Guys, has anybody used/evaluated together J, can u pl. high light your experiences in comparision to rose and other tools. I am kind of impressed by the tool, but doesn't know what it stands for. avijeet
I guess you must love Java (or you won't be here), so I can tell you that Together/J is way better than Rational Rose (for Java projects). What I like most about Together/J is the real-time sync between UML diagrams and Java codes. Rose also claims to be round-trip engineering, but it's not an easy job to achieve that. Some new features of Togehter/J are also very attractive, such as its J2EE support, hot-deployment of EJBs to various most popular application servers (WebSphere, WebLogic, etc.). Download the 5.0 version and play with it. I'm sure you'll love it. ------------------ Tony Chen SCJP, ICS & ICSD (WebSphere)
SCJP Java 2 - IBM Certified Enterprise Developer - WebSphere V4.0 & V5.0 - IBM Certified System Administrator - WebSphere Network Deployment V6.0 - Canadian Securities Course (CSC) Certification
posted 19 years ago
Thanks! Tony, I surely love java, i got struck when i found that together J has integrated Design Patterns support. I have downloaded it and should evaluate it, but again the Morphy's Laws are making there shows. anyway, how do u see this coming up in futures, can it compete with Rose? Avijeet
Actually I think Together (i use control center) is really in a class by itself. the code that is generated is just the interface (non java inteface) that is important to keep in sync. it is just soooo easy not to go back and redo the object model. and TCC reversed engineered code is a lot better. actually what i really like about TCC is that it is methodology independent: i can use it with RUP, XP, FDD or a lot of other development processes. rational was made for RUP (and it kind of expects RUP). now there is nothing wrong with RUP but we use a lot of different methodologies. if we were strictly a RUP shop then i would probably go with rational, but since we are all over the board TCC seems to fit a little better.
Also, we're finding Together a bit klunky for analysis purposes. For example, if you create a sequence diagram and assign it to a package then the only way to move that sequence diagram to another package is to recreate it. In analysis when these things are still being discovered this limitation is not really acceptable.