This is indeed a tragedy, but frankly it does not surprise me.
I've been lurking around the mozilla mailing lists (mirrored in the mozilla newsgroups) and the mozilla.org website basically since they were created. Watching the dynamic of the mozilla communication mechanism over the period of a year, a number of things became unsettling. (I've only ever lurked, primarily because I'm not a sophisticated programmer, and I would have little to offer either the mozilla or linux kernel mailing lists. The development processes in both fascinate me, however.)
First, it amazes me that anyone in the Mozilla project was able to communicate with another at all. From the moment the mozilla mailing lists were created and mirrored, it was apparent that 80-90% of the mail/posts were, and would -always- be, irrelevant fluff. The primary reason for this is that while the Netscape 4.5 support newsgroups were not public - available only through nsnews.netscape.com, the mozilla newsgroups were public, and contained those compelling words "netscape", "misc" and "mail-news". From the perspective of a user with little knowledge of the significance of the word "mozilla", there was no reason to think that the most obvious place to ask Communicator questions.
The result was/is that despite the good efforts of Dan Mosedale and Jim Cape (each of whom made valiant strides to keep the mailing list topical), the vast majority of discussion was/is about 3.0, 4.0x and 4.5 problems. Combined with another 10% of posts of the "I want my 5.0 and I want it NOW!" variety, and a further 5% of the "Now that I can order you, I demand the following 50 stupid features that I have no idea how to program myself" variety, the mozilla mailing list, to the best of my observation, became a completely inhospitable place to have useful techinical discussions such as are seen almost exclusively - by contrast - in the Linux kernel mailing list. The latter, despite the fact that its content is usually way beyond this law student, is a pleasure to read. The Mozilla list is not.
I will leave remarks about the daunting complexity of the source as a major factor to jwz and other programmers/contributors. I'm simply not qualified. But another result is that because most of the contributors (as pointed out by jwz) were still Netscape employees, communication via the mailing list for the purposes of solving localized problems was (I assume) unnecessary.
I must, therefore, put a caveat on jwz's "fishbowl" analogy. These are two huge disincentives to communicating publicly about the source tree, and the lack of consistent communication on the nitty-gritty details of development may have played its part in the failure of the mozilla project to capture the imagination - and effort - of the programming community.
Aside from that caveat, though, jwz commented that such outside observation, combined with mozilla's independence, motivated the project to redesign the layout engine, and thus the UI, from scratch. It is unfortunate that the choice to rebuild a project thoughtfully and correctly, at the obvious cost of time, is considered a 'failure', or even a bad thing. Yes, some idiots have complained and threatened that if "Netscape doesn't come up with my browser now, I swear I'll move to IE5", not understanding that (a) the development regime has changed radically, (b) the project was rebuilt, (c) it's better to ship the right thing "late" - inasmuch as there is any such thing as 'late' in an open source project - than the wrong thing when users demand it.
For someone who has lurked and gained some familiarity with the dynamics of the project and the cast of characters, none of these conditions indicate 'failure' to me. It is unfortunate that jwz does, but he'd probably know better than I.
You've misunderstood the court's ruling. From the article:
The ruling stems from a trademark infringement suit filed against a Canadian corporation, 3263851 Canada, Inc., by Umbro International, a manufacturer of soccer equipment. In the suit, Umbro claimed the defendant infringed upon its trademark when it registered umbro.com in 1997. When the defendant failed to show up in court, the judge issued a default judgment, turning over umbro.com to the company and awarding it $25,000 in attorney's fees.
The court held that despite the fact that 3263851 Canada Inc. had paid for the domain, the name Umbro was trademarked, and therefore the domian name umbro.com was an extension of that trademark.
The problems with this ruling are daunting. As I understand it (insert disclaimer), the same name can be trademarked concurrently, as long as the name refers to different contexts. i.e. there can be simultaneous Umbro trademarks owned by Umbro the sports equipment manufacturer, and Umbro the (hypothetical) dance company. Think about Windows(R) for a moment to see why this must be so.
So what happens where there are two or more legitimate trademarks of the same name (e.g. Windows) and there is only one whatevername.com? Obvously the idea that a domain name is a natural extension of a trademark cannot possibly be logically consistent.
I'd need to read the actual decision to see what the judge's reasoning is though. Further, all this is subject to an appeal.
Slashdot Mirror
Again, in case anyone is interested, there are two excellent critiques of Schildt here. First seebs':
http://www.plethora.net/~seebs/c/c_tcr.html
and (on Schildt's Annotated ANSI C Standard - even worse):
http://www.lysator.liu.se/c/schildt.html
I.
This is indeed a tragedy, but frankly it does not surprise me.
I've been lurking around the mozilla mailing lists (mirrored in the mozilla newsgroups) and the mozilla.org website basically since they were created. Watching the dynamic of the mozilla communication mechanism over the period of a year, a number of things became unsettling. (I've only ever lurked, primarily because I'm not a sophisticated programmer, and I would have little to offer either the mozilla or linux kernel mailing lists. The development processes in both fascinate me, however.)
First, it amazes me that anyone in the Mozilla project was able to communicate with another at all. From the moment the mozilla mailing lists were created and mirrored, it was apparent that 80-90% of the mail/posts were, and would -always- be, irrelevant fluff. The primary reason for this is that while the Netscape 4.5 support newsgroups were not public - available only through nsnews.netscape.com, the mozilla newsgroups were public, and contained those compelling words "netscape", "misc" and "mail-news". From the perspective of a user with little knowledge of the significance of the word "mozilla", there was no reason to think that the most obvious place to ask Communicator questions.
The result was/is that despite the good efforts of Dan Mosedale and Jim Cape (each of whom made valiant strides to keep the mailing list topical), the vast majority of discussion was/is about 3.0, 4.0x and 4.5 problems. Combined with another 10% of posts of the "I want my 5.0 and I want it NOW!" variety, and a further 5% of the "Now that I can order you, I demand the following 50 stupid features that I have no idea how to program myself" variety, the mozilla mailing list, to the best of my observation, became a completely inhospitable place to have useful techinical discussions such as are seen almost exclusively - by contrast - in the Linux kernel mailing list. The latter, despite the fact that its content is usually way beyond this law student, is a pleasure to read. The Mozilla list is not.
I will leave remarks about the daunting complexity of the source as a major factor to jwz and other programmers/contributors. I'm simply not qualified. But another result is that because most of the contributors (as pointed out by jwz) were still Netscape employees, communication via the mailing list for the purposes of solving localized problems was (I assume) unnecessary.
I must, therefore, put a caveat on jwz's "fishbowl" analogy. These are two huge disincentives to communicating publicly about the source tree, and the lack of consistent communication on the nitty-gritty details of development may have played its part in the failure of the mozilla project to capture the imagination - and effort - of the programming community.
Aside from that caveat, though, jwz commented that such outside observation, combined with mozilla's independence, motivated the project to redesign the layout engine, and thus the UI, from scratch. It is unfortunate that the choice to rebuild a project thoughtfully and correctly, at the obvious cost of time, is considered a 'failure', or even a bad thing. Yes, some idiots have complained and threatened that if "Netscape doesn't come up with my browser now, I swear I'll move to IE5", not understanding that (a) the development regime has changed radically, (b) the project was rebuilt, (c) it's better to ship the right thing "late" - inasmuch as there is any such thing as 'late' in an open source project - than the wrong thing when users demand it.
For someone who has lurked and gained some familiarity with the dynamics of the project and the cast of characters, none of these conditions indicate 'failure' to me. It is unfortunate that jwz does, but he'd probably know better than I.
The ruling stems from a trademark infringement suit filed against a Canadian corporation, 3263851 Canada, Inc., by Umbro International, a manufacturer of soccer equipment. In the suit, Umbro claimed the defendant infringed upon its trademark when it registered umbro.com in 1997. When the defendant failed to show up in court, the judge issued a default judgment, turning over umbro.com to the company and awarding it $25,000 in attorney's fees.
The court held that despite the fact that 3263851 Canada Inc. had paid for the domain, the name Umbro was trademarked, and therefore the domian name umbro.com was an extension of that trademark.
The problems with this ruling are daunting. As I understand it (insert disclaimer), the same name can be trademarked concurrently, as long as the name refers to different contexts. i.e. there can be simultaneous Umbro trademarks owned by Umbro the sports equipment manufacturer, and Umbro the (hypothetical) dance company. Think about Windows(R) for a moment to see why this must be so.
So what happens where there are two or more legitimate trademarks of the same name (e.g. Windows) and there is only one whatevername.com? Obvously the idea that a domain name is a natural extension of a trademark cannot possibly be logically consistent.
I'd need to read the actual decision to see what the judge's reasoning is though. Further, all this is subject to an appeal.