While national digital archiving strategies in Australia move slowly, some agencies are taking things into their own hands. The latest is the Australian Broadcasting Corporation, which has put slabs of archival material under a Wikimedia license. The move is part of the ABC’s 80th anniversary celebration, centered around its 80 …
Good on the ABC, been a leader in online media since the early days of the 'net.
I remember reading a few years ago that they were in the top 10 of raw online content, pretty sure that's changed with all the video sites out there these days, but it's good to see them releasing this stuff, far better then sitting in some dusty vault or server room like most broadcasters around the world do.
Since our taxes have paid for em
It's about frikken time
since WE do actually OWN the content.
The ABS got the message 10 years ago when they made available for free all datasets.
What about the rest?
It's actually the non-digital material thats of more interest. The ABC doesn't have the resources to digitise their archive, and there is significant risk that they'll license it to commercial parties to do so.
@ Glen Turner 666 - Re: What about the rest? -- An excellent point which I omitted in my post.
Your point is very well made.
From 1932 to say 2002 when the ABC got serious with digital--there's a 70 year block of analog material that mostly remains to be digitized. It's my understanding a few items have been digitized and these are the result of the archive material being used in recent programs.
This large block of analog data is of great concern for two reasons:
- The commercial licensing that you cite, and;
- The possible loss or deterioration of at least part of the archive if digitization is not undertaken soon.
Commercial licencing is a specific worry as the archives will be locked up again and only available at a cost. This is a huge disincentive for ordinary or casual users, it means that for all intents and purposes, most archive usage will be commercial. If there is to be any public access it will be difficult and or inconvenient and possibly expensive.
My second point is worrying too as I've no clear idea of what state the archive is in, especially the older stuff that's in that intervening period between the transcription disks and reliable tape storage. This covers a time frame of somewhere between 1940--the wire recorder era and the 1970s. Even the period between the late 1970s and late 80s is of considerable concern, for that was the time when Ampex had major binder problems with its recording tape formulation--the binder would stick and or liquify with age.
(I've seen the mess it causes firsthand, it's nasty. Even when the binder is just beginning to break down it clogs the tape heads in a major way. Mastering tape type 456/457 was the most prone. Being a mastering tape--as opposed to the general workhorses, types 406/407--456/457 is the most likely to be used for archiving. There's a photo of the problem at Wiki under 'Ampex'.)
Back to commercialization of the archives. There's a huge precedence for commercialization of government film archives in the US, where the US government and military allowed huge swathes of military and other films (including WWII German newsreels captured by the US) to rot through for the want of maintenance. Rather than maintain the films, the military sold them all off in a block/job lot to a commercial company called Periscope Film (at one stage the military was actually planning to dump them all).
Periscope have packaged many films onto DVDs which can be purchased for a considerable price. Periscope also makes some titles available on free sites such as the Internet Archive but these have been nobbled with a timer/counter-ticker right in the middle of the image. They also now come with Periscope opening credits.
Here's two nobbled Periscope films from the Internet Archive:
Here's the 'Periscope Film' home site followed by their story of the film archive :
This must not be allowed to happen to the ABC archive under any circumstances. In fact, if I get wind of it, I'll immediately start whingeing to pollies about it. So too should everyone else.
Very mixed feelings about the ABC's so-called 'Public Domain' release of its archives.
With reference to the "gem among the releases" and link to the 1974 Clarke interview, I don't recall having seen it before but I'm very pleased to have seen it now. It only strengthens my already high opinion of Arthur C. Clarke. (I'm one of many who skived off from a uni physics lab practical in 1968 to see the Clarke/Kubrick spectacular, 2001: A space Odyssey.)
Yamal Dodgy Data correctly points out these archives are from the Australian Broadcasting Commission, which like the BBC, is a government owned organisation, thus having them unknown and locked in archives isn't of use to anyone.
I'm annoyed the quality of this clip is so poor given that it's obvious the original footage is much better. This video would have been copied from high quality 16mm film, and as it's almost certainly news footage, it's likely come from the actual camera stock, thus the original ought to be the best quality available--as 16mm goes.
This clip is only 512 x 288 px, yet STILL its image quality doesn't do justice to this low resolution raster size. It's low resolution, full of nasty compression artifacts and when viewed full screen, the film grain isn't even slightly discernible (it ought to be). The exact resolution of 16mm is subject to argument but with established parameters I'll roughly estimate and calculate what it's limiting resolution ought to be.
In 1974, the footage would likely have been Kodak 16mm 7266 Tri-x reversal B&W shot with an Arriflex camera using precision Angenieux, Schneider or Zeiss lenses. Here, high speed film was used for indoors, but high speed film doesn't have the resolution of slow film, thus professional lenses must have limiting spatial resolutions compatible with the slower (sharper) films so I've considered their effect on image resolution to be negligible (but in practice you'd allow for it).
- 7266/16mm spatial resolution/MTF is 50% at 50 cycles/mm & 20-30% at 100 c/mm.
- 16mm aperture is 10.26mm x 7.49mm.
- Assuming scan at exactly 90 deg to horizontal then allow a Nyquist factor of 2.2.
- For simplicity, ignore any cycle/pixel 'size' differences.
- Assume no meaningful response above 100 cycles/mm.
- Thus, to capture ALL image detail the film must be scanned at this resolution:
10.26 x 100 x 2.2 = 2,257.2 horiz.; round up: 2,260 px
7.49 x 100 x 2.2 = 1647.8 vert.; round up: 1650 px
- Masking aspect ratio 4/3 for screen display will change (may reduce) this a little.
Hence, roughly we have to scan the original archive film at 2,260 x 1650 px to ensure all data is faithfully captured.
Now, what has the ABC put into the public domain? Right, the clip is only 512 x 288px, and it's been seriously compressed to boot! Frankly, viewed full screen, the artifacts are ugly, distracting and annoying, and the limited resolution makes the image as soft as a baby's bum. Check for yourself!
PARTIAL 'PUBLIC DOMAIN' RELEASES BY THE ABC
What the ABC is effectively doing is to only make partial public domain releases; thus we, the poor long-suffering public, get shitty sub-standard copies. In the meantime, by retaining full-resolution copies, the ABC keeps all its options open!
The ABC is far from being alone in this practice: the BBC, UK National Archives, Australian War Memorial--a particularly bad example; Imperial War Museum UK, British Library, National Library of Australia, National Gallery UK, Australian Museum and art galleries in each Australian state are all at it--and that's just to mention a tiny few!
Above, I've only mentioned publicly owned entities, government enterprises etc. It's clear the citizenry owns the content of these archives but it has strictly limited access to them. If we physically visit these institutions we get quality, go on-line and we get almost unusable junk info. Once, when the internet was still dial-up, there was little option other than to visit these institutions or get low resolution copies on-line. Now, most of us have high-speed ADSL etc. and we've had it for years. Of course, commercial operations do exactly the same thing but that's a harder matter to resolve.
So what gives, what's their excuse? Many argue that much of their material is still in copyright but that's plain bunkum. Sure, they've copyrighted material, but except for libraries, it's often only a fraction of their collections: for example, The Imperial War Museum is awash with documents, photographs etc. that are much older than 70 years; and these materials were only ever subject to Crown copyright (which if technicalities arose, governments could fix instantly if they so desired).
There are two reasons why government/public institutions continue to hide behind copyright laws, the first is they're frightened of the internet, in the long run it has the potential to reduce their operations (and staff) by having most of their collections on-line; the second is more pernicious, which is that much higher resolution copies than those available on the internet are available FOR SALE in their shops or on-line. This is stupid logic, it's like a council that can't survive financially without levying parking fines, clearly the financial paradigm is wrong (although very understandable, especially in Commonwealth countries where public library, museum and art gallery access is usually free).
Still, this deliberate subversion and unlegislated extension of copyright is very much not in the long term interest of the public and eventually it should be rectified by legislatures. For starters, education is a looser. On-line students who've become used to high resolution video games, images etc. will simply gloss over boring low-resolution materials even if easily available, thus it's incumbent on these institutions to provide the best image quality possible. After all, these images are either already in the public domain or they're image reproductions, or they're photos of objects collectively owned by the populous at large.
For example, it takes little effort to imagine how ineffective teaching WWI history could be for students when presented with War images that are inherently very poor but which have been further reduced in quality to little more than blurs by additional compression and size reductions (as with most existing on-line histories of WWI. The internet is awash with this barely-informative junk and it turns kids off--I've seen it all too often to know it's fact). Nevertheless, many institutions that possess original photos still refuse to make Nyquist-limit quality photos available on the Web even though these images are in public ownership.
Others simply cannot get to institutions to see or research these materials firsthand, thus they must rely on internet access. In the interests of fairness alone, on-line image quality should be high. For example, many Australians live so far away from the national capital, Canberra, that they've never been there let alone had the opportunity to visit and browse through the national libraries, galleries and museums of that city.
Also, low resolution images with little detail and loaded with compression artifacts are a definite turn-off, especially so since the advent of high resolution screens (I'm using 2560 x 1140 px now). Furthermore, resolution and image quality is only going to skyrocket in the next year or two with the advent of super high resolution OLED screens. Argument about high resolution images being useless on smartphone or iPod screens is also complete bunkum as it's already easy to reformat web pages to correctly fit these small screens and do so in ways that pages are presented ergonomically.
ARE ANY PUBLIC INSTITUTIONS DOING THINGS CORRECTLY?
Yes, some are well on the way and are clearly aware of the public's needs. Two excellent examples are the Internet Archive in San Francisco (http://archive.org/index.php) and the Library of Congress in Washington DC (http://www.loc.gov/index.html).
Whilst not perfect, they're the benchmarks by which to judge others. The Internet Archive provides many different formats for each type of public domain listing whether it's sound, video or text; and the Library of Congress's 'Prints & Photographs Online Catalog' is a wonderful example of how to present an image with an extreme range of size formats from say 50kB to 150MB or larger (http://www.loc.gov/pictures/), here users can easily select the most appropriate size for their needs.
Here is a LOC image presented in multiple sizes http://www.loc.gov/pictures/item/det1994015865/PP/ It's Atlantic City, NJ about 1900, the smallest image is 57kB, the largest 154.9MB. Download the smallest then zoom in, repeat the exercise with the largest and be surprised at its incredible detail. Image resources of such high quality on the internet are truly rare beasts but it shouldn't be thus.
Access to the ABC's archives is to be welcomed but the above example amply demonstrates how inadequate the ABC's approach to quality has been. Essentially, the internet is a zone devoid of high quality. With few exceptions, almost all data is compressed to buggery and or devoid of truly high quality. The internet is or has become mediocrity personified.
The move away from paper-based books, sooner or later, will require on-line images to meet or exceed the quality of book images. Today, in El Reg's "CD: The indestructible music format that REFUSES TO DIE", many of you techies posted valid comments about CD quality being considerably better than most on-line downloads, yet few if any of you seem prepared to bat for improved on-line image quality. Frankly, I find this a little strange and inconsistent.
I'd think a good place to start would be to pressure the new ABC on-line archive to supply an image quality that befits the actual quality of the archive's content.
Reckon El Reg and its outspoken bloggers ought to be up to the task.
- Does Apple's iOS 7 make you physically SICK? Try swallowing version 7.1
- Pics Indestructible Death Stars blow up planets with glowing KILL RAY
- Hands on Satisfy my scroll: El Reg gets claws on Windows 8.1 spring update
- Video Snowden: You can't trust SPOOKS with your DATA
- 166 days later: Space Station astronauts return to Earth