Well, after my last blog entry regarding Sun and JSR 277, I received a visit from the enforcers of that fine establishment. Wearing heavy overcoats on a nice sunny day, the elevator door opened and they came rushing out. Like a scene from the Matrix, it seemingly took them forever to pull out their impossibly huge automatic weapons they had hidden under the black leather trench coats. In only several hundred milliseconds, Sun’s professional enforcers unloaded 20 or 30 pounds of lead in the form of hundreds upon hundreds of rounds of high powered ammunition into what used to be my body, splattering my physical essence across the interior of the reception area of building 200 on the Geidi Prime campus.
Or not. Rather, what did happen was Alex and Stanley showed up at my office escorted by a fellow member of House Harkonnen and after a slightly awkward moment with Stanley (I had, after all, just blasted across the internet – with the awesome power of my blog – stating that he was an idiot), Stanley and my compatriot excused themselves to work on JSR 277 related specification business leaving Alex Buckley and I in my office, as I watched him finger a menacing looking lead pipe that he pulled from his brief case.
Or not. Rather, we sat down and talked comfortably for quite some time about about everything but the issues surrounding JSR 277 and the reason why Alex even knew of my existence. Eventually, however, we directed the conversation towards the actual issues we really needed to discuss and left the safe domain of my Dr. Who toy collection and the pleasantries of traveling through Heathrow in the U.K.
Now, Alex Buckley is obviously one of the top Spicetm enhanced brains working at Sun. Very easy to talk to and generally a no nonsense individual who clearly wants things to work out for the best of both for Sun and the OSGi community. I won’t actually go into any of the discussions about technical particulars that we discussed about such subjects as versioning formats, meta data information locations and such. These things are important, but there’s a point that is, I think, far more important than any of the technical details which will have to be worked out and that is how this process is actually working out.
We as technical people invariably have this rather blinkered view of reality in that we see the “real” problems as being the “hard” issues of the technology in question. As I mentioned, the issues of how many indexes and what format the version strings will have in the module metadata; whether we should reuse the META-DATA/MANIFEST.MF entries like OSGi does, and whether that is even sufficient for the super complex, pan dimensional needs that the Spicetm enhanced brains from Sun foresee as requirements due to their ability to look far beyond the present and into the bright future where we’ll all need shades just to keep our eyes from being seared because it’s such a hot technology. No, these issues are all things that I do believe that having the Third Stage Guild navigators from Sun locked in a room with the mere mortals of the rest of the industry could figure out in a reasonable amount of time.
The problem is, of course, that there really isn’t any functioning expert group, nor is there any serious interaction between Sun, the JSR 277 expert group and the OSGi organization.
Now, while it’s somewhat flattering to have Alex show up in my office and dedicate a non trivial amount of his time discussing the technical issues with me, the problem is that I’m not the OSGi organization. True, I’m a member of the Core Platform Expert Group and the Enterprise Expert Group. But in reality, it’s just me and this wasn’t even a working session where we were expected to produce anything other than a polite blog entry such as this explaining how wonderful things really are when you look past all the politics with JSR 277.
Indeed, the problem – from my humble and obviously biased POV – is that Alex is having a substantial number of bilateral talks with various interested parties of this rather serious issue. That is, he’s talking directly with Peter Kriens, Rod Johnson, Richard Hall, Glyn Normington, Bryan Atsatt and god only knows who from Google and the zillion other people from companies who represent various parts of this grand collective we affectionately know as the Spice Industrytm.
And there in lies the nub of the problem. As has been said by many before me, “He who controls the Spicetm, controls the universe”. And what is quite clear from the conversation with Alex is that Sun believes it controls the Spicetm. And it’s quite hard to argue with this fact. One can look at the dead bodies strewn along the path of any number of previous JSRs and see how Sun dealt with technologies that predated the JSR in question, inspired the JSR in question and were gutted and left to die in the hot sun by the JSR in question. Clearly, one can look at the writing on the wall and see much the same in the near future of OSGi. Once something is in the JRE, any technology competing with it is toast. End of story.
Consequently, Sun does not want to even give the impression that they are “negotiating” with OSGi, nor working with them in any form other than talking with the various members of this organization. Whatever one may want to do as a spec lead, it’s really the higher ups in the hierarchy of Sun that ultimately define the boundaries of interactions and it’s rather clear that those boundaries do not include actual interaction with OSGi qua OSGi. Thus, we’re left with piece meal, bilateral negotiations between individuals which may work to improve things around the edges, but won’t even come close to dealing with the weighty issues that are at the heart of the matter.
And so, while I think one has reason to be optimistic about things from a technical perspective, I think that optimism has to be seen in the perspective of the political realities. The political realities of the situation are such that Sun will not give one iota and cross the line to deal with OSGi. They simply can’t do this politically. Rather, we will continue to hear whining about how OSGi will not cross the line in the form of Peter Kriens and BJ Hargrave and fully work with the 277 expert group – an expert group which, from what anyone can tell isn’t really a functioning expert group any more, given how it’s gone completely dark and all the actual work is happening in the Java Modules project without much input from anyone else.
Consequently, however little light there may be between the technical solutions that may possibly be out there, I think the simple fact that Sun will continue to dig in their heels and hold onto control over the Spicetm, which will essentially doom the entire project. No, I’m not talking about dooming JSR 277. It’s quite obvious that Sun, controlling the JRE as they do, will be able to push through whatever they want despite any grumbling from the peanut gallery and the awesome push back all three readers of this blog will be able to generate. Rather, what will be doomed is OSGi as we know it because Sun will basically add OSGi to the long list of dead bodies of other projects that JSRs have killed in the past. It won’t be done purposefully and won’t be done because of the ego of individuals as has been suggested was the case previously. Rather it will be done because of the egos of organizations who are unwilling or unable to come to the table and hammer these admittedly highly tractable technical issues out as a team – as peers.
Granted, I hope I’m wrong and lord knows I’m quite the pessimist on such matters. But one doesn’t have to put too pessimistic of an interpretation of the past wrt JSRs and the fate of external technologies that inspired them to see the progression into the future. Whatever technical issues there may be are indeed solvable and perhaps trivially so.
But the politics – ah, the politics. The politics will provide the mechanism to compensate for what individual egos perhaps accomplished in the past to screw things over. The very fact that the JSR 277 expert group is essentially non functioning should be a warning sign that shouldn’t be ignored. The very fact that Alex is involved in complicated, tedious and no doubt highly time consuming multiplicity of bilateral negotiations should also be a warning sign that shouldn’t be ignored.
Often, even with only the good intentions of everyone involved, things still go very badly. Often, despite everyone wanting things to work out, the group interactions prevent these good intentions from being realized. Perhaps JSR 277 will be the exception to the rule and we’ll all get not only all the Poniestm we want but we’ll each be given our own private Leprechaun with a personalized pot of gold to boot.
But count me extremely skeptical.
I’m sure that OSGi will be able to adapt and deal with whatever falls out. And perhaps we’ll all be better off after we pass through the fires to come. Who knows. Stranger things have actually happened. But it’s not looking like it’s going to be a fun couple of years which is the way it should – in anyone’s approximation of a realistic world – have worked out.
the sleeper must awaken