It is to laugh.
In reading about the OpenSSH issues with Debian, one of the trails of pointers led me to a comic online, xkcd. A few comics later was this:
One coffe powered spit take later...
(and fair warning Microsoft Ninjas, Eric has guns!)
At the recent Internet World conference in London, Nitot of Mozilla made several remarks that included Adobe as an threat to the open nature of the internet.
Needless to say to those who watch the space, some of Nitot’s comments don’t reflect reality. I’d like to take a moment to clarify issues about Adobe Flash Player Adobe support for Linux, and Adobe open efforts.
Adobe delivered Flash Player 7 for Linux back in 2004. Flash Player 9 for Linux is now available and in 2007 Adobe committed to release future versions of Flash Player Linux simultaneously with Windows and Macintosh. I have Flash Player on my Macintosh OSX 10.5, Ubuntu 8.04, and Windows XP. Go here to see the version you have for yourself (http://kb.adobe.com/selfservice/viewContent.do?externalId=tn_15507)
My version shows 9.0.124.0 on all of them. While I didn't test it, I'm told you can also add Solaris to that list as well.
It’s true we aren’t able to support every single Linux distribution and we know that Linux users want Flash Player for 64b Linux. But Adobe works hard to support the most popular distributions for Flash Player (and Adobe Reader) just as we do with the other operating systems.
Adobe may not deliver all of our software as open source but we firmly believe in the power of the open web. Flash Player isn’t open source. But Adobe, with a group of industry partners just announced the Open Screen Project. As our part of the project, Adobe removed all restrictions on using the file format that drives the Flash technologies, SWF and FLV/F4V. With the release of AMF in February along with BlazeDS, and an upcoming release of Flash Cast, Adobe is equally committed to making sure the web, at least our part, stays open.
Other open activities from Adobe include contributing the ActionScript Virtual Machine as open source to Mozilla’s Tamarin project. This is the same open source AVM in Flash Player 9 and Adobe AIR. Adobe is actively engaged in the Tamarin project. Providing Flex -- a free open source framework for building RIAs, which includes the source to the ActionScript components from the Flex SDK, the Java source code for the ActionScript and MXML compilers, and the ActionScript debugger from the SDK. Additionally, other major portions of Adobe AIR, such as Webkit (the HTML engine) and SQLite for the local database functionality are open source today.
The power of the web is found in its global reach, open exchange and access to all. Adobe recognizes that the extended web, reaching from devices to desktops needs to be equally open and as such is removing barriers that will enable content creation, applications and access to spread widely.
An interesting question came up in some after-hours discussions the other day.
Is it ever a good idea to open source a program or technology from within a company if the company is no longer enhancing it? Basically, is dead code ever a good thing for open source?
My immediate reaction to this was no. Then yes. And finally, a definite maybe. (Inspired no doubt by a glass or to of my new fave-rave Cabernet, Parallel.)
So, it seems wise to appeal to the wisdom of the internet. What do you think? (If you want to go comment before reading farther, feel free, I'll wait).
First no. It seems that if a company is no longer interested in a product, then the act of open source is a Monty Python attempt to run away. Companies that open source code, particularly complex code have a obligation to aid and assist the community in working with that code. One of my rules for considering open source (in fact, the principal rule) is "Who cares?". That question goes both ways. Who cares about the code inside the company as well as outside the company. What shape is the code in? Who takes ownership? It seems like the problems are big, bigger than any advantages.
But then, well, yes. If a company isn't interested in supporting or extending a product, why shouldn't the open source community (mythical though it may be) have a shot at it. Aren't there customers that are using it? What happens to them? Can't the mere act of opening good code balance against the lack of corporate support?
Finally (in vino veritas, you know), a definite maybe, surrounded by ifs.
If someone cares that the program continues, and is will to commit their resources, why shouldn't the code be released? If the code is good enough, and doesn't include other properties, why not release it? If a community is available to control and govern it, why not? If the product was good enough to build a market on its own right, why should we believe it can't continue?
So, to both of my readers <grin>, what do you think? Should dead code be released? Are there examples of it working, or failing? And what do you think are the steps that would help us decide?
I just registered for the 2008 Open Source Think Tank. It is an invitation-only event, limited to 120 people who want to discuss in depth issues around open source and commerce. In short, it fits my job, and my mind set.
I've been attending open source events since before it was called open source. I "flew" penguins in Raleigh Durham, had early discussions with maddog Hall, and even had my smirk photographed down under. I've got a Golden Penguin on my shelf... and would love to take on the chance for another. But that said, most open source conferences are just not worth (me) going to.
The OSTT, at least last year, was different. Interactive. small groups discussing in depth concepts and beliefs around commercial activities and open source. No holds barred opinions, and yet consensus. Meeting some of the major players and shakers in the industry. Meeting newcomers that both have a clue and give a damn. And darn little sitting around listening to drivel, and no marketing... well, not much.
During the event last year, we had a presentation on the-then GPLv3, rev (something). Unfortunately for the presenter, I read ahead on a slide and started expressing my opinion on certain "features". Now, I am not a lawyer (though there were a "lot" of them there last year) but I managed to get a point across that may have made a difference, and with the current flavor of GPLv3, Adobe has actually considered it, and has adopted the LGPLv3 for one of our efforts. (But I am surprised they are letting me come back <grin>).
There are certainly folks who last year pointed out how elitist this event is. I heard from folks last year that it should be open to anyone who wanted to attend, free (as in beer). And you know, those people are probably the ones that I run into at every other show, telling me how Adobe is doomed, that Adobe should just open source everything. I talked about that in a different blog, and lets leave it at that.
I want discussion and reasoned opinion. And if you want to offer me that, feel free to let me know. I'm looking forward to the OSTT 2008, and expecting to come away with a new appreciation of open source and ways to make it work with and for Adobe.
The open source head line from Adobe today is:
Adobe Announces Open Source Technologies for Enterprise RIAs.
BlazeDS Enables Developers to Productively Build Real-Time Data-Driven Applications
Now there's a lot of technical stuff. Important stuff. Things that will make the web a better place.
But I'm not the technical guru. For that I'd invite you to check out the BlazeDS stuff on Adobe Labs.
No, I'm more interested in responding to a question that comes up all the time. In short, "Why doesn't Adobe just open source everything right now?"
Sorry, not going to happen. Even with the advantages to an open source model, no large established company can merely flip the switch. And sometimes the switch doesn't want to be flipped.
Our focus for open source is targeted to developers. Developers understand open source. Coders are capable of dealing with vagarities in code. And developers provide the benefits of open source code.
Even then, the number of people who ever write a line of code is relatively small. Most people just want to something that works, first time every time. Designers in multimedia are similar. The cost of open fragmentation is directly and exponentially proportional to the testing and customer costs. If your web browser can't correctly display a web site, do you tend to blame the browser or the site author? Survey says?
Adobe will continue to work with open source, but it will be a controlled process. We've been pretty aggressive over the last year. We'll open up things as they make sense. This isn't a "cliff launch"; we're trying to get a pretty large entity airborne here. And a long runway is required.
So, keep talking to us and with us. Just don't expect elephants to act like hummingbirds
So a simple question?
To you, what's the most important open source conference around? And why?
Is it OSCON, GOSCON, Linuxworld, OSBC, OLS? or does it no longer matter (shades of Uniforum, and the old X11 conference in the 80's)?
Again, Jay of The Mows has succinctly expressed an issue in the open world of source and standards.
And there are some real issues here.
I was reading Consortiuminfo.org (as I always do when it updates). Andy Updegrove, who kept many of us informed on the machinations with OOXML, posted "Words, Standards and Torture: What's in a Name", which I highly recommend you read. No, it's not about how standards meetings are torture, but it is about how words are the tools we use to define standards. And sometimes those words mean different things to us all. Or get redefined. Or as best put by Indigo Montoya in "The Princess Bride"; " "You keep using that word, I do not think it means what you think it means." Too often we honestly or deliberately choose to redefine meanings of words, like open or free.
Now, I have my opinions, and so do thousands (millions?) of others. We hear high visibility types like Richard Stallman talk about open versus free. We hear Jim Zemlin discuss the multidimensional intertwinings of open and free. And you know what? It only means what you think it means.
Words do have power. They have the power you give them (or society gives them, in the form of laws or cultural pressures). I favor open source, and it should be my freedom to choose such. Yes, open source doesn't mean necessarily free (as in beer) nor necessarily free (as in speech), but I do expect to be free to make up my own mind.
To me, free is the ability to implement what I want when I want. (and yes, anarchy is the only form of government that works the way it was designed... extra credit to the first person identifying the source of that paraphrase). To me, my freedom should not impinge on yours, but yours shouldn't impinge on mine. I don't like people saying " Ooo, your's is cool, give it to me". I do like saying "Here, try this".
Open standards should be about the freedom to implement. It shouldn't be about me giving you my implementation for free. While that can lead us into the messy area of patents, copyrights, the new "hate phrase" of intellectual property, I should be able to build to a truly open standard without impact.
There should be no winners or losers in the openness game. Not knowing the meaning isn't a path to winning.
Asking questions until we reach a mutual, freely derived agreement is winning for both sides.
I'm quite sure Jay didn't realize he was drawing a political cartoon with The Mows but it's either that, or my mind is slightly skewed into reading between the lines. However, Jay again has managed to capture a unique view (yes, pun intended) on my worlds of open source and standards.
You often hear the phrase "Perception is Reality" used in marketing. A well-turned phrase, it indicates that what someone perceives about your product or service is (their) reality. And often, that reality is contagious, spreading rapidly to make others perceive a similar reality.
Companies often use this concept to distort reality, adding to their own ego-dollar (and real dollar) pile. You know, the ones that say "My car makes you more virile", My clothes make you look thinner", "My products make you better". We hear see and probably even taste this every day.
In open source implementation, self-delusion is not a good idea. Perception is perception and reality is reality. Yet time and again, we see a similar play in marketing of open source. I always like the one: "Microsoft is a bad company, so buy me instead". (Not to say Microsoft isn't a bad company. .., but, just sayin').
Too often open source inside of large companies is "Perception is delusion". I'm seen (and worked for) companies that see themselves as good citizens, while chaos reigns around them. A product is not good just because it's open source; a product is equally not bad just because it isn't open source. Experience the reality, if the product solves your problem, then it is good. And you should be the ultimate arbiter of what solves your problem.
A few companies back, it was decreed that the company would use Exchange and Outlook, because it was the "only calendar system that works". Note the "Perception" here. That perception was created by a failed attempt at other calendaring systems in 1997... yes, 10 years ago. Anybody got a different view? (I certainly do.)
A recent consultation started with the concept that "opening our product would reduce costs". Wrong again. Open sourcing a closed product will always increase costs, in spite of the potential improved QA.
I do admit, when I first saw this comic (and yeah, I visit every day, it's my perceptional delusion), I immediately thought of the recent spin on OOXML and the failed ISO ballot. I leave it to the observer to do the logical proof, but let me point out that a press release is not (or should not be) a reality distortion generator, despite the existence proof of Steve Jobs at Apple. As for the press release, do the math.
While here at Adobe, we have delivered the PDF specification to ISO for standardization, we being very careful to avoid self deception. Note that we haven't managed to buy countries, stack ballots, or anything like that. We are offering to have a conversation (remember that thing about openness, it's a conversation), check out the Inside PDF blog to find out more.
Perception is perception, reality is reality. In open source or open standards, the reality matters. Be it in Posh and Beckham (as Matt Asay points out) or in "The Mows", be sure you are not the one dis-floor-einted
comments welcome
There are frillions of ways to get past obstacles in embracing open source inside of companies. Unfortunately, as the comic shows, most often, the choices lead us back to where we were in the first place.
(And Thanks to Jay Dyke for letting me use this image! Go visit him and Indy and Tigey now! )
It still amazes me how many people on both sides of the open source fence end up back in the same place. Within companies, "We can't use open source because our needs are too hard" or "We can't open source that because we'll lose our customers". means business as usual. Within open source ranks: "Use
In short, frillions of ways, but only one path.
Open source is a lot more flexible than that, though. It's not one size, or one way. It's bespoke tailoring, in spite of the noise. And fortunately, we can choose our own paths.
As I've stated in an earlier blog, I think that freedom can mean multiple things. At Adobe, we don't necessarily follow the one true open source way. While open source is a great way to develop code, there is not just one way to open source. We tend to be a community ourselves, with our own cliches, personalities and opinions. We often try to make our own way. We look at open source as to what's the drive and reason. And we pick our approaches to open source with a unique review each time.
In the platform and developer unit, we're believers in open source, but even then, we're approaching it cautiously. We need to learn how the interfaces at the boundaries will work, and whether it's in everyones best interests.
You'll note that we use a number of licenses, from MPL to BSD, and even some hanger-on of the Adobe source license. Again, we'll pick the right tool for the right job, not assume that every item requires a hammer. (yes, bad humor there, please don't hit me
Anyway, keep in mind that any company, and every individual should consider that there are frillions of ways around the blockage. Find your own way, and respect the way other pick.
|
|---|
Just this week, a major commercially oriented open source product moved to GPL v3. SugarCRM announced that the Sugar Community Edition 5.0 will be available under v3.
Congrats to SugarCRM. It's the right move, the right time, and I can attest that decisions like this are difficult, time-consuming, and actually, not a lot of fun.
Licensing is the bane of existence for most commercial companies. Open source licensing doubly so. Licensing is still a black art, and the circumstances are the driving factor, e.g. what is the desired outcome?
I'm not a fan of vanity licensing. Here at Adobe, we're working to move everything to a recognized open source license, as it makes sense.
However, I'm a bit bugged by the number of folks on the web that are claiming that the GPL v3 is the solution to every problem. In many ways the GPL v3 is an improvement over the former incarnation of GPL. It clarifies a lot of formerly ambiguous material, and so far, makes lawyers a bit more comfortable. But it is not the only right answer, nor is it always the best answer.
For Sugar, and their existing business model, GPL v3 helps protect their unique value propositions while keeping the advantages of an open source development. The team at Sugar understands the diligence and governance issues that GPL carries with it, and are well positioned to walk the tightrope GPL can bring. GPL may actually be the best choice for a hosted and services oriented model, but not necessarily the best for client products with nested capabilities.
Interestingly enough, for a large corporation with lots of software (and intermixed licensed components, etc) GPL can be a nightmare. In many regards an even more open license works for certain things, e.g. BSD.
The difference may come down to your definition of open and free.
I applaud FSF telling me I can be free. I find it more philosophically challenging to be told how to be free. I want doors that open, not doors that are free.
So, explore an open door today.
And let Sugar know that you approve.
Join us at MAX 2007
In the aftermath of the energetic and fascinating Adobe Live UK event, I wanted to share some comments and thoughts.
The event, in the evening after an amazing day of presentations, demos, hands on training, was still brimming with folks ready to dive into the world of open source. And as it was, a great line up of panelists were ready to fulfill them their wants.
No panel is any fun without controversy, and I'll admit being that focus here. After all, as stated earlier, open source is not just around altruistic beliefs. And yes, there is a number of projects that may be viewed as altruistic, but they still end up making money the old-fashion way, by providing value that someone will pay for. It doesn't mean that someone buys the project, it can mean that a person gets hired because of his work and exposure on a project.
Anyway, you can watch it for yourself: http://www.adobelive.co.uk/livefeed.asp?feed=3
Next week, I get to talk about open stuff at the Developer Live event in the UK. It looks like a fun event.
My topic (on an evening panel hopefully fueled by a good crowd and maybe a pint of something dark) is about development in the future and the impact of open source. Well, honestly, they word it a bit better:
The Future Is Wide Open
Open source development tools/platforms/environments will increasingly shape future generations of rich applications on the desktop and on the web. Will any single enviroment dominate the next generation of applications? How can proprietary tools survive in an increasingly open world? Will there be a clear winner? Which technologies will win is still to be decided, but one thing is clear, the future is wide open.
Now as you might expect, I've got a few opinions on that there topic. But I'm going to withhold my thoughts until after the event. Let's just hint that the words hybrid, commodity and community will probably cross my lips.
I'm expecting a lively panel though. So far, they haven't let me know who is on the panel with me, but somehow, I suspect I'll get challenged and pushed.
Anyway, if you're around, stop by and give me a shout. Or shout at me earlier.
“When programmers on the Internet can read, redistribute and modify the source for a piece of software, it evolves. People improve it, people adapt it, people can fix the bugs. And this can happen at a speed that, if one is used to the slow pace of conventional software development, seems astonishing.” - The Open Source Initiative
The above shows why people care about open source, from a technical viewpoint. What’s missing are the companion thoughts that should go into the value of open source.
First, let’s review:
These are pretty darn clear. (And debatable, but that gets us into licensing hell).
So what are the rules for using open source from a corporate view?
They are even easier.
Give credit where credit is due.
If you use someone else’s program, work, code fragments, or anything of the ilk, give them credit. This actually should go beyond open source; it’s a good rule for life. For example, when I was back at Cassatt Corporation, we built a complex automation control system, which makes use of other programs, and thus we created a “thanks to” page. (Don’t ask why it’s in the legal section). If you notice, it’s lengthy, and yet clearly recognizes that using open source for development, for delivery and so forth, reduces Cassatt’s industry-leading datacenter automation tools development complexity.
Return value equivalent to what is received.
Note, this doesn’t say, “Open source your products” or “All your base are belong to us“. This means if you get something, give something back. It can be in recognition, in other projects, in hosting (live or web), or opening other code. At SGI when we decided to adopt Linux, we also decided to identify and release code that would add value, such as XFS. We didn’t open IRIX (the operating system) since it was encumbered with OPS (Other Peoples Stuff), like trademarks, copyrights, and licenses.
—
Today, many, many companies make use of quality code that they received from what was, in end view, an open source effort. Let’s hope that somewhere, the non-zero-sum game that makes up open source usage is being played both competitively and cooperatively.
As always, comments welcome.