Having gone to prvious Java Ones, I did find sessions useful and pracical in most cases. But I have to admit..since I dont do anything with J2ME..I skipped them all together.

The talks seemed to focus on real issues like performance etcbut I was dissapointed that we had to wait in long lines atleast on two sessions everyday..I donot know why there were no overflow rooms and it seemed the popular sessions seemd to be in rather small rooms...but they did repeat some of the sessions..

also I thought that they too many parallel tracks..so I had to miss some sessions.Now the multimendia content requires a fee and I really think that after shelling out $1700 plus other costs for travel..the conference attendees should get access to all conference content.

Was anyone else disappointed with the quality of the sessions? We were lucky if we could understand the speakers and were extremely lucky if the speakers had anything to say that was really worth listening to! I spent an hour in a Java Instant Messaging and Presence session listening to the speaker tell us why API's are important! I think we all know why we need API's and why API's are important... Many examples of sessions like that in my experience... I have gone the last two years, and am not sure if I will go again next year. :(

As in past years, I'm not able to afford the registration fee, but I'm not complaining. Heck, I feel lucky to be employed. I've always been able to catch up on the goings-on at JavaOne by browsing the Sun web content and other sites. I just miss out the cool swag, that's all.

That changed a bit this year though. Now Sun wants me to pay for a post-event recap of JavaOne content? My employer won't cough up money for this sort of thing, and my own pockets aren't deep enough for even that.

I know, to some it probably seems like I'm just cheap. But to me it seems that Sun is trying a bit too hard to milk a profit from JavaOne... and it has ceased to be Java evangelism.

It is just me or the J1 conference did not offers anything new compare to last year?

Honestly, I don't see a reason for holding developer conferences or pay for them. A well written technical article is worth much more than some power point session in a conference hall where you can't try anything youself. And then the keynotes and panel discussions. Why can't they just discuss things in a mailing list and put their keynotes behind a big blinking banner ad that says click here if you want to waste some time on marketing speak.

I am not unsatisfied. It has been a nice conference and certainly one of the better developer conferences I have been to. The focus on J2ME and MIDP on particular seems quite odd though. Having done some significant development using MIDP I can only say, it is far far from being a good environment. It lacks mandatory features, particularly telephone support. But most of all, the implementations I have dealt with are extremely error prone, so that it really is write once, test many. Given 50 or so different platforms that is far from being fun. Sun obviously failed to provide a certification program for MIDP. This all is bound to get better with mobiles having more memory and faster CPU, but just looking at the adoption rate of MIDP from the past I am rather pessimistic. Also about 90% of the showcased MIDP applications were either bound to a particular device or plain and simply unusable.

On the other hand, it is a good thing that Sun FINALLY FINALLY FINALLY understood that they need "Java Inside" stickers on mobile phones and mobile phone boxes or even Smartcards. It is outrageous that they haven't pushed for that before, and frankly, being Scott McNealy, I would consider some serious butt kicking.

The emphasis on the integration of scripting languages seems a bit strange. Java is a perfect scripting language in its own right, if used in JSPs, so I see pretty much no use opening it to PHP. Using scripting for configuration and dynamic behaviour changing is nice. But I doubt that you would need something as powerful as python to do that. If that would be the case, you will be better of writing a java class. Basic ECMA Script (or tcl) is probably sufficient for anything you might want to do.

As for the new language features that were greeted with much applause, I am not sure about all of it. For one thing, I like the generics. Having printf is pretty ridiculous, if you dropped using it when switching to c++ output 10 years ago. The for-each loop, the static import and the enum are pretty much "syntactical sugar" and I probably could live without it. Besides, since enum is a class and thus dependent on classloaders, it will probably not be usable in Server Side Environments and is quite likely to create a lot of error conditions.

I got in for free on Thursday, pavilion only. I had a piece of paper from my brother in law (that had a real pass), saying "Bring a friend to the Pavilion for free on Thursday", but they never bothered looking at it. Also, why not buy one pass and split it up during the week? Bob goes on Tuesday, Mary on Wednesday, Tom on Thursday and Jane on Friday? We've done that several times, and there is never a problem!

Personally, I think its good to see Sun promoting the 'platform' as well as the language.

Its great that 'VB level' programmers can have access to hotspot compilers, garbage colletors and J2EE compliant app servers.

Why? The more developers on the Java platform, the better the tools will get for the rest of us.

Cheers, Peter
RimuHosting- JSP/EJB Hosting Specialists

It was worrying that admission prices were raised to over $2K per person. Yet it seemed that there were as many people at the conference as last year. Apparently someone is opening their wallet to send people to J1.

In the last five years, Java has evolved into a set of specialized platforms, each tailored for a set of programming problems. With J2ME, J2SE, J2EE and JavaCard, Java now requires significant expertise in at least two of these areas (as SE is the foundation, anyone programming in Java requires significant expertise in it).

It is also a sign that Java has matured to tackle considerable, real world problems. Programming specialisation has been around w.r.t. other languages and platforms for decades. The need for expertise and experience in a specific technology and/or problem set is nothing new.

As such, a general purpose "Java developer" really truly a rarity, both because of supply and demand in software development. With the dynamics of development, one can no longer be a Java generalist - just the same way as it is with other programming languages and platforms.

On the bright side, it is an indication that Java is sufficiently rich and powerful to address a broad array of problems. As such, "dumbing it down" (in the fashion of Visual Basic or other scripting languages) is extremely unlikely. Java development would be in big trouble otherwise.

Software development is about solutions - how can an end user take a software package, and use it to meet their needs, coding only for the specifics of their business or endeavour? Java has met this challenge very nicely with it's technology stack. However, solutions require greater specialisation, because the stack is inherently complex.

Which leads to another point - ease of use is a big concern, most especially in the EE stack. Although even J2SE 1.5, with features such as generics, has a certain level of complexity, this does not compare to the at-times overwhelming complexity of, say, EJB 2.x, and the large-scale aggregation of other technologies into J2EE 1.3 and 1.4. Ease-of-use addresses support for tooling for much of the EE stack, so that it isn't so daunting to leverage it.

In fact, it was nice to see that, in each of the stacks, progress has continued, but it is not being just thrown together as when J2EE 1.2/1.3 came out. Sun and the Java community are stepping back a little bit and thinking through what exactly they want going forward and how to put it together a little more nicely. The approach to J2EE 1.4 and the latest J2ME stack is very encouraging.

I would have to agree. The single reason we were not allowed to go was price. Salaries are much lower, companies are tight on budget, and most of us can't afford to live in this area, let alone pay for the admission cost. I do realize that the JavaOne team has to pay for the pavaillion and so forth, but paying $1700 or so is a LOT of money! Especially if you have a team of developers that want to go. I guess the other thing that turned me off was the major focus on the J2ME/wireless. I don't get it to be honest. J2ME is alright right now, but with chips getting faster and smaller, memory chips like SD memory so small and getting ever smaller, why are we so gung ho over J2ME instead of finding ways to get J2SE to these devices!!?? I know J2ME has its place for now, but it wont be long before watches can fit a GB of memory and run linux. ;)

Still, you'd think with the cost of things, such as rent, going down since a few years ago, the economy in bad shape still, and so many tight budgets, they'd have lowered the price quite a bit. Could be a big reason why attendance was what, 1/5th what it was a few years ago?

I personally feel Java has more going for it now than ever, but when you see attendance drop off like that, it isn't just .NET at work. I hope next year they make it more affordable. I'd pay a few hundred to attend for 4 days. I'd have to sell my car to afford their price now though.

I was really disappointed this year about the admission policy that is adopted for JavaOne 2003 Pavilion. In this economy, not every company would spnosor everybody to attend the conference. Also, there are not that many people who can afford to pay for full conference. But, from the information, I was told that I could not visit the Pavilion, unless otherwise I pay for the full conference. You could have some nominal charges ($25/$50) for visiting the Pavilion, so that people like me could have come and visited the Pavilion.

I hope this comment could help you make the right decision about the Pavilion admission.

Thanks.

- Thangam