This is a discussion on Some thoughts about the mail on osol-discuss and the reactions to that ... - Solaris Rss ; Well ... that was an interesting day yesterday. At first the news about the lawsuit against Google. The other thing was this document that is said to be a leaked document inside of Solaris Development in Oracle. I thought a ...
Well ... that was an interesting day yesterday. At first the news about the lawsuit against Google. The other thing was this document that is said to be a leaked document inside of Solaris Development in Oracle.
I thought a while before writing something about it. I sat a while on my fingers. But after reading some reactions i just wasn't able to writing nothing. This is my personal opinion, not the one of Oracle. I don't even confirm or deny the document. In the following article i just assume, it's authentic. Okay ... enough legalese.
Why do i write this blog entry? Because i think most people didn't really read the document or just rely their anger on hearsay or articles citing from this documents. When there is only one thing i expect from everyone: that's reading all the available informations. And i don't have the impression, that's was done by everyone felt called to comment on the situation.
Of course we are not innocent on that. As this document states:As you all know, the term “OpenSolaris” has been used colloquially to refer to any or all of a collection of source code, a development model, a web site, a logo, a binary release, a source license, a community, and many other related things.And so i can just assume a lot of things got mixed up.
What's in this document? In a nutshell: No more OpenSolaris 20xx.yy distribution, but a binary distribution with a developer-license as previews with support available from them. Source code will still be released by Oracle, but after the full release.
The situation is vastly different from the one where we were a few days ago, when Illumos was introduced. The upstream - the code from Oracle/Sun - will have an delay element from now.
However the upstream isn't cut off. Despite what some people wrote in several articles, Solaris won't be closed.We will not remove the CDDL from any files in Solaris to which it already applies, (...)Opensource stays opensource. For new stuff the document states:and new source code files that are created will follow the current policy regarding applying the CDDL (simply, that usr/src files will have the CDDL, and the very small minority of files in usr/closed might not have it)This has to be seen with the following statement in regard to the distributions based on the source code:Anyone who is consuming Solaris code using the CDDL, whether in pieces or as a part of the OpenSolaris source distribution or a derivative thereof, would therefore be able to consume any updates we release at that time, under the terms of the CDDL, LGPL, or whatever license applies.So, really don't see how some people can come to the conclusion that Oracle just closed Solaris?
I can only speculate to the reasons, that lead to the decision, but i have an opinion about that. For example: The Solaris-World of Transitive is nothing more than Opensolaris. The Compellent NAS offering is based on Nexenta. I don't play down the work Nexenta put into their products, but at the kernel they took essentially all the code Sun put into the opensource and made a product out of it. Is it okay: Hell yes ... that happens when you do open-source. Do i like it? Hell ... no! Because it feels weird, when the ore you raised is suddenly the sword people use to get your money. But you have to live with it, when you want the advantages of open-sourcing.
The interesting question is in this area: Have the advantages of opensourcing outvalued the disadvantages of competing against the own technology before you are able to offer a product on your own on the basis on the code? It looks like that Oracle saw a value in it (because of the announcement to publish the source with the release) but wanted to level the playing field (by introducing the delay).
The decisions expressed in the document still enables the use and the leverage of those component by the opensource community but ... and that's important ... in the same moment Oracle is able to do it. Of course that's different to the policy to push the code almost into realtime. Is it closing down? Well ... no. At least not in my opinion. For practical purposes it looks like a large putback at the time the release product is opensourced.
There is another thing in the last quote. I want to repeat the matching part:We will distribute updates to approved CDDL or other open source licensed code following full releases of our enterprise Solaris operating system.Did someone read that? This would mean that we will release the real source code of the current flagship distribution of Solaris. Many people forgot that we didn't opensourced Solaris 10, we opensourced Nevada ... the code to be Solaris 11.
Is Opensolaris dead? For the developer community ... well ... depends on your perspective. For users ? Due to this Solaris 11 Express program mentioned in the document? For all practical purposes: No. Just thinking about it for a moment. When did you really look into the source? And when you need it because you want to do some troubleshooting you have access to it, as the source will still be open. But i have already elaborated on that.
And:We will have a Solaris 11 binary distribution, called Solaris 11 Express, that will have a free developer RTU license, and an optional support plan. Solaris 11 Express will debut by the end of this calendar year, and we will issue updates to it, leading to the full release of Solaris 11 in 2011.Sorry ... previews with support available ... that sounds pretty much like Opensolaris to me. I want to postulate at this point, that based on this quote execept for the longer wait there is no change for the normal user in regard of end of the OpenSolaris 200x.xx binary distributions. Hey ... the document states that there is even support for it. The document even states that there will be a way to get from OpenSolaris to the next longterm-supported version of Solaris.We will determine a simple, cost-effective means of getting enterprise users of prior OpenSolaris binary releases to migrate to S11 Express.So ... where is really the difference for the user? Especially the commercial user of Solaris?
That said there is just one thing that i'm really pissed about. Some mayor proponents of the Solaris community are running amok at the moment. People like Jörg Schilling and Garrett d'Amore try to play the card "Does oracle have enough developers to keep the development running?". Hell, yes ... i know that we lost some of high-profile developers. But ... damned ... there are a lot of other developers as well at Oracle. Those people aren't idiots and last time i've looked at it, engineering (and software development is a kind of engineering) was team work. I'm sad of everyone leaving Oracle ... but that isn't the end of the word. And speculating that Oracle doesn't have enough developers to drive development just because some high profile developers have gone ... well ... that's just nonsense.
However, i see a reason for such at Garretts side. Keep in mind, he is the engineering lead at Nexenta Systems, the company selling commercial variants of Nexenta. The world just got harder for his employer. The upstream with all the fixes has now a delay.
However, based on the document Nexenta can still get into an agreement with Oracle to get real-time access to the codebase. As the document states:We will have a technology partner program to permit our industry partners full access to the in-development Solaris source code through the Oracle Technology Network (OTN).andWe will encourage and listen to any and all license requests for Solaris technology, either in part or in whole.But there will be some interesting questions in the future. For example: Where is the business case for Nexenta Systems? They can't leverage on the development done by Oracle at the same moment Oracle delivers the code. Will they innovate on their own? Do they have enough developers on their own to drive the development of Illumos? How much do they rely on the community to deliver changes to Illumos? How to get bugfixing and innovating under one hat with a limited number of people?
At another location Garrett bragged about the point, they they will do so much putbacks now. And other companies talked the same way. That's fine. I welcome that. But where were this putbacks when Sun was an independent company and Jonathan was CEO? Where were this putbacks when Oracle started to get the helm. Such putback would have been a great case to show Oracle the value of the community. Why just now? I just assume we would have a different discussion now if all those putbacks would have found their way into Solaris before over the years. That's the main point that hurts somehow inside me.
My heart isn't with Sun, my heart isn't with Oracle. My heart is with Solaris and everything needed to ensure that Solaris keeps to be the best and most advanced operating system to do real work is okay for me. When the heads of the development think they need to go certain steps to ensure that, i'm okay with it. For me it's that easy.
Did Oracle lie to the community? Well ... don't think so ... based on the document the code will be still open, there will be still a binary distribution according to the document ... what do you want more? But i assume in their anger most people will see it otherwise. Any now you are free to slaughter me ...
Read More about [Some thoughts about the mail on osol-discuss and the reactions to that ......