tedster
msg:617422 | 5:37 am on Jul 6, 2000 (gmt 0) |
Yet another "study" that doesn't line up with the server logs I look at for any client. The lowest netscape percent I see is 21% and that's a site that very much caters to "out of the box" users. This kind of reporting makes me very suspicious.
|
rcjordan
msg:617423 | 1:44 pm on Jul 6, 2000 (gmt 0) |
>The lowest netscape percent I see is 21% NS is about the same here. But it's the trendline that's the killer. Now, we're dickering over whether it's 14% or 21% --not so long ago, we used to argue whether it was 60% or 65%.
|
oilman
msg:617424 | 5:33 pm on Jul 6, 2000 (gmt 0) |
we see about 23% NS. We cater to a largerly corporate clientel.
|
tedster
msg:617425 | 8:28 pm on Jul 6, 2000 (gmt 0) |
I'm primarily a site developer and secondarily a search engine optimizer to add value for my clients. I often fantasize about how the web would look today if browser development was less of a no-holds-barred war and more of a competition for full compliance with standards. Too much development time goes into compatibility testing, and the lack of W3C compliance borders on arrogance at times. This adolescent browser warfare has undermined the ability of developers to create more enjoyable, feature rich sites. The safest path is still to program basically for version 3 [that's sad] and the costs involved with anything more, both directly to the client and in lost audience, have made the web a very frustrating medium, for all its promise. One of the above article had a footnote about IE 5.5 falling short on HTL4 compliance. This just kills me, at a time when Netscape and Opera have raised the level so nicely. Right now, Microsoft certainly has the market share and power to cripple the W3C standards for a bit longer. How foolish if they actually choose that path, just because they can.
|
rcjordan
msg:617426 | 8:49 pm on Jul 6, 2000 (gmt 0) |
>The safest path is still to program basically for version 3 [that's sad] True, I've seen several publisher articles in the past that basically said just that. I'm writing V3 with an added large dose of JS 1.2 --I'll probably stop there. Beyond that, it requires a heavy-duty code hanger to keep up with the cross-browser issues.
|
cre8pc
msg:617427 | 1:11 pm on Jul 11, 2000 (gmt 0) |
It confounds most people when they hear this, but yes, many companies still program and design "back to" Netscape 3.0. The majority of the corporations I'm associated with do, although they're aware of what server logs are telling them. Less and less people use 3.0. Overall, the wisest approach would be to study one's target market. If it's known, or server logs indicate, a strong showing from Webtv and AOL users, then to keep them coming back, the pages should be welcoming and work for them. This would include ecom sites, auctions, online shopping, and malls. B2B, which is more industry oriented, is likely to find high end equipment in use. This presents a constant headache for site developers who are part creative artist and part technically oriented. The creative voice is usually screaming bloody hell for some room for expression! Kim
|
tedster
msg:617428 | 1:43 pm on Jul 11, 2000 (gmt 0) |
I feel that I'm part artist, part technician, and a third and equal part marketer. I feel that good web development requires all three areas, and I try nurture each. The marketer part of me screams the loudest for some of the more advanced capabilities. Any message can be delivered so much more effectively with judicious use of the many bells and whistles.
|
Brett_Tabke
msg:617429 | 3:16 pm on Jul 11, 2000 (gmt 0) |
In the log analysis debate. Don't forget that there are many people who use various proxy servers, web filters, or other agents that set between the browser and the net. Those agents most often configure themselves to spoof IE agent names. My guess is that 10 to 12% of the web is setting behind some 'false' agent name. There is also all those AOL'ers, and not all of them have upgraded to 5.0, yet the proxy will never reveal that fact to you due to the cache.
|
rcjordan
msg:617430 | 4:39 pm on Jul 11, 2000 (gmt 0) |
>The marketer part of me screams the loudest for some of the more advanced capabilities. Same here, but I nearly nuked my highest traffic pages in the SEs while chasing the marketing side. The short version of the story is that I had a chance to sell some sponsorships using a (then) sophisticated javascript called 'Headliner' --this js required 25K+ if I recall. Well, I dropped it in the top of my best pages, and it was bringing in very nice ad dollars --until those pages went through an indexing cycle. I went from in the top 5 to "could-not-be-found-unless-I-searched-on-the-domain-name" YEOW!!! It took months to get back in.
|
tedster
msg:617431 | 3:38 pm on Jul 12, 2000 (gmt 0) |
I dropped it in the top of my best pages, and it was bringing in very nice ad dollars --until those pages went through an indexing cycle. I went from in the top 5 to "could-not-be-found-unless-I-searched-on-the-domain-name"
_______ Was that killer javascript right in the HTML page, or an external js file? I've had very good results moving as much js and css as I can to external files. In one case -- about 10k of javascript -- the ranking boost tripled site traffic within a single round of indexing.
|
rcjordan
msg:617432 | 4:40 pm on Jul 12, 2000 (gmt 0) |
>Was that killer javascript right in the HTML page
Yes, at that time. You are right, external js files are THE WAY TO GO!
(I have some hitting 50k now, doing everything but the laundry. External.js is very addictive, I've got to cut back.)
|
GWJ
msg:617433 | 6:57 pm on Jul 12, 2000 (gmt 0) |
Question Rcjordan.... When you say an "external file" for JS do you mean use an include statement from the HTML page to call to a page with the appropriate JS? Thanks, Brian
|
rcjordan
msg:617434 | 7:06 pm on Jul 12, 2000 (gmt 0) |
>call to a page
yes, up in the header. Ex:
<SCRIPT LANGUAGE="Javascript1.2"
SRC="http://www.somedomain.com/somescript.js">
</SCRIPT>
I'm using document.write to code whole chunks of pages, nav bars, adverts, cross-links, you name it. BUT, if you do this, you'd better cover yourself in the <noframes> section, because doc.writes aren't spidered.
|
rcjordan
msg:617435 | 7:17 pm on Jul 12, 2000 (gmt 0) |
BTW, some of my pages call multiple external.js files. Then by mixing the function calls as needed, you can build a broad matrix of possible page combinations. Also, since these external.js files are cached by the browser, you can build up some huge capabilities by the time the user gets to the page you want to display all the function calls --sort of like image preload.
|
GWJ
msg:617436 | 9:35 pm on Jul 12, 2000 (gmt 0) |
Thank you very much Rcjordan will start loking at this. Brian
|
tedster
msg:617437 | 1:39 am on Jul 13, 2000 (gmt 0) |
Yes, external js files are like a drug. I started by using document.write to protect email addresses from the spam spiders. The next thing you know, I'm writing outrageous DHTML and looking into bigtime custom pages.
|
Brett_Tabke
msg:617438 | 9:21 pm on Aug 3, 2000 (gmt 0) |
Well, thank you for keeping them external. I think that is the only and best way to use JS. I run with JS off 75% of the time because most of it is such lousy/incompatable/buggy code that never runs right in other browsers. Keeping it off page really speeds up the download time.
|
GWJ
msg:617439 | 12:33 pm on Aug 4, 2000 (gmt 0) |
Question: What is document.write? Oh yeah, started playing with JS in external files. Did some benchmarking, Wow...:). Brian
|
NFFC
msg:617440 | 1:35 pm on Aug 4, 2000 (gmt 0) |
UK site, July's figures. AOL 5.x 13513
MSIE 5.x 10486
AOL 4.x 5419
MSIE 4.x 2161
Netscape 4.x 954
AOL 3.x 710
Unknown 557
WebTV 1.x 226
MSIE 3.x 126
Netscape 3.x 83
MSProxy 2.x 54
Opera 3.x 4
WebTV 2.x 1
Netscape 2.x 1
|
|
