spontaneously. My various categories are good for browsing in, but the archiving process itself is starting to spiral out of control as the number of playlists increases, to the point where the amount of time I spend tagging new material is starting to become unrealistically unwieldy. If anyone’s got any tips I’m all ears. 🙂
Have you ever lost any material that was never backed up? What and how?
Not since I was 14 and accidentally reformatted the wrong partition on the family computer so that I could install Linux because I’d heard that Linux was cool. (There are, obviously, many things wrong in this sentence.) Suffice to say that I’ve been quite fastidious about backing up my (and other people’s) shit ever since.