I already wrote about SORBS before. Not much has improved there. Today, I had a server listed in their directory again. And I wanted to know why - but you can only query the database if you are a registered and logged-in user. To become a user, you first have to answer a bunch of questions, such as phone number and address.
Ok, sorry, but you can't put it any nicer: the operators of SORBS are filter fascists. George Orwell would have had a field day with these nutcases - administrators are hindered in their work by the incompetent operation of their block lists and are then supposed to disclose their data. And of course, a user is also required for delisting.
That their web interface is a disgrace in terms of usability, I probably don't need to emphasize. Which fields are mandatory fields are only revealed in bits and pieces, which functions require registration are only known after selecting the function (and after submitting the associated form!) and various other obstacles.
And when you've jumped through their hoops, you get nonsense like this as a reason for the listing: "Likely Trojaned Machine, host running unknown trojan" - no, my machine does not have an unknown trojan installed. Presumably, the idiots have just stumbled over a port unknown to them (ports 8080, 8081 and 9999 are in use on the machine - by the way, with quite normal servers behind them) and once again brazenly claim that the machine is corrupted - but they are simply too stupid to build a usable list.
And yes, the ridiculous demand that you can only do the delisting from the listed machine itself and for that you have to run through their silly web interface again (which is quite ugly to use with Lynx - of course, on servers you probably operate graphical interfaces and VNC according to these great security experts ...).
I'll say it again: I won't go to great lengths if anyone uses this outdated and useless list on their server and emails are bounced as a result. Anyone who uses SORBS and therefore can't receive emails from me is simply too stupid for this world and can find a playground elsewhere. I've had it up to here with incompetent block list operators and incompetent mail admins who use these lame lists ...
/IE7/ is a project that teaches IE6 CSS properly using a JavaScript library. This should also make :before and :after work in combination with content: - not entirely unimportant for HTML-free rounded corners or HTML-free link identification through symbols ...
Found in Netzbuch: Cameropedia - a wiki with camera information.
The IE6 just can't handle :before when you want to insert content into the page via content: in the CSS. Quite annoying if you use it. The IE7 project that I wrote about in the previous article doesn't work reliably for me either - for example, under a Citrix server it won't execute it, probably because some security settings are missing there. Strange. Anyway, I looked at the problem myself and found a fairly compact solution, at least for my specific version of the problem: I just want to place icons before a link.
For this, links have one of three classes or no class: class="zu" defines a collapsed navigation element, class="auf" an expanded one, class="ohne" a link that should not be specially beautified, and all other links get a standard icon.
For this, I simply attach the following code at the bottom of the file just before the /body:
var links = document.getElementsByTagName("a");
for (var i=0; i
I then wrap the whole thing in a conditional comment for IE so that it is only executed by this. That's it. Simple and effective. Disabled JavaScript is not critical in my case, as without JavaScript on the system (it's a business solution with high interactivity) nothing will run anyway in the future - Ajax needs JavaScript as a component ...