Ugh. You’d think people would realise the downside to forcing people to upgrade. Nobody liked New Coke. Why on earth did they think I’d like New Blogger?
If anything on the site is broken, it’s because New Blogger isn’t quite the same as Blogger. The most obvious thing is the login box that got broken but I guess that’s just an annoyance for me ;)
What does an FTP blog look like with New Blogger? Virtually the same as Blogger. I can’t use any of the new layout stuff but I can add Labels to my posts. Labels are New Blogger’s attempt to re-implement my Keywords thing but it’s all done on the server so it’s less brittle. Good in theory but there’s a catch.
The content that New Blogger wants to put on your page because of the Labels is just dumped into the end of the post. More useful would have been the pattern established for all sorts of other things in Blogger. Then again, it’s better than nothing and I can work around it with CSS. The bigger problem is that I have no way to customise the output of the label browse page. That’s the page New Blogger generates for each label that contains the posts for that label. Now to my mind, labels work like Archives and should be handled the same way. New Blogger doesn’t seem to think so. The page either didn’t go through my template or it went through the tag instead of the tag. This means the label browse page has all the post text on it which isn’t what I want. Click the date on this post to see what my archive pages look like. That’s what I want label pages to look like! Of course, there are differences with label pages so it would be even better if there was a tag I could use to customise the output.
I’ve sent a bug but I have doubts that it’ll get anywhere. At least my Keywords thing is still working.
Update 19 February 2007
I don’t know what they did but I’m not longer able to use Konqueror to maintain my blog. Login to Google? Crash. Edit a post? Crash. Sigh. I realise it’s better to support Firefox than to only support IE but Konqueror is a real browser too. It’s even based on the same rendering engine as Safari.
I’ve whipped up some magic around the existing keywords thing so that I can start filling in the labels now and switch over to them once they’ve fixed the display problems. Of course, nothing on this blog ever goes according to plan and this was no exception. I’m using innerHTML to pull out the text of each post. I then run a regular expression over this to find the Keywords bit (which is always at the end of a post). Well, the labels have been squeezed into the post too so I have to remove them before doing the Keywords regexp (I put them back afterwards and use CSS to hide them). Here’s where it gets tricky. After getting a regexp that works I came home and noticed that Safari was borking. After much hair pulling, I finally found that Safari is re-interpreting the HTML when I call innerHTML. I’m not being given the HTML code in the file but something Safari thinks is equivalent. Luckily, it’s just using uppercase HTML elements. That’s easily solved by doing a case-insensitive search. However, it concerns me that Safari might be doing other things to the HTML…
