Sparx Systems Forum
Enterprise Architect => General Board => Topic started by: Neelam on February 19, 2002, 05:08:40 am
-
We are evaluating EA as a possible alternative to Rational Rose. Apart from the cost-savings, I need to find out what are the benefits of using EA.
Would greatly appreciate inputs from anyone on this.
Regards
Neelam
-
Look at ...
Ease of learning
Ease of use
Speed of use
Flexibility eg extensions into requirements modelling, project management
Interoperability eg import/export, APIs
Quality of support
Responsiveness to enhancement requests
Responsiveness to faults
Product architecture
User forums
Documentation
Training
Consultancy
Standards compliance
Integration with other tools
IMHO, EA wins on most of these but not all. Its important that you work out what is important and why.
-
Alan,
I would add two major things to all of your great points.
First, the communication and support we are blessed with DIRECTLY from either Geoff or Paul. The level of personal attention to the product and to us is down-right REFRESHING. In these days of large-scale, stock-based companies where inovation and service pale, using EA has been a bright spot.
The second thing is some internal stuff. I'm working on something at the moment that I'll share at the appropriate time; however, the ability to get DIRECTLY to the data stored in my project easily and effeciently has been absolutely astounding. The notion that I can bring resources (Model Authors) from EA over to a Microsoft Project without effort, or to build in dashboard alerts based off conditions in those tables, is so bloddy easy it is not funny.
Now I tried this concept out over a year ago with Rose and found it a PITA... Together was better but still not as easy as I have found with EA.
Just my .02...
Cheers,
Steve
-
Dear Alan
Thanks for your inputs.
I needed some more information on the following:
Quality of support: I do not see any support centers mentioned on the site. Is it online support over the web?
Integration with other tools: If you could name them, it will be easier for me to sell it to my clients :)
Thanks in advance
Regards
Neelam
Look at ...
Ease of learning
Ease of use
Speed of use
Flexibility eg extensions into requirements modelling, project management
Interoperability eg import/export, APIs
Quality of support
Responsiveness to enhancement requests
Responsiveness to faults
Product architecture
User forums
Documentation
Training
Consultancy
Standards compliance
Integration with other tools
IMHO, EA wins on most of these but not all. Its important that you work out what is important and why.
-
Neelan,
>>Quality of support: I do not see any support centers
>>mentioned on the site. Is it online support over the web?
As I mentioned in my previous post, support is directly from the guys who created the product: Geoff and Paul. They have responded to questions posted on this forum or replied in emails. While I do not believe there is a support "phone number", that has to be compared with those companies that put you in touch with "Support Technician" and not the developers themselves. Moreover, the electronic support is free as opposed to those companies that charge for support and end up giving "Customer Non-Service".
>>Integration with other tools: If you could name them, it will
>>be easier for me to sell it to my clients
What tools are you specifically looking at with regards to integrations. Since EA imports and exports Java code one could stretch the definition of "integration" with such tools like JBuilder, JDeveloper, et cetera. The file format can be directly accessed through the Jet Engine and I am finding it _VERY EASY_ to integrate EA into my tool suite: CVS, Microsoft Project, Dreamweaver 4 (for executive alerts), et cetera. So maybe it would be better if you give us some idea of your toolset and we can tell you the integration issues.
Hope that helps,
Steve
-
Hi Neelam,
Here's another perspective on Enterprise Architect that may help, when comparing against Rational and some comments on TogetherSoft as well.
Rational is a Big System
Rational's a Big Name. When you buy Rational products, totally independent of the quality of the products, you're paying for a brand. The argument goes: Rational's going to stick around, Rational will be there, look at all these success stories, there are thousands of Qualified Rational Consultants, you can be SAFE.
Aimed at the Corporate market where name and brand confidence is generally first and foremost, Rational Suite is what's really on the agenda. The Complete Solution. The package is: a tool for every purpose (oh wait, you want truly user-centred design, aha, well sorry buddy you have to wait until WE think its important enough), all wrapped up nicely with a big bow. I remember thinking, coming out of software engineering classes all those years ago that traceability was key, and if we had a system that could trace through changes from requirements to code, we'd be in Nirvana. However 95% of systems do NOT NEED complete automated traceability: the reality is that there will be people on the team who will, effectively, be able to track changes with project mgmnt tools and even little bits of paper at times, and -shock- even using their HEADS occasionally! Heavens above. Anyway. What I'm saying is that most project environments are not the rigorous type that a) can afford Rational and b) need it. Motivation for most Rational Companies is brand confidence these days. (I mean, that's not a trivial thing, but less important for smaller companies.)
The reality is that, when the rubber hits the road, Rational is quite a complicated system to work with. Its license management system is a friggin nightmare (hands up who have suddenly needed another license half-way through a project, but only for like 2 weeks and that's it because the project tasks were parallelized? aargh). And at IconMedialab where I worked last year the Policy was Rational, but the reality was it was WAAY too expensive even for us (a boutique ebusiness company). And its tools: Requisite Pro, ClearCase etc.. they're really complicated. They take ages to set up. You really do need Rational Consultants, because it's a big deal to work with the suite.
I was looking at Enterprise Architect at Icon (as Geoff may remember) but what suited us better at the time with the processes that were already in place for capturing use cases was Together from www.togethersoft.com, which I bought for our team. Together's a very slick UML/Java tool with real-time sync of code and UML, with C++ and C# support as well. It does JSP really well now too. But it's Rational prices: about $US7k per seat, and it too now has license management servers which are ugly: you have to install a license server on every laptop that you want to take home... gimme a BREAK. And it eats memory and CPU for breakfast. If you want to work with just Together, you really need a minimum of 256MB/500MHz machine (not so bad these days) but if you want other java apps running, add another 500Mhz, another 256MHz... it eats CPU for breakfast. Yuuum yumm. Scrunch scrunch.
NOW. EA. A big deal is that EA does UML, just like Rational Rose, and Together. The core product does exactly the same thing: there's no proprietory model to work from and so there's a major distinguishing factor removed from Rational and EA.
Support for a small company works differently to a big company. If you're more comfortable with the bows and ribbons and flash Service Level Agreements and Assurances Of Quality Con$ultan$ Available To Help then that's fine, Rational is really for you and your company, honestly. But EA is from a small company, and their support is as good as one can possibly get from a small company:
- you have easy access to the people who write the software. If you find a major bug or issue that really annoys you, you can -expect- an update within a week or two: urgent patches I know have been done within 24 hrs if there's a showstopper. That's just how Sparx works. Try getting a bug fixed in Rational. Uh uhhh! No way. Wait until we want to release it, which will be at least 6 months away. And new features? Niggles? Forget it. Totally ignored. Sparx listens. That's what I've found. The probability of a new feature appearing in EA that you suggested is infinitely greater than in Rational. That's what's cool about using a Small Company Product.
- there's an active user community. People ask questions, and they're answered. You know what? This is what people generally resort to for almost all products: Rational, Microsoft, ATG, BEA, all of them. Except in the EA forums, as I mentioned above, EA is right in there.
Will Sparx be around? This depends I'd imagine on the continued support by the user base. Sparx has been around for quite a while alread. Like any small company starting, I'd imagine Geoff has invested an -awful-lot- of his own time for this project but there are other people responding to sparxsystems.com.au addresses so obviously things are getting better.
The future? It can only get better. With code gen, EA is a seriously attractive tool.
To Geoff and the gang: I would pay THREE TIMES the current enterprise price for real-time code/diagram sync like in Together. Integration with JUnit testing would be pretty nice -- you have the test infrastructure, JUnit is the java standard for unit testing now, makes sense to integrate. If you could just sliip JSP in there that'd be really lovely too :)
I don't know what Geoff's exit strategy is, if there is one. I guess there will be some point in the future where he might want to retire, but hell, maybe he already is! And maybe this is What He Really Wants To Do With His Life. I personally can't think of anything cooler than doing what he's doing.
All I could ask of Geoff now and in the future is that if he decides to bail, release the source so the legend of EA can live on.
Anyway, my €0.02
-
Kiwi,
"Rational is a Big System Rational's a Big Name. When you buy Rational products, totally independent of the quality of the products, you're paying for a brand. The argument goes: Rational's going to stick around, Rational will be there, look at all these success stories, there are thousands of Qualified Rational Consultants, you can be SAFE."
First, you pay a higher price due to the entire support mechanism that goes with Rose. In addition, they are a stock-traded company, meaning they have to report earnings in line with what outsiders expect. Smaller companies don't have this INFLATED expectation. Second, there is no "safe" perspective with anything... companies can choose to drop a product in a second if it doesn't generate the ROI required. Now, you can "hedge" your bets better with going with a "brand name" but still, there is no guarentee. Ask Visual Fox people how safe they feel in Microsoft hands (for example).
What I think you are attempting to do is explain what is a natural force within ANY market, as explained in the book "Crossing the Chasm". You are part of the bell curve that is considered more "conservative" and the challenge for any product that appeals with the "Inovators" and those on the left-lower end of the curve is crossing over into that mainstream where I believe you are speaking from. The trouble is that the motivational arguments are different for each section of that curve. For example, when cell phones came out only a select few HAD to have them... it wasn't until they (phones) got smaller, cheaper to use, more features, et cetera did they (phone) cross the chasm. However, the argument for having one (phone) was/is different for each aspect of the bell curve. For those who HAD to have them (like investment bankers, et cetera), the cost was not a factor... servicing the need WAS the factor. For people like my mom, she finally got it because it was cheap and it was insurance for her if, for example, she was ever stranded in her car. Two DIFFERENT needs and two different perspectives on the same general product.
This leads to the general jist of your last points which are VERY VERY valid: you had a SPECIFIC need and Together was a better fit. I applaud you, or anyone for that. We, too, looked at Together and since we don't have legacy Java code and other related issues, that wasn't a problem. Everyone is different. Comparing your NEEDS to the TOOL really should be the only way to buy software. Ironically, it is the base on why people upgrade or not: are the new features worth it for the price and is it needed? Asking what tool is better in general terms is often like trap shooting in a hurricane... interesting concept to banter about but just try hitting the target <smile>.
Cheers,
Steve
-
Fascinating Steve... thanks for joining in the discussion... yes you're right of course, the tool for the job.
Rational has invested a lot into its name to present 'the trusted/safe choice'. That's all I really wanted to say. And that this idea of 'safe' is not often necessarily aligned with reality: there's the idea of Rational sales force, but if you actually try and get access to it, depending on where you are in the world, you will have limited success getting people who really do know the products well. This whole 'consulting support wing' thing is designed to reinforce this idea of safety, and indeed it's very important, as is ideas of certification etc, for companies that size.
It's amazing how the costs scale out of control as things get bigger. Like, how EA does just fine with a freeware BBS tool for its 'consulting support wing' :)
-
Kiwi,
"Fascinating Steve... thanks for joining in the discussion... yes you're right of course, the tool for the job. "
<lol> I thought I was in the discussion at the outset <smile>
"Rational has invested a lot into its name to present 'the trusted/safe choice'. That's all I really wanted to say. And that this idea of 'safe' is not often necessarily aligned with reality: there's the idea of Rational sales force, but if you actually try and get access to it, depending on where you are in the world, you will have limited success getting people who really do know the products well. This whole 'consulting support wing' thing is designed to reinforce this idea of safety, and indeed it's very important, as is ideas of certification etc, for companies that size. "
Exactly, and this follows the patterns expressed in "Crossing the Chasm".
"It's amazing how the costs scale out of control as things get bigger. Like, how EA does just fine with a freeware BBS tool for its 'consulting support wing' "
Not amazing at all... very very sensible. Let me give you an example. GE recently received an award from CIO magazine as 'best internet' site. The under-pinnings of that constituted technology derrived from Struts, Castor JDO/XML, and CVS (just to name a few) and ALL of these products are FREE. Expanding that further, there are tools like JBOSS that are EXCEPTIONAL and part of the GNU license agreement. We are using all of these tools at World-eIT. In short, price does NOT equate to quality. The trouble, as you pointed out, that factor of scale (meaning support infrastructure) often does equate into the purchasing decision. For example, a company like Rational, CA, Oracle, et cetera, have large sales forces that take day-based trips to cover their region just to have a face-to-face meeting that is nothing more than a woo job and convincing the customer that "they" are better. Sales people will fly in technical reps to meet with the client, dinners and lunches will be added to perk-up the customer, and in the end... the price tag of the software goes up as a result to cover the added "expense".
There is, unfortunately, an expectational perspective from large companies making large quantity purchases and SLA's from large software vendors who, with market pressures of their own, inflate the price. The irony is that the software, whether it be cumbersome or eloquent, is the same if valued at 5 or at 50,000... it is all that other stuff that (as you highlighted) that inflates a realistic price to one of astronomical perportions.
Steve