Town Halls - why typed?

Ok this has been bugging me for over a year.

Why are they typed? We have these 'repeaters' sending text around all over the place - and yet we not only have the functionality to stream audio to any parcel in SL - but we also have this talk of 'Linden Radio' (or *HAD* that talk)...

So the question is:

If me and my dinky computer on a Sunday have 30 folks listening to me rant for 3 hours...

... why can't LL stream Philip from a *real* microphone and give us a URL to listen to.. doubly so if 'Linden Radio' is planning on doing streaming audio anyways? Isn't that 'news content' that was being discussed - how much bigger do you need than the CEO of the company doing a 'Fireside Chat' with his online nation.

Think about it - you could tune in if your not in SL.

Crashes wouldn't be a huge problem, nor lag - folks could choose to stream it to their parcels -- or INFOHUBS (finally - a USE for these big white elephants!)

It actually USES the technology - your own dog food... so if the tech aint up to scrap for it - you'll know first hand how something is used and how to improve it.

The entire thing could then be recorded and put on a URL for folks to download that missed it.... no formatting text - no filtering out shouting or nipple ring spam.

And with the amount of interviews the dude does - I don't see how he could have a worse time of it than my word-stumbling-constantly cussing ass on a sunday.

Good thoughts Siggy! Stuck in his ways

BTW my motives on this aren't altruistic - If such were to happen I would totally edit my own voice in there for a 'weekly interview with Philip Linden' and play it back on my radio show

I think this was tried once, way back in 1.4; I recall there were some technical issues with getting it setup that spelled the doom for it.

This has been mentioned before, but I there were several overriding factors

- chat is the de-facto norm of communication in SL
- it's easier to provide a transcript
- transcripts are easier to skim, and also to read slowly for players who's native language is not English

I'd like that. I'd rather listen to whatever they're saying while I continue building on my plot, free from lag, and less prone to crash or be spammed.

I'm beginning to appreciate more the use of voice versus typing. They should give it another try.

... and text is easily searchable and transformable (colored highlights, anyone?) by very simple parsing tools available to everyone.

There was also at least one deaf person at the town hall who wouldn't have been able to participate otherwise.

I attended part of the town meeting from the computer lab at school. If it had been an audio chat, I would have been unable to attend; it would have been unfair to all the other computer users around me. (I normally have the sound muted when I use SL from there.) Surely I'm not the only SL user who is sometimes in-world from locations where sound would be unacceptable!

They could do that and type it out too since all the speaker is doing is answering questions asked in these forums, after all there appears to be almost no interaction between the speaker and the audience during this meeting or am I missing something?

Halve the questions answerd where from the audience.

I have wondered same thing for a very long time.

Good point. How about a simultaneous text repeater, similar to subtitles ?

I'm sure that any competent typist who has ever taken dictation or worked with audio transcription could do a 'subtitles' relay of the general gist of the conversation, as long as Phillip didn't speak too fast.

Lewis

Let me bring up the most likly technical reason:

Bandwidth.

It costs an ugly amount of bandwidth to stream live audio to hundreds or thousands of residents. We are talking, even at the lowest quality levels, about 4 Kb of data per second per listener.

For 1000 people, that 's four gigs per second. Considering that we have to stream all the data for SL from the same pipeline, it's a bad idea.

The second reason is this: Phillip would have to have a streaming server setup on his computer, and then that would have to be redirected to a distro server that sends it to the general listeners.
If you get enough people listening, then you have to add a third layer to have enough power to distribute the stream evenly to everyone. This system would have Phil streaming the audio to a central server. This server would then resend to a series of sub-servers...

And so on.

We are not talking about just a few people listening here, we're talking about some major work.


Thank you, and goodnight!

Shoutcast servers exist for exactly this reason. Online radio stations do it all the time. Club 977, for example, at the moment has 4487 listeners out of a possible 66312 on its 128k feed. I normally tap into 24k streams where possible to save using up unnecessary bandwidth, and for voice only this should be more than sufficient.

The technology and solutions exist already from third party sources. Nobody says LL have to set all this up themselves.

When you think of all the tinpot radio stations out there broadcasting crap to the community in some form or another, it can't honestly be difficult. I am sure that Shoutcast would have facilities to 'rent' a stream for a few hours once a month as a one off fee rather than paying for something that isn't in use 99% of the time.

The only thing standing in the way is LL's willingness (or lack of) to use third party tools, and the cost - which is probably miniscule compared to the overall operating budget.

Don't forget that normally only 5000 or so people are online at any given time, 4700 are camping, 200 are pixel humping, 75 are busy building, 20 are writing scripts, which only leaves 5 people to listen anyway But seriously, I don't think capactiy is much of an issue when you look at the amount of listeners on the Shoutcast indexes.

Lewis

I was speaking about shoutcast. I was speaking of a 32 bit stream. And I stand by what I said: They have to have the stream prepared fo the number of POTENTIAL listeners. No one knows the number of people who will listen in, and having too few could be fatal.

Generally, I don't know the cost of third-party services right now, I just have an idea of what my current server costs me for my CPU/Bandwidth. I figure my costs are probably lower than what would be charged for such services, since I'm about one tier closer to source. Either way, the point is, it's a lot less work and potential problems to use text transcripts.

IF SL were an 'audio' virtual world, of course it would make sense, but SL isn't, so it doesn't make a lot of sense.

Furthermore the use of chat, especially at the actual location of the meeting, helps in other ways. Phil is restrained from more outrageous flights of fancy by a live audience who can, when the urge takes them, rapidly get his feet back on the ground - or at least cause an abrupt change of subject!

Were he to sit in his office in front of a microphone his 'Town Halls' would almost immediately turn into nothing but propaganda broadcasts - not at all what we want.

I suspect that almost the only real feedback Phil gets from residents is at Town Halls, other than when he wanders in world in the guise of an Alt, which is another reason he might want to continue them. No matter how annoying and unruly we can be, there is valuable feedback to be gained at Townhalls which would be lost with the adoption of a disembodied propaganda audiocast.

If I wanted to listen to audio, I would've went to www.winamp.com.

But...I didn't, I went to SL...and generally in SL one communicates via text.


Welcome to the wide world of 'context'!


- Newfie

Expanding on this idea, I think that Siggy's radio show would be better if it was typed instead of spoken.

Actually, a lot of news company have transcripts still posted for people who want to read them.. There's no reason why they couldn't still have those.

And I think 1.4 still had issues with streaming music, period. Now, it wouldn't be so bad if they did do a "Fire Side Chat," kinda thing. It wouldn't have to be done "live," so that might fit better into Phil's schedule. Also, there is the nice feature of the fact that it can be played repeatedly, so that people across differing time zones could listen in, and not have to read the transcripts if they don't want to. I'm liking this idea.

Great minds think alike.

Earlier this week when I had heard there were going to be text-chat repeaters for the town hall meeting, I seriously considered doing an audio stream of the chat text in real-time using my own voice. That way, anyone who wanted could listen in rather than read. I even considered video recording the event so people could view it later.

Unfortunately, the meeting was held during my work hours so I couldn't do it.

I dunno his Aussie accent is kinda hot.

How much of a bad idea is some sort of voice recognision system?

bahahaha! Although, I do think Siggy's karaoke would lose something in the translation, don't you?