Planet Olin

Well, I almost installed the newly released Internet Explorer 7 Beta 2. Luckily, I found out just in time from the comments on a post on the IE blog that it will install over my IE6 and not uninstall. Since I need IE6 for web dev work, I can’t install it at this point, which means that I can’t try it out. I’m disappointed, but I can still read other people’s comments about the beta release.

EDIT: Apparently it can be uninstalled, based on a new post from the IE blog. I was grossly misinformed.

I’m still torn about the new IE and CSS. On the one hand, initial comments from some people make it sound as if they’ve only fixed a few CSS bugs and left other big ones (multiple-class bug, absolute positioning bug, margin bug) unfixed. This isn’t very encouraging. However, it’s only a beta release, so maybe things will continue to improve in the final release. I’m encouraged by the tone of this post by Chris Wilson on the IE Blog. They are looking for feedback; they are at least trying to fix the problem. Dean’s comment on my previous post is also encouraging. I’m no longer suspicious of the IE team’s intent, but I’m still afraid that they may not be given enough time and freedom by their superiors to fix all the bugs that need fixing in the browser. Corporate higher-ups only need another browser release to boost market-share and awareness. As long as they include enough commonly know buzzwords, they don’t need to worry about little things like CSS2.1 compatibility.

I appreciated Fred’s post at WeBreakStuff about his recent visit to Microsoft. He claims that they’re not as evil as they seem, and that, large as the company is, there are people inside it working to move it in the right directions. Good. As Ben Parker once said in a movie, “With great power comes great responsibility.”

I seriously wish the IE Team the best, and encourage them to keep working on standards-compliance. I’ll work with you guys if you work with me. You have the power to make web dev less of a headache. Go, fight win!

I’ve always thought it would be really fun to start a group blog. Writing by yourself is fun; blindly sending out messages into the black space of the internet and occasionally getting comments back. However, there’s just something more fun about posting and commenting back and forth in rapid succession with friends, letting ideas mix and evolve rapidly.

Towards that end, my friends and I have just launched alwaysBETA, a brave new foray into the dangerous world of reading things and writing stuff. It’s a group blog where we can be free to write about anything that interests us. Some of the other authors are pretty funny, so you know you’ll want to check it out.

We’re sure to be the next buzz worthy new blog. Or if not, at least we can go down in a blaze of standards-compliant glory. Come, consume our content, and enjoy.

So, a while back I posted about my Olin friends who are starting a web company called Flagr. Basically, their idea is to create a service that allows people to make bookmarks of locations (along with photos, info, etc.) all from their mobile phones and then share these bookmarks with others and interact with them from their phones and the web. They got funding from Y Combinator, and now some big names are starting to take notice.

First, the latest Techcrunch post from Mike Arrington about the E27 startup summit in California features a small writeup about them. It incorrectly states that they are “college dropouts”, which is funny since they’re only taking a semester off. Well, either the post is wrong or they’ve been lying to us.

Second, Scoble happened to be there when they pitched their startup to Mike, so they also got a mention in Scoble’s post about the event. Scoble even took their photo with Mike! So cool!

Way to go guys. Your hard work is paying off. I hope everyone signs up for their mailing list. They’ll be launching soon.

So we had our first meeting of the semester with our SCOPE advisor, which went both badly and and well - well, in that Dave gave us an honest, candid assessment of what we, and the SCOPE program in general, had accomplished last semester, and badly, in that Dave gave us an honest, candid assessment of what we, and the SCOPE program in general, had accomplished last semester.

That got me thinking, and they weren't very pleasant thoughts. Before I start though, I'll make a caveat: Olin has changed so much since the days of yore, and each class has had such a different makeup and experience, that my thoughts can necessarily apply only to my class, and perhaps only to myself. Anyway.

Even since we came to Olin, we've been told we're special, that we're the best of the best, that we're brilliant, and risk taking, and entrepreneurial, and renaissance engineers, and lots of other nice things that make us all feel really good inside. And then classes hit, and all of a sudden, we're failing, and not doing well, and have no time for anything besides classes. To counteract that, Olin sets up a world in which we're still special, and still smart, and failing doesn't matter, even though inside we know that we're still failing at lots of things. Add to the mix the fact that everywhere Olin trumpets the fact that we're engineers who can read and write and present and work on teams, and it's no surprise that the thing that we value most about ourselves is not the fact that we can design and build stuff (because we've failed at that so many times over), but the fact that we can work on teams and write reports and present pretty presentations. The fantasy world where our Olin specialness outweighs the fact that we haven't ever actually had a project succeed prevents us from doing anything about it, because according to this worldview, there is nothing the matter. Then along comes SCOPE, and all of a sudden, we're not special anymore, and there are no more excuses. The project sponsor doesn't care about all the extenuating circumstances that we use to rationalize our Olin specialness - they just want results. So what do we fall back on, what is the one thing we know for sure that we can do well? Write reports and make pretty presentations.

At the end of the day though, reports and presentations don't get the job done, and we actually have to do engineering. But because every project we've ever worked on has failed, it's reasonable to expect that this one will fail too. But the trick here is, this project can't fail, because unlike all the other projects we've ever done, this project has consequences for others - for the next class, in that none of the sponsors will ever come back if we fail; and for Olin, in that we've made SCOPE such a centerpiece that if SCOPE fails, Olin has failed too. So what do we do? We cover our butts. We make as few decisions as humanly possible, because we might be wrong, and those decisions that do get made, get made only after an incredibly exhaustive search of possibilities to make sure that if something doesn't work, no-one can ever accuse us of not having done due diligence.

Besides time, there's also the money issue: paper and electricity is cheap - it costs us practically nothing to surf the web for endless hours doing research, but it costs lots of money to buy something, test it, and realize it's not the right thing. Couple to that our fear of breaking something or burning something out because we don't know what we're doing, and it's no wonder that few SCOPE teams built and tested anything last year. As Dave rightly points out, Olin students have racked up five-figure costs in just burned out motors and sensors and controllers alone. Given how little money we have, that statistic is not likely to encourage anyone to just turn something on and play with it to find out how it works. And so we're all frozen: we're frozen because we're scared that we don't have enough information, knowledge, and experience to make technical decisions, because the wrong decision will spell disaster for the project, SCOPE, and Olin.

And now we're graduating, and it's scary, because we really won't be special anymore, because there will be no one to tell us that failure is okay, and because a lot of us feel like we're liberal arts majors masquerading as engineers while Olin expects us to be the best engineers on the planet Earth. Worst of all, I don't know how to fix it.

Biology 100% Mathematics 92% Theater 83% Philosophy 83% Chemistry 83% Engineering 83% Dance 75% Art 75% Psychology 75% Anthropology 58% Linguistics 58% English 50% Journalism 50% Sociology 42% What is your Perfect Major?

An uncertain specter looms on the horizon. I don’t know whether to run towards or away from it. Yes, it’s true, after hemorrhaging browser market share to Firefox for a while, Microsoft finally decided that they should put some people back on the Internet Explorer project, even though it’s not a big money maker for them. As much as they might like to ignore the web as a platform, the recent explosion in user-centric web apps and Microsoft’s response with (among other things) Windows Live proves that they are no longer content to ignore the area and let sleeping dogs (or crappy browsers) lie.

Yes, Microsoft is going to release IE7, and their developer’s blog shows some encouraging signs. Native XMLHTTPRequest support (without ActiveX)? Great! Support for PNG transparency? Great! Improved support for CSS? Well… It turns out that if you dig into what “improved support for CSS” really means in IE7, you’ll find mixed blessings.

For those readers who don’t know, CSS (Cascading Style Sheets) are the standards-compliant way to style web pages. The idea is to keep content (in the XHTML) separate from layout (in the CSS). This has many benefits, one of the most important being the increased ease with which you can change the presentation of a website (namely, by altering one file instead of all of them.) In the past, Internet Explorer has had absolutely abysmal support for CSS. Certain things don’t work, or only partially work, or work in totally wrong and unexpected ways. A CSS-styled page that looks completely normal in standards-compliant browsers such as Firefox will probably look totally broken in IE unless you spend hours of extra time debugging it.

How does one debug CSS for IE? Funny you should ask. A large community has sprung up dedicated to finding ways around IE’s many CSS bugs. In most cases, you kill IE CSS bugs with other IE CSS bugs. (One example of this is the well known “holly hack”.) Thus, a large community of developers committed to standards and the use of CSS have carefully crept around IE’s issues and continued to support the crippled browser with such an unfortunately large market share.

However, times are changing. The IE team has announced that they are eliminating the “star-HTML” hack (at the core of the “holly hack”) along with many other IE “band aid bugs” that many developers use to fix more serious problems. Now, assuming that they actually bring IE7 up to the level of standards-compliance as, say, Firefox, this would be great. It even sounds like they are on the right track. But when they start talking about using conditional selectors (a non-standard IE-only feature) to feed IE with CSS fixes instead (which is more of a headache and requires multiple CSS files instead of just one), I start to get a little bit nervous.

Are Microsoft and the IE Team’s motives pure? I’d like to think so. So much of my time as a web designer has been wasted experimenting with hacks to fix their humorously broken CSS rendering that I really want to believe them. However, if they were really building a standards-compliant browser, then it would just render pages the way Firefox currently renders them (correctly), so there would be no problems. So why are they complaining about developers using hacks and warning them about issues with IE7? This might suggest something more sinister: IE7 isn’t going to fix all the issues, and thousands of developers will be forced to go back and fix countless websites to make up for Microsoft’s pigheadedness.

Alper has a much darker theory about what IE7 actually means to Microsoft. He predicts that the probable partial support of CSS in IE7 is actually just part of a plot by Microsoft to further fragment and disrupt the standards community with new headaches while they continue to push the .NET platform as an alternative to the new web application boom. Sounds plausible to me, although it makes me very sad. I hope he’s wrong. For every developer’s sake, I hope that MS makes IE7 the most standards-compliant browser on the market. But something tells me that they’d rather push for ownership of the web than cooperate with standards bodies and encourage cross-browser compatibility. Drat.

Some people might argue that designers should just lay off the pretty-juice and design simple cross-browser compatible pages that are ho-hum but functional (some might say Nielsenesque.) However, I refuse to have my creativity restricted. I can and will find ways to let the designs in my head out into my browser and the web. Even if I have to start refusing to send my stylesheets to IE browsers altogether and giving a large portion of the market a purely textual experience. Or maybe I should just go against accessibility and start using Explorer Destroyer on everything I do.