browser history: a pot meets kettle story
September 14th, 2007 by Dylan
In “Tales from the Evil Empire”, Betrand Le Roy of Microsoft has a summary of the current state of of browser history management. It’s a reasonably solid summary of browsers today, except that he comes to the following conclusion:
Now, we only have IE, Firefox and Safari Mac. I just hope this is only temporary and that both Apple and Opera repair their browsers soon.
Given that the web development community has been complaining about Microsoft for not fixing the many many flaws of their browser for most of this decade, it seems ironic that he would have the gumption to say this. Furthermore, there are other history managers that manage to work around these issues. I would still bet that Opera and Apple will “fix” these issues before Microsoft will fix the issues with their browser that the world complains about on a daily basis.
Dylan, do you really think I don’t realize the irony in that (the title of my blog should have given you a clue that I’m aware of the concept)? I’ve also bitched about IE bugs and performace in the past (and reported them to the team), and I think I’m not treating anyone unfairly. I’m not sure the same can be said of your post. Let me quote another part of my post that you apparently skipped: “I’m pretty confident this will get fixed pretty soon. I’ve had pretty good experience with their reactivity in the past.”
Not only do I recognize the good reactivity of some of the other browser vendors, I also always report any problem I find to help them improve their products.
Furthermore, I don’t think being a Microsoft employee gives me the obligation to bear responsibility for what all the other teams are doing.
I think “gumption” is the right word though. It seems like the consensus in the blogsphere is that IE is the most horrible piece of software ever produced and all other browsers are perfect gems. Well, I have the gumption to say that IE in the past has done more for the advancement of web development than any other browser (I was around when IE 4 came out) and that Safari 2 is by far the worst browser for Ajax development (but Safari 3 will fix most problems).
One more thing: I’d really like you to point me to those workarounds that you’re talking about.
Bertrand,
Relax, I’m certainly trying to pick on Microsoft in general more than you in particular. I found your post quirky and was trying to post a quick take on it rather than rip you a new one. I also found your post to be informative as stated… we both agree that the current scenario is a bit of a mess.
I do appreciate that you take pride in your product. Did IE 4 and 5 significantly improve my life? Yes. Has it cost the world billions of dollars in development time over frustrating bugs that have to be worked around because there were not significant updates for years at a time. You bet. I’ve talked with several members of the IE team in the past, and the reality is, it’s a very slow process for us to see positive change. As a community, we see things like Silverlight get attention, while the JavaScript engine and VML get neglected for indefinite periods of time.
As you are well aware, working on any type of product that anyone cares about requires having a thick skin. I generally worry more about what I do about that criticism than anything else.
I think the fact that Dojo’s Dijit only officially supports Safari 3 tells you what you need to know about my opinion on that. I’m certainly not a fanboy that’s going to tell the world that Firefox or Safari are perfect and IE always sucks. That said, I personally prefer the Firefox dev and usage experience today, but I certainly don’t hide my frustration with Firefox issues. I let them know what’s wrong and what’s bugging me and push on them to make things better.
Regarding workarounds, I was thinking of the several libraries out there (Yahoo, RSH, several others) that tackle this problem. To be honest, the history problem isn’t all that interesting to me. I want a real API for adding things to the history rather than another layer of hacks and workarounds. It is quite possible that I misspoke here, and I apologize for riling you up if I did.
No problem, thanks for the clarification. I agree about the thick skin, I guess my tolerance level was a little too low this morning, so I too apologize for reacting a little harshly.
I think our library in its current internal state and the blog post reflect the state of things today and except for one thing in IE, I think I’m showing an implementation example of a known workaround for all problems that have been solved today. The one thing that I’m not showing and that has a solution is how to avoid having extra round-trips to the server in IE. RSH and many others have the same problem that we have, but I’m trying Julien Lecomte’s document.write solution as we speak and if it works I’ll include it in the library and in the article (and credit him of course).
The two problems that I don’t know any workarounds for but would be more than happy to learn about are the new problems in Opera and Safari Windows. I don’t worry too much about the one in Safari Windows because I opened a bug in WebKit for it, which worked in the past. Plus, they’re still in beta. I’m a little more worried about Opera because their process and bug database is closed (yes, I know, oh, the irony) but I’m confident it’s something they know about and will want to fix.
I think we all want the browsers to solve those problems instead of having to hack around them, and I’m sure you’ve read Alex’s post and the ensuing discussion about browser.next. I really appreciate this approach because I think we need to be pragmatic and get things moving instead of playing the blame game.
Cheers and keep up the good work in Dojo!
Dylan – of course the tragedy is that Microsoft benefits greatly from its poor implementation of the standards. I recently got “trolled” on my blog for suggesting people use Ajax – the commenter felt Ajax was dead because of cross-browser problems and so recommended using Silverlight!
I put my response, “Another Ignoramus for Ajax” on the Webguild blog at http://www.webguild.org/2007/09/another-ignoramus-for-ajax.html