On Mar 27, 2014, at 15:39, Colin A. Smith wrote: > While the GUI is being updated, it’s probably work changing any references to > ISI, which I believe is an old and now defunct name, to Web of Knowledge. > > Cheers, > > Colin >
I guess we could change it in the UI, though not in the data. But I am still confused what they want to call it. Their website says Web of Science™. So should the label be WoK or WoS or something else? Christiaan > On Mar 27, 2014, at 15:17, Fischlin Andreas <andreas.fisch...@env.ethz.ch> > wrote: > >> Dear Christiaan, >> >> I do not believe you understood what I wrote. >> >> On 27/Mar/2014, at 14:49 , Christiaan Hofman wrote: >> >> >> On Mar 26, 2014, at 11:56, Fischlin Andreas wrote: >> >>> I am not convinced fall back is the only thing to too. That may be fine for >>> a casual user, but a regular user knows which service his/her institution >>> has and would therefore profit from being able to set the type of expected >>> service in the settings. For those users I also believe that the service >>> not being available is more likely to be merely due to an authentication >>> error, but I guess the new interface allows to distinguish that case. >>> >>> To wrap up, my suggestion is if one WOKSearchPremium is not available (and >>> any other cause than no subscription could be ruled out), then the fall >>> back could be done by testing only the availability of WOLSearchLite >>> service, and if WOLSearchLite is available then asking the user, whether >>> she/he wants to change the preference accordingly. Actual searches and data >>> retrieval would always only be done according to the settings. >>> >>> Regards, >>> Andreas >>> >> >> I don't think this i a good solution. there is no way to revert this, you're >> locked in. But your situation can change. >> >> No, with my solution there would be no locked in. >> >> >> Perhaps we can add a "Lite" option in the search group settings, set by a >> check box in the empty space between the database and username, as in the >> attached screen shot. >> >> No, not a good place. My suggested location in my previous mail just below >> the 'ISI' radio button is better, because it does relate that Lite checkbox >> with the radio button choice "ISI" that it is associated with. I believe >> your suggested location would be quite confusing for users and you need to >> know a lot to be able to relate this 'Lite' checkbox with the ISI radio >> button. Therefore, if a checkbox, then only just below the 'ISI' radio >> button (as I have suggested in my previous e-mail). >> >> I personally would prefer the two radio button solution, i.e. 'ISI' and 'ISI >> Lite'. No checkbox. This seems to me to be the clearest solution, from which >> I would expect every user to immediately understand it and have a clue what >> this option is all about. No additional help needed: >> >> [cid:0571DC6A-1381-4DEE-8C14-9D9A0EC8900A@ethz.ch] >> >> My suggestions with a "fallback" testing of the Lite service in case the >> Premium one fails, would only happen IN ADDITION to the preference setting >> with the two radio buttons given above. Therefore no lock in, and you would >> get asked as a user that BibDesk would change the settings for you. Next >> time you visit above settings dialog window, you would see that BibDesk >> would have switched the service from 'ISI' to 'ISI Lite'. But note, IMHO, >> that feature would be only nice to have and by no means necessary in any >> way. I wanted only to say that in my view that would be the only good place >> where to such a fallback algorithm as suggested by Reto would be useful. My >> main reason being mostly efficiency reasons. >> >> My suggestion for the default is however, for backward compatibility >> reasons, radio button 'ISI' is default for the factory defined 'Web Of >> Science SCI' setting. >> >> >> There is also the question of the database. It seems this son't be used >> anymore, but is that what we want? Or would we support restrictions to the >> editions to search? This could perhaps also be a space-seprated list of IDs, >> and empty or some default value like WOS to search all editions. Though I am >> not sure about backward compatibility. >> >> Andreas >> >> >> Christiaan >> >> <Screen shot 2014-03-27 at 14.44.02.png> >> >> >>> >>> On 26/Mar/2014, at 10:01 , Reto Stöckli wrote: >>> >>> Sorry, I mean search group settings. However, the automatic fall-back >>> solution to WOKSearchlite in the case of an error is also intriguing and I >>> will check out that solution first. >>> >>> Reto >>> >>> On Mar 25, 2014, at 11:53 PM, Adam R. Maxwell wrote: >>> >>> >>> On Mar 25, 2014, at 15:12 , Christiaan Hofman >>> <cmhof...@gmail.com<mailto:cmhof...@gmail.com>> wrote: >>> >>> >>> On Mar 25, 2014, at 20:53, Reto Stöckli wrote: >>> >>> One more thing. >>> >>> If you can propagate a flag from the Search GUI (by means of a >>> user-accessible button) into BDSKISIGroupServer.m indicating whether the >>> user would like to use WOKSearch or WOKSearchLite, then I can also >>> implement the WOKSearchLite SOAP decoding. Most users do not have the full >>> WOKSearch access since Thomson Reuters charges extra for it since about one >>> year. >>> >>> Reto >>> >>> >>> I don't think we should do this from the search UI, it messes it up and >>> it's not really an individual search setting. If anything, it should be in >>> the UI for the search group settings. Though it would need a special UI >>> only for this search type, which is annoying. As we're not using the >>> database anymore (is that really true, is it not useful anymore?) we may >>> perhaps appropriate that for a "Lite" setting? >>> >>> Agree, this should not be set from the search UI itself. It sounds like it >>> should be a separate search group that should be predefined along with the >>> basic WoS search, unless there's an error message from SOAP that would tell >>> you to fall back to WOKSearchLite transparently. >>> >>> Adam >>> >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> Bibdesk-develop mailing list >> Bibdesk-develop@lists.sourceforge.net<mailto:Bibdesk-develop@lists.sourceforge.net> >> https://lists.sourceforge.net/lists/listinfo/bibdesk-develop >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> Bibdesk-develop mailing list >> Bibdesk-develop@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/bibdesk-develop > > > ------------------------------------------------------------------------------ > _______________________________________________ > Bibdesk-develop mailing list > Bibdesk-develop@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bibdesk-develop Christiaan
------------------------------------------------------------------------------
_______________________________________________ Bibdesk-develop mailing list Bibdesk-develop@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-develop
