A big “Thank You!” to those who help

After my last post you could the get the impression that I’m tired of running such an open WikiWikiWeb where everybody can change everything so easily. Nothing could be more wrong! When I explain the concept of a WikiWikiWeb to people most of them find it rather strange — they don’t understand how it could work.

But it does work. From time to time I see people correcting errors found on the pages here at GimpsterDotCom, making the site a little better each time. For example, after I restored [PHP Shell][], someone was kind enough to change the word “lunched” into the correct “launched” so that the page now make a lot more sense. I’m very grateful for those corrections because they’re the kind of mistakes that are hard to catch for me when I don’t speak English natively.

It’s these small and large contributions that proves to me that WikiWikiWeb~s work and I would like to thank everybody who have helped me since I launched (note, not lunched :-) GimpsterDotCom as a WikiWikiWeb.

Come on…

Don’t people have anything else to do than mess up my pages?! Yesterday I made a new release of [PHP Shell][], and 13 minutes later some idiot (with an IP address of 12.151.162.13) replaced the entire page with the text “H A C K E D B Y A L N O R 3 S”.

First of all, you haven’t “hacked” anything — you’ve used the readily available editing facilities in PhpWiki to edit the page. Now, that’s really impressive — it shows that you can submit a form through a webbrowser! What a cool way to demonstrate the great computer skills possessed by the so-called hackers from this “alnor3s”… I guess we all have to fear these guys who have such great knowledge on how to do WikiVandalism…

Secondly, the term hacker is used about a person who enjoys exploring the details of programmable systems and how to stretch their capabilities, as opposed to most users, who prefer to learn only the minimum necessary. according to JargonFile:hacker. Perhaps you confused yourself with a cracker (see the Jargon File), but I wouldn’t even call you that since you haven’t cracked anything — a WikiWikiWeb is already wide open to everybody.

Even if you have no creative skill, there’s no reason to destroy the work or others, especially when this work is given away for free in the hope that someone out there might find it useful.

PhpShell 1.8 released

This release doesn’t add any significant new functionality, instead it breaks compatibility with versions of [PHP][] earlier than 4.1.0 :-) The only new feature is that the stderr checkbox now remembers it’s state. Thanks goes to Michael Zech for sending me a patch!

Seriously, you or your provider ought to have updated your PHP installation beyond version 4.1.0 by now, considering that it’s almost 18 months since it was released (PHP 4.1.0 was released December 11th, 2001). PHP 4.1.0 was the version that introduced the new superglobal arrays: $_REQUEST, $_SERVER, and so on. I’m using them now in [PHP Shell][] because they’re the future standard and because it’s convenient.

For this release I’ve also fixed the generated [HTML][] so that it has become Strict [XHTML][] 1.0 accompanied by a valid stylesheet.

So, this release is mostly about standards complience, which is pretty important if you ask me, but which doesn’t add much new in itself.

Go download Mozilla 1.3

Mozilla Logo I’ve just updated my [Debian][] system for the first time in a long while, and part of the update was Mozilla 1.3. Mozilla was already the best browser out there, but it’s just got even better :-)

What really cought my eye with this release is how smooth everything is. The whole interface looks absolutely fantastic thanks to antialised fonts everywhere. Antialised fonts really make a huge difference in the visual appearance making Mozilla look much more professional and it’s also much easier on the eyes.

I also installed livehttpheaders which makes it possible to see the HTTP headers that Mozilla sends to webservers and the responces it get in return. I’ve already found this very useful for testing the latest CVS version of [PhpWiki][] which caches the output from parsing the WikiMarkup. This should lighten the load on the webserver, for now it’s possible for it to return a 304 Not Modified header instead of the actual data. I’m not sure how much the processing this skips on the PhpWiki side, but it does reduce the bandwidth. I’ll put that version here on GimpsterDotCom when I’ve moved my patches to the old version. Since I’m using PhpWiki as the backend of GimpsterDotCom, I’m also changing it in small ways all the time. I try to get these changes back into the CVS repository when I can, but there’s always a difference between the “stock” CVS version and the code I have.

Back from Aalborg

I’m back at [Skejbygård][] again after visiting my mom, dad, and Kristoffer in the weekend. As you can tell from my schedule I don’t have any classes on Fridays, so I could get on a train for Aalborg Thursday evening — it had been a long time since my last visit to Aalborg and then of course it was my dads birthday.

This bunny killed censorship I left my computer on while I was away with FreeNet running. I checked on it regulary over a SSH connection (I really love SSH, it’s so cool to be able to login securely when you’re away from your machine!) to see how the load was and to read some mail.

My node saw an incredible amount of requests, the average number of requests per hour was around 30,000 most of the time, but I also saw it rise to 60,000(!) once. This made my node reject all incoming requests most of the time because all threads were in use. I can raise the number of available threads and open connections, but it wont be before I get a new computer, for I simply don’t have enough RAM and CPU power to do it.

I am connected to the Internet through Bolignet-Aarhus and there was a constant load of about 90 KiB/s on the connection. This was split between 25 KiB/s downstream and 65 KiB/s upstream. And it wasn’t me who asked my node to download this stuff — my datastore had grown to over 7.5 GiB when I got home, when I left it was about 6 GiB.

So, FreeNet it definitely working. Whether or not the contents on it is interesting is another thing, but the techonoligy works and that kind of cool.