Open Locative Media standard – who/what is it for?

Home Forums Technology Open Locative Media standard – who/what is it for?

Viewing 15 posts - 16 through 30 (of 30 total)
Author Posts
#43052
Babak Fakhamzadeh
Keymaster

Yes, WLC could host files. But, even at small individual volumes, we’ll need a budget. Our budget, so far, has been 0.
But, this opens up a completely different Pandora’s box; we’d be operating in a similar space as, say, Soundcloud, which struggles to survive. And that’s not considering ‘heavier’ media files like video.

In the mean time, WLC-based file hosting would not be used by users creating a soundwalk if they are not aware of WLC, meaning that WLC-based file hosting can only solve the problem that Josh mentions, for some.

I can be convinced otherwise, but I’m not sure that the benefits of offering ‘our’ own file hosting will not outweigh the benefits.

That said, I can also see a possibility where we’d offer a relatively small amount of storage space for the files, perhaps only a particular series of types of files, and these files only, that referenced in the soundwalk.

@Hamish, one more thing about the signatures: It’s possible I was not fully understanding your underlying thoughts on the purpose and meaning of them. Before we fully dismiss them, if we dismiss them, are you able to elaborate on how you saw their purpose a bit more?

#43539
Josh Kopeček
Josh Kopeček
Participant

So we’re at an impasse as regards the hosting of files/media is concerned. Does anyone have any suggestions?

The media is absolutely integral to the walk. One might actually argue that it’s more important than preserving reams of metadata (aside from perhaps locations, even then…).

You cannot do a walk without actually having the media. So I think we could possibly turn this on its head and ask again: how and where do we store these media?

#43548
Babak Fakhamzadeh
Keymaster

I think we’re discussing two things, here; the importance/need of a standard, and the importance/need of preserving media. These are related, but not the same.

Defining a standard stands apart from preserving media. So, we can talk about one as well as the other, without requiring one for the other.

Because long-term cost-free, or nearly cost-free, file hosting is a completely different beast altogether, I don’t think this needs to be part of the discussion on a standard for locative media.
That said, I’m also very keen to discuss options for providing long term cheap storage.

#43566
Hamish Sewell
Hamish Sewell
Participant

For me the importance of the standard outweighs the need for preserving media. Standards prompted this conversation we’re having, and to date there is still no standard (but clearly that’s coming, and most likely on a number of fronts). I’m not sure WLC wants to be in the business of archiving, despite being well placed to do so in such a niche ‘industry’.

This discussion has moved away from my area of expertise. I can’t pretend to understand some of what’s being discussed, though I appreciate that much of what’s come up here is a direct result of what @kris and I originally mooted with our .stx files.

I’ve got a background in audio. I believe audio is intrinsic to the immersive site-specific digital experience. An .stx file (or whatever we might call it/them) has come out of our rebuild and it clears away many obstacles and creates a much more viable proposition: way less technology dependent; hence simpler and more creatively enabling for audio producers, more culturally and politically enabling for communities connected to places. In the face of so many of these amazing stories disappearing, this type of approach suggests a new approach to the very survival and longevity of our collective works. I’m keen to get on with it!

#43735
Josh Kopeček
Josh Kopeček
Participant

@Hamish

this type of approach suggests a new approach to the very survival and longevity of our collective works. I’m keen to get on with it!

Hopefully we agree on all these things:

  1. we must preserve the work in its entirety
  2. the audio is one of the two core components of the work
  3. the format must be plaintext

We currently cannot satisfy all of these demands.

Currently I don’t see the point of starting to establish a standard for locative media which doesn’t preserve them in their entirety, or at least provision for long-term archival. On the other hand, establishing a vehicle for moving soundwalks between platforms is a worthy cause, but most of the battle will be getting a platform to agree to an import/export mechanism. Since platforms profit from people keeping their works on their platform, that may be an uphill struggle.

Creating a standard in the first place is a monumental task, especially with such a malleable and creative discipline as locative media. This is not something we rush into or take lightly.

A literature review, or funded PhD as well as stakeholder input might be a worthwhile starting point for this.

We need more voices to weigh in on this.

#43736
Hamish Sewell
Hamish Sewell
Participant

What’s the other component, other than audio, are you referring to here @Josh: Image/audio visual?

Preserving in its entirety…Within the remit of most locative projects today (recognising that yes there are some very creative projects that don’t follow these conventions) isn’t this what a standard map/work is doing?

If such a project can be shared, why does this necessitate an archival tool? Isn’t that creating an impossibility, given the clear lack of resources.

“establishing a vehicle for moving soundwalks between platforms is a worthy cause, but most of the battle will be getting a platform to agree to an import/export mechanism. Since platforms profit from people keeping their works on their platform, that may be an uphill struggle”

A very important point and there’s a lot in this, and clearly a bigger cohort of interested parties at this stage is needed. But I’m not sure if I agree with you. To my mind folk are itching for there to be some way of uncoupling of the shareable experience from the platform and it’d be interesting to put this question to the test by asking a raft of people who are in this industry, area. Again, I’m no developer, but my understanding @kris is that .stx files are pretty easy to enable on platforms being underpinned by a .kml file.

#43764
Josh Kopeček
Josh Kopeček
Participant

The two core components are audio (or media if we’re talking about locative media) and location.

To go back to the beginning of this discussion, plaintext cannot hold audio (or any media). Therefore we have to store it somewhere else, so the question of storage becomes necessary to answer.

What @Babak raised was that actually a standard and archival of the associated media are two separate issues. Which I agree with to some extent, but that negates one of the core use cases of the standard, which is providing longevity for legacy projects.

I think this comes down to the fact that I don’t think creating a standard for the sake of it is a worthwhile cause. Also, portability between platforms is not enough for me. If you’re wanting portability, an import mechanism is enough – look at Microsoft Word, for which most major platforms some sort of import facility. Having an ‘.stx’ import would be good for the moment. I’d be happy to add ‘.stx’ import to our platform if we could get a ‘.echoes’ import on SoundTrails!

@Hamish Would I be correct in assuming this is what you’re aiming at? A standard so that you (or any other platform) could easily import an old soundwalk?

@Babak Do you think it might be worth approaching the Internet Archive to see if they’d help with the archival process if WLC is not the repository?

#43790
Babak Fakhamzadeh
Keymaster

If requiring archival before a standard, lack of structural resources might prevent moving forward. Yes, a PhD study on this and the general concept of a standard, would be great, but, who can wait four years? And who would pay for that? Half the current apps/platforms operating in this space could be effectively dead in four years.

Involving multiple potential stakeholders is a good and important move forward. But, getting them on board would see a larger chance of success if we are able to agree on a unified and clear message, first. Inviting more people can result in more disagreement and a real risk of discussions and resolutions never leaving the starting gate.
If we do not start this discussion with a certain level of agreement, and no funding, we might not get anywhere.

I’ll let @hamish-sewell answer for himself on his intentions, but it seems to me that the obvious objective, here, is interoperability. If there is a standard, any platform can convert from and to that standard, and vice versa. So, @joshkopecek, if we agree on the standard, you can create a converter for .echoes to .olm, and @kris can create a converter for .olm to .stx.
But, this needs no explanation, right? We are all clear on this.
It’s just that, at this point, it seems to me that @hamish-sewell and @kris are in a position where, whatever the standard ends up being, Soundtrails actually can use that standard natively, without requiring an additional translation/conversion layer.
So, perhaps, for the sake of clarity, perhaps we should talk about .olm files as the standard, as opposed to .stx files, to clearly distinguish between the standard, and the format used by Soundtrails, even though Soundtrails might end up choosing to use the standard as their format.

I do think that creating a standard, without also necessitating the existence of a perpetual storage medium, is a worthwhile cause. Defining a standard is a much more important step than facilitating another free storage platform, which would have its own unique challenges, plenty of which would be completely unrelated to facilitating an open standard for locative media.
That’s not to say we can not envision and/or plan for providing a kind of storage facility for media that’s part of the instances of .olm.

Yes, @joshkopecek, approaching the internet archive is a great idea. In fact, it turns out that they already host audio, for free.
I just uploaded this, and it seems to work fine:


(Note, I marked it as ‘test’, meaning it should be automatically deleted after 30 days.)

They also appear to host video, but I did not test this.

So, this means that it appears we have a working solution for perpetually storing audio. Or, rather, as perpetual as it reasonably can get.

To go back to @joshkopecek’s list of points:

  1. I’m not so sure we all agree on that we need to facilitate archiving works in their entirety. But, with Internet Archive hosting audio and video, we (or, the user) can, at no extra cost. Specifically, then, WLC could be a repository of .olm files, with linked media hosted on the Internet Archive.
    This could, at some point in the future, and with meaningful financial support, be changed to storing all data at WLC.
  2. In the sense that the work is not accessible, if the media is available, then, yes, audio is one of the two types of essential components of .olm works. However, a standard can, and does, exist, without the need of linked/indexed files existing.
  3. Yes, the format must be plain text.
#43806
Hamish Sewell
Hamish Sewell
Participant

Hi Fellas. I’m not thinking of a standard as a way of importing former locative audio works @josh. It will be easy enough for folk to import on our new backend, but perhaps this is an option down the track. I feel the interoperability of locative works is key here, whatever file name it goes by. And I believe there’s an appetite for this at large and am keen to see some movement here.
While I’m less focused on resolving the archive question, it does seem that Internet Archive is an interesting proposition. I had some trials with Internet Archive some time back. It seemed to me that all material uploaded was publicly available under some sort of CC license, whereas work I wanted to upload needed to be private. So I gave it a miss. Perhaps I got this wrong. Perhaps material is able to be locked up with a UN & PW. Anyhow, they’re a great contender for this sort of work and I think it might be a good opportunity for WLC to make some enquiries.

#43839
Babak Fakhamzadeh
Keymaster

It’s possible to attach a range of licenses to work uploaded to the Internet Archive. But, all variations of Creative Commons.

I’m quite sure you can not put content behind a login; it would arguably defy the purpose of the Internet Archive.

Either way, I think this is sufficient for our purpose and needs, here; we’re looking for a place to store media in a way that reliably provides longevity of accessibility.

#44211
Kris
Participant

Hi all,

It is good to see the discussion evolving here – and I’m sorry to be so late to join in.

Here are a couple of cents from me on a couple of the topics :

Plain text vs. binary audio files :

There is certainly a dilemma inherent in having a file format that is descriptive and having the “real” content in seperate files. Keeping the audio and image files seperate from the new file format which describes the walk as a whole has definitely been the approach I’ve assumed as best from the beginning – a single file that combines them becomes unwieldy in size and is certainly not human readable. I believe it must be this way that the (audio and image) data files are seperate from the locative-media-experience (walk) file.
(I’m going to use “walk” to refer to the new data format in this post, we have been using “.stx”, and Josh, you proposed “.olm”, which I like)
The storage of these is seperate to the “walk” file formats themselves of course, but there is still a responsibility to consider how the data files are kept with the “walk” file seeing as the latter does not work without the former.
The approach we have taken so far is to make the “walk” format be completely flexible in terms of specifying the storage location of the data files (you could have a single audio story with different images from all different locations for example, or have everything in the one location and specify a base URL, or a combination of both).
We are now working on our builder for and what we are doing with that is applying a convention over the top of how we utilise the “walk” standard – essentially keeping all the data files in a single location with rules on the folder structure based on users and file types and story identifier.
What that means is preserving a given story is just a matter of storing the “walk” file itself and the folder that contains all the associated data files.
(Actually potentially the data file URL reference could even be relative to the “walk” file location, although we hadn’t yet considered that before now)

To summarise my ramblings : I don’t believe the storage of the data files should be part of the new standard file format, but that is mostly because I don’t see a straight forward way to approach that, and because it can be worked around by applying conventions and being strict with them.

Regarding the format itself :

I want to mention here that a premise that we have started with, and that I suggest should remain, is that the foundation of the format be the map.
This is the thing that separates what we are wanting to achieve from the existing TourML format, or something else that is completely generic.
I’m also suggesting using KML as the base and adding kml-valid extensions to that, because it means that all the existing tooling out there that works with KMLs will be able to at least partially represent “walks”(/.stx’s / .oml’s) already.
I know you want to discuss high-level details Josh before the nitty gritty, but I do wish to put the map-centric, and kml-based, proposals into the discussion here.
(Being KML doesn’t need to mean only XML either, as GeoJSON is the JSON equivalent of that – and there is of course no reason you couldn’t have represent the same thing as YAML also)

—————————————————

For those coming to the discussion for the first time here, here is the link to the proposal that kicked off the idea for the standard :
View at Medium.com

#44871
Josh Kopeček
Josh Kopeček
Participant

I believe that in locative media we have to be open about what form these works take. Which is why I’m hesitant about tying us to a ‘location first’ approach. KML is strictly ‘map annotations’, whereas we are proposing media which relates to a location. Two things there – media and location. I can envisage (especially amongst legacy walks) media which might not be attached to a location but which might be integral or have different ‘triggers’ – such as manual playback, or intervention by a performer. So we can take a ‘location first’ OR a ‘media first’ approach here.

This might be nitpicking, but it’s one reason why I don’t think we should jump to any assumptions about the structure of the format.

To cut this short, I propose an investigation to examine the existing canon to establish leading legacy soundwalks and their structures as the best way to start.

#44964
Hamish Sewell
Hamish Sewell
Participant

Studying the canon of locative media: along with archiving possibilities, looks like a PhD @Josh – or several. Many iterations of the locative experience here & constantly evolving. I can’t see myself doing this. Whereas KML (or others) is ready to go now and I’d suggest, serves a real need. Where to now people?

#45032
Josh Kopeček
Josh Kopeček
Participant

Can I tentatively suggest that KML is not our solution for the reasons I outlined above? If you’re looking to establish an open standard for soundwalks it needs to be designed by the community, for the community. If you say “here’s my standard, adopt it” this benefits only you.

TourML – from what I’ve seen – looks like an interesting starting point. The ‘stops’ are the central point of organisation for the schema, and by extension could be associated with geolocations or other stops. There’s no way for KML to have an independent piece of media or other interactions defined. TourML is also designed for migrating museum audio guides, which might not be fit for purpose either.

For instance Janet Cardiff’s walking pieces don’t have a GPS coordinates associated with multiple points of audio, but they are most definitely locative audio pieces. So how do we find a descriptor for them?

#45103
Babak Fakhamzadeh
Keymaster

It seems to me there are a few strands of intent that are intertwined, here.

A standard should be broad enough, but also has to recognise it is unlikely to cover all instances and all eventualities. A standard that allows for everything is not likely to be a standard, meaning we would have no choice but to make abstractions and set limits.

A format that has a broader application than existing formats, and is more widely adopted then existing formats is beneficial to everyone involved.

It needs to be clear what the applications and limitations are for any format that we suggest.

To understand the potential challenges and limitations, we need to understand the variety of works we might want to include in being able to be described through this format.
Without being too abstract, can we think of examples of works, and what exactly sets them apart functionally, that we should consider to be able to be described through this standard?

I’ll start.

  • Pretty much all ‘tours’ in izi.travel (or rather, the ones I’ve tried) are quite conventional audio guides. That is, audio is connected to geometric shapes that are defined in a geospatial environment (that is, circles or other shapes on the surface of the earth).
  • It’s my impression that the works available through Soundtrails are, effectively, functionally, similar to izi.travel; audio/media connected to geometric shapes on the surface of the earth.

Josh, I’m not familiar with Janet Cardiff. Her website appears offline. I just checked the description of a work of hers, here:

https://web.archive.org/web/20051124075037/http://hirshhorn.si.edu/exhibitions/description.asp?Type=&ID=20

It seems to me this particular piece is, really, just one audio file.

What about Echoes? What about other, similar, tools? Do some include a kind of data that goes beyond media connected to geospatial information?

Viewing 15 posts - 16 through 30 (of 30 total)
  • You must be logged in to reply to this topic.