Reading yet another Alzheimer treatment discovery on Digg, I wondered and crawled the RSS just for fun. A Google Reader search shows that so far, as cited on Digg, there have been:

• 13 different cures for Cancer
• 5 different cures for Alzheimer
• 6 different cures for AIDS
• 4 different cures for Diabetes

Granted, those are only the ones as of January 2007, and I didn't search very deeply - limited myself to popular items and only did a search on treatment|cure. But every time I open up Google reader, I got more chances to find another Digg item about marijuana or cannabis curing AIDS than Engadget blogging about Apple getting sued.

The closest I can picture the Yahoo/MS battle to is two kids fighting in primary school over a candy bar. It always makes me smile.
That said, there's this rather big group of people who just think the world will come to an end if Google doesn't get competition, and that the Yahoo/MS deal needs to be done otherwise the world will come to an end. No kidding.
Truth is that competition keeps commerce healthy. Enterprises always try to be better than the other. They have to keep releasing new and attractive products, updating their current products, keep their prices down, and so on. A lack of competition allows the enterprise to go by its current products, their current prices or even higher ones, and above all it allows it into a monopoly situation which prevents competition and enters a vicious circle which is not comfortable for anyone but the enterprise itself.

Google has plenty of competition in a lot of areas. Yes, even the websearch market. Yahoo is, to Google, a competitor in websearch and, to a lesser extend, webmail. I won't go over them since that's not the subject. My point is that there is competition and competition.
Microsoft has been, since years, holding a monopoly position on a few markets, notably the OS and Office application market and, until recently, on the browser market. To get a common misconception settled right now: a monopoly is not a 100% market share. A monopoly is, in a few words, a situation where an entity can force a product onto its users without having to worry about serious sufferings.
If you don't believe that, go back in time a few years. Netscape used to have almost 100% market share. Microsoft released Internet Explorer and, surprisingly, Internet Explorer had that number all for itself just months later. While this post is not here to accuse Microsoft of monopoly either, truth is that Microsoft really is not the best choice of a possible competition. They were anticompetitive in the past and they still will be until Ballmer is fired.

But, this is Internet. People love to jump on the bandwagon and get back this bit of coolfactor they might have always missed while younger. Competition is good as long as it is real competition, folks. Some enterprises have a very different definition of it.

I'm annoyed at people who keep going on about how you need firefox to have a "better web browsing experience". Annoyed, because it is not true.

Let's get it done right now: Internet Explorer SUCKS. The world would be a better place without it. Now, don't get me wrong either, I love Firefox and I've made a small bunch of patches for it. That said, Firefox is not the best browser out here.

For those who don't know the basics of web browsers technical details, the code of browsers is split into two parts: The browser itself, and the rendering engine (often called the Core).
To make it simple, the rendering engine takes care of everything that is inside the "content window". IE's rendering engine is called Trident (MSHTML), Firefox's is called Gecko. But lets put it that way: Gecko sucks. Not as much as Trident, but it still sucks.

Gecko is now an old engine, that has never been entirely rewritten but been patched over and over. It's slow and still has bugs from 1997 which might be fixed in version 2.0. That's over ten years!
Firefox, itself, is a great browser. Firefox 3 has incredible performance optimizations. "Places" management is great, and most of all it has a great community building addons which I wouldn't be able to live without. However, I'm a linux user, and I don't have much choice. Epiphany sucks (well, no, it's a great browser for a lightweight one but its just not for me). Konqueror is awful for web browsing. Opera and Safari? Don't make me say anything about them. And really, if I'm going for something that will embed Gecko, I might aswell stay with Firefox because it's the best of all of them.

A young rendering engine called WebKit was born around 2003; originally a fork of KHTML (Konqueror's rendering engine) developped by Apple, it grew into something solid, lightweight and blazing fast and was open-sourced later on. Today, only very few browsers run WebKit (Apple runs a very slightly modified version of it, Epiphany will dump Gecko in 2.24... those are the "major" ones).

What a shame! It currently is by far the most actively developed of them all, with the least bugs and most performance. On top of that, it is the easiest to embed into a browser.

Maybe someone will hack it into Firefox (hoping it'll happen, since Mozilla so far understandably doesn't want to abandon Gecko). Meanwhile, don't believe what they all say, use the browser you prefer. But for the sake of anything decent left in this world, do NOT use IE. Not IE6. Not IE7. Not IE8. And my words will be the same until Microsoft seriously thinks about improving Trident. And to you all, Web Developers, do your best to ignore IE's flaws, and don't try to fix them, it'll only be that way users (or MS) will finally move on to reality.

Just an excuse to write a bit more, of course.

I haven't given updates in a while. Well, OpenWDB is still keeping me busy - I'm soon done with the DBC parser, and I'm thinking of learning the MPQ compression algorithm to make an extractor. Apart from that, I'm still in the process of moving. I also recently got in touch again with a lost close friend, one of those things that make you smile.

Oh yeah, I'm considering sending the Qzerty XKB layout upstream. More stuff to keep myself busy *shrugs*. Less writing more coding though. Happy April's Fool's day everyone, a bit late. :)

PS: To the ones who are still reading this on WW, there is an automatic copy of the blog posts on here, if you prefer an actual blogging platform.

Yup. Here is the first WDB file, following the current draft I have not-yet fully updated. Though, the column names are the same as the ones listed on the Itemcache.wdb page. Here is what I need:

• Testing. I need to know whether the values are always correct, and if there is no byte off.
• Renaming suggestions. A couple of names for the columns were just placeholders that were better than "Unknown1" etc. This is very important, it can't be changed once it's on the wiki.
• Ideas. I need feature suggestions, particularly for space optimization. If you download this file, you'll see it's 95mb big, and it's not a complete itemcache.

Once again, I have a few ideas I haven't correctly put up on the draft yet. If you are interested by the project, drop me a message on IRC.

Update: Here are the newer links to the files. Happy parsing!

I haven't posted news about OpenWDB for a while, and this is due mainly to the fact I've been working on it almost exclusively locally. Here is a quick round-up of changes/features that haven't been precised yet:
- The XMLs will be based on a DTD instead of another XML file. I didn't do this in the first place due to some problems I managed to work around only recently.
- WDBs and DBCs will be separated as much as they are separated in the game files. Meaning, they will be similar in structure, but will be different in the core. This is to make "source rebuilds" easier.
- Headers and separators will be included in the DTD. I'll make some documentation on this to make it clearer to everyone.

Hope this helps.

Happy Pi Day everyone!

My computer has been unavailable for nearly a day (damn libc6), so I haven't been that active lately. Nothing much to say, unfortunately, but I'm currently looking for a Mac OS X-ish looking theme that actually works with Firefox >= 3b4 on Linux.

Also been grumbling at Wine's support for 16-bit games. Hoping to get a couple of bugs fixed there.

I just finished a few hours ago installing and configuring an IRC bouncer. Here are a few facts for the ones interested in setting one up:
1) There is NO documentation on IRC bouncers on the net. None I used to say there is absolutely everything on the net - I didn't think THAT would be lacking.
2) Some dedicated server providers block port 6667 by default, both incoming and outgoing - OVH does at least. Luckily, you should be able to modify that somewhere.
3) The only comparison of IRC bouncers on the net is here.
4) Don't search, all the best bouncers lack an important feature. irssi-proxy lacks backlogs, ctrlproxy lacks documentation & support, znc lacks multiserver connection, ezbounce lacks TONS of stuff...

I tried the ones above except ezbounce, and I spent a LONG time trying to configure ctrlproxy, thinking it was the one suiting my needs the most. They say they have an user friendly interface, I have yet to find it.
I then came back on my choice and gave ZNC a try. This one is wonderful. I did however spend a good hour trying to configure all the channels I was in...

I also said ZNC does not support multiple networks, and this is true. Luckily, I could work that around:
Create as many users as you have networks. Adys, Adys2, Adys3... I highly recommend using the webadmin plugin, by the way. It's sickly wonderful. Give them the same password, and assign each user to a certain network.
Now, the hard part is hoping your IRC client does not suck. You need to create multiple networks of the same address on the port you assigned at first. You will have to change the Username for each server, since this is the only factor redirecting you to a different backlog.
Since XChat does not offer that possibility, I created an onconnect file on the first network, that automatically signed on the other servers. (x = ip, p = port)
/newserver x.x.x.x pppp username:password

There. I hope this ends up helping someone.

This is common sense to a lot of people, but not enough. Rant of the day is on all of you, people, still saving screenshots in JPG format. Particularly the ones saving them in MS Paint. Use PNG, it's lossless, today just as widely supported, and good programs (read: not MS Paint) don't make huge files out of them.

That is all.

As you may or may not know about, Acid3 has been released a couple of days ago.
Just like Acid2, Acid3 is released during the fine-tuning of the latest Gecko engine (1.9), so it won't make Firefox 3. It will however make Gecko 2's release - which will be integrated into Firefox 4. Let's hope Acid4's release will be nicer with Mozilla.
The WebKit guys (WebKit is use are doing an amazing fast-paced development on the bug, and I'm glad to see that. Hopefully, Safari will soon be Acid3-compliant.
As for the IE team, I wish them luck and hope IE8b1's rendering is not nearly as bad as IE7's.
Oh, yeah, IE8 beta1 is available for download as of today if you haven't heard yet. I'm personally glad - I'll be able to tell off fanboys that want my sites to be IE-compatible. What else.

PS: While we're talking about future browsers, it's probably worth mentioning that the about:memory core just got finished in Firefox 3 trunk. Here is an early screenshot of what it should look like when it's done. Should rejoice all the performancecheck addicts.

It's true I was not supposed to come back working on any WoW-related projects, particularly now I have other very long non-WoW ones. Though, a couple of days ago, I got my hands on the client again (hey, against my will!). Figured it was time to complete a 1-year old project.

Creating a database on the wiki is nothing specially new. With the help of Zeal and Tekkub, we had that started during the BC betas. The ones who were around at this time will remember all the slowdowns, crashes, errors... well, it was horrible. The hacked templates were slowing the wiki down in a horrible and painful way. Eventually, we stopped.
A couple of months later, the parser functions were installed on the wiki. Well-needed, we got around finishing a lot of neat templates. But we had given up on creating a db back then, including me.
Well, it's time to get the project up again. During these months, I had largely the time to design a solid ground-up MediaWiki-based WoW all-around database (SGUMWBWOWAADB for short). The ones who don't know how a db works will want to read up on the soon-to-be-updated process page, which will explain what will happen and how (and eventually why). More knowledgeable folks will want to read up on the OpenWDB syntax and definition as well as the Item default and an Item example.

New month, new plans. I've already started packing up for the move to Greece - Not leaving France until September though. All of February's planned work is done, everything is fine!
In other, more specific news... I will probably attend the WWI this year (Paris). If anyone else is planning to come there, give me a poke.

You may or may not know about Google Trends. It's, in short, a neat tool that lets you see the trends in Google search. And it just occurred to me that it is a much more reliable alternative to a lot of statistics on the web.
When users search for a word, they are interested by it in a way or another. It means they know or want to know (more) about the search item, for one.

First example, using trends as a replacement to Alexa. Let's take the example of WoWWiki and Thottbot. And now, lets compare the Alexa results to the trends. Scarily similar for two totally different stats, no? Well, if you think about it for over 3 seconds you think it's totally normal. Not even mentioning the fact that a lot of browsers do a Google Lucky Search when only a word is typed in the URL bar.

Second example is Firefox and Internet Explorer. Understandably, Firefox is well over IE. It does not mean it is more used, but that users are more interested in it than they are in IE. A lot of IE users are users who don't visit the web a lot, and don't often search Google - or got no reason to. Let's not forget the default search engine is MSN Live Search on Internet Explorer 7+.

Now, I'm definitely not saying trends are accurate. Users search Google more than once on the same subject. Not everyone uses Google. People mistype words, abbreviate them... But it's still a very good insight into popularity comparisons. Remember there are millions of searches every day.

This will certainly not interest everyone, but Matt Cutts posted an interesting article on NOINDEX, poll coming with it. Starting webmaster, or owner of a high-profile site, check it out, it concerns you.

For those who didn't read up on it yet, Microsoft is embracing open source for Windows Vista, Windows Server 2008, SQL Server 2008, Office 2007, Exchange Server 2007 and Office SharePoint Server 2007.

Well, not exactly, anyway. They are just releasing API documentations, most of them already available elsewhere. Though people are not wrong in that Microsoft is not leaving out the OS path completely out of the question. And this is within reason.
In the "greater ends", and as Internet gets even more popular and diversified even to John Neverusedacomputer, Open Source is and will be more popular. It's the matter of Free To Use vs Pay For It. In most case, this big difference gives a better opinion of it.
Microsoft is afraid of Open Source, this is no secret. It has the power to bring down any paying version of what is made, with a right tactic. Hell, in the next years, Linux will be more and more adopted as companies understand Ubuntu is far beyond the point of "desktop-ready" and vista is just not "notebook-ready".
Microsoft wants popularity. They want their part of the FOSS cake. Their move just proves they are scared, just like the Yahoo! bid did a few weeks ago. I'm not blaming them for doing it, just taking it with a grain of salt to demonstrate the obvious.

Well, only a few hours late. And what a day! After a couple of years of membership, I finally got around making a quick bloggish news feed. Call it laziness, because it is. Stole a fair part of the design from Zeal.

I hope it's been a good day for everyone, with or without someone else to kill time with. Spirit's up!