Summary
There's stuff to do at JavaOne without spending a dime

I decided not to go to JavaOne this year. Didn't go last year either. My company doesn't see a lot of value in seminar things unless they're very, very specific to what we're doing, and I couldn't really justify the cost of going on my own nickel. However I did manage to go to a couple of Jini related things that didn't require a J1 badge.

Monday was an informal meet up of Jini interested folks at a bar called the Thirsty Bear. Completely misjudging traffic I actually arrived about 5 minutes early, driving in from the east bay. A cruise through the rather crowded bar revealed no faces I knew. A hell of a lot of J1 badges were in the place with the attendant "I have no idea why I'm here" folks, the "Hot damn, gimme a beer" folks, and the "Hmm, what's this all about" folks. Rather standard conference fair. What I think was some sort of javablogs get together was happening and it looked to me like adequate quantities of beer were and were going to be consumed by that bunch. Anyway, since I was there I figured I'd grab a beer and hang for a few minutes and see if any Jini types showed up. The beers there are named after bears and, after being informed there was no more Polar Bear beer, I ordered a Brown Bear beer. Having a beer surrounded by geeks. Nice. I stood out of the way and watched the ebb and flow of things.

About 15 minutes later a bunch of people cruise in and I happen to recognize someone as a Jini person. I go up and say hi, how are ya, nice to meetcha, and like that to these several folks. I was chatting with someone about a Jini job they landed, trying to decide if I killed him and then sent my resume in really fast would I be able to get his job. I decided against it in the end, largely because I didn't feel like putting my beer down just then. This guy will never know how close he came though (I was close too Sean :-). Just about the time I decided I'd rather drink than, well, kill someone up pops Bill Venners. After a little small talk I somehow get introduced these two other guys who seemed not to be particularly involved in the whole Jini thing. They weren't really, but I ended up spending the next two beers with with these guys.

One of them turns out to be Geir Magnusson who is some muckety muck with the Jakarta project. And, demonstrating concretely how bad I am with names, the other guy is working on Apache Geronimo. These were interesting fellows. I asked Geir why there was no Apachefied Jini (in jest, the outrageously stupid Jini license doesn't allow this) and was really surprised when he just kind of said "jeez I dunno. Go ahead.". Trying to explain the fact that Jini licenses wouldn't let that happen he started not so gently grilling me with questions I couldn't answer. Luckily for me I noticed Jim Hurley finishing up a conversation and, unfortunately for Jim, told Geir to talk to Jim. I didn't pay much attention to their discussion but it seemed, well, rather directed. Not harsh, just very to the point.

Well anyway, I spent some time talking to guy-whos-name-I-dont-remember and discover the company he founded a few years ago was recently funded. I had to chuckle to myself, ruefully, when he mentioned the company was profitable until it had taken VC money. Good for him he got his money and I hope it works out. I've been through this gotta-make-money-by-spending-the-VC-money, no profits until success thing before. It can be not pleasant. There were a few other things we all chatted about, extreme programming, why SMS wasn't as popular here as in Europe and some other things.

At the end of my evening I'd spent about 10 total minutes talking to Jini folks and the rest talking to these 2 guys. On the way home I was thinking about what had transpired and realized a couple of things. First, both of these guys said they found the idea of Jini appealing. This from even the guy building the Apache J2EE container. This is consistent with the fact that virtually everyone I speak with (ok, every geek I speak with. My wife and kids and almost all the managers I know excluded) sees some value in Jini. This can range from the, "yup, Jini is interesting" folks all the way up to the fanatics who think Jini can do anything.

Second, I left when some guys came over and started a conversation with Geir that went something like, "Hey, this is Joe (whatever). He wants to ask you a question". Joe follows on virtually immediately with something like, "What the fuck is going on with the JDO JSR?!?". He wasn't agitated with Geir (I think) but I gather Geir had some answers for him. Being more or less entirely not interested in JDO I said my goodbyes and started for home. What I thought was really interesting about this is how well it sort of defined the "us" of Java. So here's this guy who has a question about the direction of the Java platform. And he's talking to someone not from Sun. He's interested, and Geir's interested (or at least informed) and off they go. And it's even possible that something about the platform might change based on some conversation in a bar, between these entirely not Sun folks. I just thought that was cool.

That was the first night. If you get a chance you should try to talk to this Geir fellow. Nothing quite like talking to competent (or so I surmised) and pleasant geeks like this. Oh yeah, I got a Jini T-Shirt. I've coveted a Jini logoed something-to-wear for a while. It's the little things I guess.

So the next night I was again off to the city for the Java Communities in Action thing. This held much promise because a) you didn't have to be a J1 attendee to get in and b) Sun paid for the beer and food. There were sure to be so of the Jini crew there, and I'd done some small bit of work helping let people know about it. Really bad name though.

So I get to the Argent hotel, where this free beer fest is to take place, and immediately see John McClain and someone else whose name I forget working on getting some what I suppose were Jini services talking between their Apple laptops (John had an iBook, the other was a 12" Powerbook). Also, surprise, Jim Waldo was sitting on the end of a table kibitzing with these guys. After standing there for a minute I took off upstairs. That's where the beer was after all.

Jim Hurley was there, of course. We chatted for a few minutes and he sort of dragged poor Brian Murphy over. Brian seems just a nice guy. We talked for a while and I ended up with some new insights regarding Jini and Sun. I don't think I'm supposed to talk about the specifics. I do think now that some of Suns senior tech crew seriously has no idea what it is to deal with a live system. Not all of them of course. Right? We spent a couple of minutes talking about BS, and other Jini sorts of things. One thing we talked about was how many people in the world are or truly aspire to building/build distributed systems. It seems there is theory within the Jini group that this number might be rather small, in the 10s of thousands (note, I said theory, not belief. Big difference). If this is true (and given what I've seen of rather a few systems it might be), Jini is incredibly popular. This lead to the question as to whether issues of partial failure and the other hard distributed systems problems are things that all programmers should work on, or just some few specialists. I gather that there are at least a few places where the opinion is that it should be just the specialists. Brian didn't offer an answer or an opinion either way. For myself, I tend to think that everyone should understand this stuff. If people did actually think about all of the things that will (<= carefully chosen term) go wrong with a system that has a lifetime of decades and we started building substrates to address those things, I can't help but think that we'd end up building cooler things, having to spend less time fixing the messes we left for ourselves already.

I talked to a few other folks, the editor of OnJava and java.net for a while. He seemed roundly and well cynical. Not in a particularly bad way, but in that sort of "yup, me too" sort of way. Hard to describe really, but if you talked to him you'd know what I mean.

Oh, and Jim Waldo. Jim is always a treat for me. You know that saying, "if you're not a rebel at 21 you have no balls, and if you are at 30 you have no brains"? Jim always comes across to me as someone who is definitely older than 21 is a rebel, and basically can get away with "yah, yah, no brains. Fuck you". It's just a breath of fresh air to hear someone say, with some frequency, "that's wrong" or "that's stupid" and then rather convincingly outline why it is so. I pinged him about his thoughts on ease of use in Jini and was totally caught off guard when he said he thought it was important and then outlined some specific things that the Jini folks should do. Given my perception of his minimalist design tendencies I really didn't expect this. We sort of ended the conversation with him saying something like, "if they don't do it, I will". As I say, a treat.

There wasn't too much to note beyond that. Talked to John McM for a couple of minutes, watched Jennifer Kotzen (sp?) the Jini marketing person rattle off what so-and-so and such-and-who was doing with Jini so fast I thought someone had wired her brain or something, (had some beers), talked with someone from Valaran and Majitek. Both of these guys are doing what seems to be a SOA platform based on Jini. If you recall the app server days before J2EE (Kiva, Tenga, and the like) it reminded me a lot of that. I'm not sure that's a bad thing, but from a technology purchasing perspective it would make me nervous. As utterly shitty as EJBs happen to be (he said with due respect) there is some promise of source level portability across vendors there. I can't help but think that the Jini world would be well served crafting a series of service specs well beyond those in the world now. Logging, config, management and so forth. Sort of like what's happened with the ServiceUI stuff.

Oh. One interesting comment from Jim Hurley the night before (at the bar). When asking about why one would do an Apache licensed version of Jini, he asked, "but if the Sun versions were licensed that way, wouldn't you start with them?" or something like that. I took that to mean that they're working on freeing up the Jini license. No idea if this is exactly true, or what it might look like, but it sure would be welcome.

So that's my J1 without the badge experience. Much fun I thought. I do get so incredibly energized talking geek to geeks. It's still not worth the money to me though (well, maybe it is. It's not worth it to my wife for sure). I think Jini is as healthy as it's been for rather a great long while and it might be getting healthier.

Talk Back!

Have an opinion? Be the first to post a comment about this weblog entry.

RSS Feed

If you'd like to be notified whenever Rick Kitts adds a new entry to his weblog, subscribe to his RSS feed.

Digg | del.icio.us | Reddit

About the Blogger