Forum Moderators: open
Should I even care about NN4 users and just include a sniffer at hte top of my page and say something to the effect of
"your browser is a load of poo, go get a new one"
;-)
Any resources that detail the current user stats for browser.
Cheers
The trick though is to make sites usable in all (or at least most) browser, although it might not be necessary to provide all with the very same look.
If you can't achieve this, at least say sorry to those using a browser with incapabilities. This depends on why your site looks ugly in N4. If it is because of extensive use of CSS you might look for "@import" in site search and you'll get the trick. If it is something else, you might want to sniff. If it is because your code is invalid and you just don't care, don't go write something like "your browser is a piece of sh*t, go away", because then the fault is yours.
Whatever you do, be kind to N4 (and other browsers-) users and write something polite.
That was a tongue in cheek remark, politeness is essential at all times. My problem is the state of the UK academic institutions who still see fit to use NN4, but hopefully with time it will become less widely used.
Regards
CSS Rounded Corners [albin.net]
Then follow the link at the top of the page.
Seems polite to me.
However much we designers are eager to get CSS off the ground properly, if you tell your visitors their browsers suck and they should get new ones, they won't. They will go to your competitor's cross-browser compatible site and you will never see them again.
That may not be nice, but it's a fact of life.
dedicated to the downfall of NS4..
Sign me up! ;) ;)
On my sites, if it can be done cross browser, that's what I strive for.
Otherwise I make sure the site degrades reasonably.
I personally wouldn't link to webstandards in this way, preferring to wait it out.
Related:
[lists.w3.org...]
rewboss said....
it's actually very impolite, and very silly.
Okay, I can buy that and accept that it's reason enough not to do it, but I think there might be another problem too. If you do this, then the SE spiders are going to parse it similarly to the way it's rendered in NS. So if it extracts the first bit of text from your page to use along with what you have in your meta description, it's going to use text of your upgrade message. It's like all of these sites where the summary extract says "You do not have a frames-compatible browser". Now the summary will say "Your browser stinks, please upgrade."
Couldn't that be a problem? Wouldn't it be better if it said "Your internet source for widgets"? Is there a way around it? You can't put the browser upgrade message lower in the html and use absolute positioning to get it higher on the page, because that would pretty much defeat the purpose. I can think of two possible ways around this problem (neither of which address rewboss' concerns).
1. sniff for the browser. In that case, what do you sniff for? Just NS4 or also Lynx, palms and so forth? This brings you back to sniffing and serving up different pages for each browser and depends on proper ID and so on and so on. Too complex for me honestly unless I had a high-traffic site with a huge budget, but even then I don't really like this approach.
2. Make your upgrade "text" a gif image with a blank alt attribute, so that it would validate, would not show up in SE or stylesheet compatible browsers, but would show up in anything that ignored the CSS. Of course, now you've added several KB to your page with the sole purpose of suggesting that your users leave.
In any case, it still means that any user who accesses the site, for whatever reason, with something other than a full-blown modern browser gets told to leave as the first thing....
As far as I know, there's no way to find out whether a given CSS property/attribute is available by sniffing is there (like how you can test to see whether a DOM object is available)? In other words, you need to know which *browser* supports the feature and then sniff for the *browser* not the *feature availability* correct?
I'm glad this topic came up - can't wait to see what others have to say.
Nick
You need to make sure that you position your html content properly so that it degrades respectfully and presents the user with the core content first. They could care less about whats at the bottom of the page.
When you start using all the latest bells and whistles in CSS is where the problems come in. The @import command was a savior in our case. We feed NN4.x users a site that looks almost identical with a few exceptions. Those exceptions probably would not be noticed by the visitor unless they had IE and NN open at the same time.
Since late last fall and throughout the winter and into early spring, NN4 was maintaining a global presence of slightly over 4.5% (as reported by several large Web statistic reporting resources).
The same sources have been reporting slow, but steady growth for NN6.x (until recently) and an ever accelerating adoption rate for the Opera Browser.
IE6 has shown a very high adoption rate, though it still trails IE5.x by approximately 10%.
Suddenly in May, the relatively steady trends spiked in several areas: first, IE 5.0 was reported showing an actual increase; it would not seem likely that M$ users would be upgrading from IE4 to IE5 rather than the current v.6, nor did the IE4 graph indicate the possibility. Neither would it seem likely that those migrating from other browsers to IE would chose an older version. One possibility that does come to mind is that Opera is set to ID as IE5.0 by default (Windows version). In any case, it is an interesting spike.
The big surprise however is a sudden and sharp downturn in NN4's reported stats. One large stats resource reports NN4 has fallen to slightly over 3% (11.1 million visitors) as measured against 348.8 million (total reported) visitors for the month of July.
Also for the first time, Opera has surpassed NN6.x (2.8 million) and is now statistically, just under 1% total visitors (Opera 3.3 million / 348.8 million). Do keep in mind however, that Opera may also be contributing to the IE5.0 upsurge, so the actual Opera increase may be significantly under reported.
What is also interesting is that a general category, Netscape Compatibles, is also losing ground. M$IE's 91% share has remained somewhat stable with the majority of movement in that category hase been the rise in IE6 numbers and the mysterious increase in IE5.0 stats (IE5.5 reveals a steady downward trend).
A graph of NN4.x's stats illustrates the sudden downward spike. From a purely speculative and reflective perspective, I can't help but wonder if the steady increase of Web developers who have migrated to advanced CSS techniques hasn't had a marked influence on NN4's decline. Once you taste the power of CSS and the logic of seperation of presentation and content, it is hard to go back.
Obviously and realistically, there will be pocket's of NN4 users with us for quite some time yet, but, at least for now the rank and file seem to be leaving the old ways behind.
- aural browsers and CSS-p. How will they treat "display: none" since presumably they don't display anything at all? How about absolute positioning?
- WAP - what happens? I've looked at some WAP simulators and they seem to parse CSS-p correctly (though the whole first screen is taken up by a fairly small logo). I don't know how typical that is.
Anyone have any comments/insights?
Tom
I have a site that works well in IE, Nescape 7 and Opera, but when I view it in NN4 it looks like a bag of nails.
Hee! Thanks for the visual.
Have you looked to see how many of your users are using NN4.x? This guides my own consideration of NN4 in site design.
Currently about 5% of my customers use NN4 ... for me, 5% of my daily customers is a pretty decent number. I don't want to alienate those folks, so I'm going to continue to design for NN4, even though it KILLS me to do so.
:)
As soon as that number drops below 2% for 3+ months, though, I'm going to get stinking drunk, do a jig, and start tweaking my site.
I'm very close to the stinking drunk myself. It surely will make work a lot easier if they can be punted.
www.gics.gov.uk
Not sure exactly how they do it or how practical it would be over large scale sites.
Ergophobe: As far as I know, there's no way to find out whether a given CSS property/attribute is available by sniffing is there (like how you can test to see whether a DOM object is available)?
[xs4all.nl ]
3% NN4 may seem small, but if you're getting 10million page requests a month thats 300,000 chances to display an optimal page for selling/generating leads/whatever to a customer that you're forgoing.
I try not to pass up 1 single chance to make money for my clients :)
"your browser is a load of poo, go get a new one" - you must put a polite message. I was looking around at new shopping cart software and looked at a major site with a high page rank. I decided to look at some of their example sites and the first one gave me the message (in a huge graphic) "F*** Netscape".
I emailed the shopping cart software firm and requested that they looked at the site in NS. Within 24 hours their link had been dropped from a high page rank page and the shopping software firm emailed me to thank me. If you are going to be offensive, even to a small amount of people, you may lose important inbound links. (It wasn't a competitor, but I don't see why they should insult me, cause another companies reputation to be harmed and get a high page rank when I can do something about it)
Ergophobe: As far as I know, there's no way to find out whether a given CSS property/attribute is available by sniffing is there (like how you can test to see whether a DOM object is available)?[xs4all.nl...]
You got my hopes up, but I think you misunderstood what I was looking for. What I was trying to say in the last post is that while it is possible to check for DOM/JS objects and then adjust scripts accordingly rather than sniffing for browsers, as far as I know there is no way to do this with CSS properties and attributes.
I went to the page that you suggested, but all I saw was a how-to on testing for JS/DOM objects. Did I miss something?
Thanks,
Tom
