(ROSE) Rose - Basic usability issues and features missing

[Ari Vi�ar J�hannesson]

I am a seasoned Rose user and I like using Rose. For sure it has improved a lot over the years and it is defenitely the best tool on the market. However I have wondered why a number of basic usability issues and minor features haven’t been fixed/implemented yet. Is Rational managing their risks to such an extent that they do not dare fix some basic things in fear of the risk it involves?   Here are a few things from the top of my head in no particular order:   * Multi-select and drag/drop items in the tree view not possible. You can multi-select the items but if you drag the selected items and drop them into another package, only a single item is moved.   * "Resize-sensor-area" is way to small, i.e. you have to click on a tiny point to resize an element so that you always need 2 to 3 clicks before you succeed.   * It's silly to have fixed package names, i.e. Use Case View and Logical View. There are so many variations you can set up a model that this should not be forced upon the users.   * You cannot order/sort packages manually in the browser tree.   * You cannot order/sort by <<stereotype>> in the browser tree, even though that is usually the most appropriate ordering of elements.   * No object diagrams!   * Deployment diagrams extremely limited and far from being UML compliant. At least you should be able to depict components inside nodes.   * You can only do one deployment diagram for the whole model.   * It should be possible to filter the tree browser so that you only see stuff your are currently working with. It's quite confusing to work in big models and find stuff in the tree, especially if they are far a part in the tree.   * Only one-step undo! No history!   * The last-diagram back-arrow has no history. You can only toggle between current and last.   * Tools menu is too big with all the Add-ins listed and sort of strange compared to the Add-ins menu which only has one menu item, i.e. "Add-in Manager".   * No "File/Close" menu item - inconsistent with other software and no reason not to have it.   * You cannot change a path-map if a read-only .cat file references it. You have to unload the model (not by "File/Close" but by "File/New") open an empty model to change the path map and then reload the model original model.   * The path-map concept is too confusing in general.   * “Format/Automatic resize” doesn't work, i.e. you have to manually take it off and then put it on again for it update.   * It's only possible to move items in the browser tree by dragging, which is quite cumbersome if the browser tree is big.   * When you drag something (e.g. a class) into a closed/collapsed package, the package is always opened/expanded, instead of remaining collapsed.   * Automatic resize of notes as you type! I don’t think I have ever added a note to a diagram without having to resize it afterwards (using the tiny dots in the corners of the note).   * Not possible to duplicate packages or elements, i.e. so that if you duplicate a class a new one is created with a new ID.   * It is not possible to view the properties of a file attachment in the model, i.e. if you insert a Word into a package you cannot check its file location.   * You cannot name URLs in a model. If you insert a URL reference into a package the whole URL is shown in the three, which might be very long and cryptic. You should be able to give it a descriptic name.   * Graphical items change in size and shape without the diagram being modified. Name changes shouldn't affect graphical representation such that when you chance the name of an element all the diagrams it appears in get messed up.   * Not possible to show same diagram in more than one package! Maybe a UML restriction but I can't see the harm. At least it would be good idea to be able to have "symlinks" or "shortcuts" to an item in another package.   * You cannot depict classes inside components in component diagrams.   * Not possible to attach colors to elements, stereotypes, package elements (you have to do it on each graphical representation of an element).   Best regards, Ari Johannesson