ASCII by Jason Scott

Jason Scott's Weblog

What’s New at the Ol’ Salt Mines —

Hey there! Haven’t really talked in a while. Been pretty busy.

What with? Oh, speeches, presentations, editing and shooting some movies. You know, the usual?

How’s that archive.org thing doing for me? Oh, it’s doing fucking awesome, trust me. Best job I have ever had. Great folks, great stuff.

What have I been up to there? Oh, this and that.

Other than that, not much. Oh, but my buddies have been SUPER busy.

Everyone’s got their nose to the grindstone. You know.

Hey, gotta go. Good talkin’ with you.

4 Comments

Sony Vegas Media Bins: Not Ready for Prime Time —

Let me explain what’s going on with this entry. Summary: Most of you can ignore it.

Perhaps it’s a reflection of how things have gone with online discourse, but I’m mostly posting this as a way of gaining search engine attention to the fact that Sony Vegas, the video editing software, has something called “Media Bins”, and those media bins kind of suck. In this way, maybe Sony will make one of their guys at Madison Software (now called Sony Madison) fix this part of the software.

I’m doing what is perhaps a bit of an edge case with Media Bins, but one would hope in a world where Sony Vegas brags about being worthwhile of professional projects, something like Media Bins would be basically usable.

Media Bins are ways to declare clips or parts of clips in Sony Vegas to be entities. For the DEFCON Documentary, I decided to do this because with 1.4 terabytes of film footage, and thousands and thousands of potential shots, my old trick of rendering out clips seemed quaint and unwise.

So here I am with a big ol’ mess of folders:

…which should be fine and good. When I do the final edit (I think, I hope), I’ll be able to traverse the folders and see the inside clips, which I’ve worked hard to describe, like this:

(There’s two clips because one is the video track and the other is the separately-recorded audio track. It’s weird but it would take forever to bind them up, and sometimes I want to choose which one I use.)

And this all should be great except:

  • No export function. Once you make a bin of clips, you are owned. You can’t produce a .cbin file and put it into another Vegas project. You can’t even really cut and paste.
  • At 5,000 clips, it’s already slow to do any work in the main Media bin. Terrifyingly slow, really, considering how many cores and how much memory is available. This smacks of no optimization whatsoever.
  • Random scrambling of subfolders. I’ve had cases where I find the subfolders have MOVED TO OTHER FOLDERS. That is CRAZY. That’s grade-school.
  • No export function. Saying it again, I am now trapped in this specific .veg.
  • Can’t re-adjust the order of the folders. Why? When you create them, that’s the order they stay in, forever. WHY? Some people have come up with a hack of moving ALL of the folders to a sub folder and then, by hand, putting them back up one level in the order you need. Seriously? Seriously.
  • Drag-and-drop…. well, it sometimes works.

All of this points to a rather immature feature set, which is odd considering it’s been in there for years.

Fix it.

Thanks.

5 Comments

The Last of the GET LAMP Batch —

I figured you’re my loyal audience so you should know first – I turned around and realized I was down to the last 130 or so copies of GET LAMP. In other words, get your orders in as soon as you can to guarantee you get a copy of the original box+coin combination.

I haven’t yet decided if I will sell more, and I’m almost certainly not going to make more coins. The coins are awesome but that is a lot of money to spend on an item. The fans who were most directly affected by the existence of text adventure games to want an artifact bought them, while people of the “that’s nice” variety are the ones trickling in. Coins are expensive, friends – between 50 and 60 percent of the unit cost! (And also amazing).

On a more forward-looking scale, I do believe we’re just on the end of possibility with physical copies of media except for very high-end premium packages provided as one-offs for really special fans. That’s the sad part (although it’s much more convenient to point at something and go “NOW SHOW ME” than to click a button that says “AND MAKE IT EVENTUALLY APPEAR WHERE I LIVE). We’ll see how I feel in a week and after my duplicator lets me know how much another print will cost me.

I am delighted to have sold over 3,800 copies of GET LAMP around the world and that the image of my cousin smiling is in thousands of homes, and that this isn’t the sum total of his existence on the internet.

Hurrah for selling out GET LAMP!

3 Comments

Javascript Hero: The New Batch —

Dormant while my attention was focused on a few other matters, the Javascript MESS Project is coming back. And if you are ready, willing, and a coder, we need your help.

If you don’t know what the Javascript MESS Project is, I wrote entries here and here about it. Oh, and here and here. I guess I’ve written quite a bit about it.

In essence, a group of people have been porting MESS, the Multi Emulator Super System, to Javascript. The reason to do this is to turn all of computer history into an embedded object, like a movie or audio or other information. Here’s a song, here’s a movie trailer, here’s a PDP-11 running Adventure so you can just try it out like any other thing. That’s what we’re up to here. It’s kind of a big deal, if it all works.

As an example that is not guaranteed to work in all browsers at all, here is KC Munchkin playing in a window at the end of a historical overview of the KC Munchkin case. Imagine talking about software and there you are, trying that software out. That’s the dream we’re going after, here.

And we have it working really well in some aspects; for example, nearly the entire MESS backend of information and modification is available in the window!

Exciting, really. A full Colecovision emulated in a window. It’s slow, to be sure, but we’ve done absolutely no optimization on it.

I have a theory about volunteer coding projects. There are two main kinds of programmers who will contribute some coding time to a volunteer project – both need to think the goal is worthwhile, of course, but they each have different preferences.

The first likes the crazy challenge of “it’s completely impossible, there’s no proof it’ll work, there’s not even a guarantee you’ll get halfway before you’re mired in misery and failure”. These guys are as rare as can be, especially for volunteering. I was lucky enough to find Justin De Vesine, an old friend, who did the months-long effort of hacking through two systems (MESS and Emscripten) that had completely different outlooks on life and which he forced together into playing nice. Justin worked on this whole thing for nearly a year and got a ton of the pieces in place.

And then there’s people who are comfortable seeing it all works in practicality, even broken practicality, and who want to bring some optimization and cleanup and improvement to the thing. In this way, you know SOMETHING will come of it all – after all, it already has.

So Justin Kerk took the main reins and helped bring the whole thing working to the point we see now, where you could actually bring up various Colecovision cartridges, then Odyssey 2 and we’ve even made inroads to computers like the original Commodore PET 2001 (although it doesn’t boot up 100%).

Let me stress that at least a dozen other people have been involved, not the least of which is Alon Zakai, maker of Emscripten, who’s been helping with a bunch of optimizations, and Gil Megidish, who hand-optimized javascript so we could see speed improvements before we go into more involved compiling. Tons of people have helped us get to where we are; and remember – ALL these changes and results are open-sourced.

It works, that’s right! But we need more help.

If you’re comfortable with Makefile, C++, and Javascript, please consider joining up with this project. You could really make a different contributing your energy and thoughts, and  if we get this JSMESS working to an expected level, actual millions of programs will come alive again in browsers around the world. Programs that will teach, inform, surprise, entertain and challenge. That’s quite a reward, sitting there.

The programming and project channel is on EFnet on IRC: It’s the #jsmess channel. If you want to write me about this project first, I’m all for that as well. We’re looking to:

  • Start making it easier to compile the 600+ machines that MESS can support (!)
  • Add a pre-loader that will bring up a JSMESS window AFTER a keypress, and with info
  • Make the whole thing as fast as possible by nailing down issues
  • Enjoy booting some crazy stuff in a browser

I hope it sounds like fun. Amazing things await. If you can help, please do.

Thanks.

Comments Off on Javascript Hero: The New Batch

Let’s Just Solve the Problem Month —

I had really good success when I put out the call for the Archive Team, so let’s try that again, with an entirely new idea.

I would like to declare November 2012 the very first Let’s Just Solve the Problem Month.

JUST SOLVE THE PROBLEM

Here’s how it works, and what problem I want to solve.

As that sexy pontificator Clay Shirky has said on several occasions, instead of getting hung up on whether Wikipedia is great or not great, instead realize that Wikipedia represents a massive expenditure of energy recovered from not watching television. Not only that, but Wikipedia is one of what could be many different things happening that benefit the world. All you need is a dash of organization, a clear set of principles, and off you go.

I buy into this.

I also buy into the idea behind National Novel Writing Month, which has at its core that everyone has at least one (incredibly shitty, possibly unreadable, vogon-level-quality) novel inside them, and by setting aside one month of you being encouraged, forced, guilted and tortured, you will blow out one 50,000-word novel in that time. What happens next is up to you – burn it and move on, take it aside and polish it until you’re the next JK Rowling (or Hunter S. Thompson), or whatever tickles your fancy. But at the end, YOU WROTE A NOVEL BEFORE YOU DIED. Not bad.

What I know to be true is that there are a number of “problems” out there that need to be solved, that need one single thing to push them from “impossible” to “solved”, or, at least, “1.0”. And that thing that it needs is a lot of human thinking. Often rote, often boring, but necessary, to slam that thing out.

JUST SOLVE THE PROBLEM

So since I got to come up with this idea, let me declare the first month, November 2012, to be SOLVE THE FILE FORMAT PROBLEM MONTH.

Here’s the problem, in more detail:

In the last couple centuries, we’ve created a number of self-encapsulated data sets, or “files”. Be they letters, programs, tapes, stamped foil, piano rolls, you name it. And while many of those data sets are self-evident, a fuck-ton are not. They’re obscure. They’re weird. And worst of all, many of them are the vital link to scores of historical information.

Everyone knows this problem. It’s why old novelists cry they can’t pull their first novel out of Wordperfect. It’s why someone who used U-matic tapes to record the first meetings of a famous protest group goes “oh well”. It’s why, in all things, someone looks at anything older than five years, and goes “bye”, figuring there’s nothing they can do.

And I’ve had to listen to the mewings about this problem for at least 20 years now, in various forms. A lot. And then the person lights up about maybe solving this problem, and then dims and says “well, we can’t really solve the problem”. Because they know – it’d take an army of people to do it.

Let’s make that goddamned army.

And before I give you a battle plan, let me say: This will solve a major issue. This will give thousands, later millions, access to a whole range of materials now shut off from each other. Stuff being made after 2012 will be scrutinized to see if it has made ways to access it clear. Stuff made before? We’ll have docs, or a thread, or even a few first steps towards understanding what it was. People writing modern software will be able to make filters or plugins that use these standards – it’ll drop from being a needless rathole to being a simple matter of writing a perl library or a javascript routine to pull the data in and make it work with the new thing. That will be very helpful indeed.

Battle Plan:

  • In October, I’ll be making noise about this happening. We’ll have a logo, and we’ll have some preliminary work done.
  • It’ll be a big wiki, with people taking various roles of the exciting and boring parts, working on a structure, yanking in what we need.
  • We’ll scour the internet, and online and offline worlds to pull in every potential format ever. If it sounds like a hierarchy issue, yes indeed it is… but classification’s bugbear is a distant second to acquiring the wealth of formats now extant.
  • We’ll acquire examples of the formats, links to programs that deal with the formats, known variations or problems with the format, and so on.
  • We’ll keep doing this from the low-hanging GIF and JPEG and PNG documentation, to the aforementioned piano rolls, microfiche, obscure barcode formats and disk layouts of Cray platters. We’ll just keep doing it.
  • At the end of the month, having had our knees on the chest of this problem for 30 days, we’ll be dragged off the problem, kicking and screaming and still punching, and see where we are.
  • The resulting work will be open-licensed and available to anyone.

Now, if you just read all this, let out a big “pffffff” and are having your fingers twitching with the urge to write about how this is all impossible, just get the fuck out now. The project doesn’t need you, now or ever. Just enjoy the summer, grasshopper, and come knocking on the ant’s door in December when we’re at 1.0.

But if you read this and said “Well, I could take a shot at it, might be worth a few hours”, then you’re EXACTLY what is needed.
Think what giving a month every year will do for a problem like this. There’s plenty of others – but this is one that has vital meaning to the work I’ve done with Archive Team and to the hundreds of archivists and historians I’ve met over the past few years. If this problem is in some way handled, if an OED of formats is blown out, lives will change – projects thought undoable will be doable, and the flood of old information saved will be incalculable.
JUST SOLVE THE PROBLEM

So who’s with me? SEE YOU IN NOVEMBER.

42 Comments

The Archive Team Kickstarter Prize Round-Up —

So, Archive Team has become very good at downloading sites before they go down. Way good. Some might almost venture too good. Someone trying to make a point, like me.

Take, for example, Tabblo, a photo-sharing site that HP stabbed in the neck because it wasn’t getting the same internal love as Snapfish, which HP also owns. Instead of doing the reasonable thing and providing users of Tabblo with an upgrade/switch path to Snapfish, HP simply told people it was going down and thanked them for existing. Except, of course, HP sent it to out of date e-mail addresses, or it got caught in spam folders, or people didn’t check that folder for a month, because when Tabblo went down, people came out of the woodwork screaming, many of them finding out Tabblo was down when they tried to upload new photos of their kids to accompany the old ones. Nothing like finding out your entire child’s early life in photos is dead and gone for no obvious reason.

Enter Archive Team’s download of Tabblo. We got the news about Tabblo shutting down a bit late – about a week before the shutdown. We activated our machines, ran our awesome new clients, and 39 people downloaded all of Tabblo in 36 hours. Then, we did a quality check, found some hundreds of broken/timed-out downloads, and re-ran them, cleaning THOSE up in just a few hours more. If you had a photo up on public in Tabblo, Archive Team grabbed a copy, and we’ve been making them available back to people crying about their lost photos – and getting grateful responses.

Tabblo worked out to just about a terabyte and change of data. Not small, but not terrifying either – just a lot of individual pictures, which we tucked away on archive.org. In fact, we tend to tuck a LOT of data on the Internet Archive’s servers, and they’ve been very generous.

But we crossed a line with the brand new MobileMe download. The Archive Team MobileMe download is over 275 terabytes. That’s over a quarter of a petabyte.

Even in today’s crazy storage amounts, this is significant. A lot of drives at the Internet Archive are storing MobileMe for future generations, and that was space allocated for a lot of other just as worthy projects and creations. Nobody’s said anything to me, but I feel this is going to strain the nice situation and relationship, just like staying at someone’s house and you park six cars out front. It’s just not the best thing to do.

So, I’d like to begin arranging a Kickstarter to donate money to the Internet Archive. And I’d like your help.

Yes, we could just keep printing the Internet Archive donation link, and people most certainly use that… but folks often like to contribute as part of an “event” or a goal. So that’s what I’ll be doing at some point soon – putting together a Kickstarter to have people contribute towards a goal, and then that money (minus fees) goes to the Internet Archive. I checked, and it appears your donation would be tax deductible, which is a nice bonus.

The way I’d like to get help is this.

Kickstarter works best when there’s prizes or rewards at various amounts. Donate $100, get a hat, donate $1000, get a dinner with someone famous, etc. I’d like to have some of those. I’m going to chat up people about this, see if I can get friends and old contacts to make themselves available, provide tutorials or dinners or items, and so on.

If you think you have some item, or action, or contribution that could be used as a prize, please contact me. I’m shooting to have this kickstarter begin in August, so we can bundle up some good prizes, get things going, shoot the video, and so on.

Help Archive Team be a slightly better houseguest, although we’ll never been able to explain off what we left in the fridge.

Thanks.

6 Comments

The GET LAMP Raw Interview Flood —

When you overshoot your documentaries as much as I do, you end up with a lot of perfectly good, probably movie-worthy footage that just doesn’t make it into the final cut. When I edit down the dozens of hours of interviews, I have thousands of clips of people saying on-topic, coherent things, all labelled, and then build the movie out of that. It works well, with my subjects completing each others’ sentences and a smooth narrative going from one topic to another. But that leaves hours and hours of interview unseen.

My solution, which I started with the BBS Documentary, is to post up the interviews for everyone to see. I do this thanks to the Internet Archive, which has the space and bandwidth to provide the gigabytes of interview I have on my hard drives.

Recently, we finally set up a situation where I fedex a drive into the Archive from my home in New York, meaning that I can dump in hundreds of gigabytes at once and not wait for a California trip or an anaemic cablemodem to do the work.

So, here we go: 50+ interviews from GET LAMP are now up. They all have some level of description of the circumstances of the interview, including why I interviewed the person. It even includes interviews for which no footage showed up in the movie. (So far, that’s Keith Nemitz, Richard Hewison, and Paul Meyers.)

For the insightful, motivating portion of the audience who fat-finger out a “WHAT TOOK U SO LONG” on the release of these, it mostly came down to time and opportunity. The previous desktop machine, which GET LAMP was cut on, could take between 10-20 hours to render out a video-noise-reduced, full-hour interview. The new machine does same in less than an hour. Also, I was working on a lot other priorities before this, so it took a while to set aside the process of putting these interviews together. So that’s why. GET LAMP was released in August of 2010, and here it’s roughly a year and a half since that time, so it’s not too bad. But since the interviews were done in 2006-2008, it does mean that up to six years have gone by since these nice people have sat down and talked to me. I’m sure their lives and opinions have shifted and changed since some maniac wanted to talk to them extensively about text adventures.

If you’re not sure where to start, I suggest starting with Dave West, the lead of the cave expedition filmed for GET LAMP. Then if you want some clearly-stated nostalgia, Rob Griffiths goes over what it was like to be at the beginning of text adventures as a young player. From there, you can just click around, listen to people. I uploaded the HD files as MPEG2 renders, and then the system redid them into H.264 versions. I suggest looking at or downloading the H.264 versions, unless you’re looking to cut a new version of the movie – then you can just use the original MPEG2.

These interviews are just the answers given by the people, not my questions. Why? Well, two reasons, one bitter, one not.

BITTER: With the BBS Documentary, I started uploading the full hours, just going from one end to the other, dumping out the entire tapes. What I got, frankly, was a lot of shit from a lot of cheeto-spitters about my style of interview, one which results in very good footage. It comes down to the fact that these aren’t just interviews about a new app someone just came out with, or what people think of a current political situation – these are intense mining situations where the interviewee and I are working really hard to bring back memories that have been absolutely dormant and untouched for, in some cases, decades. I tell jokes, I give them common-ground stories about myself or what I’ve heard.. really, whatever it takes to get the footage, the emotion, the sense of “being there”. That’s how the sausage is made, and I got tired of people ripping into the interview footage and just stopped uploading the fuckers. I don’t want to walk away from that for GET LAMP and so this new approach sans Jason is underway. I am sure I’ll get back to the BBS Documentary and do it this way as well.

NOT BITTER: Besides the fact that going back and re-ripping the full interviews would delay this project another couple years, the fact that I went through the months of pulling the gems from these interviews and putting together clips I thought were GET LAMP worthy means that the clips should be able to stand on their own. Cases where I would say “Was it worth the trouble to code this feature into your game” and the person would answer simply “yes it was” are not present in the final clips – the clips had to impart information and be useful in the film. That filter means these interviews are pure fact – not chumming around, not going off into wild tangents about where it’s good to eat around here or what happened down the street last week. And they’re not filled with me telling the same stories over and over to help dredge footage. They’re “cooked”, not “raw”.

In some cases, this means the person says something, the clip stops, and the clip plays with an overlap, as it goes to a new idea. For these minor hiccups, I think they’re very watchable, and informative.

So enjoy!

There’s still copies of GET LAMP available for sale, with gold coin included.

9 Comments

Kickstarter Endorsement: GIT ANNEX ASSISTANT —

UPDATE: This kickstarter hit $20,000, meaning Joey will be working on this new project for a whole year. I can just imagine the possibilities as it matures and gets done.

Over the course of this year I will occasionally push you to kickstarter projects related to things I think are important to Archive Team‘s goals and mission. I’m not as likely to write these for just any project I like – if you want to check THAT sort of thing out, it’s probably a better idea to just follow me on Kickstarter and see what I’m up to.

Today’s suggested project is GIT-ANNEX ASSISTANT, by Joey Hess. At this moment, there’s about 11 days left to get in on it. He asked for $3000, he has about $13,000, and he wants $20,000. I want him to have it too. This is why.

To understand the power of GIT ANNEX, you need to understand GIT. To understand the power of the GIT ANNEX ASSISTANT, you need to understand neither. Let’s do what we can.

GIT is, at its heart, a version control system, allowing multiple people or entities to work with code and ensure nobody steps on anybody, changes can be reversed, and that progression, rather than mistakes and screaming, are the order of the day. Programmers love git. They love it enough that a site that does nothing but make git respositories public, called GITHUB, is a hugely popular site. But you’re not a programmer, or you work alone and don’t need this. Fine. Stick with me.

Normally, with GIT, you need to “check in” everything you’re coding into a GIT Repository. This is fine when we’re talking about a few dozen or hundred of megabytes of code. But, and we’re making a jump here – GIT isn’t really just for coders and developers. GIT can track any data. And that any data can be photos, videos, music, CD-ROM images, you name it. And with that comes the power of tracking changes, of sharing updates, and of undoing silly things done by others or by yourself. But you really don’t want some poor GIT respository to hold every single point of data being tracked – not when you’re using, say, GITHUB or another public repository unwilling to deal with that level of stuff and preferring to stick to the “keeping track of how it is being handled” job. Well, GIT ANNEX does this very thing – instead of the repository having to keep a copy of, say, every file in all your hard drives, you can instead use GIT ANNEX to say “just keep track of the description of these files” and that’s all it’ll do.

Am I losing you? Keep with it.

I am not a developer, not a programmer. I don’t use GIT at the moment for any of my projects (although plenty of Archive Team members do, and I and many others benefit from its use). What excites me about GIT ANNEX is how it fundamentally tracks the backup and availability of any data you own, and allows you to share data with a large or small audience, ensuring that the data survives. 

This is where we get to GIT ANNEX ASSISTANT.

GIT ANNEX ASSISTANT, built on top of these rock solid building blocks that already exist and which are in use around the world, will bridge that last final and critical gap from programmer-nerd-wonkery to next-generation-data-availability for the end user. It will enable you to tell this graphic interface to “just do it”, and ignore all the amazing stuff underneath that’s making sure stuff works for you.

If you’re stuck back on the “woah, hold on, what exactly is GIT ANNEX there, bucko” phase, then feel free to browse around the GIT ANNEX site. Joey Hess has done an amazing job explaining this utility.

But GIT ANNEX ASSISTANT will allow you to use all this cloud technology for what it promises, instead of what it delivers.

Now Hold Up.

Did the dude who wrote Fuck The Cloud just endorse cloud storage?

Well yes, yes I did.

Because GIT ANNEX (and therefore GIT ANNEX ASSISTANT) makes it so that you can make sure your data is on a bunch of clouds, and local storage too, meaning that you don’t have to trust any of those fuckers, including yourself. Fuck the Cloud before it Fucks you! GIT ANNEX ASSISTANT will be tracking things for you, asking you to plug in drives when needed, and retrieving your data.

Oh, but even better. It allows you to declare data in such a way that a bunch of you can share file repositories, not unlike Dropbox or an FTP site, so that everyone gets access to the data, changes are tracked, and so on.

Joey is fantastically cheap. I went over this whole project and his lifestyle with him, and $20k gets him full-time on this project for a full year. He’ll be going through the list and coding up greater platform access, improving GIT ANNEX ASSISTANT to be even easier, and a whole host of related projects. This is a fantastic bargain.

As I write this, he has $13,000 and 11 days to get to his goal. I am endorsing him and telling you this will help the world get away from obfuscated, cheap-ass and unregulated cloud storage, and back to you being in control of your shit without your shit taking control of you.

Here’s a link to the kickstarter page. He has my full endorsement.

Thanks.

4 Comments

UNDO DEL IUMA —

The best projects cause a lot of people to shrug or go “oh, that’s nice” and then a much smaller percentage of people to drop whatever they’re carrying and stagger forward in disbelief. Here’s one of those.

For some people, just bringing up the phrase “The Internet Underground Music Archive”, or “IUMA”, summons memories of all sorts of cool bands who were technically savvy enough to make their music available online for a small but growing audience. Through the IUMA main site, you’d browse hundreds, later thousands of bands, and within those bands would be thousands, later hundreds of thousands, of songs. It was huge, and it was long lived – founded in 1992, we’re talking a site that brought you music through .AIFF, .MP2, RealAudio and MP3 formats because it wasn’t clear which of those would be dominant.

Trust me – IUMA was the place to be, one of the rock-solid sites on the net, as powerful and as well-known a force as Hotwired/Wired, Suck, Drudge, Salon, The Well. A mainstay.

If you want an awesome overview of what the hell IUMA was, please zip on over to this YouTube video of a CNN story on IUMA from 1994 – all the founders are there, the power and strangeness of audio in the 1990s is there, and best of all, the sky is obviously still the limit and clouds have not yet appeared. (There’s a similar story from 1994, but it’s on MTV and even by 1994 MTV news is pretty awful.)

So what happened? Well, like a lot of other such endeavors, it was sold in the dot-com boom years, Y2K edition, where it fell under the retarded purview of multiple owners, and then entered a state of living death through the mid 2000s until undergoing a sad, dull little shutdown around 2006.

With that initial 2000 buyout came increased resources and bandwidth and reach, no doubt, so a lot of bands signed up with the place, dropping example tracks and then asking you to buy their CDs or come to their concerts. The entity that ate its lunch as a profitable venture was likely Myspace for many bands, with others being sucked directly into things like iTunes and Amazon and so on. IUMA didn’t keep up, and now it’s just a distant memory.

Well, until now.

6 years ago, John Gilmore (yes, That John Gilmore) saw that IUMA was in a zombie state and very unlikely to ever get out of the ICU. So he grabbed a copy of all he could – which wasn’t all of it, of course, but it was a hell of a lot of it. He stored them on some backup tapes, and as the site went down, disappeared, and faded into the mists of memory, he looked for a chance to have someone get a copy up somewhere. I was that person.

What I just spent most of the last week doing was taking some dupes of John’s backup tapes, writing scripts to ingest them into archive.org’s servers, and now I am going to tell you that I put IUMA back up.

Oh, you are in for a treat and a hell of a lot of modern musical history just got saved. This is over 45,000 bands and artists, and over 680,000 tracks of music. That number sounds made up, but I’m not kidding – six hundred and eighty thousand songs are in this collection. I did a back-of-the-google-calculator check and came up with 243 days of music – solid, 24-hour days of songs. You could leave it running now and look up in 2013 as your playlist ran out.

Where possible and where they were grabbed, I added descriptions from the HTML files for the site. Pictures do not appear to have been saved due to a quirk of the download (they kept the photos on a weird server) – but you have enough to go on. Compare, for example, this page from the wayback machine and what I got ingested into the site. Not perfect, but something.

This should all be considered 1.0 – if I find more ways to pull in information properly, I’ll do so. And naturally I’ll ensure the original, before-jason-messed-with-it data is stored safely away so the next set of folks can try better techniques to get it back.

It once was lost, and now it is found.

Let the music play.

 

21 Comments