Is coding for NS4 a waste of time???

[ChuckRW]

Am I wasting my time coding for NS4?
Just looking for some feedback or to get a feel for how many people really still use that browser...
It seems in all of my few projects that I spend at least 60% of the total project time trying to make NS4.6 do what IE and NS6+ do easily.
What do you think?..........
Chuck

[beetle]

Yes. Without a shred of a doubt in my mind you are wasting your time. I've been in this business over 2 years and have not once accomodated NS4. It's been a long while, but NS4 is going down.

Check this (http://www.scottandrew.com/ns4.html) :D

[whammy]

Yes. Unless the client absolutely insists on it (and I will try to talk them out of it!).

[Vladdy]

Can not help but throw my YES in here.
Here is a good one I told a few clients of mine when they asked for NS4 support (I modify the numbers based on your 60% time)
"Does less than 5% of internet users justify 250% increase in the project cost and delivery schedule?"
Have not heard a positive response yet :D :D :D

[whammy]

Heh... I like your signutare is totally appropriate to this thread, as well Vladdy. :D

[ChuckRW]

EXACTLY what I wanted to hear.
Thanks guys!
If I have to I'll use Vlads %s and quote on the client, though I'm sure it wont be necessary :cool:
Chuck
p.s. Beetle, nice link

[CitznFish]

I DON'T think so. NS 4.7 users are the majority of NS users on our site. Here is a break-down of visitors browsers and OS's for the site I work on. We get millions of visitors and it is a family oriented site. These figures are for the first 6 hours of today, but are representative of the overall breakdown of browser type and OS type....

Netscape 4.7 is still leading the charge for the NS crowd.

[CitznFish]

OS war

[beetle]

CitznFish

Yes, and those numbers seem consitent with what I've seen at thecounter.com. However, the total number of Pre-gecko Netscape users you have is still around 5%, which fits with Vladdy's quote quite nicely. I don't think the question was "Is NS4 the most used NS browser?" but rather "Is NS4 a major concern for development resources?".

P.S. What program do you use for those stats...those graphs look nice! :D

[CitznFish]

Originally posted by beetle
CitznFish

Yes, and those numbers seem consitent with what I've seen at thecounter.com. However, the total number of Pre-gecko Netscape users you have is still around 5%, which fits with Vladdy's quote quite nicely. I don't think the question was "Is NS4 the most used NS browser?" but rather "Is NS4 a major concern for development resources?".

P.S. What program do you use for those stats...those graphs look nice! :D


Those are from Hitbox Enterprise edition. While pre-gecko users are still lower than 5% if a site has 1,000,000 visitors a week that is still 50,000 people who will have trouble viewing a site. In our case that is 50,000 less impressions used to fill ad garauntees. :)

[Vladdy]

Originally posted by CitznFish



Those are from Hitbox Enterprise edition. While pre-gecko users are still lower than 5% if a site has 1,000,000 visitors a week that is still 50,000 people who will have trouble viewing a site. In our case that is 50,000 less impressions used to fill ad garauntees. :)
Having a site that has the same look and functionality when viewed with NS4.7 as it does in mordern browsers is COMPLETELY different from having a site that is accessible by NS4.7 and other non-compliant browsers.
My point is that trying to achieve the same looks and functionality is a waste of resourses. With correct coding the site can still be accessible by anything else from Lynx to Opera (even if in a basic text mode)

[zoobie]

Yes...NS6 & 7, too. :D

[beetle]

Originally posted by Vladdy

Having a site that has the same look and functionality when viewed with NS4.7 as it does in mordern browsers is COMPLETELY different from having a site that is accessible by NS4.7 and other non-compliant browsers.
My point is that trying to achieve the same looks and functionality is a waste of resourses. With correct coding the site can still be accessible by anything else from Lynx to Opera (even if in a basic text mode) Absolutely. Consider my new site (not done) as an example

http://www.peterbailey.net/new/site.php?page=why
Fully featured site. Should render identical (mostly) in IE5.5+, Gecko, Opera 7 (maybe 6, dunno), and IE5+ for Mac

http://www.peterbailey.net/new/site.php?page=why&css=0
Same content, no CSS for browsers like NS4

http://www.delorie.com/web/lynxview.cgi?url=http://www.peterbailey.net/new/site.php?page=why
How the same page looks in Lynx, a text-only browser

http://validator.w3.org/check?uri=http%3A%2F%2Fwww.peterbailey.net%2Fnew%2Fsite.php%3Fpage%3Dwhy&charset=%28detect+automatically%29&doctype=Inline
All this, and it's valid XHTML :D

[whammy]

I totally agree... the site must be ACCESSIBLE in any browser including NS 4.x (even a PDA, hopefully!), but to make it look "pretty" in NS 4.x for the sake of making people happy that obviously will have all kinds of errors and ugly websites whenever they surf the 'net...

Don't bother.

I'm sticking with the standards - that's the only way to convince $MS, NS, and other browser "makers" to quit coming up with proprietary garbage.

[jkd]

Originally posted by whammy
I'm sticking with the standards - that's the only way to convince $MS, NS, and other browser "makers" to quit coming up with proprietary garbage.

You'll find that all non-MS browser makers are doing bang up jobs of implementing standards. Netscape started Mozilla, the most standards-compliant web browser in existence. Opera Software has Opera 7, which supports CSS2 even better than Mozilla (just barely ;)). And the KDE team has Konqueror, which can hold its own.

Then you have MS's IE6/Win which is currently ruining the potential of applying these awesome web standards, but even they are improving bit by bit. IE5/Mac is an excellent product, and I can only hope IE7/Win can at least match IE5/Mac.

[whammy]

Yeah, I'm optimistic too... it looks like things might be finally coming together a bit. :)

It beats the heck out of what was going on 2 years ago, anyway. :D

[jkd]

Originally posted by whammy
It beats the heck out of what was going on 2 years ago, anyway. :D

:: groan :: ;)

[realisis]

"I'm sticking with the standards - that's the only way to convince $MS, NS, and other browser "makers" to quit coming up with proprietary garbage."

Whammy, in this and other posts, I get the impression you would label anything that is proprietary as being ipso facto garbage - purely by definition. If so, that's just another way of saying that only the W3C can come up with valid ideas, or that no-one should be allowed to proceed with an innovation unless they first have permission (ie, standardization) from the W3C. And Political Correctness arrives on the web-scene.

I have no problem with independent innovation - keep the good ideas coming everyone.

I also do not believe the W3C is infallible, or the sole guardian of insight. Where do you think the W3C would be today if IE hadn't invented styles in the first place?

[jkd]

Originally posted by realisis
Where do you think the W3C would be today if IE hadn't invented styles in the first place?

IE didn't invent Styles. The W3C did. IE was the first browser to provide a minimally decent implementation however.

The W3C Committees have industry representatives. Employees from Microsoft, Netscape, Sun, etc all contribute to the specs. Tim Berners-Lee just doesn't say, "Let there be CSS!" and expect the browser makers to agree. Instead, its more along the lines of, "There should be some way to do this! What are your ideas?" :)

And there is always room for proprietary innovation, as long as people who use it are made clearly aware of the fact, and what sacrifices it implies.

But I don't see the reasons behind using proprietary "garbage" over a well-defined web standard, such as VML over SVG, or document.all over DOM for example.

[realisis]

jkd, thanks for your good-natured response. Without turning this into too much of a polemic:

"IE didn't invent Styles. The W3C did."

Well, unless I am mistaken, IE first used styles in IE 3, and the technique was a direct offshoot of Microsoft's MS-WORD, where the technique was first used, and which predated its acceptance by the W3C and its subsequent use on the web, no?

"IE was the first browser to provide a minimally decent implementation however. "

No - IE was the first browser to provide any implementation of styles. Period.

...

"And there is always room for proprietary innovation, as long as people who use it are made clearly aware of the fact, and what sacrifices it implies."

Always room perhaps, but then that is your personal position on the issue, not necessarily the one I originally contested (which is whammy's).

But your statement sets up a false alternative: no sacrifices are necessary - a browser can both conform to a standard AND provide alternate techniques for accessing or manipulating a given behaviour - something at which I believe IE generally excels. And in cases where no standard yet exists, obviously no sacrifices are needed, since the "new" spec represents an advance and a possibility which is not otherwise possible.

"or document.all over DOM for example."

Straw man again. Since document.all chronologically *preceded* DOM, I'm not surprised it exists. But again, you are giving a one-or-the-other scenario which is not in fact the case: MS doesn't stipulate doc.all *OVER* DOM - it responds to both (as of IE 5 at least). It implements the standard and allows for a proprietary alternative.

...

"Tim Berners-Lee just doesn't say, "Let there be CSS!" " (etc)

Yes you're right of course. I have no beef with the W3C itself. But I see a fair number of posts (on this board and others) which are entirely uncritical of any W3C pronouncement, while at the same time being hyper-critical of any alternative or deviation from the "allowable" norms. "You're not supposed to do that!! It's against the rules!!!"

When I see opinions expressed as discussed above this way, by yourself and others who obviously have a *lot* of expertise to offer - the posts give me the effect of being simply an attempt at polarizing opinions on a given issue among a readership, and at elevating personal preferences into political crusades. This sort of evangelism adds a cultish flavour of religious fundamentalism to the discussion which seems suspicious and misplaced to me, given that the act and style of coding a web-page is an entirely voluntary one, that a surfer's choice of browser is an entirely voluntary one, that their decision on whether or not to visit a given web page is an entirely voluntary one, and the visit is also free of any charges, to boot.

Polarized opinions like those being discussed are best left to more important issues in life: you know, things like murder, political repression, exploitation, etc - situations in which voluntary participation and compliance is NOT sought. Folks we're talking web-pages here.

[brothercake]

It's certainly true that opinions vary. There are those who reject IE's proprietary features (such as VML) out of hand, purely because they are MS, while embracing other proprietary features (such as XUL).

Similarly, with the w3c versus proprietary hacks argument. The w3c specs have many holes, many omissions, and many aspects which are just plain stupid; inevtiably browser vendors will attempt to "fix" them - after all, it would be insane to support an unworkable standard, while refusing to provide a functional but proprietary alternative, simply because its proprietary.

That kind of evangelism is counter-productive. But as with anything - only to oneself.

I tend to the view that the only standards that matter are the de-facto standards that people actually use. I'm not overjoyed that MS is dictating the de-facto standards, but that's largely irrelevant. As long as most people use IE, its standards are more important than those which are proscribed by the w3c.

I think as long as we all remember not to take anything personally, polarised views can continue to flow without fear of offense.

[jkd]

Originally posted by realisis
"IE was the first browser to provide a minimally decent implementation however. "

No - IE was the first browser to provide any implementation of styles. Period.

Just for the sake of this interesting debate, I'm going to argue that NS4 was the first browser to bring a decent styling system. JSSS (Javascript StyleSheets), if anyone recalls. And NS4's CSS support was basically a (bad) conversion over from JSSS. IE4 was the first bare minimum CSS implementation I could have worked with.

I am unaware of any styling that IE3 supported, but an informative link would be much appreciated to the contrary. :)

[realisis]

brothercake, thanks for the input, and also to jkd.

Your last point reminded me of something: I haven't been hanging around at CF as long as you veterans obviously, but I am continually impressed at the utter lack of flames and personal attacks found on this board. I also gladly admit that the expertise shared in these threads is top-notch - I have learned so much already in my time here, and CF is my favourite board.

So in that sense jkd and the other moderators and long-time posters do a very good job of providing a proper example, and the board remains focused on the exchange of valuable knowledge.

When I originally posted I was just musing on the existence of perceived attitudes and issues. No offense intended, and none taken I hope.

Cheers -

PS: jkd "I am unaware of any styling that IE3 supported" ...

So was I until about 2 weeks ago. I was hunting around for something on the MSDN site, and noticed references to styled elements in IE3... I'll see if I can hunt up any relevant references.

[jkd]

Originally posted by realisis
So was I until about 2 weeks ago. I was hunting around for something on the MSDN site, and noticed references to styled elements in IE3... I'll see if I can hunt up any relevant references.

Quick search on google turns up:
http://www.endoframe.com/css/ie3.html
and
http://www.w3.org/Style/CSS/msie/

Apparently IE3 did have very limited CSS support. It didn't cascade well, nor support much, but at least it was enough to cut down on the <font> tags running rampant...

But CSS was already in existence before IE3 implemented part of it, so I still disagree with the statement that MS invented the concept of styling webpages. ;)

[brothercake]

Interesting ... I noticed from my own testing that IE3 supported font-family for common elements like <p> and <td>, and text-decoration for links. Makes things a bit prettier ... If only we'd known then what we know now :)

[jkd]

Originally posted by brothercake
If only we'd known then what we know now :)

Then I'd be complaining how IE3 doesn't support any of DOM1, DOM2, CSS2, most of CSS1, XHTML, etc etc ;).

[brothercake]

:D :D

[krycek]

...just to add another flame to the fire of this riveting discussion (no sarcasm intended - I have been following this thread with great interest! :D) I would like to point out mouse buttons.

What about them?

Well, just that the ONLY logical means of dealing with mouse button values is the Microsoft way. The W3C way is totally, utterly stupid... for a start, it cannot detect more than one button being pressed, whereas the MS model can.

Not that I am saying Microsoft is brill or anything, but I do occasionally wonder where the W3C have put their brain cells and logic caps :)

::] krycek [::

[whammy]

Heh... no offense taken here. I just don't like proprietary "garbage" when it ends up the way javascript was for the last few years... having to code for all kinds of different things, i.e. document.all, document.layers, still no good actual browser detection method to make it a bit easier to deal with, etc.

No biggie to me really, a solution can be found for anything just about - but it's a pain if you're on a schedule sometimes. ;)

It really doesn't bother me personally too much (since I have a tendency to do things as simply as possible and avoid proprietary things - if you don't count the fact that I use ASP (and shortly ASP.NET) server-side), but stuff as simple as "marquee" tags (yech) are an endless source of confusion to newbies who ask "why doesn't this work in all browsers"? etc.

P.S. I actually like some of Microsoft's proprietary server-side "garbage", though. VBScript may be weak, but it's also very useful at times - and the Date functions in VBScript beat the heck out of JavaScript, although nowhere near as handy as the Date class in .NET.

.NET is not weak at all from what I've messed with and heard from the people in the classes ahead of me, though (finally it looks like MS did something right (only over half a decade later than Sun) as it's true OOP, everything is compiled, and from all comparisons I've seen, much faster than anything else out there).

The only thing that bugs me about .NET is the majority of the books, etc. are either in Visual Basic .net or C# - both proprietary languages... sigh. But it pays well, and at least some things (like XML Web Services readable by any platform you can think of), are very easily done with it...

:D

[jkd]

Originally posted by whammy
The only thing that bugs me about .NET is the majority of the books, etc. are either in Visual Basic .net or C# - both proprietary languages

Actually C# is undergoing standardization by ECMA if I remember correctly. And there are 3rd party C# (well, .NET in general) implementations for Linux. So in reality, C# isn't proprietary ;).

[whammy]

That's a plus. :)

[jkd]

Originally posted by krycek
Well, just that the ONLY logical means of dealing with mouse button values is the Microsoft way. The W3C way is totally, utterly stupid... for a start, it cannot detect more than one button being pressed, whereas the MS model can.

And why would button-mashing be a required capability for scripts? Firstly, not all pointing devices are capable of such things, secondly, nor are all users (read: disabilities).

Secondly, when the left and right mouse buttons are clicked simultaneously, it is typically considered a middle-click for mice that don't have a middle mouse button. And middle clicks are handled just fine in the W3C DOM2 Events interface.

If you're going to call it stupid, first please form some valid arguments. And then second, why don't you add your (constructive) input to W3C mailing lists? Your input can influence future drafts. As proof of concept, something I brought up about the SVG DOM implementation and how you can't progmatically create SVGZoomEvent's was mentioned to the group, and lo and behold you'll probably see that change in SVG 1.2.

[whammy]

I know how that works, as I absolutely LOVE TextPad over any text editor I have seen. But I was baffled that I couldn't find where to "repeat the last action performed" anywhere!

When I emailed tech support about it, they admitted it wasn't currently possible, and promised to put that in their hat for future improvements. Hope they do, it's about the only power that text editor lacks (that I use anyway) at the moment.

Constructive input is a great thing.

:)

[krycek]

Originally posted by jkd


And why would button-mashing be a required capability for scripts? Firstly, not all pointing devices are capable of such things, secondly, nor are all users (read: disabilities).

Secondly, when the left and right mouse buttons are clicked simultaneously, it is typically considered a middle-click for mice that don't have a middle mouse button. And middle clicks are handled just fine in the W3C DOM2 Events interface.

If you're going to call it stupid, first please form some valid arguments. And then second, why don't you add your (constructive) input to W3C mailing lists? Your input can influence future drafts. As proof of concept, something I brought up about the SVG DOM implementation and how you can't progmatically create SVGZoomEvent's was mentioned to the group, and lo and behold you'll probably see that change in SVG 1.2.

OMG... :D See how an*innocent* ;) comment can cause another round of debates!

Well, firstly, jkd, I do actually let the W3C know what I think, as do many other people. So, when you said "why don't you..." you were assuming, incorrectly, that I don't :)

Second, I didn't list everything bad about the mouse handling simply because I was throwing it in as an example; it was not meant to start another debate. But, seeing as it did...

...There are two properties for finding out which mouse button has been clicked: 'which' and 'button'. These properties don’t always work on a click event. To safely detect a mouse button you have to use the mousedown or mouseup events.

'which' is an old Netscape property. Left button gives a value of 1, middle button (or mouse wheel) gives 2, right button gives 3. No problems, except its meager support (and the fact that it’s also used for key detection).

Now 'button' has been fouled up beyond all recognition. According to W3C its values should be:

Left button – 0
Middle button – 1
Right button – 2

According to Microsoft its values should be:

Left button – 1
Middle button – 4
Right button – 2

No doubt the Microsoft model is better than W3C’s. 0 should mean “no button pressed”, anything else is illogical.

Besides, only in the Microsoft model button values can be combined, so that 5 would mean “left and middle button”. Not even Explorer 6 actually supports this yet, but in the W3C model such a combination is theoretically impossible: you can never know whether the left button was also clicked.

In my opinion W3C has made some serious mistakes in defining button.

Oh, and the thing about button mashing? It is always best to present the widest possible (logical!) choices to the programmer, so the he/she can set up error handling etc. because otherwise they may want to do something that is either impossible, treated the wrong way, or assumed to be an error etc.

...does that answer your questions? :D

::] krycek [::

[realisis]

jkd: thanks for the links you posted. According to material presented therein, Hakon Lie was the one who initiated the STYLES format... I shoulda known, and so I stand happily corrected.

I see this thread is moving right along, and very interesting too. I'll check in again later though, I'm in sort of a rush at the moment.