# Fileset

[cmb_apss.pdf](https://mdr.nims.go.jp/filesets/77873cde-accc-4e02-b66e-d2594ce99563/download)

## Creator

[TODOROKI, Shin-ichi](https://orcid.org/0000-0003-3986-1900), KONISHI, T, INOUE, Satoru

## Rights



## Other metadata

[Blog-based research notebook: personal informatics workbench for high-throughput experimentation](https://mdr.nims.go.jp/datasets/73c31445-7434-4734-892c-43692e2d779f)

## Fulltext

Blog-based research notebook: personal informaticsworkbench for high-throughput experimentationShin-ichi Todoroki∗, Tomoya Konishi and Satoru InoueAdvanced Materials Laboratory, National Institute for Materials Science,Namiki 1-1, Tsukuba, Ibaraki 305-0044, JAPANReceived 8 December 2004; accepted 24 March 2005AbstractIn this age of information technology, many researchers are still conservative in keepinga log of their activities in paper-based notebooks. This style of log-keeping brings aboutthe situation that our experimental data and their descriptions are recorded separately intohard disks and papers, respectively. Such a data separation is likely to be a serious rate-limiting factor in high-throughput experimentation from the view point of getting feedbackon each researcher’s work from what he has done. We propose to utilize a blog (Weblog)as an electronic research notebook and discuss technical requirements for maintaining it,on the basis of the blogging experience for four years by one of the authors. We needa user-installed blog server with authentication function for personalization and networkinfrastructure enabling us to “blog anytime, anywhere”. Although some knowledge-sharingsystems have similar electronic notebooks as their front-end, the present blog system isdifferent from these because it stores personal information which is not meant to be sharedwith others. This blog-based notebook cooperates with these e-notebooks by promotinghyperlinks among their contents, and acts as a personal informatics workbench providingconnections to all the resources needed.Key words: blog, informatics, electronic notebook, high-throughput experimentationPACS:07.05.Kf, 07.05.WriPublished onAppl. Surface Science. Available online 28 September 2005.http://dx.doi.org/10.1016/j.apsusc.2005.03.235http://www.geocities.jp/tokyo1406/node2.html#Todoroki05ASS∗ Corresponding author.Email address:TODOROKI.Shin-ichi at nims.go.jp Tel:+81-29-860-4348 Fax: +81-29-854-9060 (Shin-ichi Todoroki).URL: http://www.geocities.jp/tokyo 1406/ (Shin-ichi Todoroki).Preprint submitted to Elsevier Science 28 September 2008http://dx.doi.org/10.1016/j.apsusc.2005.03.235�http://www.geocities.jp/tokyo_1406/node2.html#Todoroki05ASS�http://www.geocities.jp/tokyo_1406/�LOG BOOKLOG BOOKNetwork(a) (b)Fig. 1. An illustration comparing two styles of logging researcher’s activity, old (a) andnew(b).1 IntroductionIn this age of information technology, many researchers are still conservative inkeeping a log of their activities in paper-based notebooks. This style of log-keepinginvolves one problem, which we have overlooked for a long time since the widespreadproliferation of personal computers (PC); our experimental data are separately recordedinto our hard disks apart from our logbook which stores the rest of information suchas experimental strategies, data evaluation, supplemental notes etc. (see Fig. 1).This separated data storage is likely to be a serious rate-limiting factor in high-throughput experimentation because of its bad traceability of the context of pastexperiments; we have to consult both our PC and logbook at the same time andto reorganize the information by the aid of some keywords of dates, filenames andother knowledge in our patchy memories. Such a task prevents us to get a quickfeedback from past experiments, especially for the case of treating large amount ofdata.The core of this problem is that the information stored in paper-based media needsmanual operations for data retrieval and referencing to other information, in con-trast to the ones stored in electronic media. The solution to this problem is verysimple. All information should be stored electronically in one place and discardpaper-based logbooks. It is, however, easier said than done.Here we propose to utilize a “blog” as an electronic research notebook since a blogsystem provides user-friendly interface compatible with web browsers, easy-to-useauthoring tools and full-text retrieval. A blog-based research notebook is differentfrom original blogs in one important way; its readers are restricted to its author (andhis/her colleagues, if necessary). We discuss technical requirements for maintain-ing it, on the basis of blogging experiences for four years by one of the authors. Inthis paper, after summarizing the blog technology briefly (section 2), we discuss therequirements to use it as our research notebook (section 3) and give tips to enhancethe traceability between our experimental data and blog-based research notebook(section 4). Finally, we compare the present notebook with existing electronic note-books[1,2] (Section 5). Such a discussion is useful not only for researchers but also2for the system vendors willing to carry on their business in informatics tools.2 Popularization of blogThe word, blog, comes from “web log”, a publicly accessible personal journal foran individual. This is similar to a personal diary, but is shared with other people overthe web. Blogs appeared in late ’90 and became popular since some free servicesfor blogging appeared, such ashttp://www.blogger.com/andhttp://blog.goo.ne.jp/.Basic functions of blogs are the followings; (1) displaying the contents in orderof time, (2) displaying specific contents, responding to user’s requests of dates,topics and keywords, (3) registering newly edited contents including graphic arts,multimedia and links to other resources such as past blog contents, web pages,local data files and remote databases, and (4) all the functions are called via webbrowsers without any preliminary knowledge.3 Requirements as our research notebooksSince a blog is basically a communication tool, its contents are to be open to thepublic, but it is not appropriate for our research notebook. Thus, user authentica-tion function is to be added. Secondly, the contents should be browsed and updatedanytime and anywhere during our research activities. Lastly, all the resources to beincluded in the notebook should be accessed via network wherever you are brows-ing it. In this section, technical requirements to realize the above features under theexisting blog systems are discussed.3.1 User authenticationSince this feature is generally not available in the existing free blogging services,we have to set up at least one computer acting as a web server including a bloggingservice. The software packages needed are a HTTP (Hypertext Transfer Protocol)server and CGI (Common Gateway Interface) tools for blog service. The authenti-cation function is provided by the HTTP server. There are variety of user-installedblog software packages (for example, see [3]) and we can find some which meet ourspecific requirements, such as cost, operating system, skills etc. The authors con-structed a Linux-based system including Apache HTTP server[4] and tDiary[5].3http://www.blogger.com/�http://blog.goo.ne.jp/�OfficeInternetLANHomeFirewallLab(2) Terminal A (4) Mirror server A(1) Blog server(3) Terminal B (5) Mirror server BFig. 2. An illustration showing accessibility of a blog server from its clients.3.2 Accessibility to the blog serverIn order to realize “blogging anytime, anywhere”, networked terminals need to beavailable in the office, laboratories, and at home. If the server has to be accessedover a firewall blocking HTTP requests, it is necessary to place a mirror serveroutside the firewall (see Fig. 2(5)). The next best solution is to construct such amirror server in a notebook-size computer and carry it along when venturing out-side a firewall(see Fig. 2(4)). Wireless access via PDA (Personal Digital Assistant)or cellular phone can be preferred by people who do not care for carrying them allthe time and their restricted usability due to their compactness.3.3 Accessibility of the contentsSince the contents of the research notebook are sent over the HTTP protocol, allthe data to be included, such as text, photographs, figures and links to other re-sources, have to be stored in HTTP servers including the present blog server andother servers speaking the HTTP protocol. The latter involves database servers for aclosed user group, managed by research projects in order to share data among theirmembers (for example, see [1,2]). All the contents mentioned above are embeddedin the main text of the research notebook by HTML (Hypertext Markup Language).Figure 3 shows an example of blog output.4 Enhancement of traceability between a blog and data filesOnce this blog-based research notebook is set up and kept for some time, we canretrace the past experimental data from this blog notebook as long as we have em-bedded them in it (see see dashed arrows in Fig. 4 pointing to (2) from (1)). How-ever, it is more frequent for us to refer in the opposite direction, that is, we refer tothe page of the notebook corresponding to the date on which the data we are seeingwere collected (see dashed arrows in Fig. 4 pointing to (3) from (2)). It is therefore4Fig. 3. A screenshot of a blog-based research notebook managed by one of the authors.necessary to have an automatic means of producing such references instead of typ-ing the date in the blog every time it is needed. This function has to be provided bythe software for data-visualization and also the database servers mentioned above.We developed a simple program to visualize our data and to open the correspondingpage of the logbook as shown in Fig. 5. The user can choose one datafile in thelower half of the window. This software then shows the date of data storage in theupper half of the window. The selected data is visualized by clicking “Visualize by”button located near the center of the window. The corresponding page of the blognotebook is shown by clicking “Display the Diary” button in the top of the window,just like in Fig. 3.By the aid of such a utility, we can access all the information through our PCwithout any stress for managing separated data storage as mentioned in Section 1and the right side of Fig. 1.5• 2004/12/8 • 2004/12/9 • 2004/12/10Automatichyperlinkingto Blog pages• 2004/12/10Datafile1Datafile2Datafile3Manualhyperlinkingto datafiles...T...T ...T(1) A blog page(2) Visualizing software(3) blog pagesFig. 4. An illustration showing hyperlinks between blog pages and data files. Roundedrectangles are pages of the blog-based research notebook or visualizing software. Dashedarrows are hyperlinks between these pages.Fig. 5. A screenshot of a utility program for visualizing datafiles and showing correspond-ing pages of the blog-based research notebook. It is written in an object-oriented scriptinglanguage, Ruby[7], with Ruby-GNOME2 library[8].5 Comparison with former electronic notebooksUntil now, several “electronic notebooks” have been proposed as a front-end ofknowledge-sharing systems[1,2]. There is no big difference between these note-books and the present blog-based notebook from a viewpoint of functions of pre-senting the stored data. They are different, however, in practical use. The contentsof the former are to be shared among the users of the knowledge-sharing systems.6The users upload to the server only what other users are interested in. On the otherhand, the present blog stores every information the blog owner will refer to in thefuture, that is, what we used to store in a paper-based research notebook.A practical advantage in the blog-based notebook is that each individual can startblogging in their own fashion even if existing electronic notebook service is notavailable. The people who are already using an electronic notebook provided byone of the knowledge-sharing systems can find benefit in starting their own blog-based notebook even if they worry about another separated data storage. As longas each of their data servers allows the contents to be linked to each other, theuser does not feel the border between different systems. These notebooks shouldbe properly used depending on the purpose; for data sharing or for private use. Inaddition, a sophisticated technique to make a crosslink between two blog servers isvery popular in the blog community, called “trackback”[6]. It is possible to makea link to specific page in any page of a remote blog server accepting trackbackthrough CGI. This function is essential for existing knowledge-sharing systems inorder to cooperate with their users’ blog systems.6 ConclusionBlog-based research notebook enables us to manage all the personal informationelectronically, which was formerly recorded in our paper-based notebooks. Thisservice requires user authentication function in a user-installed server and infras-tructure for “blogging anytime, anywhere”. Once these are satisfied, the blog acts asa personal informatics workbench; a gateway to all the information needed whichis traceable and retrievable. Although some existing knowledge-sharing systemsalso have electronic notebook service as their front-end, the present and existingnotebooks should be properly used depending on the purpose; managing personalinformation or promoting knowledge sharing.References[1] S. J. Ludtke, L. Nason, H. Tu, L. Peng, W. Chiu, Object oriented database and electronicnotebook for transmission electron microscopy, Microscopy and Microanalysis 9(2003) 556–565.[2] S. Meguro, T. Ohnishi, M. Lippmaa, H. Koinuma, Elements of informatics forcombinatorial solid-state materials science, Meas. Sci. Technol. 16 (1) (2005) 309–316.[3] O. Winkler, Blog software breakdown,(http://www.asymptomatic.net/blogbreakdown.htm) (Apr. 2004).[4] The apache software foundation, (http://www.apache.org/).7http://www.asymptomatic.net/blogbreakdown.htm�http://www.apache.org/�[5] tDiary, (http://sourceforge.net/projects/tdiary/).[6] Mena and Ben Trott, A beginner’s guideto trackback, (http://www.movabletype.org/trackback/beginners/, Japanese translation:http://kotonoha.main.jp/weblog/000138_trackbackguide.html) (Mar. 2003).[7] Ruby Home Page, (http://www.ruby-lang.org).[8] Ruby-GNOME 2, (http://sourceforge.net/projects/ruby-gnome2/).8http://sourceforge.net/projects/tdiary/�http://www.movabletype.org/trackback/beginners/�http://www.ruby-lang.org�http://sourceforge.net/projects/ruby-gnome2/�