John Willkie wrote: >So, the question of the day is: are those sites lacking in W3C compliance, >or is it the browser that is not W3C compliant? > > Mozilla is more compliant than IE. Microsoft people nowadays claim that no one wants compliance (that is, they claim no browser USERS do). >Back a decade ago, it was IE that was not fully W3C compliant. > A decade ago is 1994. That is a fairly early date in the browser wars, and periods of 6 months to a year make a huge difference. For example, MSN was in beta test in April of 1995 along with Windows 95. MSN went live in the fall of 1995, Windows 95 having released in August. Some copies of the Windows 95 beta given out in April of 1995 had temporary free MSN access thrown in. MSN at that date was using something called Blackbird to enable third-party content providers. In early 1996 Microsoft decided to get with the Internet. By late 1996 MSN was using MSIE, HTML and ActiveX instead of Blackbird, which burned lots of small businesses that had invested in Blackbird. In 1996 I was trying out Draft CSS1 on my personal web site using MSIE, the only browser that I could easily get that handled CSS1. > These days, >I believe it's Mozilla/Netscape that lacks W3c compliance. It's been a few >years since I designed a web site and ran pages through the W3c validator, >then tested them against all common browsers, and obscure ones like the >Netscape implementations. I found that fully W3c validated web pages would >create problems with Netscape, and not with IE. Netscape, I was told, was >unhappy with the W3C, and so held up any "Microsoft inspired" changes in the >spec. > > Netscape was never "obscure". In the pre-open-source time frame, when Netscape was the market leader, Netscape may have been more non-compliant. Microsoft was smart to beat Netscape over the head with standards when MSIE was the minority player. The situation has changed dramatically since then. In many respects the exact opposite situation pertains. Mozilla's Gecko engine is certainly more compliant than MSIE now. >For me, the tipping point was with IE 3.0, the most obvious helpful feature >it offered was "auto-complete" where it remembered web sites and pages you >visited, and would complete the common ones, or offer a selection of >choices. Apparently, Netscape, with the Mozilla and Gekko engines, never >tried to see what new things were in IE. Because, it wasn't until Netscape >6.0 (bloatware, like all Netscape installations after 1.2) some five years >later, they offered this helpful feature. > > Here's the part where you see where I'm coming from. I was initially, in '95 and '96, gung-ho to use Netscape over IE because Microsoft was clearly the entity to avoid. Being an early adopter I tried to use the public beta version of the about-to-release Netscape browser (I think 3.0 Gold-- Gold as in branding not release status). The browser itself worked okay, except that uniquely among all the FTP transactions I did, the biweekly update of that Netscape beta was impossible to download over FTP using Netscape. So after a few rounds of using IE 2.0 (or sometimes other methods) to download the upgrades, I gave up on Netscape. I didn't try a "Netscape" browser again until the open-sourcing in 1998. (One of my machines is full of Mozilla milestone release browsers) I don't remember the autocomplete feature so much, different features are must-have for different people. I actually was pretty solidly using MSIE starting with 3.0 beta. Netscape Navigator 6.0 was the first open-source-based Netscape browser based on the Gecko rendering engine and the Mozilla project. Because of its origins, its standards-compliance proposition was a huge improvement over Netscape 4.x. NN6.0's release roughly coincided with when the little STB contract-software company I worked for was bought by AOL. (I no longer work for AOL) Mozilla.org has always been run by standards-oriented, open-source-oriented people. The Gecko engine was started from a clean slate by Netscape folks, to comply with standards and meet some other goals not being met by the standard Netscape offerings of the day. Sometime after the Netscape browser source was thrown open in the Mozilla project, the Gecko engine became the engine for the Mozilla browser. The Netscape division had standards religion (everyone of every rank everywhere was a believer, not just the few with mozilla.org email addresses) by the time I transferred to Netscape in 2002. In the summer of '03, Mozilla.org became completely independent of AOL. Netscape-branded packaging of Mozilla is pretty irrelevant today. There has been talk of a Netscape 7.2 release, but I don't know anyone that that release matters to. Influentials have migrated to Mozilla Firefox and recommend it over the Mozilla suite (sometimes referred to as "Sea Monkey"). There is still effort going into the suite, but I can't remember seeing anyone recommend it since Phoenix 0.5 or so. The Netscape packaging of Sea Monkey includes an AOL Instant Messenger client and some other baggage. It is unclear to me what the difference between Netscape 7.2 and Netscape 7.1 would be. Mozilla Firefox is just a browser, using the Gecko engine. I have been using it exclusively since the version called Phoenix 0.5. >I had been Netscape all along, until IE 3.0 made life easier. I may want to >encourage usurpers, but not at ALL costs. > > Even from my position I didn't recommend Netscape to technical friends until slightly after I transferred to Netscape and saw that Netscape 6.2 was useable in my own daily work. Until that point I had not used Netscape or Mozilla as much. >Now, Doug, to bring it back to religious differences, you are a Mac user, >no? And, so, your computer lacks Active X compatibility, does it not? > > > Technically, an ActiveX implementation has existed for Microsoft Office/Mac, but I've never heard of MSIE/Mac interoperating with it and obviously it would be tricky for MSIE/Mac to download and run x86-based ActiveX objects (in other words, it doesn't happen), so MSIE/Mac would not be vulnerable to ActiveX-vectored malware. MSIE/Mac is an independent project from MSIE/Win and has had slightly better standards compliance. I believe MSIE/Mac is an orphan now. New Mac users get Apple Safari, derived from an open source project. >John Willkie > > I would also say that putting the blame on ActiveX is just wide of the mark. Many apps can and do host ActiveX objects, but MSIE is unique in the extent to which it provides mechanisms for causing trouble with ActiveX objects. If you assume that MSIE is implemented perfectly (for argument's sake, please) and that ActiveX is implemented perfectly, there is a configuration of the two working together that creates unique vulnerabilities for end users, via silent downloads of ActiveX and via access to ActiveX objects that can be addressed from a page in the browser that you wouldn't expect to be part of the browser bailiwick. The essential piece of this configuration is MSIE. NSAPI is the interface between browser plug-ins and Netscape browsers. MSIE could have stuck with reverse-engineering that interface. If it had, I think that *some* of the problem we are talking about today could have still existed. That underscores how essential the browser's design and implementation are in the situation we actually do have. > <snip> -- Mike Enright mail: michaelt@xxxxxxxxxxxx AIM: enr Yahoo: michaeltenright Beautiful Cardiff-by-the-Sea, California ---------------------------------------------------------------------- You can UNSUBSCRIBE from the OpenDTV list in two ways: - Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word unsubscribe in the subject line.