Comments from respondents: "I'd also want really good stability, equivalent search, and archives from a paid FF clone." ($10); "Maybe not FriendFeed itself, but a similar service that I actually trusted? Would definitely pay a small fee." ($10); " It would need to improve, though!" ($50); "Even a $0.01 fee will stop people signing up." ($0) - Bill Hooker

I'm one of the $100 datapoints. - Bill Hooker

I was a $10 datapoint, but so much depends on the details. I could go higher, but it would have to be pretty damn cool to justify me spending $100 on it. - Mr. Gunn

Just noticed, there are two datapoints missing. Those are the totals Surveymonkey gave me, but they add up to 26 and it claims n = 28. - Bill Hooker

Bill, I'm also another one of of the $100 datapoints (although maybe I'm not since 2 datapoints are missing). - Graham Steel

Just wanted to clarify my $10 vote was based on friendfeed exactly as it is now, with search that goes back further into the archives, and with assured stability. I'd pay more for something more tailored to research and with more professional analytics features. - Mr. Gunn

The missing two datapoints are both $200 -- either I simply missed them the first time I looked, or they have just caught up between the counter and the analysis page. - Bill Hooker

So, $1030/28 approx = $37 per person if we let everyone pay what they think is fair. TLS has about 1300 users, so 1300 x 37 is around $48K per year. This is, of course, nonsense for many reasons, not least of which is n=28, but still. I'm getting the impression that we couldn't raise enough to roll our own, but maybe we could get buyin to an existing service? I'm not sure how that would work though... would we have to form some kind of legal entity in order to invest? The legal etc fees would eat up anything we could raise! - Bill Hooker

This is just a hunch, but I figure that a much larger percentage of the non-respondents would be in the $0 camp due to increased likelihood of apathy about which services they use. Also, don't forget the non-TLS scientists, e.g. 878 subscribers to Science2.0 who don't completely overlap. - Matt Leifer

I agree about the apathy -- if they can't be arsed to respond they're not likely to be willing to pay! It's probably just silly to do calculations like mine up there without at least gathering a much larger sample. I can't bump the survey because that seems to have stopped working, so I'll post updates with links back to the OP every so often this week, and see just how many respondents we can get. - Bill Hooker

I would not pay - I think any charge would kill the participation to a level below critical mass for proper networking. I think we could find free hosted alternatives if necessary - Jean-Claude Bradley

agree - Claudia Koltzenburg

This is the problem in a nutshell. Unfortunately, it takes money to support the service. ($100 datapoint) - Walter Jessen

I was one of the lazy buggers who didn't get around to answering - I'd be prepared to pay for some form of service but my suspicion is that a more imaginative business model is required. Some form of freemium or pay for privacy system might work. Other suggestions include paying for archive access or paying to make your archive available...but I'm not sold on any of these I have to admit. I worry that the free hosted services are going to shrink radically over the next few years - leaving just a few dominated by the big players - Cameron Neylon

@Cameron I didn't think of this before, but your comment made me think of the Evernote business model. I think that could work quite well. The basic service would be free; additional features such as analytics and archiving would require a monthly or yearly fee. Article here: Using ‘Free’ to Turn a Profit: http://www.nytimes.com/2009... - Walter Jessen

I think if we want (master/PhD) students and post-docs involved (which I think we do), we should also consider institutional fees... - Egon Willighagen

@Egon, or "lab" fees (such as 10 licenses for substantial discount). I sort of lean towards agreeing with everyone (ha ha). It's worth more than $100 to me right now. But as Jean-Claude says, FF wouldn't be anywhere near as good as it is now if it had had a fee originally. A big grant (such as the NSF one OpenWetWare obtained) would be appropriate. But even those aren't stable enough. So, probably commercial or private foundation would seem best ownership. - Steve Koch

It's also worth somehow making a very clear presentation to facebook as to how valuable friendfeed is for scientists around the world. It's not impossible that they couldn't keep it running, perhaps in collaboration with scientists. I'd tolerate instrumentation-related and other ads around this site. - Steve Koch

Considering that social networks for scientists are often labeled "Facebook for scientists", I think asking Facebook directly to let this one turn into one that fits this description may have chances of success. And, as Steve points out, manufacturers of scientific instrumentation might well be willing to advertise in this space, thus providing a way to cover the operational costs. - Daniel Mietchen

I think Steve and Daniel are onto something there. How about a letter to Facebook making exactly that case? -- i.e. you have got something here that scientists are using and don't want to lose, that could grow into a real "Fb for scientists", and that we'd be willing to let you stick some advertising on; if you make a commitment to it we'll help spread the word. - Bill Hooker

So another concept that has been kicking around is the idea of a non-profit foundation that is supported by and integrated at some level with a range of commercial suppliers who build on an open source(ish) base to provide paid for services but where users could choose to run up their own installation(s) or if that's not appropriate use a hosted service. So Wordpress.com/org Apache or Mozilla.com/org are examples of this kind of thing. Not sure whether such a central organization would deal with software or interchange standards here but it might be a way of both getting commercial partners in place but not having the landscape dominated by one of them? - Cameron Neylon

I'm liking the sound of Cameron's idea much more than the idea of making a plea to Facebook. I worry about Facebook being so in bed with marketers and not respecting privacy. - Mr. Gunn

I hear you on the privacy concerns, MrG, and I also like Cameron's idea a lot. A whole lot. That non-profit could do other things besides run our FriendFeed replacement... - Bill Hooker

I worry about Facebook being curators of a site that is widely used by the science community. In the past, they have shown themselves to be purely driven by the profit motive, with concessions to users only being made under duress. If we were talking about Google then that would be one thing, but I don't think we'll hear a peep out of Zukerberg if we write to Facebook. - Matt Leifer

I worry about the Facebook/google approach in two ways. One is that in such a big pond you'd only ever get a small voice. The other is the risk of disappearing under the need to make money as Matt mentions. Something that can leverage the high quality work done and vast resources committed to consumer services but that balances that against the need for researchers to have a strong (but not necessarily the only!) voice would be the aim or whatever we can put together. - Cameron Neylon

I agree with your comments, Matt and Cameron, but still think it may be good to contact them and point out that there is a sizable community of researchers who would like to run (and improve) their own installation of Friendfeed and whether they could (a) make it open source entirely (don't think they will, but asking is probably OK) or (b) give the code to a scientific institution - e.g. http://www.tubit.tu-berlin.de/ (partner in Björn's DFG proposal) or http://www.stfc.ac.uk/ (Cameron may have heard about them) - for non-commercial use, perhaps with an option to release the code entirely after an embargo period. There may well be a (c) here too. - Daniel Mietchen

Agreed. Worst case is there's no response. - Walter Jessen

My experience with getting site sponsors is that biotech companies do not like to sponsor sites unless they have significant reach. I'm only now starting to get people interested in advertising on LabSpaces. I don't know what kind of support you're looking for from instrument companies, but my pleas for $100 a month to host on a dedicated server and support a minor ad budget were turned down because I was only serving 30,000 pages. Now I'm up to 150k/month and I have interest but not enough to support a dedicated server or advertising. I'd say go the private investor route if you're serious about this. Or buy into sciencefeed since the groundwork's already put down there :P - Brian Krueger - LabSpaces

Brian, that's very useful feedback, thanks. - Bill Hooker

I agree site sponsors or advertising isn't either going to pay enough nor be reliable enough to give people confidence. I would certainly agree if the question to Facebook/Friendfeed is "would you give us the/some source code and if so under what circumstances" - I'd be happy with the code as it was prior to the UI change even - I can think of a few ways it might be possible to get it hosted - few different organizations probably under different conditions. - Cameron Neylon

I also wonder how much money you'd need from the get-go. My site is hosted on a $10 a month virtual host box and although they say its unlimited bandwidth it can only handle 30 simultaneous connections. Bandwidth has only ever been a problem for me when the site gets hammered by Reddit, Fark, et al. So at the outset, you guys could definitely run this off of a small budget, assuming you're not looking at a high bandwidth operation (and I don't think you are, even if all 800 life science room subscribers decide to migrate). Your major costs will be in coding the site, unless you plan on doing this as a side project with a couple of you coding on the weekends. - Brian Krueger - LabSpaces

Its the setup and maintenance costs (I'm kind of assuming this is a full time job for one person at minimum) that are the big issues. - Cameron Neylon

@Cameron, what kind of maintenance are you expecting? Once the site reaches critical mass, then maybe it will require constant attention, but if it stays on the sub 1000 users level I don't think it would be a huge effort. I maintain my site all by myself. It takes an hour every morning before work and an hour every night to set-up the news queue and clean up the spam. Most of the maintenance issues can be delegated to the group, or even in this case, the community. This is very doable on a small budget, and if it does reach critical mass, then it's a non-issue because by that point sponsors will be beating down your door. I'm more than willing to help this effort with coding time, but I won't jump in to something unless there's a lot of support here and a definite user base. I don't need two LabSpaces on my hands :P - Brian Krueger - LabSpaces

Would US$50k do for a start? http://ff.im/gXoNX - Daniel Mietchen

That's a fantastic offer, Brian. I guess we should check into that. - Mr. Gunn

I thought I would just mention that AOL has a friendfeed like service at http://lifestream.aol.com I don't think it has been mentioned yet in our "what to do if friendfeed implodes discussions". It's pretty similar to Google Buzz in that it allows you to import from a limited number of services and doesn't support anything like groups/rooms yet. Still, playing with it for a while I found it to be a better user experience than Buzz, so it is another one worth watching. - Matt Leifer

It's hard to take anything.aol.com seriously - Mr. Gunn

I know, but post Time-Warner, AOL actually has one or two good sites, e.g. http://spinner.com. Most of their successes are not branded AOL for obvious reasons. - Matt Leifer

Bumping this as a reminder that two weeks are left to put in a proposal for US$50k on an Open Friendfeed, see http://ff.im/gXoNX . - Daniel Mietchen

@Brian -- how about that $50K? Would that buy six months of your time, plus trimmings, to set up and maintain an Open Friendfeed? If so, I think the proposal could be put together on that basis: we have someone willing to be Benevolent Dictator for the coding and do the maintenance, we have X users who want this NOW, we think it's worth doing to get FF out of the hands of the for-profits and into the hands of the user community... - Bill Hooker

That would be more than enough money to run the site for a couple of years. I can probably use a lot of the labspaces backend for authentication, so it shouldn't take more than a week or two of coding to get off of the ground. Let me know if you need help writing the proposal! - Brian Krueger - LabSpaces

Actually, it looks as though they require institutional backing. Also, I registered for the site but cannot find any mechanism for submitting an application -- the "apply" links all take you to the privacy statement. - Bill Hooker

Bill, I had reported that problem a while ago and it appears to have been fixed. I am stuck at a different step - getting a Grant ID - because the form would not accept any phone number I entered. The likelihood of small bugs like these seems to be one of the reasons why they start out small, and I consider it a Beta anyway, as long as the review process is not open. - Daniel Mietchen

Update on the fineprint: It's basically like the usual NIH grant application. Specifically, affiliation with a US institution is required (any volunteers?), and so is agreement with a "no indirect cost" stipulation once an award is being made. More via https://apply.fundscience.org/user_ap... . - Daniel Mietchen

That same FAQ page, however, also states a severe thematic restriction: "we encourage doctoral students to apply for FundScience support of projects pursuing hypotheses related to the pathogenesis or modeling of diseases including Crohns and Familial Mediterranean Fever, and diseases predominantly affecting the developing world." Bye-bye, FF clone? - Daniel Mietchen

Daniel - yes I ran into the same problem - Jean-Claude Bradley

We could probably work around the institutional affiliation part, but given the disease focus my feeling is that they would not be interested in funding a FF clone. - Bill Hooker

OK, so we will have to say good bye on this one. Another option for a grant is at http://ff.im/hGFHZ - Europe and Global Challenges - and I think we could give the transition to open science a try. Funds up to Euro 1M are possible, so the FF clone may fit in as part of a broader concept. - Daniel Mietchen