I promised some of you I would do this a while ago and I simply haven’t got to it. But enough of the excuses. There has been a huge number of launches in the past few months of sites and services that are intended to act as social network sites for scientists. These join a number of older services including Nature Network, OpenWetWare, and others. My concern is that with so many sites in the same space there is a risk none of them will succeed because the user community will be too diluted. I am currently averaging around three emails a week, all from different sites, suggesting I should persuade more people to sign up.
What I would like to do is attempt a critical and comprehensive analysis of the sites and services available as part of an exercise in thinking about how we might rationally consolidate this area, and how we might enable the work that has gone into building these services be used effectively to build the ‘next generation’ of sites. All of these sites have good features and it would be a shame to see them lost. I also don’t want to see people discouraged from building new and useful tools. I just want to see this work.
My dream would be to see an open source framework with an open data model that allows people to move their data from one place to another depending on what features they want. Then the personal networks can spread through the communities of all of these sites rather than being restricted to one, and the community can help build features that they want. As someone else said ‘Damnit, we’re scientists, we hold the stuff of the universe in our hands’ – can’t we have a think about what the best way to do this is?
What I want to do with this post is try to put together a comprehensive list of sites and services, including ones that get heavy scientific use but are not necessarily designed for scientists. I will miss many so please comment to point this out and I will add them. Then I want to try and put together a list of criteria as to how we might compare and contrast. Again please leave comments feel free to argue. I don’t expect this to necessarily be an easy or straightforward process, and I don’t expect to get complete agreement. But I am worried if things are just left to run that none of these sites will get the amount of support that is needed to make them viable.
So here goes.
Sites
Blog collections: Nature Network, ScienceBlogs, Scientific Blogging, WordPress, Blogspot, (OpenWetWare),
Social Networks: Laboratree, Ologeez, Research Gate, Epernicus, LabMeeting, Graduate Junction, (Nature Network), oh and Facebook, and Linkedin
Protocol sharing: Scivee, Bioscreencast, OpenWetWare, YouTube, lots of older ones I can’t remember at the moment
Others: Friendfeed, Twitter, GoogleDocs, GoogleGroups, Upcoming, Seesmic,
Critical criteria
Stability: funding, infrastructure, uptime, scalability, slashdot resistance, long term personnel committment
Architecture: open data model? ability to export data? compatibility with other sites? plugins? rss?
Design: user interface, ‘look’, responsiveness
Features: what features do you think are important? I don’t even want to start putting my own predjudices here.
How to take this forward?
Comment here or at Friendfeed, or anywhere else, but if you can please tag the page with Fb4Sci. I have put up a GoogleDoc which is visible at http://docs.google.com/Doc?id=dhs5x5kr_572hccgvcct (current just contains this post). If you want access drop me an email at cam eron ney lon (no spaces) at googlemail (not gmail) and I will give anyone who requests editing rights. Comment and contributions from the development teams is welcome but I expect everyone to make a conflict of interest declaration. Mine is:
I blog at OpenWetWare and use the wiki extensively. I have been known to ring into steering committee meetings and have discussed specific features with the development team. I am an irregular user of Nature Network and a regular user of Friendfeed and Twitter. I have a strong bias towards open data models and architectures.
[ducks]
SciLink would probably be under social networks as well, and is one of the more established ones (though this can be a bit of a turnoff as it sometimes seems too commercial). Would JoVE fit under protocol sharing? It could also be argued that Labmeeting is not primarily a social networking tool. I had a conversation with one of the core developers last week and he stressed that it is first and foremost meant to be a tool that is useful even if you’re the only one using it (PDF organizer); the network effects will come later and naturally. I plan to blog more extensively about this idea since I hadn’t thought about things in that way before, and it really makes a lot more sense, especially when your target users are scientists. It will be much harder to build a network if you bill it as such, and perhaps much easier to build a _tool_ that doesn’t depend on a network (but has the potential to become much more once the network is already there).
SciLink would probably be under social networks as well, and is one of the more established ones (though this can be a bit of a turnoff as it sometimes seems too commercial). Would JoVE fit under protocol sharing? It could also be argued that Labmeeting is not primarily a social networking tool. I had a conversation with one of the core developers last week and he stressed that it is first and foremost meant to be a tool that is useful even if you’re the only one using it (PDF organizer); the network effects will come later and naturally. I plan to blog more extensively about this idea since I hadn’t thought about things in that way before, and it really makes a lot more sense, especially when your target users are scientists. It will be much harder to build a network if you bill it as such, and perhaps much easier to build a _tool_ that doesn’t depend on a network (but has the potential to become much more once the network is already there).
I’d like edit access: cwhookerATgmailDOTcom. No promises but I’ll try to help out, I think this is a great idea. One other possibility: use the biogang wiki rather than Google Docs?
I’d like edit access: cwhookerATgmailDOTcom. No promises but I’ll try to help out, I think this is a great idea. One other possibility: use the biogang wiki rather than Google Docs?
I guess you could have another category for bookmark/reference sharing with Connotea, CiteULike and Delicious. As for the social networks, Scholarz.net should definitely be included!
I guess you could have another category for bookmark/reference sharing with Connotea, CiteULike and Delicious. As for the social networks, Scholarz.net should definitely be included!
Protocol sharing and you miss myExperiment (http://www.myexperiment.org/) social networking, workflow sharing, remote enactment of scientific workflows..
Protocol sharing and you miss myExperiment (http://www.myexperiment.org/) social networking, workflow sharing, remote enactment of scientific workflows..
How about stuff like researcherid and biomedexperts?
How about stuff like researcherid and biomedexperts?
Way to go, Cameron! I’d suggest adding somewhere, the Science Advisory Board (though under what category)? This somewhat social, somewhat commercial site – http://www.scienceboard.net/ – led to my starting a blog, “meeting” Richard Grant, ending up on NN, from there to FF and Facebook… (the LinkedIn and OWW stuff are other lineages, but it’s sometimes funny to follow one’s temporal links).
Also, Protocol Online (http://www.protocol-online.org/), on which I left a complimentary comment in (ouch) 2000…
I’ve used a couple other protocol fora and can’t remember either. While searching, came across this site: http://www.molecularstation.com/
Way to go, Cameron! I’d suggest adding somewhere, the Science Advisory Board (though under what category)? This somewhat social, somewhat commercial site – http://www.scienceboard.net/ – led to my starting a blog, “meeting” Richard Grant, ending up on NN, from there to FF and Facebook… (the LinkedIn and OWW stuff are other lineages, but it’s sometimes funny to follow one’s temporal links).
Also, Protocol Online (http://www.protocol-online.org/), on which I left a complimentary comment in (ouch) 2000…
I’ve used a couple other protocol fora and can’t remember either. While searching, came across this site: http://www.molecularstation.com/
Yes, I realised on Friday night that I had missed myexperiment which is a bit embarrasing given I was talking to people about it on Wednesday :) And yes, citeulike, connotea, delicious, should be there as well obviously.
@Bill I thought googledocs might be better as it is ‘neutral’ ground. Writing a criticism on one of the sites being looked at seemed inappropriate. Although a wikispaces site would be fine I guess. Might be easier.
Biomedexperts is something to include, not sure sure about researcherid. I’ll need to look closer at the services they have.
From Martin Fenner on Friendfeed, one question should obviously be whether there is an API.
Yes, I realised on Friday night that I had missed myexperiment which is a bit embarrasing given I was talking to people about it on Wednesday :) And yes, citeulike, connotea, delicious, should be there as well obviously.
@Bill I thought googledocs might be better as it is ‘neutral’ ground. Writing a criticism on one of the sites being looked at seemed inappropriate. Although a wikispaces site would be fine I guess. Might be easier.
Biomedexperts is something to include, not sure sure about researcherid. I’ll need to look closer at the services they have.
From Martin Fenner on Friendfeed, one question should obviously be whether there is an API.
If you’re making a list of web2.0 services for researchers, you could add the ones we run –
http://publicationslist.org (site for managing your own publications list online / embedding it in your homepage) – and –
http://a.nnotate.com (annotate / discuss PDFs online).
They both have free versions or you can pay for extra online storage space. Neither is a facebook clone though…
If you’re making a list of web2.0 services for researchers, you could add the ones we run –
http://publicationslist.org (site for managing your own publications list online / embedding it in your homepage) – and –
http://a.nnotate.com (annotate / discuss PDFs online).
They both have free versions or you can pay for extra online storage space. Neither is a facebook clone though…
Fred, thanks for that – will include them on the list. Its not necessarily purely facebook clones that I’m interested in (in fact that would be rather dull really) but anything that fits in that kind of space of sharing materials within some sort of network.
Fred, thanks for that – will include them on the list. Its not necessarily purely facebook clones that I’m interested in (in fact that would be rather dull really) but anything that fits in that kind of space of sharing materials within some sort of network.
I think this effort deserves a wiki of its own. I recommend wetpaint or wikispaces. (Both accept OpenID, btw.)
I think this effort deserves a wiki of its own. I recommend wetpaint or wikispaces. (Both accept OpenID, btw.)
I am one of the builders of the Laboratree (http://laboratree.org/) and faculty researcher in biomedical informatics. The data problem you allude to is non-trivial and pervasive throughout biomedicine. To ameliorate the other issues you mention, our approach with Laboratree is to support open standards, such as OpenSocial and OpenID. (OpenID is not quite live yet, but will be shortly). The apps that we build are developed using OpenSocial and can be used in any other compliant opensocial container (Orkut, Google, etc). If you log into Laboratree you can see them under ‘Browse Apps’. We will likely open source our app code, to aid in development of new ones. For data standards, we will likely encourage the use of NCBO’s ontologies for standardization of stored data or defer to the big biomed informatics cyberinfrastructure projects such as caBIG or whatever comes out of the CTSA informatics projects. Again, this is a nontrivial problem that isn’t coming — it already exists (some people use R and some people use MatLab, for example).
I am one of the builders of the Laboratree (http://laboratree.org/) and faculty researcher in biomedical informatics. The data problem you allude to is non-trivial and pervasive throughout biomedicine. To ameliorate the other issues you mention, our approach with Laboratree is to support open standards, such as OpenSocial and OpenID. (OpenID is not quite live yet, but will be shortly). The apps that we build are developed using OpenSocial and can be used in any other compliant opensocial container (Orkut, Google, etc). If you log into Laboratree you can see them under ‘Browse Apps’. We will likely open source our app code, to aid in development of new ones. For data standards, we will likely encourage the use of NCBO’s ontologies for standardization of stored data or defer to the big biomed informatics cyberinfrastructure projects such as caBIG or whatever comes out of the CTSA informatics projects. Again, this is a nontrivial problem that isn’t coming — it already exists (some people use R and some people use MatLab, for example).
Sean, thanks for your comments, and I hope I get to meet with you later today! All of this sounds good and is the type of thing I would like to support. I agree that there are very big non-trivial technical issues involved here, but I think they are worth tackling. My answer is partly we should worry less about data standards and more for standards for describing data and analysis standards. The problem with research is that there will always be non-standard areas, indeed that is precisely where the interest is likely to be so building in coping mechanisms for that will be key.
e.g. MatLab versus R, both can be wrapped as webservices and the underlying code opensourced with full version control. So just put it up somewhere in a wrapper and then point at the underlyng code in the appropriate repository.
Sean, thanks for your comments, and I hope I get to meet with you later today! All of this sounds good and is the type of thing I would like to support. I agree that there are very big non-trivial technical issues involved here, but I think they are worth tackling. My answer is partly we should worry less about data standards and more for standards for describing data and analysis standards. The problem with research is that there will always be non-standard areas, indeed that is precisely where the interest is likely to be so building in coping mechanisms for that will be key.
e.g. MatLab versus R, both can be wrapped as webservices and the underlying code opensourced with full version control. So just put it up somewhere in a wrapper and then point at the underlyng code in the appropriate repository.
well I guess part of the problem is to get scientists to use the web 2.0 sites. Most of them are so drowned in grant writing and fighting for a place under the tenure sun, which makes them not have time for those. Seems like the solution to that will be convincing scientists of the added value of using those sites – despite the initial learning curve, which is not so steep as I discovered myself. A good idea about a “sign-up fest”:
http://blog.pansapiens.com/2008/08/01/a-web20-for-scientists-week-signup-fest/
well I guess part of the problem is to get scientists to use the web 2.0 sites. Most of them are so drowned in grant writing and fighting for a place under the tenure sun, which makes them not have time for those. Seems like the solution to that will be convincing scientists of the added value of using those sites – despite the initial learning curve, which is not so steep as I discovered myself. A good idea about a “sign-up fest”:
http://blog.pansapiens.com/2008/08/01/a-web20-for-scientists-week-signup-fest/
I’m from Nature Network. Cameron, very glad that you’re doing this. We’re looking forward to the results of your analysis.
About the categories you’ve put us into…yes, our blogs and forums are active, but we are at the core a social networking site (ie user profiles, users can create personal networks, etc). We see lots of connections being made between individuals on the site, so I hope that in your analysis, we get evaluated as a SN site.
Thanks again for doing this.
I’m from Nature Network. Cameron, very glad that you’re doing this. We’re looking forward to the results of your analysis.
About the categories you’ve put us into…yes, our blogs and forums are active, but we are at the core a social networking site (ie user profiles, users can create personal networks, etc). We see lots of connections being made between individuals on the site, so I hope that in your analysis, we get evaluated as a SN site.
Thanks again for doing this.
Now, that some of the web browsers (Ensembl;NCBI) have ways to create user profiles and personalize settings, they should really incorporate some variety of networking for scientists.
I think Ensembl has a blog that is put out by the site itself, but it lacks interaction from users.
Now, that some of the web browsers (Ensembl;NCBI) have ways to create user profiles and personalize settings, they should really incorporate some variety of networking for scientists.
I think Ensembl has a blog that is put out by the site itself, but it lacks interaction from users.
“My concern is that with so many sites in the same space there is a risk none of them will succeed because the user community will be too diluted.”
Isn’t the whole idea behind capitalism and evolution that from a struggle only the few great will survive. I don’t think you have to worry about the community being diluted and therefore having no purpose, eventually the best will succeed while the others will give up. That’s just how business works. Use Facebook and Myspace as an example.
I’ve been running my social network, Labspaces.net, since the spring of 2006. I’ll be the first to tell you that it is VERY hard to get the majority of older scientists interested in sites like this. The scientists I talk to are afraid of the data sharing/collaboration aspects some of these sites are purporting as “features.” They fear that their data and ideas won’t be secure on social networks, and they don’t think these services, from a collaboration standpoint, are a significant improvement over e-mail and telephone communication.
Bringing scientists together with these sites offers a great opportunity to spur collaboration across multiple disciplines. As far as my site is concerned I think it offers the widest variety of content you will find. It hosts free science blogs, protocols, a science forum, science news (Featured press releases from universities and mainstream news outlets), lab profiles, user profiles, and groups. You can view my profile here: http://www.labspaces.net/view_profile.php?ID=15
The time is also ripe to get the young community of scientists involved in sites like this. I don’t think you’ll see a conversion of the over 40 group of scientists to web social networks, but younger scientists who already used to checking their facebook 4 times a day will be more willing to join a secondary niche network.
“My concern is that with so many sites in the same space there is a risk none of them will succeed because the user community will be too diluted.”
Isn’t the whole idea behind capitalism and evolution that from a struggle only the few great will survive. I don’t think you have to worry about the community being diluted and therefore having no purpose, eventually the best will succeed while the others will give up. That’s just how business works. Use Facebook and Myspace as an example.
I’ve been running my social network, Labspaces.net, since the spring of 2006. I’ll be the first to tell you that it is VERY hard to get the majority of older scientists interested in sites like this. The scientists I talk to are afraid of the data sharing/collaboration aspects some of these sites are purporting as “features.” They fear that their data and ideas won’t be secure on social networks, and they don’t think these services, from a collaboration standpoint, are a significant improvement over e-mail and telephone communication.
Bringing scientists together with these sites offers a great opportunity to spur collaboration across multiple disciplines. As far as my site is concerned I think it offers the widest variety of content you will find. It hosts free science blogs, protocols, a science forum, science news (Featured press releases from universities and mainstream news outlets), lab profiles, user profiles, and groups. You can view my profile here: http://www.labspaces.net/view_profile.php?ID=15
The time is also ripe to get the young community of scientists involved in sites like this. I don’t think you’ll see a conversion of the over 40 group of scientists to web social networks, but younger scientists who already used to checking their facebook 4 times a day will be more willing to join a secondary niche network.
Brian, I don’t disagree at all with either of the two statements you are making here. I am just not convinced that the population of ‘possible users’ (i.e. tech savvie scientists with a bit of time on their hands, and in a position to benefit from social networks, under the age of 30) is big enough to spread between the 15 or so current services and actually create a real market. That what will happen is consolidation based on PR not on product quality.
Personally I worry about bringing all of these services onto one site because I think this isn’t the way forward so variety of content is less important IMO than quality. But that’s a question that hopefully we will be able to address in terms of both what the users want and whether this is actually what they need in the longer term. I look forward to the conversation about this.
Brian, I don’t disagree at all with either of the two statements you are making here. I am just not convinced that the population of ‘possible users’ (i.e. tech savvie scientists with a bit of time on their hands, and in a position to benefit from social networks, under the age of 30) is big enough to spread between the 15 or so current services and actually create a real market. That what will happen is consolidation based on PR not on product quality.
Personally I worry about bringing all of these services onto one site because I think this isn’t the way forward so variety of content is less important IMO than quality. But that’s a question that hopefully we will be able to address in terms of both what the users want and whether this is actually what they need in the longer term. I look forward to the conversation about this.
I think the internet has proven time and time again that quality content is what drives users to a website. No amount of PR can make a terrible site viable. For example, Nature Network has a huge amount of PR behind it, but the activity on the site is pretty low (~3 new blog posts a day and 10 comments on those blogs). They’re obviously doing something wrong if they’ve been around for a year, have a huge name and a large PR budget, and still don’t have a fan base.
For another example, take a look at Friendster and Facebook. Frienster, at the time, had a gigantic network, better content, and more street cred than facebook, but friendster was stuck in the 1990s using slow cgi/perl causing massive network problems and everyone migrated over to facebook. People choose content …
Additionally, I think that the population of possible users is pretty large for a niche social network. My estimates come to about 500,000 users. I say give this sector another year or two to develop and a few top players will come out. What’s going to make any of these networks stick out is listening to user feedback and developing the site based on that feedback. That’s how I’m trying to structure my site. I laid a basic groundwork of things that I would like to have in a network and now I’m trying on my tiny graduate student ad budget to get user to the site to help me build it into something they’d like to use everyday. Unfortunately, I don’t have the constraints of proprietary software, free APIs, an angel network, or a fat cat journal CEO weighing down my ability to expand my site in whatever way my community of users decides, I just need to get those users ;)
I think the internet has proven time and time again that quality content is what drives users to a website. No amount of PR can make a terrible site viable. For example, Nature Network has a huge amount of PR behind it, but the activity on the site is pretty low (~3 new blog posts a day and 10 comments on those blogs). They’re obviously doing something wrong if they’ve been around for a year, have a huge name and a large PR budget, and still don’t have a fan base.
For another example, take a look at Friendster and Facebook. Frienster, at the time, had a gigantic network, better content, and more street cred than facebook, but friendster was stuck in the 1990s using slow cgi/perl causing massive network problems and everyone migrated over to facebook. People choose content …
Additionally, I think that the population of possible users is pretty large for a niche social network. My estimates come to about 500,000 users. I say give this sector another year or two to develop and a few top players will come out. What’s going to make any of these networks stick out is listening to user feedback and developing the site based on that feedback. That’s how I’m trying to structure my site. I laid a basic groundwork of things that I would like to have in a network and now I’m trying on my tiny graduate student ad budget to get user to the site to help me build it into something they’d like to use everyday. Unfortunately, I don’t have the constraints of proprietary software, free APIs, an angel network, or a fat cat journal CEO weighing down my ability to expand my site in whatever way my community of users decides, I just need to get those users ;)
it is first and foremost meant to be a tool that is useful even if you’re the only one using it (PDF organizer); the network effects will come later and naturally. I plan to blog more extensively about this idea since I hadn’t thought about things in that way before, and it really makes a lot more sense, especially when your target users are scientists. It will be much harder to build a network if you bill it as such, and perhaps much easier to build a _tool_ that doesn’t depend on a network (but has the potential to become much more once the network is already there).
it is first and foremost meant to be a tool that is useful even if you’re the only one using it (PDF organizer); the network effects will come later and naturally. I plan to blog more extensively about this idea since I hadn’t thought about things in that way before, and it really makes a lot more sense, especially when your target users are scientists. It will be much harder to build a network if you bill it as such, and perhaps much easier to build a _tool_ that doesn’t depend on a network (but has the potential to become much more once the network is already there).
I think Ensembl has a blog that is put out by the site itself, but it lacks interaction from users
I think Ensembl has a blog that is put out by the site itself, but it lacks interaction from users