Netscape 7.0 Final News and ReviewsSaturday August 31st, 2002Though Netscape releases no longer get the coverage that they used to, the launch of Netscape 7.0 was reported in several places on the Web. Articles can be found at AOL Time Warner, BBC News, CNET News.com, Geek.com, HindustanTimes.com, Internet Magazine, internetnews.com, Internet Week, MacCentral, MacSlash, NewsFactor, PCWorld.com, Reuters, Slashdot, The Register, Yahoo! Finance and ZDNet News. Both CNET and eWEEK have reviews of Netscape 7.0 while TechTV features an editorial. If you're after user opinions, Download.com has a wide selection. UPDATE! PC Magazine gave Netscape 7.0 four out of five, calling it "a sleek, fast, friendly browser that may make you consider switching from Microsoft Internet Explorer as your default browsing tool." Finally, here's the solution to the biggest complaint about Netscape 7.0. I've been running Mozilla nightlies for the last year or so - but the latest Netscape release has become my default browser. What's most impressive to me is the memory management - especially under Windows (win 2k sp3 specifically). When I minimize Netscape to the taskbar the memory used drops to around 1.2 or 1.4 Mb's from around 25 Mb's. Netscape has been stable - with little or no crashing. I'm set up now to use Netscape by default, and Mozilla when I need to do or test something specific. Good job. Fix a couple of little bugs - and this is a winner. I'm even using the built in AIM client to connect with people from work. It's a well-known phenomenon that Windows NT and 2000 report that applications use far less memory when you minimize the application window. The application is in fact not using any less memory, but Windows reports the the memory used is less. You need to look at the VM size or something like that to see how much memory the application is really using. Doing the same thing in 1.1 shows less then 2MB of memory usage under XP so this wouldn't be a big reason to switch would it? The only feature I miss from the Netscape release is a spell checker but I hear that might make it into 1.2. Go get it at http://spellchecker.mozdev.org/ Works great for me in Moz. At that's left now is moving ICQ over... hmmmm?? After I read around 10.000 (or more?) comments about Mozilla 1.1 and Netscape 7 this are the biggest problems, which prevent people from using them: 1. we must find a way to deal with broken profiles as soon as possible! Maybe the Mozilla version should be saved in the profile, and if the versions are incompatible, users should get a warning, maybe with the possibility to wipe out all critical files (localstore, /chrome, cache). 2. the startup performance must be umproced very very much. Even rather expirienced people only see, that Opera is faster and don't understand, why it is faster. Joe User won't ever understand why MSIE is faster... I have no idea, how we can manage this, but as I see the hard disk performance matters very much and the CPU nearly doesn't matter at all (Doron's P3/800 sometimes outperformes the P4/1600 Netscape "normally" uses..), maybe we can do something based upon this. 3. TE problems. Maybe even adding a message to the status bar, if an error in the JS console appears will help the prople to understand, that the page is broken and not the browser. 4. we need big pages using features MSIE doesn't have - why aren't mozilla.org or netscape.com or at least "my netscape" using position:fixed, attribute selectors, rounded borders, translucent PNGs andgenerated content? Currently they are "optimized for Netscape 3", with unscalable table layout! sorry for the lost newlines - damn this form! subject says it all. The best feature from NS7 that I want to use in Mozilla is the icq sidebar. Please please please can god come down from heaven, just to make this happen for me? Jabberzilla is still being worked on. Also, I believe the Jabberzilla code has been sent over to some Sun programmers to be worked on. Hopefully it will soon hit a state in which it actually becomes usable. Does anyone know if there is documentation on Netscape7 on how to do customizations, e.g. for coporate installs - LDAP configs, proxy configs, pre-loaded digital certificates, corporate bookmark (preferably folder of bookmarks) added to existing bookmarks, different default preferences (for corporate standard), Corporate theme (if we spent the time to make one), pre-loaded .xpi's (like ad-blocking, themes, etc.) - all to be applied after a migration from netscape 6.x or less (mostly less, like 4.79)? "P4/1600 Netscape "normally" uses.." Where do you come up with these things? I was actaully just looking at a spreadsheet less than two weeks ago that listed all of the Netscape developer and QA machines and I'd wager that 95% of Netscape employees are using a machine less than half that speed. Not to mention that QA (and especially performance QA) teams have a wide variety of equipment with the specific reason being testing Netscape/Mozilla on machines at the low end of the spectrum (far more machines at 500MHz and below than above). Are you just making things up or have you been misinformed by some Netscape person that "Netscape 'normally' uses" P4/1600s? --Asa that's, why there are the "" around - don't ask me who said it, but it was several times, that at least some of the developers have a P4/1.6GHz. Software developers typically have the faster computers to do fast compilations and generally fast computing. They need to be productive and do little testing. The testing/QA teams of a software development company are the ones that will have slower, older computers, and access to even slower still computers to do testing on more real-world equipment that the user base will be using. FWIW my P3 600mhz win2k box runs Mozilla 1.0 very snappy, including new window creation and even launching without quicklaunch enabled. So I don't know what the hoopla is about slow performance at all, unless you're on a 128megs or lower machine or a 300mhz or lower cpu (Intel). Then I could see performance with Mozilla or Netscape 7 becoming less than ideal. I have a AMD K6-2/300, 128MB RAM a Celeron 433 Laptop, 128MB RAM a K6-2/500, 256MB RAM and an Athlon 1000, 512MB RAM ....the Celeron is the slowest of this 4, I won't use Mozilla on that, if I had a better alternative on Linux. I still have an old pentium 166mhz. I found some old SIMMs in various places, so I have 64 megs of ram in it now. mozilla is a little sluggish in linux, but a little better in windows. Galeon runs a little better still. This didn't used to be the case. the 1.0 series seems a lot faster to me than the early 0.9's. I'm wondering if 1.1 would be comparable to galeon based on 1.0.0 or not. Anyway, I'd suggest dropping the performance req. Maybe recommend what is listed now, but a 166 mhz processor is enough to run mozilla w/o pain. That's actualy one of not many popular pages that is designed to use mozilla features. You can drag and drop panels to customize it. You have to register and sign in to use this feature. you really want to know, what I see there...? Oh, better not.. 1. the small images have an alt=""-Text, which contains some info, for what this button is good - appears as a tooltipp on MSIE and NN4, no info on Mozilla -> on this point a Netscape page is optimised against Netscape (!) 2. I see GIF images, which display everywhere the same, bot no PNG images, which are THAT argument for Mozilla all over 3. on that very long CSS rule list (somebody haven't understood the "C" in CSS or?), I can't find a single rule, which doesn't work on MSIE. I duno, if the Drag&Drop works on MSIE, but everything else works there as good or better. THIS page is more an argument not to use Mozilla, "because it can't show tooltipps on images"! The drag and drop on my.netscape.com works fine in IE 6.. Any browser that shows ALT text as a tool tip is wrong. Period. If you think that netcape.com (or any other site) should have tooltips on the images, then you should suggest that the site use the TITLE attribute. I know. the text shows a tooltipp on MSIE, which is an enhanced functionality, no matter (!), if it has to do with standards. Even worse, that it additional shows, that Netscape's Web Authors doesn't understand the standards nor understand how to code for their own browser! > which is an enhanced functionality, no matter (!), if it has to do with standards. No, because a proper alt text makes sense as a replacement for a graphic, while a proper title text provides additional information that helps someone looking at the graphic understand what it means. Alt text is primarily an accessibility feature, for e.g. blind users (though they help a lot with Lynx, too); the title text is primarily a usability feature. If you don't understand the difference, I'd suggest doing a little reading at the W3C WAI site; also, I seem to remember that one of the Mozillians has a good intro to this (Hyatt, maybe?). Showing the wrong text is not an enhancement. It just seems like an enhancement if you have never seen the way it should work. And it's an "enhancement" that gets in the way of things working the way they should, which would be a real enhancement. I know the difference between alt="" and title="", and this text makes sense as both. But as it's the _only_ alt="" attribute on the whole page, no title="" is placed anywhere and MSIE + NN4 display an alt="" as a tooltipp, it looks, like the author of that page doesn't know this, and wanted a tooltipp with the alt="". It is also possible that the web designer did not want any tool tips at all and that is why there are no title attributes. It is hard to determine what the web designer intended in this case. If the web designer intended for the alt attribute to be a tool tip on the netscape.com website then he/she should be severely reprimanded or perhaps fired. I know it. But in that case the tooltipp is an enhancement, so it should be there :) A tooltip might be an enhancement, but if it should be there then the developer should write the proper code to create it. Web browsers should not improperly display a web page just because bad web developers can not figure out how to use the title attribute. Maybe the tooltip would be an enhancement - to the *web page*. The browser has no way of knowing whether what is in the alt attribute was intended as alt text or as title text, so it has to follow the spec, not engage in profitless mind-reading exercises, no? (BTW, the tone will be missed on this: it's amusement, not hostility to TheK. It's worth hashing these things out now and then.) > the text shows a tooltipp on MSIE, which is an enhanced functionality No it is NOT enhanced functionality. It is an extremely annoying bug in IE and older versions of Netscape. I want to make web pages that are easily accessible to the blind using alt, but when I do so, the websites are littered with excessive tooltips in IE. Every signifigant image on a web page should have alt text, but not every image should have a tool tip. Also, the appropriate alt text and the appropriate tooltip text are often different. The issue of alt text unwantingly popping up as a tooltip also comes into play when you have a website where you want to use fancier CSS/Javascript generated tooltips. On IE, the desired CSS/Javascript tooltip appears along with an unwanted tooltip based on the alt text. Sure I can remove the alt attribute so that IE does not screw things up, but then I am denying usable access to the blind and to people using text browsers. I should not have to exclude users from accessing my websites just so that IE can continue to violate the standards; that is not an enhancement. workaround for MSIE: empty title="" - only bad luck for NN4. Your workaround does not resolve the issue of proper alt text for blind websurfers or text only browsers. The real problem is that too many web designers are not using the alt attribute properly and IE facilitates that improper usage. Providing a workaround for web designers who are trying to do the right thing does not help to get the real problems fixed. Correct me if I am wrong, but your position seems to be : 1) web designers who have properly coded valid HTML should code their pages with a workaround in order to compensate for a flaw in IE 2) the flaw should be added to Mozilla in order to allow the web designers who improperly code their pages can continue to do so Making it harder for the good guys and easier on the bad guys really does not sound like the way to a better web. no. 1. Mozilla has to stay as it is. 2. Web Developers who do not want a tooltipp should add an empty title="" for now (instead of leaving the alt"" away and so breaking the spec), until MS fixes this bug and add an alt="" with a replacement text to _every_ image and an empty alt="" to "useless" images. 3. pages which want a tooltipp have to write this into the title="". 4. in this specific case the text makes sense as both, as it's a good short descripting and a good tooltipp text, it should be added for alt="" (where it is) _AND_ title="". I think the idea of a status message informing about broken pages is interesting: of course this depends on how and if it can be done technically, but Mozilla/NS *do* get a lot of blame because of badly doen web pages. This wont help with MS passport registration and similar sites of course :/. Suggestion: A small icon between the progress bar and the online icon might make sense: a yellow exclamation point with a title text saying "page loaded with errors: the (x)html code on this page does not validate" or "page loaded with errors: there are errors in the javascript code on this page," etc.; and other icons for other kinds of page loading messages (a red exclamation point for HTTP error codes, with the HTTP error code number and explanation as a supplement to the error page that Mozilla provides in the document window; some other kind of icon for a page that loads properly with no errors, etc.). not "does not validate", only display errors, which really matter for the page display _in Mozilla_ Unfortunately, "display errors" is a subjective thing. Take the image alignment issue: I don't know about current versions of Mozilla, but earlier versions hewed so closely to the W3C specs on CSS that images tended to be aligned in such a way that whitespace appeared between images that were intended by the developer to touch. The code would validate fine, but the page would look like a pos. Now, how does Mozilla distinguish between pages with such sliced images that do not have the proper img[[.classname]] {vertical-align: bottom;} rule and pages which do not have sliced images and so do not need that css rule? I can think of ways of doing it, but they would be pretty computing intensive (map out the layout of the page and look for image elements that are adjacent on the page layout, then check the css page for the required rule, then send the error to the page: problem here is that you have to check each image to see if it is adjacent to other images on the page, and then make a kind of value judgment about at what density adjacent images become sliced images - and there will always be exceptions). ( [[ above is not CSS, but to indicate an optional selector. ) Sorry if this isn't clear, I'm just having a hard time explaining this. Maybe one of the coders could chime in? Unfortunately, "display errors" is a subjective thing. not really, if there's a JS function, which returns errors, it's quite shure, that something went wrong. It it's a CSS lenght without unit in sloppy mode, this is shure invalid, but it doesn't matter, as Mozilla (and btw all browsers I know) know what this guy means. 3 examples: font-site:12px <<valid || font-size:12 << invalid, but browsers know what to do || font-size:12 pt << invalid and only MSIE reads it, so only this should return an error. An Enhancement request already exists in Bugzilla for this feature http://bugzilla.mozilla.org/show_bug.cgi?id=47108 [bug 47108]. Note that full validation in the browser window has been marked 'invalid' http://bugzilla.mozilla.org/show_bug.cgi?id=109881 [bug 109881]. #43 User notification of improper HTML/Javascriptby SubtleRebel Tuesday September 3rd, 2002 10:46 AM I think it would be useful even if it only pointed out that the website was not displaying properly because it was using outdated, improper, non-standard, and proprietary code. It does not have to be a 100% validator in order to be useful. Just noticed something that appears to be ever so slightly amiss with the review of Netscape 7 on the Internet Week site, " . . . the new Netscape browser is a genuine milestone, the first Netscape-branded version of the open source Mozilla code that the company set free years ago." Errr, I don't know about you, but that's news to me. Just exactly which planet has the writer of that article been living on since April 2000, when Netscape 6.0 Preview Release 1 was released? maybe he wanted to say "the first _usable_ Netscape-branded version..." ? :=) > the first Netscape-branded version of the open > source Mozilla code that the company set free > years ago Or, better, the first version of Netscape based on Mozilla code that was released after the Mozilla people felt it was ready for prime time. I can't find it. Is N7 in fact build on 1.0.1RC2 or are Mozilla.org just slow getting 1.0.1 available for download? Does anyone know if there is documentation on Netscape7 on how to do customizations, e.g. for coporate installs - LDAP configs, proxy configs, pre-loaded digital certificates, corporate bookmark (preferably folder of bookmarks) added to existing bookmarks, different default preferences (for corporate standard), Corporate theme (if we spent the time to make one), pre-loaded .xpi's (like ad-blocking, themes, etc.) - all to be applied after a migration from netscape 6.x or less (mostly less, like 4.79)? baffoni, Netscape has historically always provided CCKs - Customization Kits. I'm not sure whether they've updated the CCK for 7.0 yet, and if not, I'm sure they are going to in the very near future. They are very interested in corporate accounts and might even produce a customized build for you, if your company happens to be a high-profile prospective customer. http://wp.netscape.com/browsers/6/cck/ The CCK for version 7 appears to be available from http://channels.netscape.com/ns/browsers/partners/cck.jsp http://australianit.news.com.au/articles/0,7204,5017135%5E15306%5E%5Enbv%5E,00.html From the Australian IT Add it to the list already there if you like. Michael Is there a way to install it in Mozilla? The Netscape themes page http://wp.netscape.com/themes/7_0/index.html says that I need Netscape 7, but I believe it should work with Moz 1.0.1 as well if they would only allow it to install. |