Re: Proposal for non-normative example section (Was: RE: How is it possible to devise such a feeble system?)

On Thursday 25 October 2001 13:57, Andy wrote:
|   Vadim Plessky wrote:
|   > I hope that Konqueror will pick up soon, and overgrow Mozilla/NS in
|   > visible future. Than things will be easy. Konqueror emulates MS IE's
|   > "quirk" mode pretty well (but implementing this consumed a lot of
|   > resources), for standards-compliant (Strict) mode you don't have such
|   > development overhead. You just need to implement standard. So I would
|   > say in such scenario we will get standards-compliant web space just in
|   > case only Microsoft (but not Netscape) implemented CSS3.  KHTML will be
|   > there as well, so both leading browsers will fully support CSS3 (core).
|
|   I must admit I'm missing your logic here somehow. If the two leading
|   browsers were Konqueror and MSIE and both implment IE's quirks mode (I
|   assume you refer not just to those places where they differ from the CSS
|   spec but also their DOM features like document.all) isn't there a

yes, document.all is supported by Konq.

|   tremendous risk that everyone will simply author to MSIE's quirks mode
|   and ignore standards? That's pretty much what we've seen with the 2
|   quirks modes of Netscape 4 and MSIE in the last 2 or 3 years. In that

come on, all web sites are tuned for MS IE! (I am not counting mozilla.org 
here)
Is it any kind of surprise for you?
Do you think that somebody, in good mind, will follow MS IE 4.x messy 
DOM/DHTML? It's somewhat of business necessity. :-((

|   scenario, about the only thing that would prevent that from happening,
|   would be for Microsoft to sacrifice backwards compatitibility with their
|   own quirks mode and "force" people to standards. This makes no business
|   sense, indeed the practical side of me sees it as irresponsible, much as
|   the romantic side of me might like it.
|
|   This also ignores the fact that Konqueror doesn't run on even as many
|   platforms as MSIE let alone the 30 or so Mozilla supports. (yes, I know
|   support is not equal across all platforms). Perhaps I'm wrong - with the

Well, Konqueror runs on a lot of platforms which Microsoft doesn't support.
I start counting (in no sepcific order)
1) i586 Linux
2) Itanium Linux
3) PowerPC Linux
4) Alpha Linux
5) ARM Linux
6) i586 FreeBSD
7) i586 OpenBSD
8) i586 NetBSD
9) MIPS - IRIX (SGI)
10) PowerPC - AIX
11) PA RISC - HP-UX
12) Alpha - Tru64
13) SPARC - Solaris
14) SPARC - Linux
15) i586 - BeOS  
16) PowerPC - BeOS
17) IBM s/390 mainframe (wow!)
18) upcoming AMD 64-bit architecture (sledgehammer)
19) Atheos (Konqueror/Embedded)
I am not speaking about Konqueror/Embedded here *in general*, but you can run 
it on any platform where QT/Embedded can be compiled. Plus, with any variant 
of  Mobile/Embedded Linux, where you can run TinyX server. (like Compaq iPAQ)
 
|   port of the Qt toolkit to Mac OS, perhaps Konqueror can at least match
|   Mozilla's three core platforms, but its going up against some very
|   strong niche competitors in the Windows and Mac worlds (Omniweb, Opera,
|   iCab) even disregarding the heavyweights Mozilla and IE.

Yes, you are correct, you can expect Konq on Mac.
But note that Mac is somewhat niche market - it is less than Linux (on 
desktop) here, in Russia.
Apple is still strong in US, but most likely US is the only strong market for 
Apple at a moment. So, there is no way how I can consider Mac platform 
*important* nowdays. Well, it's still here, I wish everything the best to 
MacOS X, but it's niche market.
As about Omniweb, Opera, iCab - they are just *too buggy*.
No offence to developers of any of those browsers, but that's how the things 
are going.
Opera is catching up - but due to closed-source nature of their product, and 
$$$ required to get rid of annoying ads, I doubt it has feasable feature in 
mass market. Anyway, they can be successful in embedded market - like recent 
deal with Canal+ shows. 
I just wonder what can be the reason for commercial company to pay license 
fees to opera for their browser in embedded market, when you can have 
Konq/Embedded for free.
(and yes, I am aware of several companies which use Konq/E)
 
|
|   > Now question is how fast Microsoft can implement CSS3, in case standard
|   > is "fixed". We can only guess here :-)
|   > There is an opinion (from some Konqueror developers)  that some parts
|   > of CSS2 are very difficult to implement, and implementing those parts
|   > just doesn't worth efforts.
|   > Compact, clear, non-misleading CSS3 specification will allow KDE
|   > developers to skip CSS2, so we will get standard-compliant web space
|   > *faster*. As I said, CSS3's modular concept is Good. We just  need to
|   > define good set of modules, and fix all bugs in those modules ASAP.
|   > And get commitment from Netscape and Microsoft to implement those
|   > modules. But it's already another story ...
|
|   I'm curious as to why you appear to think its more likely Konqueror will
|   implement CSS3 than Mozilla. Note I say "mozilla" rather than
|   "netscape". Its hardly as if Mozilla's future plans are any secret. Now,
|   you might be arguing Konqueror will make quicker progress than Mozilla
|   towards full support for CSS-2 or 3, which may be true, but I think you
|   will have to look long and hard to find an organization who has commited
|   to try to deliver standards support than Mozilla. If you believe they'll
|   fail, that's one thing, if you don't believe they will try, I'm curious
|   as to why you might think that.

You mean, wether I think that Mozilla will fail? I have mixed opinion of it.
In its current state, I doubt Mozilla can get significant market share.
But project will not be shutted-down. Most likely, it will continue to 
develop on its own, even when Netscape completely abandoned its support and 
cutted all Netscape-related paid jobs.
Good example is Eazel/Nautilus. I was monitoring that company from the 
beginning, and what happened was quite in line with my predictions.
Eazel was bancrupted, Nautilus is open-sourced, so it has some on-going 
development. Additions are minimal, but it still has some improvement.
Will it catch up on Konq? I hardly belive so. But Redhat commited some fixes 
to nautilus, just to make it usable for RH 7.2 release.

As about CSS3 support - I have strong intention to support its adoption, and 
in particular, CSS implementation in Konq/KHTML.
Well, current CSS support in Konq is better than in Mozilla, while Mozilla is 
about 4 years under developement and Konq - still not 1 year old.
Biggest problem for Mozilla is not standards support (where it's pretty good 
IMO) but :
* startup speed
* open new window speed
* close window/browser speed
* memory footprint
* time to restore from minimized state

Can Mozilla hackers fix those things? I don't know. Only time will proove.
If they can fix it - why they didn't do so during last 3 years?
  
what Mozilla also lacks is a strong Vision.
Why they are doing browser, after all? To strike Microsoft? Hah, what a goal 
;-)
With such goal, they will fail.
To get best standards-compliant browser? Well, then where is *completed* CSS2 
support in Mozilla? I am not interested in 6-year old technologies (CSS1)

Most likely, Mozilla also lack strong leader. I don't know why it is so, and 
*how* it happened. But without strong leader, they will fail.
I can tell you that my ex-colleague was meeting with Steve Balmer 
(Microsoft), and was very excited about Steve. In contrast, his opinion about 
other famous leader, B.Gates, was not soo good.
So I guess while Steve Balmer rules MS, there is nothing that Mozilla/NS can 
do to make MS bleeding.

-- 

Vadim Plessky
http://kde2.newmail.ru  (English)
33 Window Decorations and 6 Widget Styles for KDE
http://kde2.newmail.ru/kde_themes.html
KDE mini-Themes
http://kde2.newmail.ru/themes/

Received on Thursday, 25 October 2001 13:26:29 UTC