A couple days ago, I posted a little conversation I had with netscape 4 while trying to fix a browser-compatibility issue. I’m happy to report that I solved this issue. And how? Well, I downloaded Macromedia’s Cold Fusion Studio 5 finally (I own a subscription to the product). It comes, like previous versions of Studio with a little function called ‘Validate code’. I had run this on the page in version 4.5.2, with no real effect. It still didn’t work in Netscape 4. I ran it again under Studio 5, and guess what? I handily listed the 58 invalid lines of code in the document, told me how they were wrong. So I went, item by item, and fixed them all (well, not all. I ignored the HTML 3.2 incompatibilities, because I don’t care about HTML 3.2 anymore). And then, with a little ‘Hey Presto!’, the page worked. Beautifully. In netscape 4, IE 6, Opera 5, netscape 6.1. So yay. And everyone, if you are looking for a kick-ass environment to develop pages it, be they static HTML, CFML, PHP, ASP or JSP, I’ll recommend Cold Fusion Studio 5. It rocks.
So one of the things I do is design data-systems. In particular, I have been working on a set of custom content-management tools that both I and Digitopolis can resell. Because I wanted to make this system scalable and universal, I’ve tried to abstract all the systems in it. To this end, every object in the system has a unique ID. And object could a table, a row in a table, or a field (column). I keep track of relationships in a special index. However, this didn’t suffice for some cases, and so I had to create a new way of noting relationships, which I have called normalization tables (as they are all used in data normalization). This has proven useful in one particular case from which I am hoping to derive a universal law. The case is thus:
I can’t say that I’m a huge Larry Ellison fan, or that I’d ever even considered using Oracle except when forces by circumstances, I’ll say one thing – they have interesting marketing practices:
Well, not entirely:
The existence of such a technology, the amendment states, “may seriously intrude on the privacy of these users. The use of such devices should therefore be prohibited unless the explicit, well-informed and freely given consent of the users concerned has been obtained.”
In other words, sites must be open and explicit about their cookie practices. And they won’t be allowed to set any cookies without users’ consent. Which doesn’t seem like such a bad thing to me. The downside that The Interactive Advertising Bureau U.K sees is a world where people have to re-enter certain data with each visit. Like a username and password? Like what you do every time you turn on a computer? Gee. That’s rough.
I like the proposed law, and hope that something like that comes along to this side of the pond soon.
TO INSTALL WINDOWS XP
New to Tannock.net : Radio!
That’s right kids, you can now listen to the same music I do! You can be cool by proxy! Or if you’re cooler than me, you can be a nerd by proxy!
So, point your ears here..
Over the next couple of days I’ll pop up a section that’ll detail who you might find on here. And, as a note, as this is semi-random, I’ve excluded all my continuous mix music. Which means that this is more rocky than my personal collection is, percentage-wise, but you’ll just have to live with it. Currently, you should be able to experience Al Green, Fiona Apple, Cake, Bebel Gilberto, Bassment Jaxx, Bruce Springsteen, Chemical Brothers, Beck and possibly some others. I can’t remember right now. But enjoy! Oh – I’m not sure of the bandwidth on here, so if it’s really bad, let me know, and I’ll complain to the powers that be (and here too!).
Looks like the Fusebox 3.0 Spec will be released this week. It looks great. I downloaded the demo app from Hal Helms’ site, and it appears that the new ‘engine’ tags address the main issue I’ve been having with using my hybrid FB1/FB2 app – which is loss of scope when calling an application as a custom tag. This has been a major stumbling block with my development of a custom content manager application. Ideally, the custom content manager would sit in the Custom tags directory on a webserver, and any number of sites on the same server call the same codebase. This will make installation on new sites just that much easier, and perhaps more importantly ensure that everyone is using the same code. Currently, to call this as a custom tag there’s some really, really ugly code that must be put in the calling site’s directory. Which is just no good at all…I’m only at version 0.9 right now, an already the two sites that make use of the demo app are using mutually incompatible versions of the code. The public site, fortunately, is using the ‘real’ code, while my playground buried deep within Digitopolis‘ website is not. When I have some time I’ll update the DB so that it can run the new code.
By the by, the word of the day is popinjay. I don’t particularly have a use for it, but it sure does sound silly as it rolls off the tongue. Say it a few times and just examine how it feels exiting your mouth. It’s an odd one!
New image for the Archives section now…and slightly different layout as a result. Use the link at lower left to find it!
So I just discovered that this site works in NS 2 & 3 much better than it does in NS 4. Why? Because earlier versions of netscape simply ignore all the CSS-P stuff that I do in here. It ends up with every element just appearing in a list, one after the other. Which is great! Netscape 4, however, with it’s buggy, limited support of CSS-P, correctly positions the elements, but then as it doesn’t know what to do with the overflow attribute, it can’t have that funky scrolling thing that I do. So it looks like I’ll need to have two DIV declarations : 1, for CSS-P-compatible browsers that know what to do with the overflow attributes. These ones will scroll, so that they’ll match to the existing box length. Then for NS, I won’t include a height attribute, so the box should just extend for as long as there’s text. Not sure when I’ll get to this, but sometime I will, I swear. 🙂
Apparently, Netscape 4 does not support < /P > tags within a < DIV > tag. At the Vancouver Chamber Choir website, I had been trying to come as close as I could to using full XHTML syntax. I didn’t think this would break older browsers, as they should just ignore any unknown tags, no? But apparently not. Several layers were not positioning themselves correctly in NS 4.08. As soon as I removed the < /P > tags from these pages, the layers started positioning themselves as they should. At least on the PC. Unfortunately, I have neither a Mac nor a Linux box currently test this at home, but I’ll have a look-see on Monday.