Re:Good for the average joe
on
Netscape 6.2
·
· Score: 1
Does the page use document.all ? If so, that would be big news.
Re:Another reason stopping people from using Netsc
on
Netscape 6.2
·
· Score: 1
OK -- I just put Netscape 6.2 on this Citrix box and I see what you mean. IE uses the normal (bad) Windows dithering, but something is just _wrong_ with Netscape.
Re:I have a suggestion for the internet.
on
MS DOS: A Eulogy
·
· Score: 1
My suggestion is to find the pile where the IT guys throw the old MSDN and TechNet CDs. It will be in there along with Excel 2000 for Alpha, MS DirectAnimation, and other wonders.
Re:Rather than whine about Mozilla...
on
Netscape 6.2
·
· Score: 1
Part of the perceptual speed problem with Mozilla was goofyness caused by the incremental renderer. On some sites, you would see a malformed version of the page for a half of a second before the real page would appear. I don't think it was actually slower in rendering, but it seemed like it.
Since I upgraded to 0.95, I can't find a site to reproduce this, so maybe it's been fixed.
New Window speed is still a little too slow for comfort.
Re:Another reason stopping people from using Netsc
on
Netscape 6.2
·
· Score: 1
IE doesn't really dither it's interface -- it uses different icons for low-color installs. As usual, Windows dithering still sucks -- browse the web on a low-color Mac if you want to see the difference.
Is there something in Bugzilla requesting a low-color theme?
Of note, LINK was in HTML 2.0 (1994?), but has never been implemented in a mainstream browser until now.
It's a key piece of functionality (Prev Next Up Contents) that 'groupware' applications like Notes and Outlook have had for a long time. I recall reading about it way back when and thinking "Cool -- I can't wait until browsers implement this!" Now if Microsoft gets on the stick, a big rational for Frames and other page-oriented crap can be removed.
Re:Netscape? no thanks.
on
Netscape 6.2
·
· Score: 2, Troll
Back in 1996, Netscape was running around touting "standards-based platforms" to their customers, but they were in fact very anti-W3C.
CSS was barely supported because Netscape had developed something proprietary called JavaScript Style Sheets (which CSS is internally transformed to). That's why NS4 ignores all CSS if you turn JavaScript off.
Netscape also developed a completely different proprietary document object model (document.layers). Which could theoretically could do cool stuff except that it crashed 90% of the time. They blew off the W3C's work on DOM, which was roughly tracked by Microsoft.
The end result of this standards split is that most of the WWW is stuck on 'common' pre-1996 standards. Ugly HTML 3.2-type markup, very little CSS, and Netscape 3 DOM-type JavaScript.
The bad thing is there's 10% of the userbase that seems to be holding out for good on Netscape 4.x -- they aren't interested in IE, they aren't interested in Netscape 6. That essentially means that modern HTML authoring will never really come into vogue, and we will be stuck in 1995 until Microsoft actually finally gets the balls to 'fork' the WWW so that their stuff only works on their platform.
Re:why is mozilla engine so slow?
on
Netscape 6.2
·
· Score: 1
Well, I'm sure "makes Netscape slow" was considered a feature by Microsoft, and not a bug.
Part of the legend is also based on the fact that a long time ago (~1995) microsoft.com had special slow-down code for Netscape browsers. MS even fessed up to that one.
Re:netscape cares about the details...
on
Netscape 6.2
·
· Score: 5, Insightful
Actually, I would say that your management was right on. As mozilla.org says: We make binary versions of of Mozilla available for testing purposes only!. We provide no end user support.
This is something that's missed by the "Mozilla advocates" that hang on Slashdot and Mozillazine and other places. Mozilla is not an end-user browser. It's for voluntary developers and voluntary QA people only. No non-nerds even know what Mozilla is, so if you try to encourage people to use it, the funny looks they are giving you are well grounded.
So, if you are worried about a MS-dominated WWW, encourage people to try Netscape 6.2. Don't even mention Mozilla -- it detracts from the message. Unfortunately, lots of (normal) people took a look at the horrific 6.0PR releases and the terrible 6.0 final and need some encouragement to take another look at the releases that actually work.
Re:why is mozilla engine so slow?
on
Netscape 6.2
·
· Score: 2, Informative
Because of Netscape 4.x's only partial HTTP 1.1 support, it does become very slow with some HTTP servers, one of them being IIS (but also WebLogic and others).
I haven't noticed any particular problem with Moz, although it can be kind of clunky with pages with lots of form elements.
Re:Not only does XP have the command prompt
on
MS DOS: A Eulogy
·
· Score: 1
The "DOS" environment in NT was forked off of OS/2 which was forked off of DOS 5.0. That's why you get strange things like edlin pre-installed. (Edlin was removed from DOS 6 for disk space reasons, I think -- it's still downloadable somewhere on ftp.microsoft.com last time I checked.)
Re:I have a suggestion for the internet.
on
MS DOS: A Eulogy
·
· Score: 1
Either that or try to find Internet Explorer 5.0 for Windows 3.1 (yes, it was released, but it's not for download anymore). Seemed much better than any 16-bit version of Nutscrape.
Re:Quick and Dirty Interrupt Handler
on
MS DOS: A Eulogy
·
· Score: 2
A great book about the the DOS-Windows relationship is "Undocumented Windows 95" by Andrew Shulman. In it, he describes how Win95 was pretty much just a fancied up Win3, and how to hack Win95 into a 386 Protected Mode version of DOS.
As to shani's comment, Microsoft did ship a 286-mode version of DOS with memory protection and preemptive multitasking. It was called OS/2.
Re:Little content, little meaning...
on
MS DOS: A Eulogy
·
· Score: 2
(2) Removing MS-DOS from XP breaks backward compatibility
The semi-compatible "Virtual DOS Machine" was pretty much unchanged through MS-OS/2 1.3 to Windows NT 3.1 to Windows 2000. It's good enough to run most DOS business applications, but not good enough to run most games or anything that requires funky drivers.
DOS compatibility was actually the "killer feature" back in the late 80s -- as in the lack thereof killed OS/2 1.x as a viable desktop OS. IBM significantly improved it for OS/2 2.x, but Microsoft (probably because they were still making a gazillion dollars off of MS-DOS) never really took fixing the NT emulation very seriously.
I understand that with XP, it now emulates a SoundBlaster and VESA. This is pretty amazing considering that absolutely no work has been done on the thing in 10+ years. Anyway, OS/2 shipped in 1987 and it's not like people haven't been warned for more than a decade that DOS is going away and would be replaced by some virtual emulator.
Re:anti-Microsoft conspiracy theories
on
MS DOS: A Eulogy
·
· Score: 2
One of the interesting things about most of the anti-Microsoft conspiracies is that they all involve settlements covered under Non-Disclosure agreements. This way there is no way to validate the authenticity of the story.
This Kildall Magic Keystroke story is always one of my favorite Internet legends. It's sources are that "MS the Company" page (from a guy with an obvious chip on his shoulder), and a bunch of old pre-Deja Usenet posts which only exist in my head:) Anyway, the story is not just out of the blue.
The folklore side went a little futher -- The QDOS guy actually wrote the OS, but was running out time so just he translated some of the CP/M utility software. Something obviously had an easter egg in it.. maybe even DEBUG itself.
Soon after IBM started pouring money in for the substantial DOS 2 rewrite.
The funny thing is that they all started arguing over Unix paths and nobody really addressed your point.
System and System32 were invented becauase it was possible to install a NT 3.x installation directly over Windows 3.1, and then boot back-n-forth between them. That broke with 95 and NT4, but MS has told everyone for years to hardcode System32, so that's what we get.
I don't think WOW is hack at all -- at least compared to 9x -- it's just an app that intercepts 16-bit API calls and reissues them as 32-bit. I agree that it will start to look hacky if System64 appears. Do 64-bit Unixes have two versions of their system libs?
Re:Only problems DrDos had was with Win3.1+
on
MS DOS: A Eulogy
·
· Score: 2
Get out of your conspriacy cave, man. He said "Alot of software took a shit with it.", and I'll expand that:
DR-DOS sucks, sucked, and always will suck! It's in incompatible piece of shit that doesn't work with barely anything. I could give a shit if Windows 3 worked or not -- the fact is that DR had problems with Lotus and Borland and every other DOS app company on the planet.
I remember when "Novell DOS" (as it was called at the time) had a bug which prohibited NetWare from booting under certain circumstances (something to do with EISA.) Even Novell was telling people to use MS-DOS.
I even tried to download the Caldera DOS Web browswer thing a couple years ago, and THAT didn't even work well on DR-DOS. Switched back to MS-DOS and it was fine. P-A-T-H-E-T-I-C
I'm not one to defend Microsoft -- but they were probably right to ban that piece of shit for support reasons. In fact they were right, because DR/Novell spent the next couple years working all the Windows-related bugs out.
Easier said than done -- how do you test for the fact that Netscape 4.x claims to implement certain functionality that is in fact totally broken? Also functional compatibility doesn't say anything about what the end result looks like (a real concern for those guys).
Mozilla is only supposed to attract attention from developers and testers -- it's not supposed to be an end user browser. That's what Netscape is for.
Note that Netscape 6.x wasn't blocked -- that indicates that there was no great consipriacy towards the #2 browser (or it's dev branch).
More like some marketing shmo said "There's this thing called Mozilla, I have no idea what it is, but we don't have time to test it, so let's block it." Typical stupid decision you see all the time when you get non-technical people making technical decisions.
(To some extent, mozilla.org brought this on themselves by holding the pretention that the Mozilla browser is something entirely disconnected from their employers over at AOL/Netscape. If Moz sent a Netscape 6-like user-agent string, they wouldn't have had this problem.)
The document.form object (introduced with Netscape 3 and emulated in virtually every other browser) is a "vendor extention" that was never W3C ratified.
But yet it's a defacto standard and it's virutally impossible to do client-side form validation or manipulation without it.
You say "vast majority", but in my experience, the few weirdos left using Netscape (on Windows/Mac) are doing so because they've standardized on the mail client.
Of course, releasing buggy crap like the NS6 mail client was probably worse than not releasing a mail client at all.
Yeah, and Apple took the "cute" wrap very badly -- they started shipping with a dull gray desktop, toned down their icons, and told their games developers to buzz off. Poor Apple was shipping machines with the hardware to do 16 thousand colors and a nearly monocrome desktop, while Windows was looking cheery on a 16-color VGA. Terrible decision because, as recent OSes show, people actually like cute.
Although, most of the "cartoon machine" guff I heard about Macs was from Novell admins about to get smacked upside the head with a 10 pound NT 3.5 resource kit.
Slashdot Mirror
Does the page use document.all ? If so, that would be big news.
OK -- I just put Netscape 6.2 on this Citrix box and I see what you mean. IE uses the normal (bad) Windows dithering, but something is just _wrong_ with Netscape.
My suggestion is to find the pile where the IT guys throw the old MSDN and TechNet CDs. It will be in there along with Excel 2000 for Alpha, MS DirectAnimation, and other wonders.
Part of the perceptual speed problem with Mozilla was goofyness caused by the incremental renderer. On some sites, you would see a malformed version of the page for a half of a second before the real page would appear. I don't think it was actually slower in rendering, but it seemed like it.
Since I upgraded to 0.95, I can't find a site to reproduce this, so maybe it's been fixed.
New Window speed is still a little too slow for comfort.
IE doesn't really dither it's interface -- it uses different icons for low-color installs. As usual, Windows dithering still sucks -- browse the web on a low-color Mac if you want to see the difference.
Is there something in Bugzilla requesting a low-color theme?
Of note, LINK was in HTML 2.0 (1994?), but has never been implemented in a mainstream browser until now.
It's a key piece of functionality (Prev Next Up Contents) that 'groupware' applications like Notes and Outlook have had for a long time. I recall reading about it way back when and thinking "Cool -- I can't wait until browsers implement this!" Now if Microsoft gets on the stick, a big rational for Frames and other page-oriented crap can be removed.
Back in 1996, Netscape was running around touting "standards-based platforms" to their customers, but they were in fact very anti-W3C.
CSS was barely supported because Netscape had developed something proprietary called JavaScript Style Sheets (which CSS is internally transformed to). That's why NS4 ignores all CSS if you turn JavaScript off.
Netscape also developed a completely different proprietary document object model (document.layers). Which could theoretically could do cool stuff except that it crashed 90% of the time. They blew off the W3C's work on DOM, which was roughly tracked by Microsoft.
The end result of this standards split is that most of the WWW is stuck on 'common' pre-1996 standards. Ugly HTML 3.2-type markup, very little CSS, and Netscape 3 DOM-type JavaScript.
The bad thing is there's 10% of the userbase that seems to be holding out for good on Netscape 4.x -- they aren't interested in IE, they aren't interested in Netscape 6. That essentially means that modern HTML authoring will never really come into vogue, and we will be stuck in 1995 until Microsoft actually finally gets the balls to 'fork' the WWW so that their stuff only works on their platform.
Well, I'm sure "makes Netscape slow" was considered a feature by Microsoft, and not a bug.
Part of the legend is also based on the fact that a long time ago (~1995) microsoft.com had special slow-down code for Netscape browsers. MS even fessed up to that one.
Actually, I would say that your management was right on. As mozilla.org says: We make binary versions of of Mozilla available for testing purposes only!. We provide no end user support.
This is something that's missed by the "Mozilla advocates" that hang on Slashdot and Mozillazine and other places. Mozilla is not an end-user browser. It's for voluntary developers and voluntary QA people only. No non-nerds even know what Mozilla is, so if you try to encourage people to use it, the funny looks they are giving you are well grounded.
So, if you are worried about a MS-dominated WWW, encourage people to try Netscape 6.2. Don't even mention Mozilla -- it detracts from the message. Unfortunately, lots of (normal) people took a look at the horrific 6.0PR releases and the terrible 6.0 final and need some encouragement to take another look at the releases that actually work.
Because of Netscape 4.x's only partial HTTP 1.1 support, it does become very slow with some HTTP servers, one of them being IIS (but also WebLogic and others).
I haven't noticed any particular problem with Moz, although it can be kind of clunky with pages with lots of form elements.
The "DOS" environment in NT was forked off of OS/2 which was forked off of DOS 5.0. That's why you get strange things like edlin pre-installed. (Edlin was removed from DOS 6 for disk space reasons, I think -- it's still downloadable somewhere on ftp.microsoft.com last time I checked.)
Either that or try to find Internet Explorer 5.0 for Windows 3.1 (yes, it was released, but it's not for download anymore). Seemed much better than any 16-bit version of Nutscrape.
A great book about the the DOS-Windows relationship is "Undocumented Windows 95" by Andrew Shulman. In it, he describes how Win95 was pretty much just a fancied up Win3, and how to hack Win95 into a 386 Protected Mode version of DOS.
As to shani's comment, Microsoft did ship a 286-mode version of DOS with memory protection and preemptive multitasking. It was called OS/2.
(2) Removing MS-DOS from XP breaks backward compatibility
The semi-compatible "Virtual DOS Machine" was pretty much unchanged through MS-OS/2 1.3 to Windows NT 3.1 to Windows 2000. It's good enough to run most DOS business applications, but not good enough to run most games or anything that requires funky drivers.
DOS compatibility was actually the "killer feature" back in the late 80s -- as in the lack thereof killed OS/2 1.x as a viable desktop OS. IBM significantly improved it for OS/2 2.x, but Microsoft (probably because they were still making a gazillion dollars off of MS-DOS) never really took fixing the NT emulation very seriously.
I understand that with XP, it now emulates a SoundBlaster and VESA. This is pretty amazing considering that absolutely no work has been done on the thing in 10+ years. Anyway, OS/2 shipped in 1987 and it's not like people haven't been warned for more than a decade that DOS is going away and would be replaced by some virtual emulator.
One of the interesting things about most of the anti-Microsoft conspiracies is that they all involve settlements covered under Non-Disclosure agreements. This way there is no way to validate the authenticity of the story.
:) Anyway, the story is not just out of the blue.
.. maybe even DEBUG itself.
This Kildall Magic Keystroke story is always one of my favorite Internet legends. It's sources are that "MS the Company" page (from a guy with an obvious chip on his shoulder), and a bunch of old pre-Deja Usenet posts which only exist in my head
The folklore side went a little futher -- The QDOS guy actually wrote the OS, but was running out time so just he translated some of the CP/M utility software. Something obviously had an easter egg in it
Soon after IBM started pouring money in for the substantial DOS 2 rewrite.
The funny thing is that they all started arguing over Unix paths and nobody really addressed your point.
System and System32 were invented becauase it was possible to install a NT 3.x installation directly over Windows 3.1, and then boot back-n-forth between them. That broke with 95 and NT4, but MS has told everyone for years to hardcode System32, so that's what we get.
I don't think WOW is hack at all -- at least compared to 9x -- it's just an app that intercepts 16-bit API calls and reissues them as 32-bit. I agree that it will start to look hacky if System64 appears. Do 64-bit Unixes have two versions of their system libs?
Get out of your conspriacy cave, man. He said "Alot of software took a shit with it.", and I'll expand that:
DR-DOS sucks, sucked, and always will suck! It's in incompatible piece of shit that doesn't work with barely anything. I could give a shit if Windows 3 worked or not -- the fact is that DR had problems with Lotus and Borland and every other DOS app company on the planet.
I remember when "Novell DOS" (as it was called at the time) had a bug which prohibited NetWare from booting under certain circumstances (something to do with EISA.) Even Novell was telling people to use MS-DOS.
I even tried to download the Caldera DOS Web browswer thing a couple years ago, and THAT didn't even work well on DR-DOS. Switched back to MS-DOS and it was fine. P-A-T-H-E-T-I-C
I'm not one to defend Microsoft -- but they were probably right to ban that piece of shit for support reasons. In fact they were right, because DR/Novell spent the next couple years working all the Windows-related bugs out.
Easier said than done -- how do you test for the fact that Netscape 4.x claims to implement certain functionality that is in fact totally broken? Also functional compatibility doesn't say anything about what the end result looks like (a real concern for those guys).
whoops! Thank you for correcting my ignorance.
The LAYER tag was never even recognized by the W3C, much less deprecated. It was a proprietary Netscape-ism.
Instead of going their own way, Microsoft followed the W3C work, which is why they support DIV and Netscape 4 never really did.
So, it seems that in your hated of Microsoft, you've lost sight of the standards and become a Netscape collaborationist dupe.
(I'm sure the discussion of why P was depricated can be looked up.)
Mozilla is only supposed to attract attention from developers and testers -- it's not supposed to be an end user browser. That's what Netscape is for.
Note that Netscape 6.x wasn't blocked -- that indicates that there was no great consipriacy towards the #2 browser (or it's dev branch).
More like some marketing shmo said "There's this thing called Mozilla, I have no idea what it is, but we don't have time to test it, so let's block it." Typical stupid decision you see all the time when you get non-technical people making technical decisions.
(To some extent, mozilla.org brought this on themselves by holding the pretention that the Mozilla browser is something entirely disconnected from their employers over at AOL/Netscape. If Moz sent a Netscape 6-like user-agent string, they wouldn't have had this problem.)
The document.form object (introduced with Netscape 3 and emulated in virtually every other browser) is a "vendor extention" that was never W3C ratified.
But yet it's a defacto standard and it's virutally impossible to do client-side form validation or manipulation without it.
It looks like a replacement for the commercial program "System Commander", which runs for around $75 last I checked.
(Although, System Commander does other neat things like remapping drives, multiple DOS/Win installs, and boot support for some obscure OSes.)
You say "vast majority", but in my experience, the few weirdos left using Netscape (on Windows/Mac) are doing so because they've standardized on the mail client.
Of course, releasing buggy crap like the NS6 mail client was probably worse than not releasing a mail client at all.
Yeah, and Apple took the "cute" wrap very badly -- they started shipping with a dull gray desktop, toned down their icons, and told their games developers to buzz off. Poor Apple was shipping machines with the hardware to do 16 thousand colors and a nearly monocrome desktop, while Windows was looking cheery on a 16-color VGA. Terrible decision because, as recent OSes show, people actually like cute.
Although, most of the "cartoon machine" guff I heard about Macs was from Novell admins about to get smacked upside the head with a 10 pound NT 3.5 resource kit.