Next Big Language--UML?

I took a class at work called, "Real-Time Software Design with UML 2.0." The instructor seemed to think that it was destined to be the Next Big Language. I have my doubts.

When I design with code, I can type it very quickly and even somewhat sloppily into a code editor. The editor will then take care of formatting it all nicely, indenting, spacing, presenting with an nice font, colorizing the various code constructs, and so forth. I just have to get the ideas expressed and the tool takes care of most of the formatting for me. Even more, it's saved in a simple text format that is universal. Any text editor on any operating system will understand it. Once I've written the code, I then have a compiler and/or linter that will tell me if I've left off any key constructs, made any syntax mistakes, or any of a number of common logical errors. Basically, with the good tool support that is out there, I can write code as fast or faster than I can dream it up.

With UML, from what I can gather, it seems like you get none of that. You either draw on paper or a whiteboard which obviously makes the formatting a manual process. At least a whiteboard is easier to erase and redraw something than paper, but I'm still very slow at drawing stuff out, and I find myself either spending too much time making sure my boxes and lines are neat and nice, or I end up with an incredibly messy drawing. With Visio or other diagramming tools, you get nice straight lines, but you still get to worry about the niggly things like where to place your bubbles and boxes, how to arrange your arrows. Should it flow from left to right or top to bottom or just have a spaghetti mess of boxes and arrows all over the place? Not to mention there is nothing to verify that you are using proper UML syntax, and that you haven't left anything basic to a coherent design out. I find that having to think about all these details distracts me from the actual design work. Then, once you've painstakingly drawn it all out in one tool, you are stuck with that tool. Want to share your design with someone else? Sure, you can print it (did you draw it so that it fits nicely on letter size sheets of paper?), or export it to a PDF, but if someone else wants to edit the diagrams, they need to use the same tool you did. Let's not even mention collaborative design editing and revision control.

Despite all this, Glennan Carnie from Feabhas told a very convincing story this week. I came into this course very skeptical, and for the most part he had an answer for all of my concerns. And pretty good answers too. In a short, intense, week-long course it's hard to evaluate whether everything he says is really practical and useful in the real world, but the way he tells it I'm at a point where I'd be willing to give it a try...I mean, have a go at it (did I mention Glen is British?).

This course was not at all what I expected. Glennan freely pointed out all the flaws, idiosyncrasies, and shortcomings of the UML standard, other UML proponents, and even his own lecture slides. He made sure to point out more than once that UML is no silver bullet and that design is still difficult to get right. Instead of presenting UML in glowing worshipful tones, he presented it as a useful tool that can help you systematically design an entire object oriented software system. A tool which can be used to design first at a high level, ignoring implementation details, and then which can be used to refine the design down in a number of steps all the way to the nitty-gritty details of deciding how big a message queue is needed to communicate between two threads, and what the format of the message should be. It's very impressive.

Still, I wonder how it would work to go from an ideal solution to the heavily constrained designs that are needed for my work. Low-level ASIC driver stuff would be an interesting case-study. Too bad we didn't have more time with him to get into some of our own stuff like that.

Course description.

Comments

piyo said…
Hi, I saw your post on Planet Emacsen. :-o

The course description: "The course specifically does not make use of a CASE tool. From our experience a CASE tool distracts delegates from learning design issues and UML. However, the workshops clearly demonstrate when CASE tools are helpful and when they are not, thus aiding CASE tool selection."

Can you tell me more about the criteria for selecting a CASE tool? And do instructors really refer to learners as "delegates"? :-/

I'm not an expert at all on UML, but for my work documents I use the Jude community edition (freeware) application to draw UML sequence diagrams, etc. It verifies that you are drawing UML correctly it seems, which should be better than a non-constraining environment. For example, there's an option for "always draw return arrow from synchronous call", which is handy.

Tell me if there's something open-source though, since this one doesn't seem to be.
December 8, 2007 at 5:29 PM
Bryan said…
Thanks for your comments. I'm not really one to comment on how to select a CASE tool, if you couldn't tell from the blog entry, I haven't liked any that I've tried. The course instructor seemed to like Enterprise Architect the most:

http://www.sparxsystems.com/products/ea.html

But he said he uses everything, even PowerPoint to draw UML.

Here's a pretty extensive list of tools:

http://en.wikipedia.org/wiki/List_of_UML_tools

ArgoUML seems to be the most prominent open source UML tool.
December 8, 2007 at 9:37 PM
Post a Comment

Popular posts from this blog

SystemVerilog Fork Disable "Gotchas"

This is a long post with a lot of SystemVerilog code. The purpose of this entry is to hopefully save you from beating your head against the wall trying to figure out some of the subtleties of SystemVerilog processes (basically, threads). Subtleties like these are commonly referred to in the industry as "Gotchas" which makes them sound so playful and fun, but they really aren't either. I encourage you to run these examples with your simulator (if you have access to one) so that a) you can see the results first hand and better internalize what's going on, and b) you can tell me in the comments if this code works fine for you and I'll know I should go complain to my simulator vendor. OK, I'll start with a warm-up that everyone who writes any Verilog or SystemVerilog at all should be aware of, tasks are static by default. If you do this: module top; task do_stuff ( int wait_time); #wait_time $display ( "waited %0d, then did stuff" , wa...
Read more

'git revert' Is Not Equivalent To 'svn revert'

I just learned that if you have some changes in your working tree that you want to get rid of, you don't type 'git revert' like you might guess. No, that's what cvs, subversion, mercurial, and bazaar (to name a few) use revert to mean, but not git. With git, revert is used to undo actual commits. Thankfully, you can undo your revert with another 'git revert', I just learned. So let me repeat to be clear, if you have changes to your working files that you want to abandon, DO NOT do this: git revert HEAD That will undo your last commit. Do this instead: git reset --hard HEAD I'm glad I have that straightened out now. I'm wondering if /etc was really a good place for me to start out playing with git. UPDATE: Nearly two years later and I'm still getting comments on this. I'm glad I've been able to help people out this way. The discussion in the comments is good, and one thing I'd like to point out is that I now always use...
Read more

SystemVerilog Streaming Operator: Knowing Right from Left

SystemVerilog has this cool feature that is very handy for converting one type of collection of bits into another type of collection of bits.  It's the streaming operator.  Or the streaming concatenation operator.  Or maybe it's the concatenation of streaming expressions (it's also called pack/unpack parenthetically).  Whatever you want to call it, it's nice.  If you have an array of bytes and you want to turn it into an int, or an array of ints that you want to turn into an array of bytes, or if you have a class instance that you want to turn into a stream of bits, then streaming is amazing.  What used to require a mess of nested for-loops can now be done with a concise single line of code. As nice as it is, getting the hang of the streaming operator is tough.  The SystemVerilog 1800-2012 LRM isn't totally clear (at least to me) on the details of how they work.  The statement from the LRM that really got me was this, "The stream_operator <...
Read more