Comments on: More on online tools for scientists http://www.cshblogs.org/cshprotocols/2008/02/25/more-on-online-tools-for-scientists/ Sat, 17 May 2008 01:10:23 +0000 http://wordpress.org/?v=2.3.3 By: David Crotty http://www.cshblogs.org/cshprotocols/2008/02/25/more-on-online-tools-for-scientists/#comment-14915 David Crotty Tue, 26 Feb 2008 16:23:15 +0000 http://www.cshblogs.org/cshprotocols/2008/02/25/more-on-online-tools-for-scientists/#comment-14915 Antony--thanks for the comments, and particularly for the links to your site. Although chemistry is not my particular field of interest, I'm sure the lessons being learned there will be valuable and applicable elsewhere, and I look forward to digging deeper through your site. A few responses to your comments: ---This is a very interesting statement and I do sense that some of this is going on, yes. The opposite also happens…the small players trying to make a difference get slammed pretty hard--- This is something I don't really understand. I think perhaps the motivation here, the reason why smaller efforts from within the community are less appreciated is that there's a strong desire to see sweeping changes that encompass an entire field. The big corporations seem to be the only ones with the financial backing and the attention reach to accomplish something like this, hence they end up favored. I'm not sure if they're likely to succeed, as it kind of makes more sense to think of these sorts of sites as appealing to niche communities, rather than a "one-size-fits-all" approach. Participation in social networks for the general public continues to decline: http://news.bbc.co.uk/2/hi/business/7257073.stm It's likely that this behavior will become less mainstream, or at least the behavior will fragment into smaller interest niches, rather than everyone being on Facebook/Myspace. The question from a publishing point of view is whether smaller niches like this can financially support such efforts. Which is why they may end up better served from within the community (although this will require grant funding). You also have the corporate control of data, with terms of service that can be altered at any moment, as I mentioned in my posting. ---I run three blogs. It’s a tight community. Many people don’t comment even though they read. Public comments exposes peoples opinions and most are shy to speak in public. The community of vocal participants is VERY SMALL.--- Yep, agreed. That's a point I made in a talk given to a publishing association: http://www.cshblogs.org/cshprotocols/2008/02/14/why-web-20-is-failing-in-biology/ Some of the commenters on that particular post felt that my depiction of the science blogosphere, as a largely circular self-referential community was inaccurate. Antony–thanks for the comments, and particularly for the links to your site. Although chemistry is not my particular field of interest, I’m sure the lessons being learned there will be valuable and applicable elsewhere, and I look forward to digging deeper through your site. A few responses to your comments:

—This is a very interesting statement and I do sense that some of this is going on, yes. The opposite also happens…the small players trying to make a difference get slammed pretty hard—

This is something I don’t really understand. I think perhaps the motivation here, the reason why smaller efforts from within the community are less appreciated is that there’s a strong desire to see sweeping changes that encompass an entire field. The big corporations seem to be the only ones with the financial backing and the attention reach to accomplish something like this, hence they end up favored. I’m not sure if they’re likely to succeed, as it kind of makes more sense to think of these sorts of sites as appealing to niche communities, rather than a “one-size-fits-all” approach. Participation in social networks for the general public continues to decline:
http://news.bbc.co.uk/2/hi/business/7257073.stm
It’s likely that this behavior will become less mainstream, or at least the behavior will fragment into smaller interest niches, rather than everyone being on Facebook/Myspace. The question from a publishing point of view is whether smaller niches like this can financially support such efforts. Which is why they may end up better served from within the community (although this will require grant funding). You also have the corporate control of data, with terms of service that can be altered at any moment, as I mentioned in my posting.

—I run three blogs. It’s a tight community. Many people don’t comment even though they read. Public comments exposes peoples opinions and most are shy to speak in public. The community of vocal participants is VERY SMALL.—

Yep, agreed. That’s a point I made in a talk given to a publishing association:
http://www.cshblogs.org/cshprotocols/2008/02/14/why-web-20-is-failing-in-biology/

Some of the commenters on that particular post felt that my depiction of the science blogosphere, as a largely circular self-referential community was inaccurate.

]]> By: ChemSpiderMan http://www.cshblogs.org/cshprotocols/2008/02/25/more-on-online-tools-for-scientists/#comment-14913 ChemSpiderMan Tue, 26 Feb 2008 16:02:38 +0000 http://www.cshblogs.org/cshprotocols/2008/02/25/more-on-online-tools-for-scientists/#comment-14913 David, Regarding your comments stating "Usually those most enthusiastic about Web 2.0 and such are also those heavily involved in the Open Access, Open Data and Open Science movements. While they decry the policies of the big corporations and publishing conglomerates, they seem to be very willing to jump into bed with them for the creation of new online tools. Why the willingness to trust your data and your time-consuming efforts to for-profit corporations in this arena? Is it just that the big corporations are the only ones with enough cash to fund big experimental sites?" This is a very interesting statement and I do sense that some of this is going on, yes. The opposite also happens...the small players trying to make a difference get slammed pretty hard http://www.chemspider.com/blog/another-response-to-constructive-feedback-from-peter-murray-rust.html The article you refer to is missing a lot of the issues around QUALITY of data in my opinion. Robots fail..and they can often fail in a big way. Manual curation of data is still key and crowdsourcing should be encouraged. There are quality issues everywhere. Some example posts include: http://www.chemspider.com/blog/struggling-to-scrape-crystaleye.html http://www.chemspider.com/blog/how-big-is-the-challenge-of-curation-and-what-is-the-structure-of-ginkgolide-b.html http://www.chemspider.com/blog/will-the-correct-structure-of-taxol-please-stand-up-part-3.html and the efforts going on here: http://www.chemconnector.com/chemunicating/dedicating-christmas-time-to-the-cause-of-curating-wikipedia.html Moving large amounts of data around without care for quality is going to be problematic. I admit to being part of the problem myself. We are aggregating millions of chemical structures and it is not easy to clean it all. However, we are trying. We have offered online curation of data to the users of ChemSpider as detailed here: http://www.chemspider.com/docs/The_Process_of_Curating_Identifiers_on_ChemSpider.pdf . This type of annotation of data I believe is critical to generating data of sufficient quality to be used by others but this is limited to chemical structure datasets. Labs do generate terabytes of data and while disk space is cheap etc I wonder whether the saturation of available data will cause other issues. Regarding “I’ll also point out that in my last posting, I stated that science blogs are mostly read by other science bloggers, and most comments on science blogs are made by science bloggers. Please note who left comments on the post.” I run three blogs. It’s a tight community. Many people don’t comment even though they read. Public comments exposes peoples opinions and most are shy to speak in public. The community of vocal participants is VERY SMALL. David,
Regarding your comments stating “Usually those most enthusiastic about Web 2.0 and such are also those heavily involved in the Open Access, Open Data and Open Science movements. While they decry the policies of the big corporations and publishing conglomerates, they seem to be very willing to jump into bed with them for the creation of new online tools. Why the willingness to trust your data and your time-consuming efforts to for-profit corporations in this arena? Is it just that the big corporations are the only ones with enough cash to fund big experimental sites?” This is a very interesting statement and I do sense that some of this is going on, yes. The opposite also happens…the small players trying to make a difference get slammed pretty hard
http://www.chemspider.com/blog/another-response-to-constructive-feedback-from-peter-murray-rust.html

The article you refer to is missing a lot of the issues around QUALITY of data in my opinion. Robots fail..and they can often fail in a big way. Manual curation of data is still key and crowdsourcing should be encouraged. There are quality issues everywhere. Some example posts include: http://www.chemspider.com/blog/struggling-to-scrape-crystaleye.html
http://www.chemspider.com/blog/how-big-is-the-challenge-of-curation-and-what-is-the-structure-of-ginkgolide-b.html
http://www.chemspider.com/blog/will-the-correct-structure-of-taxol-please-stand-up-part-3.html

and the efforts going on here: http://www.chemconnector.com/chemunicating/dedicating-christmas-time-to-the-cause-of-curating-wikipedia.html

Moving large amounts of data around without care for quality is going to be problematic. I admit to being part of the problem myself. We are aggregating millions of chemical structures and it is not easy to clean it all. However, we are trying. We have offered online curation of data to the users of ChemSpider as detailed here: http://www.chemspider.com/docs/The_Process_of_Curating_Identifiers_on_ChemSpider.pdf . This type of annotation of data I believe is critical to generating data of sufficient quality to be used by others but this is limited to chemical structure datasets. Labs do generate terabytes of data and while disk space is cheap etc I wonder whether the saturation of available data will cause other issues.

Regarding “I’ll also point out that in my last posting, I stated that science blogs are mostly read by other science bloggers, and most comments on science blogs are made by science bloggers. Please note who left comments on the post.” I run three blogs. It’s a tight community. Many people don’t comment even though they read. Public comments exposes peoples opinions and most are shy to speak in public. The community of vocal participants is VERY SMALL.

]]>