Forum Moderators: open
Last year I posted that most of my sites statistics were showing Netscape 4
users around 30%. Well, a year later, they are still running about 30% - 29% to
be more accurate. Granted, most of my sites are informational in nature:
Municipal information, Transit Authority, theater resource, etc., and my draw
more on users such as schools and libraries. But I think it is important for
everyone to keep backwards compatibility in mind.
I still do not think you can force people to upgrade. People stick with what they
are most comfortable with. I can also say factually that some sites I deal with in
my “day job” require Netscape 4. Granted, these sites deal with government
related information, but due to the high cost of upgrading, most do not change.
So when our office upgrade to Windows XP, we had to install Netscape 4.
Take this rant for what it is worth - just an FYI.
(Please don't take this as an opportunity to list how great various browsers are :) If they are that great why aren't folks upgrading at a faster rate?)
You didn't miss it, as it isn't there - if the browser doesn't understand the <link rel="stylesheet"> tag the page just looks like the barebones version - no tables, just <hr> for structure, and all link and text formatting is the default.
The direct link is just a basic on/off demo and i haven't really gone through their stylesheets and js-includes, so they might even cater to specific browsers within them. I only posted it as an example of a page that did exactly the same things but looked extremely different to a particular (non-CSS) browser.
/claus
These four UA's were the total amount of non-IE and non-spiders:
- Mozilla/5.0 (Macintosh; U; PPC Mac OS X; da-dk) AppleWebKit/85 (KHTML, like Gecko) Safari/85
- Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.2) Gecko/20021216
- Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4b) Gecko/20030516 Mozilla Firebird/0.6
- Mozilla/5.0 (compatible; Konqueror/3.1-rc4; i686 Linux; 20020222)
The grand total for all four was 1.6%. Like all other stats, this is just a sample and your figures may be different.
/claus
IE dominates now and I agree that we must cater for it. But to do so exclusively breaks the idea of the web. It is perfectly feasible to write code that degrades so everyone, including screen readers for the blind, can access the content.
You are all are so busy justifying your desire to code for the stupidest common denominator(IE) and nothing else that none of you are willing to help someone actually make something work in that browser that WOULD work if IE wasn't such a shoddy piece of garbage. I'm referring to the thread entitled "onchange works for Mozilla, nowhere else". I have experimented further with that code and so far I have made it work in all the browsers mentioned except IE.
Have any of you ever considered that if you aren't getting a lot of Netscape/Mozilla visitors it's because you have done something that works in IE but NOT in a standards-compliant browser, and that Netscape/Mozilla/Opera users have simply written off your site as being coded too ineptly for them to bother with?
Anyway, I don't care anymore about getting my js to work in IE. I'll put a message on my site saying that if you're using IE and something doesn't work then get a real browser and take that stupid "Hack Me" sign off of your forehead.
</rant>
DA
>> everyone here seems so enamored with the idea of promoting IE
- i've posted a few lines in this thread and also read all the others. I really don't see everyone promoting IE - on the contrary, there are quite a few posts claiming that standards compliance is important.
It's a fact that the overall market share for the alternatives is currently low, but it's not certain that it will be so forever. As the thread headline (and also quite a few posts) says, there's still areas where cross-browser support is really crucial and the alternative share is very high.
/claus
I have recently taken on an informational site based at a university, dealing with a very specialised subject. As you can imagine, with a high proportion of users from educational institutions, there is a much higher number of NS4 (on Unix!) users than in the general audience. This notwithstanding, on such sites, the content is the key, not just the layout, and I'm catering for my NS4 users by giving them completely unstyled content, by using the @import method. The result - happy users all round. Those with modern browsers appreciate the design, and those with older browsers / computers, etc. love the speed and the accessibility. The site works great in Lynx, on a phone, whatever. I don't consider that I'm abandoning support for NS4 - those who keep such a browser aren't interested in the layout, they just want the facts - and I make sure they are well catered for.
Incidentally, one big advantage of NS4 is it's small memory footprint. Why should I have to have the latest Pentium 4 just to surf the web? Checking what RAM resources are used by browsers (with just one window open), I get:
Mozilla Firebird 0.6.1: 25Mb RAM
Opera 7.11: 31Mb
and Netscape 4.79: 8Mb.
I'm a Firebird user, but on a lower-powered machine, NS4 remains my first choice (or Links/Lynx if it is a really basic machine). NS4's email client is great too. When I use NS4, I don't want style, just substance. Dump the CSS, Javascript and such, and feed it to me as unstyled text on a gray background, and I'm happy. From my experience, most NS4 users would agree with me too.
Many of my sites are informational in nature: governmental type stuff. I design them similar to what you said. I make a nice css layout and a corresonding "Text Only" page using a different style sheet, usually with a larger font setting, in percentages, so the viewer can adjust it. I simply take the content from the "standard" page and paste it in the "Text Only" page. I convert all graphic links to text links and the page works on any browser - plain, simple, straight forward text - no table, div, etc., and I put a link rel tag in the head. If the viewer wants fancy, they can go to the standard page. If there is an image that may be important, such as a map, I put a note on the Text Only page to view the Standard Page to see the image. Works well and it's fully ADA accessable. And it only adds a few minutes of work.
If anyone is wondering why I still support NN4, one site in particular averages 50,000 visitors a month, 27% using NN4. That would be a large chunk to loose.
If anyone is wondering why I still support NN4, one site in particular averages 50,000 visitors a month, 27% using NN4. That would be a large chunk to loose.
Now we're talking. Percentages are all very well, but raw visitor numbers tell you if you need to support NS4 and other "minority" browsers. If you've got 1 million page views a month and 1% NS4 users, that gives you 10,000 potential customers a month that you'd lose if you didn't give them reasonable access to your site. This is why Google and Yahoo both work just fine in NS4, however low the percentage of pageviews in that particular browser.
Stick to your NS4 support and look after your particular customer base.
<added>And WebmasterWorld seems to work just fine to in NS4!</added>
When I use NS4, I don't want style, just substance. Dump the CSS, Javascript and such, and feed it to me as unstyled text on a gray background, and I'm happy. From my experience, most NS4 users would agree with me too.
Be aware that there are several ways to do @import, with different browsers compatible with each one.
On the rare occasion that I use my old Windows 98 machine (with 32MB of RAM etc.), I use Netscape 4.0. It's either that or IE 4.0, and it's not too hard to decide which to go for. I have tried using more recent browsers on there, but they just grind to a halt.
Modern browsers need a lot of memory. I use Mozilla and Win98SE on a 350MHz laptop. With 64MB it was unusable with a couple of windows open or if trying to save something. With 192 MB it fliiiiiieeeeeesssssssss. NN4 is quick, but falls over on almost any page with CSS in the head, or inline, or linked.
That's not entirely correct. If you've got 1% NS4 users you've got one percent NS4 users no matter how many pages they are viewing. By using percentages (as in 10K of 1M pageviews) you simply assume that the users of one particular browser will tend to view as many pages as the ones using other browsers, on average.
In your example, if you have 10K NN4 pageviews out of a total of 1 million, then you probably have 1% NN4 users, but that's only equal to 10K users if all users on average view only one page per month.
In stead you should take your stats for Unique Visitors (if you are tracking using cookies) or the stats for Unique Sites (if you are tracking using IP-based log files) and multiply this total number by 1% to get your raw total number of NS4 visitors.
If there are three pageviews per visitor on average, your 10K pageviews will translate to 3.3K visitors (which is still a lot).
/claus
I use the @import for CSS too. It hides the CSS from NN4. Gives them quick loading unstyled pages.
Isn't that the whole point of using CSS: to deliver stylized content to the browsers that support it and if older browsers don't support CSS they can still access the content?
So are those of you on the anti-NN4 podium saying "to hell with stylization for NN4" or are you saying "to hell with accessibility for NN4". The former I agree with, the latter misses the point IMHO.
Because most folks are largely uninformed about the features of other browsers. ...But what is really sad is that any attempts to inform them is generally frowned upon and even out right demonised, such as "best used with" icons. It's a catch-22 situation
I don't think this is demonised, I think it's just pointless and ineffective, and (rightly) regarded as a waste of space. If a person can see the site it's unlikely they'll immediately go off and do a massive download because of an overwhelming desire to see dashed borders ;)
DeadAgain:
get a real browser and take that stupid "Hack Me" sign off of your forehead
All browsers have security issues. Opera 7 started out with some fairly horrendous ones. Security is a process, not a product...
hartlandcat:
90% of posters on this forum (myself included) seem to use some form of Netscape/Mozilla browser.
Wow, really? I use IE because that's what 99% of my clients and customers use. But (of course) I do support all browsers and always have done right since NN2 and Mosaic days (he adds hastily ;)).
Maybe this is the core of the problem with getting users to upgrade: alternative browser makers have focused on what webmasters and techies want (e.g. standards compliance), rather than what normal people want(?)
That's a very good point. In fact, two hours ago i was explaining my girlfriend that to use my PC to go to her favorite site for backgammon, she'd just have to pick whatever of these options she felt like:
1) click the little "e" icon, OR
2) click the little "dragon" icon, OR
3) click the other little "dragon" icon, OR
4) click the little "o" icon, OR
5) click the little "star" icon
I demonstrated it with three different browsers (IE, Moz, Opera) and she responded: "why should i use those others? i'm used to clicking the little 'e', can't i just do that?"
I just had to reply "of course you can." I saw no benefit for her that i was able to tell her about, as the "features" that i saw in the other browsers were just plainly annoying to her ("...no pictures? tabs? why? what is it good for? - if the only difference on the page is the dragon picture in the corner in stead of the e why should i use it?")
/claus
I saw no benefit for her that i was able to tell her about
You might have explained that a lot of improvements are under the bonnet. Imagine two cars that look the same but one has a more advanced engine.
Familiarity though is likely to be the biggest barrier against people changing from IE. The same goes for NS4. Why would someone used to it want to learn a whole new program? Of course the change is worth it IMO.
You might have noticed that there's no little "n" icon in my list. It's because it has load times that are even worse than those of Moz, so the shortcut is not found the same place as the other five, but i didn't really want to start telling her that as well.
I don't think this is demonised, I think it's just pointless
and ineffective,
and (rightly) regarded as a waste of space.
If a person can see the site it's unlikely they'll immediately go off and do a massive download because of an overwhelming desire to see dashed borders ;)
Granted the end user isn't going to download a new browser for just one site. But convincing users to switch browsers has to be done as a collective effort by a large number of webmasters.
...convincing users to switch browsers has to be done as a collective effort by a large number of webmasters
I don't mean to be harsh but any such effort seems imho doomed to failure! Zeldman et al tried actually blocking access and even such extreme/ludicrous tactics only had a negligible, if any, effect.
The only collective effort we as webmasters could do that would have any effect would be a black-helicopter opperation, going round individually updating everyone's PCs when they're not looking ;)
We shouldn't have to make a collective effort. The benefits of upgrading should be self-evident...
"This site/page optimized for HTML 4.01 and CSS 2 compliant browsers."
I don't think there are any fully CSS2-compliant browsers, are there?
Plus, software *always* has bugs: there will always be a necessity to test unless there were a strict standard browser html and css test suite, and all browser makers /had/ to achieve compliance before releasing browsers. This would also require clarifying all the grey areas in the w3c specs.
I don't mean to be harsh but any such effort seems imho doomed to failure! Zeldman et al tried actually blocking access and even such extreme/ludicrous tactics only had a negligible, if any, effect.
The only collective effort we as webmasters could do that would have any effect would be a black-helicopter opperation, going round individually updating everyone's PCs when they're not looking ;)
We shouldn't have to make a collective effort. The benefits of upgrading should be self-evident....
I don't think there are any fully CSS2-compliant browsers, are there?
there will always be a necessity to test unless there were a strict standard browser html and css test suite, and all browser makers /had/ to achieve compliance before releasing browsers.
I simply don't support IE below 5.0
I don't do anything to support Netscape 4.x & 6.x - both are dead and gone
Netscape 7.x is the same as Mozilla.
IEmac 5.x is only supported by coincidence - with the advent of Safari that's in my opinion a rather safe bet.
Safari aims to be just as standards compliant as Mozilla so if it's working in Mozilla it's working in Safari.
I develop using Mozilla augmented with at lot of web-developer tools from mozdev and others so I can mess around with what happens with javascript on/off, stylesheets on/off, analysing html-headers, etc. I'm careful to!NOT! use any html, css or javascript that doesn't hold for the smallest set of common features. I validate everything and make sure to have a dtd.
When I've got it working under Mozilla, I get hold of WinTel computers running IE5.0 , IE 5.5 and IE 6.0. Using these, I test to see where the css breaks, and using the IE-only conditional comments I use these to post-load css-files that 'correct' the standard stylesheets.
My browserstats say: 16% IE5.0 , 16% IE5.5 , 66% IE6 with the IE5.x portion steadily diminishing in numbers. IE4.x is at 1% and seems to be mostly search engines, web-proxies and such things. Finally, Gecko-based and -like browsers make up the last 1% of the audience.
Then, I've also got my own, completely private non-commercial website that's just a pet-project that I don't particularly care whether people can access or not. It's coded in XHTML served as application/xhtml-xml which leaves *any* version of IE in the dust. I've got a browser-check on every page in serverside PHP that simply re-directs to a browser-upgrade page if the browser isn't capable of handling application/xhtml+xml. My take on this is that if people want to see my private web-site they can bloody well upgrade to a (fully) standards-compliant browser ;-) And if they are interested enough in my private musings they will!-D
Safari aims to be just as standards compliant as Mozilla so if it's working in Mozilla it's working in Safari.
This is a bad approach. Safari is known for odd bugs which are due to its newness. Pages that work in Mozilla can foul up in Safari. Likewise, pages that work in Safari can foul up in Mozilla. Each browser acts slightly differently and has its fair share of bugs. Even if each browser was 100% CSS2 compliant, there are so many combinations of code you can write that problems can still occur.
...I get hold of WinTel computers running IE5.0 , IE 5.5 and IE 6.0. Using these, I test to see where the css breaks...
Why not just note the common faults such as the broken box model and add hacks to allow for it? That would save you testing.
Safari aims to be just as standards compliant as Mozilla so if it's working in Mozilla it's working in Safari.This is a bad approach. Safari is known for odd bugs which are due to its newness. Pages that work in Mozilla can foul up in Safari. Likewise, pages that work in Safari can foul up in Mozilla. Each browser acts slightly differently and has its fair share of bugs. Even if each browser was 100% CSS2 compliant, there are so many combinations of code you can write that problems can still occur.
Also, I do keep my coding as simple as possible. Some of the things I've recently done have been completely broken in IE5mac, and I didn't do a damn thing to compensate even though I certainly could have. One thing that'll break in IE5mac is having a <div> containing absolutely positioned <img>s postioned on the top of the corners of the <div>
...I get hold of WinTel computers running IE5.0 , IE 5.5 and IE 6.0. Using these, I test to see where the css breaks...Why not just note the common faults such as the broken box model and add hacks to allow for it? That would save you testing.
