Can you think of other people to invite to this discussion? If so, tell me or use the system to invite them.
Also, should we schedule a voice/video conference to discuss this?
Monday, November 23, 2009
Paper discussion won't work!
Let me try to be controversial... I don't think you can get people to read and discuss papers ahead of time. Maybe someone will read some papers ahead of time if you publish the proceedings on-line a week before the conference. Can you do that?
If not, you will depend on people to post discussion at the conference. It might be worth experimenting with a live multiple person stream during the session. I agree with the comment about wanting to protect the speaker. I think the way to do that is to have a moderator. Doesn't have to do much, just push through or hold back a comment.
One way to seed the discussion might be to put a link to the paper and a list of links to referenced works in the paper - or maybe just a collection of <5 links.
What are the high value, low cost items to collect? I think it is slides, demo video, and pictures.
Hence, my suggestion is "do less but make it easy and efficient to use."
If not, you will depend on people to post discussion at the conference. It might be worth experimenting with a live multiple person stream during the session. I agree with the comment about wanting to protect the speaker. I think the way to do that is to have a moderator. Doesn't have to do much, just push through or hold back a comment.
One way to seed the discussion might be to put a link to the paper and a list of links to referenced works in the paper - or maybe just a collection of <5 links.
What are the high value, low cost items to collect? I think it is slides, demo video, and pictures.
Hence, my suggestion is "do less but make it easy and efficient to use."
Meeting with J-School student tomorrow
I am currently teaching a Web Programming course and one of the students is a Journalism major / CS minor. I had previously agreed to advise him for some independent study credits next semester. I am going to see whether or not he would be interested in contributing to this effort as part of that indep. study, as I am meeting with him tomorrow. Will report on that tomorrow as well as respond to Wu-Chi's comments about different directions that we might go.
Having thought about it for the last few days, I think that having a local conference server is not going to be particularly helpful. In the end, we will want to move all of the content on to something more archival anyway, so having a server on-site doesn't really help us much unless we think that bandwidth to an off-site server is going to really be that bad. I think much of that problem can be mitigated by limiting rich content (at least at first) to just audio of the presentations, slides, and photos that people want to upload.
Content is going to be key. I think Wu-Chi's option C where we invest in getting good content on the site in advance as well as during the conference will have the most impact. Participant-added content is probably going to be pretty minimal and probably limited to a few discussion posts on a few papers. We may be able to get some key participants who are interested to add a lot of content, but those are probably going to be just a handful. I would suggest that we recruit one or two "editors" who are charged with finding interesting related content (local information, related papers, related conferences, etc.) and drawing it in. Hiring a student or two to do this for maybe 3-4 weeks worth of part-time work might be the way to do that.
Having thought about it for the last few days, I think that having a local conference server is not going to be particularly helpful. In the end, we will want to move all of the content on to something more archival anyway, so having a server on-site doesn't really help us much unless we think that bandwidth to an off-site server is going to really be that bad. I think much of that problem can be mitigated by limiting rich content (at least at first) to just audio of the presentations, slides, and photos that people want to upload.
Content is going to be key. I think Wu-Chi's option C where we invest in getting good content on the site in advance as well as during the conference will have the most impact. Participant-added content is probably going to be pretty minimal and probably limited to a few discussion posts on a few papers. We may be able to get some key participants who are interested to add a lot of content, but those are probably going to be just a handful. I would suggest that we recruit one or two "editors" who are charged with finding interesting related content (local information, related papers, related conferences, etc.) and drawing it in. Hiring a student or two to do this for maybe 3-4 weeks worth of part-time work might be the way to do that.
Friday, November 20, 2009
A Facebook Group for MMSys
Another possibility would be to use the MMSys system for a forum per paper where people could add comments/discussions and presenters could add other content. We could add to that a facebook group for MMSys and have admin approval for people so that they can add photos and such.
Friday, November 13, 2009
Some brainstorming ideas… some not entirely thought out.:)
It seems there are two main areas of concentration: content and systems. Ketan and I had discussed making the proceedings available on the mmsys website before and during the conference (via password protected pages). We had also planned on making a discussion page / forum per paper to allow discussion regarding each paper. The goal was to create an archive where future multimedia systems researchers could get up to speed on a particular paper / area. What we want to ensure in the long run is that the mmsys website is used as an integral part of research, rather than a historical artifact.
Given this basis, how can we make the conference experience richer with a conference server. I think it would be great if we had an area to post slides, images, and video. Perhaps, it would be best in terms of brainstorming to start throwing ideas out for what the conference server might look like and iterate based upon what people think they need out of such a system. So here are three possibilities:
Advantages: simple, minimal dev.
Disadvantages: bandwidth could be a serious problem with richer content; it may be kind of clunky; may be difficult to do access control across a bunch of different services
Advantages: Let’s us control all the content
Disadvantages: work.:); it may be kind of clunky given that we’d probably have less polished interfaces
---
c) Local conference server – make the conference server a more integral part of the conference experience. In addition to the content in (b), we could work ahead of time to have people with interesting systems, etc to upload video demos on the conference server. You could also imagine having a lot of virtual tour type stuff of Scottsdale already filtered and organized so that attendees wouldn’t have to search all over googleland to enrich their conference experience.
Advantages: Big plus for attendees
Disadvantages: Lots of work
Some caveats:
*) Somehow we will need to ensure presenters don’t get dumped on while they are presenting… perhaps moderating somehow needs to be enabled.
*) We might have issues with people being video recorded. Larry, thoughts on the video recording?
Some other things to get started:
*) Do people have favorite systems (parts of systems) out there for blogging / social interaction?
*) Do people have favorite systems for capture / playback of video that we could create locally at MMSys?
Monday, November 9, 2009
Other responses
Several more people responded including:
Ramesh Jain
Klara Nahrstadt
Ketan Mayer-Patel
Gerald Friedland
Gene Golovchinsky
saying that it sounded like a good idea.
Ketan asked “what do you suggest as the best way to move forward?”
We need to put together a list of services that would be useful to include and a plan for implementing some of them for MMSys. Then we need to figure out who will implement what and how it will be done. And finally, identify what will be tested/trialed at MMSys.
To start off with, we need to brainstorm some ideas. I doubt we can all get together to do this, so we need to do it remotely and probably asynch. I suggest we setup a limited access blog with discussion threads that we can all post ideas to – we should make it possible to mail an entry to the blog too.
Let’s spend a week or two allowing people to suggest ideas. Then, we can hold a video conference with audio participation by folks who do not have the necessary equipment to work out a plan.
We need to figure out who wants to participate – no watching the activity without contributing. We should also decide if there are other people we should invite to participate. I don’t want to do a general mailing since we’ll get lots of people watching and talking but not doing anything. Right now – “doing something” is much more important than “figuring out the answer to everything.” Our time and resources are limited, so let’s keep it easy on ourselves.
Ramesh Jain
Klara Nahrstadt
Ketan Mayer-Patel
Gerald Friedland
Gene Golovchinsky
saying that it sounded like a good idea.
Ketan asked “what do you suggest as the best way to move forward?”
We need to put together a list of services that would be useful to include and a plan for implementing some of them for MMSys. Then we need to figure out who will implement what and how it will be done. And finally, identify what will be tested/trialed at MMSys.
To start off with, we need to brainstorm some ideas. I doubt we can all get together to do this, so we need to do it remotely and probably asynch. I suggest we setup a limited access blog with discussion threads that we can all post ideas to – we should make it possible to mail an entry to the blog too.
Let’s spend a week or two allowing people to suggest ideas. Then, we can hold a video conference with audio participation by folks who do not have the necessary equipment to work out a plan.
We need to figure out who wants to participate – no watching the activity without contributing. We should also decide if there are other people we should invite to participate. I don’t want to do a general mailing since we’ll get lots of people watching and talking but not doing anything. Right now – “doing something” is much more important than “figuring out the answer to everything.” Our time and resources are limited, so let’s keep it easy on ourselves.
Wu-chi Feng's Response
I'm interested. As you know from our discussions at MM '08, I'm interested in trying to create a community through the MMSys web site.
Our original plans were to:
*) Have papers on the site with discussion forums before the conference
*) Have archives contributed by industry (e.g., 360 degree surveillance footage)
*) A place for people to get the gist of papers (e.g., what people thought). In the long term this could be like a "product review" site for the papers. Hopefully, our papers will be viewed as excellent... keeping fingers crossed.:)
What you are proposing is great and in line with what we are trying to achieve.
Our original plans were to:
*) Have papers on the site with discussion forums before the conference
*) Have archives contributed by industry (e.g., 360 degree surveillance footage)
*) A place for people to get the gist of papers (e.g., what people thought). In the long term this could be like a "product review" site for the papers. Hopefully, our papers will be viewed as excellent... keeping fingers crossed.:)
What you are proposing is great and in line with what we are trying to achieve.
Conference Server for 2010 MMSys
This posting opens the discussion about the proposal Ramesh Jain and I discussed at ACM Multimedia 2009 in Beijing. The problem we discussed was the difficulty of enhancing the conference attendance experience, producing live feeds from a conference that can be viewed remotely, and capturing artifacts from a conference for historical access.
The background to the idea goes back to producing live webcasts from conference presentations and the effect that Twitter and Facebook are having. At MM09, Ramesh and I lamented that we could not easily access material from the conference through our smart phones and laptops. For example, we could access the electronic proceedings but we couldn’t access slides from the various talks nor could we easily share pictures that different people took at the conference. Moreover, some of us created Facebook (FB) tweets during the conference but the only people who saw them were “our friends.”
We discussed many ways to attack these problems, but eventually came up with the idea of setting up a server at the conference site with interfaces that allowed people to generate FB-like feeds and post slides, photos, and videos. The idea was that any presenter could post his or her slides before or after the presentation, anyone could post photos, and we would have a blog for discussions about the various papers and sessions. We also discussed the possibility of making the server available to people who are not at the conference – probably if they are SIGMM members.
Then, after the conference, the idea is to take the server back home and upload the content to the SIGMM website as a document of the conference. The presentations and blogs could be linked to the papers in the ACM DL and the photos could be linked to various event descriptions.
Eventually we’d like to capture more media live (e.g., audio, video, and rgb (for slides and demos)) and webcast it, but for now, doing something limited would provide a convenient way for people to communicate and contribute material to the conference archive and make it easier for people to access that material.
We think this approach would be much better than having someone collect material and post it after the fact. In fact, if the server software was developed appropriately, you could sync it to the SIGMM website as the conference happens. The only issue is network bandwidth to the conference site.
The background to the idea goes back to producing live webcasts from conference presentations and the effect that Twitter and Facebook are having. At MM09, Ramesh and I lamented that we could not easily access material from the conference through our smart phones and laptops. For example, we could access the electronic proceedings but we couldn’t access slides from the various talks nor could we easily share pictures that different people took at the conference. Moreover, some of us created Facebook (FB) tweets during the conference but the only people who saw them were “our friends.”
We discussed many ways to attack these problems, but eventually came up with the idea of setting up a server at the conference site with interfaces that allowed people to generate FB-like feeds and post slides, photos, and videos. The idea was that any presenter could post his or her slides before or after the presentation, anyone could post photos, and we would have a blog for discussions about the various papers and sessions. We also discussed the possibility of making the server available to people who are not at the conference – probably if they are SIGMM members.
Then, after the conference, the idea is to take the server back home and upload the content to the SIGMM website as a document of the conference. The presentations and blogs could be linked to the papers in the ACM DL and the photos could be linked to various event descriptions.
Eventually we’d like to capture more media live (e.g., audio, video, and rgb (for slides and demos)) and webcast it, but for now, doing something limited would provide a convenient way for people to communicate and contribute material to the conference archive and make it easier for people to access that material.
We think this approach would be much better than having someone collect material and post it after the fact. In fact, if the server software was developed appropriately, you could sync it to the SIGMM website as the conference happens. The only issue is network bandwidth to the conference site.
Subscribe to:
Posts (Atom)