Bump

I'm afraid that this has become a public secret :(

That is rather uninformed and mean minded. Check the 2.6x GIT to see what is being worked on and when stuff changes. There is shedload of dev going on in 2.6x currently.

Most of Bob's points (some valid) are because this is a scheduler feeding the content queue into Liquidsoap and thence to Icecast.

Have you filed your concerns as formal bugs in the bug tracker? If you do not - nothing will happen. That is how their system works.

Voisses' newest liquidsoap AutoDJ fallback script is really awesome. Implement it and maybe extend it - and most of your issues could go away.  Here

It seems you are expecting AT to be a live playout system. It isn't really, its a scheduled playout system. You can, of course, feed a live DJ system like MIXXX into AT as a stream - and do most of what you want.

John Chewter said:

this is a scheduler feeding the content queue into Liquidsoap and thence to Icecast.

And Sam Broadcaster is not a "scheduler"...airtime looks like a stripped
down version of SAM to me and they both do "scheduling". If the demand is there, which it is, then that's the
direction the project should take. I really don't see your point.
There's an overwhelming demand for Airtime to do everything SAM,
Rivendell, Station Playlist does and more. I really have to say, your
mindset is really limited and so what if Airtime feeds content to liquid soap. The potential is there for it to do what people want it to do and it's not there, and I suppose anybody who writes constructive criticism is "mean minded". :D

John Chewter said:

That is rather uninformed and mean minded. Check the 2.6x GIT to see what is being worked on and when stuff changes. There is shedload of dev going on in 2.6x currently.

This gets mentioned a lot; but not everyone knows where the git actually is (or what it is) and because of the its not initially that easy for non developers to find it; even if they do know a bit about computers; in exactly the same way that club/rave DJ's are initially freaked out when they see a proper radio station console when it is merely a bigger version of what they already use.

I did eventually find the link on the website; curiously I'd never noticed it before; but as it is I confuse Sourcefabric with Sourceforge and get git and svn mixed up as well..

https://github.com/sourcefabric/Airtime

it does seems some users are trying to deploy Airtime in a way that is not designed for which is going to amplify some existing bugs. It also processes a lot of data in real time and either does need a more powerful virtual server; or using the local icecast and liquidsoap to create a stream which can be processed with another instance of liquidsoap elsewhere (which is what I now do).

Airtime does actually work reasonably well (I've gone back to using it) provided you are aware of some issues and keep a sharp lookout on what is happening ahead; (and don't try and use crossfades!) hence why it remains popular with college/community broadcasters who have the resources to monitor playout (and with these broadcasters the Communications Ministry and their funders/advertisers often give them a bit more leeway than a small commercial station in a competitive environment).

Although the demand for the extra features is certainly there; there is already a culture in all software development (both open and proprietary) where more and more features are added instead of concentrating on getting the core of the software resilient; and development teams are becoming leaner and more fragmented (it was getting bad 15 years ago when I was last a broadcast engineer).

Hardware is cheap these days ("obsolete" kit can even be obtained for next to nothing); analogue broadcast consoles can often be picked up cheaply by anyone who has the transport to haul these heavy items; why not just set one up and use a combination of different software and hardware to keep your station going?

Airtime does have some bugs and issues (I can understand Barts frustration as if he is who I think he is I quite often listened to his station in their previous incarnation, and he might not want to trawl through yet more sourcecode when he has to already do that as a day job)

I wouldn't use Airtime alone it in its current incarnations for a station which has greater obligations to funders and/or the Communications Ministry; although it could be usable.

I wouldn't say the project dead but progress is slow; simply because the core team is very lean and funding harder than ever to get. I found Airtime by chance via a half abandoned blog about online broadcasting tools; which had a news article dated 2001 about how some big German company had gone bankrupt with the loads of jobs lost after a failed product launch for radio automation; if the Germans end up in this situation you know it is a challenging environment...

John Chewter said:

I don't see much constructive critism here. If you read Bart's history of posts, its just one moan after another and slurring Sourcefabric. Go and look in the Git and see all the work being done before spreading rumours that the project is effectively dead. It is far from that.
Sourcefabric is not a democracy, is a commercial company, following the demand of people who pay for their services and its very popular. That is how it works.

If patching a simple is script to add AutoDJ (script elsewhere in the forum) which is a 10 minute job, is one of the biggest gripes, maybe you would be better off with something else.

Why do you not use Sam - it seems ideal for your purposes?

Well obviously Bart sticks around because he cares about the project and wants to see it go in the right direction. I haven't seen any "slurring" of Sourcefabric. As far as spreading rumors that the project "is" dead, that wasn't said by me. In terms of using Airtime in a way in which it wasn't necessarily designed for, I don't understand that ideology when right on their website you find this:


It sounds to me like what Airtime is supposed to be is
exactly how we're using it, and features people have asked for haven't
been implemented yet. Airtime is clunky, slow, takes up a lot of CPU
time and isn't software that's very intuitive. I'm used to it now, but
I've been using software across different platforms for many years;
Linux, BSD, Windows, DOS, and I know a thing or two about usability and
UI design. Like I said earlier, Airtime is suffering from the
"developer's mentality syndrome", and it needs to grow up and graduate
by way of fixable problems such as the absurd 24 hour show idea, strange
placement of features and buttons (some things which are practically
hiding), smartblocks not populating playlists for 24 hours, and bugs
like the audio dropping out or 2 songs playing at once. It's hard to
write exactly what's wrong with the UI in detail as that's a
conversation that would be needed either verbally or in person to point
out what's wrong with it, but it's obvious that there's something wrong
with it as other people who have looked at our server agreed that they
didn't exactly love working with Airtime. Here's just a small example:
"Add Media" doesn't need its own menu on the toolbar. It should be an
option within the "Library" GUI. And in its current state, this huge ridiculous "Select Files" bar and drag and drop window take up the entire window. That's just silliness -- software that's not grown up.

I don't use SAM because I don't
run any proprietary software on our server (besides SHOUTcast DNAS but this will change in the future), you know, you never know
what has a backdoor or not so I just outright refuse to use closed
source software on my workstation or server and do not use WINE. We run
a Linux server and just switched over to Airtime from a BSD+SHOUTcast
Transcoder setup as there aren't many available options as far as an
automation system for Linux. Airtime seems to be the right solution for
us at this time. I'm just hoping it gets better and they listen to
feedback.

They did a lot of things right and there are some great aspects to it, and I'm surprised that it even worked at all when we installed it.