 |
 |
Thomas, it’s always good to have a couple of proverbs and formulas at one’s disposal. In the context of the Barry article I remembered Lasswells formula for content analysis „Who says what to whom with which effect“. Let’s start with „who“. Christine Barry tells us in her article that she is an „ex-technophobe“ and that she is three years using Nudist. The German Automobile Assoc (ADAC) often makes surveys among car owners and asks them three or four years after they have bought a car if they would buy this car again from today’s perspective. The astonishing thing is that about 90% of the car owners answer that they would buy the car again, even if they have a car type that had many problems and defects (according to the statistics of the ADAC). I think, probably, the same mechanisms are working concerning the use of software. So, it was easy to predict, it would hardly be possible for Atlas to „win“ the comparison. It is interesting that Christine Barry compares „Atlas versus Nudist“ with the earlier „DOS versus Mac/windows“ and she attributes nearly the same characteristics to DOS as to Nudist (Chapter 6.3). Finally, she declares the choice between DOS/Mac to be „ a matter of personal preferences“. What then happens in the article is something Luhmann would call „reduction of complexity“, and I can understand very well that you are dissatisfied with the misinformations published in the article. I think the author should make a good effort to correct this in case of the „Boolean search“ and the „Merge projects“ statements. In case of the authors remark that Atlas has „no option to automate commands“, things are more complicated, since this is in deed true (at least partly). We would have to discuss about functional equivalence, and this of course would again be „a matter of personal preference“. Anyway, I am convinced that the correction of these misinformations would not change Christine Barrys judgments, particularly the overview given in Table 4. To me this seems to be the more serious problem of this article: the usage of terms like „complex project“. Terms like this are not clearly defined and you will not find anything about it in the literature. It is a rather folkoristic term. But people read it and they keep in mind: „Atlas is not good for the analysis of a complex project“. And, of course everybody wants to classify his research as „complex“ and not as „simple“. This reminds me on the use of the term „theory builder“ that played a role in the Weitzman/Miles book (years later I meet people telling me that Nudist is a „theory builder“ and Winmax is a „textbase manager“!). Of course, everybody wanted to buy a theory builder and not a „text retriever“ or a „code-and-retrieve-program“ (in this case good for Atlas). As Lasswell I will end with the „effects“: The effects seem to be the most important problem of the Barry article. In the row „complex projects“ you will only find NUDIST, NUDIST, NUDIST. Everybody is free to have a personal opinion and may say „I prefer DOS“ or „I’m still working with WP 5.1“ (I know many colleagues doing this). Problems arise if someone thinks this could be an assistance for researchers making their choice of software. Regards Udo Thomas Muhr - Scientific Software Development schrieb: > Udo, > read -> think -> reply! (Wish I could always comply with this rule). > 1. I have no objection against comparisons, even if they are not in favor of > my "favorite". > 2. I never wanted to discuss Barry's personal opinion. I did not even want > to discuss, if misinformation should be published. > 3. You are quite right with your description of Winmax!-) > 4. You and many others know that ATLAS.ti and Nud*ist can do X and can do Y, > but the Barry article is directed towards those who often have no idea at all. > 5. I would like to support your last statement concerning preview by > developers but this could be an obstacle to get anything written by free > authors. > > Some topics that could be worth discussing: > > * Appropriateness of QDA tools for specific methodologies or purposes > * Helpful and not so helpful metaphors and artifacts > * Which are the essential and real-world predictors for choosing a software > in the end? > * Are there "user types" relevant for choosing software ? > * What should be improved ? > * What is missing ? > * How do the different tools handle complex projects? > * What comprises complex projects ? > > Remember, to discuss such topics is one legitimation for this list. > > Regards > - Thomas > > At 16:56 12.10.98 +0200, you wrote: > >Christine Barry has written an article that compares Atlas and NUDIST and the > >result is that one of the developers is not satisfied with the her > judgements. So, > >what! This seems to be quite normal and of course every developer wants to see > >his/her package on the top (by the way winMAX is far better than NUDIST and > easier > >than Atlas !-). To me it seems not very fruitful to discuss if Christine > Barry's > >opinion is right or wrong. Comparisons of software packages have always > subjective > >elements. I would recommend to anyone to download the demo versions of the > >programs and make their own experiences. > >Of course, most of us know that is completely nonsense that Atlas only allows > >simple boolean searches, but should we "discuss" about it? To avoid > >misunderstandings the publisher of such papers should give a draft of the > articles > >to the developers and get their feedback before printing. > >So far > >Udo > >---------------------------------------------- > >>>> Priv.Doz. Dr. Udo Kuckartz > >>>> Free University Berlin > >>>> Department of Education and Psychology > >>>> Arnimallee 12 > >>>> 14195 BERLIN > >>>> Tel. +49 30 8385539 Fax +49 30 83875494 > >>>> http://www.winmax.de > ___________________________________________________________________ > Dipl.-Psych. Dipl.-Inform. Thomas Muhr > Scientific Software Development - Internet: http://www.atlasti.de > e-mail: [log in to unmask] TEL [+49 30] 861 14 15 FAX 864 20 380 > ATLAS-TI mailing list (join, leave): www.atlasti.de/joinlist.html %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%