Slashdot Mirror
Apple Terminates Safari Seed Program
coolmacdude writes "This morning Safari beta v67 was leaked to the Internet. Because this is the third time it has happened (v62 and v64 were leaked), Apple has apparantly had enough and decided to terminate the seed program that provided unreleased beta verisons to selected developers. In a email sent to all developers and posted on Mike Wendland's blog, Apple says:
'Due to Safari 67 postings to the internet, we have closed the Safari Seed project. We know that the majority of you are not responsible for the leaks to the internet, and we sincerely appreciate your feedback, time and effort with this project.'"
More testers = more bugs found = better product.
I know that Apple probably has good reason not to make the various beta releases of Safari available to the public, but I think they are missing out on a good marketing opportunity here. These Safari releases are keeping everyone interested, or they're keeping me interested anyway. Plus, people can see the new features as they are implemented and maybe once a favorite feature is added, emails about getting that feature will reduce.
Smeghead every day of the week.
I consider this to be a small ethical violation on the part of the individual who leaked the beta, at least compared to many other things. Nevertheless, Apple had placed their trust in a group of developers, and some jerk decided to violate that trust. However small of a violation that this may be, as compared to, say, murder, I would still like to see that individual publically flogged.
Surprisingly, not everyone follows the open source mantra. There are legitimate reasons for not wanting to provide constant releases. One is confusion among a less educated (some might also say intelligent) user base - people getting what is effectively a beta and don't know it end up bitching at Apple. This makes them look incompetent, and can cause problems for their image.
Apple has good reasons for wanting to keep their stuff under wraps until they ship. This doesn't make them wrong, unenlightened, or the enemy.
-Looking for a job as a materials chemist or multivariat
There is a big trade-off between getting a decent sized community to test a product and allowing a not-yet-debugged product out into the wild.
As a developer is it very valuable to have a willing group of people willing to test and feedback on not-yet-ready-for-market products. Unfortunately if these releases then get a wider distribution to people who don't understand that the app us a work in progress (as has happened with safari), any problems (which would be solved before an official release) reflect badly on both the product and the developer.
Given that the betas are being leaked, and Apple's reputation for quality of its products, I don't think they had any option but to cancel to program. I also welcome their move for other reasons:
As a web developer, one of the major issues I face is not just making a site compatible with the major browser releases (which in itself is a problem), but also with all the betas that are still being used. Many beta releases (or should have been betas) have quite significant bugs which are *very* difficult to work around. For example, I still see hits from people using betas of Netscape 4.
Once a pre-release product makes it into the wild, many of the initial users will continue it use it since 'it works for me'. Of course, if this browser doesn't work with a site due to bugs or incompatibilities in the browser, its the sites fault - from the users perspective - and my clients if the user complains. These almost-right products seem to persist almost forever.
Actually, they already do. By default, Safari has a toolbar button that sends a bug report, along with an optional screenshot/code snapshot, to Apple. In fact, this is one of the reasons why they choose to release the beta, so they could iron out all the bugs without having to the test all of the pages out there.
However, there is no need to get bug reports for a product that they know is unstable or incomplete (the post-v60 builds). If they posted one of those publicly, not only would they get a backlash for releasing an extreemly unstable build of their product, such as the first beta, which had a nice "feature" that would automatically delete ~/ for you, but all of their bug reports would be for a build which is still incomplete. Instead, they could just post their more complete, milestone builds, and get feedback which is much more beneficial to the developers.
No, someone must go through every bug report, and eliminate the ones that are for already known problems (with a public beta you potentially could end up with thousands of people thinking they are the first to find some problem that seems obscure). Then you have to eliminate the ones that aren't really problems (the copy and paste shortcuts are confusing by design since those who use them will use them often enough for the pain of memorizing strange key combinations is less than the pain of having to easy to remember short cuts that are harder to use on the keyboard). Next deal with the miscolanious problems (user didn't plug computer in, got a corupted download, has no net connection, and other problems that are either stupid user, or other stupid problem not related to the program).
Really what it needed is a few QA testers who can test everything, but that isn't possiable. Not even Apple with control of all supported platforms can do it. A public beta might seem like a hope that the gain is less than the costs. In reality a public beta is generally a way for marketing to get a almost working version out before it is ready for release, and the bug reports that might come in are worth much less than the hype.
I'm actually surprised that they didn't use steganography to uniquely identify each copy of the browser that they release to their individual, "select" developers. That way they could release the dogs of war on that poor soul.
:)
On another note, it's sad to see something like this ruined by what is probably a small number of bad seeds.
My blog
Seriously dude, the average person on the street isn't like us. They talk about the war, and reality TV, and 401ks and the like. They don't care when a developer's beta release program gets cancelled. For God's sake, I just read this whole message board and I could care less.
It may or may not conform to GNU's idea of 'free'... and we can argue back and forth about whether that's basically because GNU has always been determined to hate Apple, from day one, and will always be, no matter what Apple does.
But it doesn't matter, because he didn't say 'free'. Not as in beer, not as in speech, not as in political prisoners.
No, he said 'open source'. And, why lookie here... Apple's license is on the official list of 'open source licenses'.
http://www.opensource.org/licenses/
So stop with the trashing already. He said it was open source, it is open source.
-fred
Sign #11 of Slashdot overdose: You see the phrase 'moderate Republican' and you wonder if that would be a +1 or a -1.
Contrast the way Apple has "managed" the Safari development with the way the mozorg folks have done with Camino and Mozilla.
Apple releases a couple of "beta" releases, fires up interest and demand, and then nothing happens (from a public perspective) for a relatively long time. Given that it is beta software, there are a lot of things that need fixing -- the more people liked the initial rollout, the more demand there is for improved releases. But only frustration is available.
OTOH, look at the Mozilla camp. There are milestone builds on a frequency on months wherein an attempt is made to level-set at a certain level of stability, and nightly builds that are expected to be fraught with bugs, but steadily progress towards the next milestone build. This method serves the people who want stability and predictability above all else, the bleeding edge lunatics who want the newest thing out, bugs and all, and the developers, who benefit from having the largest group of testers that is practical.
How many people sent in bugs or suggestions for Safari? How many have seen even one of their personal hot buttons addressed? Virtually zilch, because Apple has been so stingy with new releases. OTOH, I personally have had several bugs looked at in Mozilla/Chimera(Camino), and feel a much stronger involvement with those products as a direct result of this.
I think Apple is missing the point about Open Source software -- it's not just that it's cheap, it also has closer ties to the user community, and as a result, probably better fits the needs of that community. You can take Open Source, develop in behind closed doors with an army of people, and still release it as an open source product -- but it's the dumb way to do it. It's how Microsoft would do Open Source.
A little knowledge of human nature and a smattering of statistics should tell you that this was almost guaranteed to happen, no matter what precautions (statutory or technological) were set against it.
Is Apple stupid for thinking this wouldn't happen, or did they plan on it?
It's rare that you're presented with a knob whose only two positions are Make History and Flee Your Glorious Destiny.
If KHTML would be distributed exclusively under the GPL license, Apple would have to provide any code for released software that links against KHTML.
KHTML must be using LGPL (or at least something similar).