Header Leaderboard Ad

Collapse

SEQanswers Experimental Wiki!

Collapse

Announcement

Collapse
No announcement yet.
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • SEQanswers Experimental Wiki!

    THREAD UPDATED 9/17/2009!

    Based on some recent experience, and some feedback from this thread, I've just installed and configured a MediaWiki instance for the community.

    Find the wiki here: http://wiki.seqanswers.com

    Forum credentials *should* work on the wiki, so you won't have to re-register. Also, Semantic MediaWiki (SMW) has been installed so the smarter amongst you can help us organize the data.

    My first goal at this point is to capture and organize the information in the above thread. It's grown to the point where a single forum thread cannot contain it, and sci_guy and I cannot maintain it alone. Not to mention that we would all benefit from a format that could be used to slice the entries to get what we want efficiently (ie...SMW will allow quick queries to get say all the open source programs that support illumina data and miRNA).

    Anyway, stay tuned and please post here if you're a SMW whiz, a MediaWiki whiz, or have suggestions. =]

  • #2
    Nice work!

    Can you please install the 'semantic bundle' extension? There are lots of crucial extensions for working with SMW, such as semantic forms and semantic drilldown.

    You can see a subset of the bundle in action here:

    http://www.bioinformatics.org/wiki
    http://www.bioinformatics.org/wiki/Special:BrowseData
    Last edited by dan; 07-20-2009, 06:49 AM. Reason: Added a more appropriate link.
    Homepage: Dan Bolser
    MetaBase the database of biological databases.

    Comment


    • #3
      Will do this evening Dan! I really like the ngs sandbox you started...the main difficulty is the tagging/properties used...

      Comment


      • #4
        Originally posted by ECO View Post
        the main difficulty is the tagging/properties used...
        Do you mean, this is the thing that is most difficult to model? Or do you mean this is the thing that is the most difficult to get people to do? Or the most difficult thing for people to decide? Or do you mean this is technically the mostly tricky thing with the wiki? Sorry for being a pain!

        I'm guessing you mean some of the above... it's 'the most difficult to decide', and from there, hardest to implement 'properly'...

        Do you know any ontologies related to software? SMW can be set up to work with ontologies, but this is one feature I haven't looked into in detail. i.e. if we could tag software programs with terms from well defined ontologies, that could help a lot with organizing the content of the wiki. Thinking about it, something related to this must be being developed by people interested in 'standard operating procedures' (SOPs) and web service 'resource descriptions', pipeline frameworks, etc. etc. I wonder how technical we can (usefully) get ;-)

        This is kind-of what I mean when I say we don't want to duplicate work... I'm sure people at O|B|F / BioCatalogue could provide input, and there is so much on Wikipedia already... This site has a good focus, which is useful for any project, but it would be nice to be able to actually make use of the semantic web features of SMW with the proper use of ontologies (and it should be a big help with the vague idea of software 'tagging' that we have). Basically, the risk is that we go off and do something different, and miss the opportunity to 'plug-in' to wider efforts.

        However, the scope of what I'm talking about is probably a full time research project! Whatever we do we should keep it as simple as possible.

        Dan.
        Homepage: Dan Bolser
        MetaBase the database of biological databases.

        Comment


        • #5
          Originally posted by dan View Post
          Do you mean, this is the thing that is most difficult to model? Or do you mean this is the thing that is the most difficult to get people to do? Or the most difficult thing for people to decide? Or do you mean this is technically the mostly tricky thing with the wiki? Sorry for being a pain!
          Not a pain at all, I was being unclear as I was running out the door to work. What I was trying to say is basically captured by what you said below...to maximize the usefulness of any data entered here, it is important to carefully choose the metadata that is entered around every software package.

          Do you know any ontologies related to software? SMW can be set up to work with ontologies, but this is one feature I haven't looked into in detail. i.e. if we could tag software programs with terms from well defined ontologies, that could help a lot with organizing the content of the wiki. Thinking about it, something related to this must be being developed by people interested in 'standard operating procedures' (SOPs) and web service 'resource descriptions', pipeline frameworks, etc. etc. I wonder how technical we can (usefully) get ;-)
          I'm not aware of any, but that certainly doesn't mean they aren't out there, I'm just not the right person to ask.

          This is kind-of what I mean when I say we don't want to duplicate work... I'm sure people at O|B|F / BioCatalogue could provide input, and there is so much on Wikipedia already... This site has a good focus, which is useful for any project, but it would be nice to be able to actually make use of the semantic web features of SMW with the proper use of ontologies (and it should be a big help with the vague idea of software 'tagging' that we have). Basically, the risk is that we go off and do something different, and miss the opportunity to 'plug-in' to wider efforts.
          I certainly don't want to duplicate effort, and agree with all your points above. I would love to identify such an ontology and use it to build the database of packages.

          However, the scope of what I'm talking about is probably a full time research project! Whatever we do we should keep it as simple as possible.
          Exactly. I think some reasonable first goals would be to look into such ontologies, make some preliminary decision about categories/tags (either derived from an ontology we find, or from our own agreement), and get a template/form combination up which would allow people to enter packages tagged properly into the wiki. If anyone has experience with that type of wiki form, please let me know!

          Comment


          • #6
            Originally posted by dan View Post
            Can you please install the 'semantic bundle' extension? There are lots of crucial extensions for working with SMW, such as semantic forms and semantic drilldown.
            Done! (I think! )

            Comment


            • #7
              Originally posted by ECO View Post
              If anyone has experience with that type of wiki form, please let me know!
              http://www.bioinformatics.org/w/inde...ction=formedit

              :-D
              Homepage: Dan Bolser
              MetaBase the database of biological databases.

              Comment


              • #8
                Very cool, and exactly what I had in mind for getting users to help submit the packages to the wiki.

                Any chance you could help me set the thing up here? Or do you think this again falls into the category of unnecessary replication?

                edit: I found the holy grail....http://www.mediawiki.org/wiki/Extension:Semantic_Forms
                Last edited by ECO; 07-21-2009, 10:24 PM. Reason: I googled it for me.

                Comment


                • #9
                  Originally posted by ECO View Post
                  Very cool, and exactly what I had in mind for getting users to help submit the packages to the wiki.

                  Any chance you could help me set the thing up here? Or do you think this again falls into the category of unnecessary replication?
                  Yeah, I can duplicate that system on the SeqAnswers wiki.

                  I am worried about duplication... here is my chain of thoughts: Any successful wiki project depends on a community of users. Users are motivated for many reasons, but one strong motivation is peer 'pressure' (or peer recognition) within a ... community of users! So building up a community is a big part of building up a wiki. So if we create many similar wikis we kind of split the community and furthermore it doesn't feel good (from a user perspective) to not really know where one should edit...

                  Thinking about it though - This forum is very successful, so I don't see why we shouldn't try to build on that success with a nice well defined and focused wiki project :-)


                  Originally posted by ECO View Post
                  That is the mother lode ;-) But don't forget the basics!

                  http://semantic-mediawiki.org/wiki/Semantic_MediaWiki
                  Homepage: Dan Bolser
                  MetaBase the database of biological databases.

                  Comment


                  • #10
                    Good work ECO and dan

                    I'll chip in somewhat when I get the time... presently I'm furiously analysing some SOLiD reads before a conference abstract submission date rolls around.

                    Comment


                    • #11
                      Beta version of the SEQwiki software database up and running...

                      Check it out...

                      http://seqanswers.com/wiki/SEQanswers
                      http://seqanswers.com/wiki/Software


                      Dan.
                      Homepage: Dan Bolser
                      MetaBase the database of biological databases.

                      Comment


                      • #12
                        Thanks to an immense amount of work by "dan" for which I am eternally grateful, the wiki is ready for more widespread use.

                        Please browse the software collection, and most importantly enter and update any new software packages with the new handy form: http://seqanswers.com/wiki/Software
                        Last edited by ECO; 09-17-2009, 10:46 PM. Reason: dan put so many features into the wiki i could not possibly capture with my first attempt

                        Comment


                        • #13
                          Originally posted by ECO View Post
                          Thanks to an immense amount of work by "dan" for which I am eternally grateful, the wiki is ready for more widespread use.
                          The real heroes are users like "Krobison", who has made over 180 edits, including the addition of many new packages and references! Thanks to all the users who have contributed to, and who continue to contribute to the wiki!

                          Currently (without DB access) I can't make a detailed summary of the 'top contributors', but I'll try to set that up ASAP. See here (with the help of "ECO" ;-)

                          Please do use the feedback page to discuss how the NGS software archive on the wiki can be improved. For example... do we need a dedicated forum for the wiki?


                          IMPORTANT NOTE: I put some notes about what I have been doing on the wiki here, which includes attribution for the source data.

                          Please note that the work I have been doing is purely through the 'wiki interface', so any user can login and change / expand what I have been doing.
                          Last edited by dan; 09-17-2009, 11:47 PM. Reason: Added links.
                          Homepage: Dan Bolser
                          MetaBase the database of biological databases.

                          Comment


                          • #14
                            Wiki forum now active!

                            Yes, I'm aware of how silly it may seem to have a forum discussing a wiki that came from a forum...but mediawiki's discussion possibilities are downright awful. May I never edit another User:Talk page.

                            Comment

                            Working...
                            X