View Full Version : Netscape 6 Coding
I've found out, that N6 dosn't like any underscore either in Div ID's, nor in Variable Names <img src=icon_smile_shock.gif border=0 align=middle>
I need to convert a heavy DHTML Site to Netscape 6, so if anyone knows other JS-Code Bugs in NS6 please post it.
Greetings,
Awytok
Yea? I always use underscores and I haven't had any problems with that at all.
Yeah I use underscores religously always have i only use the interCap convention for function names.
Maybe you could post what you're working on and we coudlhelp you spot what's causing you grief.
function Mike(name, mail) {
*****this.name = "Xanth";
*****this.mail = [url:8jaas0gokc]belgedin@earthlink.net[/url:8jaas0gokc];
*****return this;
}
Hmmm... I've had problems with underscores in Div names... very strange that you guys don't.
Another thing, I'm redesigning the front end to my China/Vietnam photos and NS4 renders it wrong, can't figure out why (I'll post a link later this week).
--
Dan
www.pupius.net
Ok, here are the examples:
http://www.pms-web.ch/__test/home_NS6.htm
and
http://www.pms-web.ch/__test/home.htm
home.htm with _ works only in Explorer, home_NS6 in both.
Another NS6 Problem:
<div id="Letter">
<table border="0" width="100%" height="450" cellspacing="0" cellpadding="0">
<tr>
<td width="130" height="450">images/Couver_L.jpg</td>
<script language="javascript"><!--
if(!document.layers){
document.write("<td width=\"100%\" height=\"450\">");
document.write("<img border=\"0\" src=\"images/Couvert_M.jpg\" width=\"100%\" height=\"450\">\n");
}else{
document.write("<td width=\"100%\" height=\"454\" background=\"images/Couvert_M.jpg\">\n");
document.write("*");
}
//--></script>
</td>
<td width="120" height="450">images/Couvert_R.jpg</td>
</tr>
</table>
</div>
The Middle Pic should be streched to the ScreenWidth here the solution for NS4.75 and IE 5/6, Netscape igores the 100% in the table and in the IMG Tag, someone had already a similar problem or even better a solution?
tnx
There is a little problem on a HP made by myself with your coolmenu. The script works fine in ie5 and nn4x, but not with ns6. You can look at it, but on mouseover the menu will be hide by the place where the submenus should be.
It's an include homepage (php). May this be the reason for trouble?
May be you don't understand my english cause I'm stupid. But you can visit the page:
http://www.axus-ag.de
Would you help me?
Thanx a lot!
rwien
I don't have trouble with underscores either, though I generally avoid them because I've read in several places that certain releases of NN4 will choke (especially on Macs). If I remember right, an underscore in a selector was one of the bugs that was causing Jeffrey Zeldman's site to crash NN visitors.
Awytok:
NN6/Mozilla has had a ton of trouble with its image rendering, especially within tables. Here's a partial list (LONG nasty URL -- I don't know if it'll wrap until I post, so watch out):
http://bugzilla.mozilla.org/buglist.cgi?bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&field0-0-0=product&type0-0-0=substring&value0-0-0=images&field0-0-1=component&type0-0-1=substring&value0-0-1=images&field0-0-2=short_desc&type0-0-2=substring&value0-0-2=images&field0-0-3=status_whiteboard&type0-0-3=substring&value0-0-3=images
The good news is that they've just inserted a completely rebuilt rendering engine into the latest builds, so your example displays the same in Mozilla 9 and IE. Hopefully it'll be in the next update to NN6.
rwien:
You're using a version of coolmenus that hasn't been updated for NN6. You can try this one:
http://home.attmil.ne.jp/a/nikzhowz/coolmenus_borders2.js
or better yet, switch to Coolmenus3. The one above still has some dodgy screen paint problems in NN6.
OFF-TOPIC (SORRY) =)
Can anyone see what DCage posted? I have to scroll for about 1000px to see what else he wrote.
Anyone else?
Rad
[url="http://www.e-syed.net/viewsrc/"]viewsrc[/url:di8jolgsig]
Uh-uh. It fits within the screen for me. And I didn't use any formatting in the post. Odd...
I never thought about it; But it seems like Netscape is not able to wrap long lines that doesn't have any spaces in them. I did some testing and as far as I can see netscape just won't do that :(
I assume your using netscape Rad, that's probably why you've had this problem on some other posts as well.
I am not sure what to do about it though. I could place in a space if the lines where to big, but that could ruin links and code examples.
-Thomas Brattli-
http://www.bratta.com/
http://www.dhtmlcentral.com/
You could edit the ASP code that prints the links. I presume the code looks something like this
<pre id=code><font face=courier size=2 id=code>
' Look to see if someone is using <%=linkText%> ( method
if linkText = "" then linkText = linkUrl
www.pupius.net
____________________
</font id='Tahoma'>
Hey Dan,
There's a semi-regular who posts to comp.lang.javascript with updates to his AI project. I don't pretend to know or understand the stuff, but I always figured that attempting AI in client-side javascript would be like trying to stuff 60 kilos into a 20-kilo bag. Anyway, he usually gets sober appraisals from better minds than mine, so I guess what he's doing has some merit. Here's his page if you want to visit:
http://www.geocities.com/mentifex/jscript.html
This page:
http://www.dhtmlcentral.com/forums/topic.asp?TOPIC_ID=134&FORUM_ID=4&CAT_ID=1&Topic_Title=DCAGE+%2D+%3Canimated+foldoutmenu3%2F% 3E&Forum_Title=DHTML+Scripts
extends beyond the screen width for me in IE5.5. How about you guys?
hmmm... I've seen that before, he posted a message on comp.ai and got ripped to pieces.
I don't really know what the backend is meant to be doing but the results are far from impressive. And I don't know why he's calling it an "artificial mind".
Basically it takes a simple 3 word sentence and parses it, rearrages the words, tences and subjects:
"you went there" -> "I went there"
"you see me" -> "I see you"
"John loves you" -> "John loves I"
etc.
What I was thinking was more using AI principles to aid the solving of problems. For example, there are some interesting solutions to the problem of search, planning and it would be relatively straight foward to implement a simple Neuarl Network or Genetic Algorithm in Javascript.
I just don't know why I'd want too. There just doesn't seem to be any practical use for them on the web. Most of the ideas I have would just be experimental examples of what can be done (e.g. stacking boxes, maze solving etc.). But they still don't have any real use.
<font face='Tahoma'>-------------------------
Dan
www.pupius.net
-------------------------</font id='Tahoma'>
Icestorm
20.05.2001, 11:13
I remember seeing an interview with an AI researcher on the discovery channel once... the poor man gets nightmares about the end of the human race at the hands of robots with superhuman capabilities... he was sure it's gong to happen but still he keeps researching it and furthering the field of AI.
So heh now I'm interested, what about you Dan, what does your vision of the future look like? Hopefully it's a good one for all of us!
btw: that's an intriguing link DC :) even though I don't understand much of it!
ps: DCage, yes that page extends for me too... it's because of the [ code ] [ /code ] tags around the stuff I posted. Should I cut it up?
Edited by - icestorm on 05/20/2001 10:18:39
Ah, that answers that for me. I never saw much to his demo either, but I figured some demonstrations of AI wouldn't be impressive except as demonstrations of something deeper than meets the eye. Since I don't know about AI I figured maybe it would take a fair number of generations before any novelty would appear. But I guess within his noun/verb/object constraints, it really couldn't spit back much more than what it does.
Hey Ice,
<snip>ps: DCage, yes that page extends for me too... it's because of the [ code ] [ /code ] tags</snip>
Oh, so that's it. No, no need to edit the post, thanks. I just wanted to know how to avoid causing it.
OFF-OFF-OFF-OFF-OFF TOPIC:
Watch out if you're using more than one window and the "active topics" link. I keep coming up with posts that were neither listed in the Active Topics pages or flagged purple on the front pages.
Wha...? Thomas hasn't done the [ i ] fix? Or is it done differently now?
Icestorm
20.05.2001, 11:48
DCage -- yes I've asked Thomas about the italic tags already, but it must have slipped his mind because he didn't answer it or fix it. So I dunno.
Ice... The main problem is defining intelligence. When I say I do an AI degree people immediatly think Terminator 2 and computers that "think" like HAL from 2001.
But really what I study is Weak AI.
Weak AI uses principles and methodolgies observed in Humans and animals to solve problems that computers generally find hard and to improve interaction between humans and machines.
Some branches would be Computer Vision (processing both 2D and 3D images to gain understanding of what the structure represents), Natural Language Processing (building systems that *understand* natural language, or at least can interpret it), Mobile Robotics (finding ingeneous methods of solving problems in a noisy environment with limited sensory input - this is what I specialised in).
Strong AI aims at producing an artifical conciousness. There is a problem with this as how do we quantify, whether something is concious or not.
The Turing test is flawed (can an observer tell the difference between the response from a compter program of a human user when having a conversation over a terminal). Given a big enough knowledge base you could create a system that could converse as well as a human, but would it be truelly conscious - probably not! Then again, its the black-box scenario, if you know how something works it's not as clever as it appeared before.
If you knew how the human brain worked would you still consider us conscious. I have an idea that we are less conscious than we believe we are <img src=icon_smile_wink.gif border=0 align=middle> (for example, when you are talking or typing, do you REALLY know what's going to come out until after you have pressed the keys :)
Anyway, Ice... I think we are a long way off from producing artificial Conciousness (if it is even possible). As regards computers taking over the world, I think that's a little stupid. It isn't hard to program a computer to protect itself - just look at the Russian missle system, if the silos in the Ukraine loose contact with Moscow they assume Moscow has been destroyed and launch their missiles. So the main problem is what we, as humans, do with the technology.
Anyway, that is so off-topic it is unbelievable.
<font face='Tahoma'>-------------------------
Dan
www.pupius.net
-------------------------</font id='Tahoma'>
Edited by - Dan on 05/20/2001 10:53:22
Icestorm
20.05.2001, 12:14
Ehm yes sorry for the off topic blabber that just got blurted through this thread <img src=icon_smile_sleepy.gif border=0 align=middle>
...the problem is defining intelligence...
Indeed, there's been almost 100 years of modern research to that now, but there's no definitive answer yet, every time they find that there's more factors involved and then a new theory is formed. And well consciousness is a whole other chapter ten times as big yet and it has been thought about for thousands of years. I can't answer either of these things, but I have my own small opinions of course :-)
Anyway, thanks for the background info mate, I always wondered what exactly you did with that AI stuff :)
vBulletin® v3.6.7, Copyright ©2000-2008, Jelsoft Enterprises Ltd.