Forum Moderators: open
Judging by the continuing downtrend [upsdell.com] in v4 browser usage - and that XHTML 1.0+ is now the defacto-markup standard - I was very interested to know what other developers' thoughts were on this subject.
Personally, I'm leaning towards full XHTML 1.0 compliance using external stylesheets and JavaScripted DHTML events. How about anyone else? Have you taken the "next step" and moved over to Dynamic XHTML? What was the response like from your v4 customers, and were there any issues you hadn't anticipated (or couldn't work around)?
[edited by: NetGrease at 10:50 am (utc) on Nov. 25, 2002]
When presented with a new project, look at the target audience.
Working with the target audience you can guestimate how many people will be affected by using scripting techniques (incl. CSS) that are not understood by old browsers. It really depends on your client and their market. I'm reluctant to ditch support for old browsers, because if your site is a hit and you're getting 1000 unique visits a day and 2% of people can't see your site, that's 20 potential customers lost...
Our customer base uses a slightly larger percentage of v4 browsers than average, but after having coded some extensive (and occaisionally nightmarish) workarounds on a project last year, we weren't 100% happy with the end result. The main criticism was that we weren't able to implement everything we had originally wanted to, or the best results of the workaround needed weren't satisfactory.
My strongest "Against Coding for v4 Browsers" argument is that by encouraging [webstandards.org] the user to update, aren't you doing them a favour in the long run - not only in terms of your site, but the web in general (not to mention the plethora of extra, more interesting dynamic content they could use)?
Also, many organisations that still force their users to surf on old browsers will not allow them to download software (such as new browsers). I accept that these are mainly educational institutions and some draconian businesses - but if a site is hoping for these people as a client-base, then ignoring the constraints of their browsing environment is immediately disenfranchising them.
Alas, I believe that we still have to be stoical when it comes to old browsers and, in certain circumstances, bite the bullet and code for them.
But, like you, our debate on the matter went over and over the arguements both for and against, and in the end we felt that the time is not quite yet here for totally ignoring them.
I'm looking forward to the day when it does come though! ;)
My bank does not support Gecko browsers. If I try to log in to my account using Mozilla, I am directed to a page which explains that because they have not had time to do security testing, yada yada yada, they had no immediate plans to support them, we couldn't do transactions but we could view the pages if we wanted to or use IE or NS4. Fine, I can deal with that.
I came to one shopping site using Mozilla and was redirected to a page entitled Your Browser Is Obsolete. This was about three screenfuls of arguments why I needed to dump Netscape 4 and why they refused to support it. Well, I was looking for a hard drive, not technology evangelism, and certainly not from someone with an outdated sniffer who wouldn't even let me see their home page much less order anything. Needless to say, I did not bother to revisit.
[edited by: choster at 4:51 pm (utc) on Nov. 25, 2002]
Unfortunately, a lot of the time they dig in their heels and insist on v4 support - why, I don't know! It's an uphill battle!
Part of the rationale for keeping things readable and usable in N4 is that for various reasons that have been re-hashed in many threads here, N4 is still going to be used for a while. Another part of it, and the reason I also test in a text-only browser, is that there are lots of non-standard technologies that get used to view web sites, and they can render quite differently. I know people who have to browse using text-to-speech screen readers. I know a few people, including myself, who at least occasionally browse using a cell phone. Cell phone browsers are pretty much universally less capable than desktop browsers. They may not be common, but they are likely to be used by people with disposable income and a high degree of comfort with the internet, and therefore with internet sales. Lynx in particular seems to give me a good idea of how my pages are likely to be rendered on a cell phone browser.
As for DHTML, I'm exceedingly wary of it. Most of the time, I don't think it adds usability to a site. It won't work in my cell phone, in my text-mode browser, if I the user has Javascript turned off, or consistently across even the browsers that claim to support it. It can *occasionally* provide a nice touch on a page, but if you use it for important functionality I think you're shooting yourself in the foot.
I agree there, but I'm not fond of JS and DHTML for the reasons mentioned above. I do like to serve up a version of sites that can be rendered in 4x browsers but there will be no more convoluted work-arounds catering to 4x.
We'll do a couple of prototypes, some with dhtml and some without and they will nearly always go for the dhtml proto because they think that a little js magic will draw users by the squillions to their site. Then they also want full v4 compliance...
As they say, a little bit of knowledge is a dangerous thing.