It’s not going too well, though. I do have sites for the majority of letters, but some are proving to be really hard to find. I’m trying my darndest to stay away from blogs, but some managed to slip in there anyway (you find something for Z that isn’t Zeldman!) My method of finding sites has largely involved going through bookmarks and my extensive browsing history in Firefox.

Here’s the list so far:

• A List Apart
• bash.org
• Ctrl+Alt+Del
• deviantART
• Engadget
• Facebook
• Giant in the Playground
• H_____
• IMDb
• J_____
• K_____
• Last.fm
• Mugshot
• N_____
• Overheard In The Office
• Penny Arcade
• Questionable Content
• Rotten Tomatoes
• Slashdot
• Threadless
• U_____
• Virb
• Wired
• xkcd
• YouTube
• Zeldman

From this list you can tell a lot about me. For example, I love Threadless. I’ve spent probably $200 there on myself and others, and the shirts I’ve bought are amongst my favorites of those I own. You can also tell which social networks I belong to, as well as more or less every webcomic I read.

What’s disappointing with this whole exercise is what needs to be omitted due to overlaps. For example, both Penny Arcade and PureVolume begin with P, so I can’t include both. For U I’m thinking Uncyclopedia or Urban Dictionary. I dislike both of those sites, but I need a U! Quite a predicament.

Schools usually associate this email address with their portal, which is generally a horribly designed union of school news and student services, such as applying for housing or signing up for a meal plan. Drexel’s has a lot of room for improvement.

The first problem happens as soon as I load the site: I get one of those browser not supported! pages. On Windows, they recommend using Firefox 1.5.0.1 or 1.0.7. I’m using 3.0b5 (edit: according to other people I’ve talked to, this page doesn’t show for newer stable builds of Firefox). I don’t get why they’re doing user-agent checking in the first place (the only things they’re worried about lack of support for are Java, JavaScript, and cookies) but why they’re checking for specific versions of a browser is even more of a mystery. If they supported all three back in Firefox 1, why wouldn’t they support them in Firefox 3? At least there’s a link at the bottom to continue.

On a similar note, requiring JavaScript, Java and cookies to be enabled worries me. There are legitimate reasons for disabling any or all of those three. They can all be exploited for malicious purposes. Drexel’s website is not one I’d expect to do something malicious to my computer, but if it were ever compromised users will be very vulnerable. I know people who browser with JavaScript disabled, and I’m sure it wouldn’t take much asking around to find people who disable Java or cookies as well.

Turn off JavaScript or cookies and the site won’t let you in—probably because they use JavaScript for all their navigational links. I tried going in with Java disabled and the site told me that it’s not needed, but strongly recommended. I’ve been browsing around for a bit but haven’t found anything that’s worse off for lack of Java. Oh well.

Once I actually got in, I wanted to change my default password, which was eight random characters, or so I assume. Oddly, I couldn’t find the account settings link, which the help pages assured was there. I searched the source. It wasn’t. Luckily there’s a separate page for managing all my accounts on their servers, which is wholly unnecessary. I had to set up the password recovery wizard, which helps me recover my password if I ever forget it. The site asked me to specify a question which it would ask me if I ever called up the wizard, and if I answered it correctly it would reset my password to one which I’d specified when I set up the wizard. This has one giant flaw: what if I don’t remember the password that I specify? The best way would be to change my password to a randomly-generated one which is emailed to a different email address. I don’t understand why they’ve decided to forgo the system that works so successfully for so many other sites out there, including sites for entities like banks, which require far more security.

The next thing I tried to do was change my password. My initial attempt failed, because, according to the site, my password was too long. Passwords must be between six and eight characters in length. The security implications of this are staggering. For one, it vastly reduces the amount of work an intruder has to do to brute force a password. Perhaps this is the reason that account names are students’ initials and a pair of random numbers—and if it is the reason, it’s a bad one, as security through obscurity is a proven bad technique.

I’m not sure which operating system is on the server they use to manage accounts, but if it’s Windows, the short length of the passwords is an even greater concern. Windows uses LM hashes to store passwords with fewer than 15 characters. There are two large security flaws present here: first, that characters longer than seven characters are split into two seven-character pieces and hashed separately, and second, that all characters are converted to uppercase (a more detailed description of the flaws of LM hashes can be found here). If you are reading this and you have a DrexelOne account, I strongly urge you to create an eight-character password, the longest allowed by the system.

I’ve had my account for probably a day and a half now. Hopefully this is the worst of it. We’ll find out in time.

Update: Drexel runs UNIX on at least one of their servers. They also give students a very, very nice thing: shell access! As dissatisfied as I am with DrexelOne and their accounts system in general, I have to give them a lot of credit for this one—I haven’t heard (although I may easily be mistaken) of a single other school that does this (edit: apparently the Rochester Institute of Technology does as well). Either way, I’ve already set up PuTTY for it, and I’ve started doing something with the webspace they gave me. Props to Drexel for an awesome feature!

Those who know me know that I was not a fan of Super Smash Bros. Melee.  Yes, it had many more characters, many more play modes, many more stages and items—my problem was with the game play.  Melee looked a lot better on the surface, but at its core the fighting just didn’t feel as good to me.  The gameplay was faster (which I liked) but the hits felt chintzy, like they didn’t do anything.  Sort of like how Deus Ex: Invisible War felt to me, actually.

Brawl fixes this.  The combat is both fast and solid, and the characters are more well balanced (I was a fan of Ness from the original Smash Bros., but they nerfed him in Melee).  The Final Smash moves provide an interesting element to the gameplay.  For the uninitiated, each character has an attack called a Final Smash, which is a sort of crazy attack that can easily damage or defeat multiple opponents.  This is triggered by hitting a floating item which occasionally appears (the Final Smash Ball) and pressing the B button.  Before, the main object of Super Smash Bros. at all times was to knock your opponents off the stage.  When the Final Smash Ball appears this object shifts, and the game becomes a sort of mad dash for the item.  This shift in objectives is fun, as it requires a completely different style of play than what a player is used to.

In addition to the traditional multiplayer and classic modes, the game features a single-player mode called The Subspace Emissary, in which gamers play through the characters’ fight against Subspace Army.  Clearly the developers worked harder on this mode than on past single-player modes in the series, but it still feels like an afterthought.  The CG cutscenes are cool, but the platformer-esque levels quickly grow tedious, especially later on when the player is forced to go through every single room.

The multiplayer mode offers many new options.  In the game options, players can control the speed of the game.  There is now a stage builder with which users can create their own fields to play on.  It is in multiplayer mode that the gameplay really shines—fast like Melee, solid like the original.  The characters are better balanced (although I see too many people playing as Pit) and I’ve found that I can do just as well with Olimar and Lucas as I can with some of the more serious characters.

Overall, the game is excellent.  The one-player mode is a bit dry, but the gameplay is fun and the plethora of different modes and options is enough to keep one occupied for a long time—especially if they have three friends to play with.  If not, there’s always online play.

I’ve been trying for a while to find a good way to have a bookmarks feed on my site.  At first I used del.icio.us, but they used Javascript to parse the feed… and to display the contents.  I’m staunchly opposed to using Javascript for anything important, because disabling Javascript would then disable that important bit of content or style.  I had tried and had problems with Magpie RSS (an RSS parser written in PHP), and also with Wordpress’s blogroll feature (if only I could sort the links chronologically…)  Luckily, in a recent blog post Jeffrey Zeldman linked to Alex King’s Twitter Tools, which uses that cool Twitter API to send updates to your blog.  I installed it, and, when it wasn’t working the way I wanted it to, dove into the PHP and hacked it up a bit.

So here they are: my Twitter updates.  I’ll end up using it more for bookmarking than for microblogging, so it’ll end up being more of a del.icio.us type thing than a Twitter type thing.  This way, though, I can be sneaky and throw in a microblog post when I’m feeling sneaky.

We really couldn’t have asked for a better group of developers on the IE8 team - if for no other reason, because they’ve made a huge effort to listen to what people wanted and to deliver a product that satisfied the largest number of people it could. In doing so, they made one of their most controversial decisions as of late: to use version-targeting to determine how a webpage would be rendered, and to default to a non-standards-compliant rendering engine.

I personally don’t see the point of having three rendering engines in a browser, but that’s another story. Writing for A List Apart, Jeremy Keith makes a good argument for making the standards-compliant rendering engine the default one. To clarify the situation, I asked Chris Wilson what would happen if IE8 were to encounter a valid, well-formed document with a strict doctype he says. My worst fears were realized when he confirmed that the browser would behave exactly as if it were its predecessor.

What the IE team wanted to do was use a meta element to tell the browser how to render the page; if no such element were encountered, the browser would default to the IE7 rendering engine. This is a problem because we already have a way of telling browsers to render a standards-compliant version of the webpage: the doctype. If I’m developing a webpage, I don’t want—nor should I need—to include an extra element just to make it display correctly in IE8.

I won’t worry any longer, because the IE8 team has reversed their decision and decided to make the standards mode the default one. Now come IE8, our webpages can work in all browsers without hacks or conditional comments. This is a milestone for web designers and developers, and for the first time I’m eagerly awaiting the release of an Internet Explorer browser.

What we’re aiming for is this:

Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Morbi leo pede, bibendum nec, accumsan at, bibendum quis, nisl. Integer pulvinar eleifend pede. Sed ligula. Vivamus ultrices auctor urna. Etiam laoreet. Phasellus interdum, nunc vitae porttitor auctor, sapien nulla ultrices mauris, id ornare quam dui non ligula. Pellentesque facilisis risus non lorem. Nunc tempus porta dolor. Suspendisse dapibus tortor. Donec non lorem vel lacus mollis fermentum.

The reason that the way shown on Web Designer Wall fails validation in XHTML 1.0 Strict is because <blockquote> can’t contain character data on its own—it needs a <p> to wrap it. XHTML 1.0 Transitional accepts this error, which is why I don’t like transitional doctypes, but that’s another story. The code for the fancy blockquote looks like this:

<blockquote><p>Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Morbi leo pede, bibendum nec, accumsan at, bibendum quis, nisl. Integer pulvinar eleifend pede. Sed ligula. Vivamus ultrices auctor urna. Etiam laoreet. Phasellus interdum, nunc vitae porttitor auctor, sapien nulla ultrices mauris, id ornare quam dui non ligula. Pellentesque facilisis risus non lorem. Nunc tempus porta dolor. Suspendisse dapibus tortor. Donec non lorem vel lacus mollis fermentum.<p><blockquote>

Looks like a normal blockquote, right? That’s because it is! There’s no extra markup involved, so it’s lean and nice semantically. Here’s the (essential) CSS:

blockquote {
padding-right: 22px;
background: url(”location of closequote image”) bottom right no-repeat;
}
blockquote p {
padding-left: 22px;
}
blockquote p:first-child {
background: url(”location of openquote image”) top left no-repeat;
}
A quick glance through the code reveals that we need two images: one for an opening quote, and one for a closing quote. Luckily, they’re easy to make (just pop open Photoshop, use the Type tool to make a quotation mark, and rasterize it).

I have a 17×16 pixel image for the opening quote, just big enough for it to fit. The closing quote, while the same width, is much taller: 24 pixels, and the quote is pushed all the way to the top. Why? It’s because of the way CSS handles background images. We can say that we want the image at the bottom right, or at the top left, but we can’t say we want it at the bottom minus a certain amount. We would use a percentage, but we since for each line of text we add the image moves further from the bottom, it won’t necessarily be aligned with the bottom line of text. Our only option: fake the space at the bottom by adding pixels at the bottom, and positioning the image all the way at the bottom right. The opening quote is already at the top left corner of the top line, so we don’t have to do anything else to get that one in position.

You may be wondering what the pseudo-element first-child does. The answer is simple: it applies only to the first element contained by a parent element. In this case, that’s the first <p>. We apply it to the first <p> element and not each one contained within the <blockquote> so that we don’t have more than one opening quote.

The padding is present so that the text doesn’t run into the images. Since my images are both 17px wide, I’ve set the padding at 22px, which gives us a nice 5 pixel gutter between the images and the text. Although the background is only applied to the first, the padding is still applied to all of the <p> elements so that they’re all indented the same amount.

That’s how to position the images. For the blockquotes I use here, I changed the color, the font and the font size. I didn’t show those in the CSS, because they’re trivial changes.

This is cool, but doesn’t have much practical application. I doubt that most users want to be updated on every facet of their friends’ lives (that you can update from your cell phone indicates that the service is meant to be used often). Furthermore, 140 characters isn’t really enough space to say much of anything—good for typing from your cell phone keypad, but bad for getting across a meaningful message. If the “status” feature of Facebook (which I have both used and seen used extensively) is any indication, the general signal-to-noise ratio will be fairly low.

I pick on Twitter, but this is the problem with micro-blogging in general: there isn’t really a situation in which you would need to let everyone know what you’re doing. Planning an event or an outing seems like it would be made easier, but text messaging or calling will accomplish the exact same task. You could use the service to let everyone know that you’re not at home, but is that really necessary?

All things considered, though, Twitter is pretty nifty for one reason: a public API. There’s a whole bunch of third-party apps using Twitter, some of which are really cool (a big list can be found here). On Slashdot today there was a link to an article from Botanicalls with instructions for a DIY project that monitors the moisture level in the soil of a potted plant and, when there’s not enough moisture, makes a Twitter post telling the owner to water the plant.

I had been having problems with my previous installation for a while. For one, I somehow got the Windows search utility to not index the hard drive, which made searching for anything a pain. I installed Google Desktop to alleviate that, but that came with another whole sidebar of gadgetry (widgetry? Someone should tell Microsoft that they wouldn’t be ripping off Apple just by calling them “widgets”—they’re the new big thing). That was something else I’d let get out of hand: the “dirtiness” of the installation. I would save random stuff on the desktop, install software that I ended up uninstalling five minutes later, etc. There were a whole lot of files scattered around that weren’t needed. Maybe I’m the only one who doesn’t like that (it’s the same way with my music library: everything has to have a correct ID3 tag) but it had to go.

The final straw, which I discovered a few months ago, was that HP had my laptop in some sort of grasp. I had bought Supreme Commander: Forged Alliance on Black Friday and was trying to play it. It barely played: even with all the graphics settings on low it was still a slideshow, which is odd because my computer passed the recommended requirements according to System Requirements Lab. A friend suggested I update my drivers, which I tried to do through nVidia’s website. After trying a bunch of different drivers, I eventually got a message something to the effect of your computer vendor requires that you download drivers from their site. From this I deduced that HP has some sort of software installed that prevents you from installing third-party drivers. Which was great, except that when I went to HP’s site, the drivers weren’t up-to-date, meaning no SupCom.

It’s been about twelve hours since I finished the installation and I’ve been slowly repopulating my computer with drivers and software. I have Firefox, Pidgin and Winamp, but not yet SupCom. There’s a whole list of stuff to get before it: Thunderbird, the entire Adobe CS3, tablet drivers… Eventually I’ll update my drivers and install SupCom. It should work this time.

I think the turning point with the old site—mlingojones.com—came when I realized that it was hard to understand when spoken, and the translation from hearing the name to typing in the domain name was a confusing one. It wasn’t until an associate mistyped the domain name as “lingojones.com” that I figured this out. I wanted a new domain name, a new online identity, which would come with the freedom to do something I’d wanted to for a while: launch a web design studio. I’d tried to do something like this on a more personal level with the old site, but had completely failed.

Ultimately, that was the problem with the old site. It was too personal. The persona I wanted to create was something like that of Dave Shea on mezzoblue: professional, but not totally removed. Unfortunately I fell into the same trap that doubtless many new bloggers do. Meta-entries (entries about the blog itself) comprised a large number of my posts, and the signal to noise ratio was low, which I now realize is possible even in a one-way method of communication.

Now I’m a year and a half wiser, both in terms of design skills and maturity (especially as far as blogging goes). I’ll try to keep my updates at least tangentially related to the general topic of the blog (web-related technologies and design) but I know I’ll probably end up branching out into other topics too. As long as the general direction of the content doesn’t revert to the same sort of juvenile commentary I was putting up before, I should be fine.

Let’s you and me both hope I can keep this up.