Ree's musings

(Been keeping notes in this private entry. Making it public now that I think I've found all the whitespace errors. I suspect I'll be picking tiny nits out of it for some time to come, though.)

LJ crossposts from Tabulas have some extra code at the button, making a large space at the foot of the entry. It should drop the last <p><br/><br/></p> at least. It puts a BR at the end of each line, plus a line return which LJ translates into a linebreak. This results in way too fucking much whitespace for each ENTER when typing the post.

LJ posts are dated according to time on the user's computer (Xanga's are by server time, adjusted for user timezone) but since crossposts use Tabulas' server time, they don't line up correctly with LJ-native posts. This can result in an error msg for the user if they try to post to LJ and their current time is earlier than the timestamp on their last, crossposted entry. (It's not a serious error, but it requires the user to backdate the new entry -- which means it won't show on LJ friends pages, which IS important -- or manually rearrange the entry timestamps.) Tabulas stores the user timezone offset, so it should accordingly correct the timestamp on LJ crossposts or send the timestamp the user sets on the Tabulas entry; either should work nicely. This should be minor to fix, but then I'm no programmer either, so what do I know?

Xanga also sticks extra space at the end:<p>[two line breaks -- translated BRs?]</p> With that removed, formatting is otherwise fine. I tested with Xanga Classic (free) and Premium in case there were differences; I didn't notice anything (and I was looking) but I also didn't have Premium's extra posting screen crap enabled either.

I personally wouldn't use STRONG for entry titles (I'd go for H4 myself), but that's semantics. The look of it is nice, I admit, and B is deprecated, so that works.

Tabulas itself crossposts to Xanga and LJ, *then* previews the Tabulas post, if the Preview checkbox is selected when crossposting. Um... there's ways around this (post as private without crossposting, preview, then edit and select the crossposting you want), but it'd be gravy if it handled itself instead of requiring special user tricks.

(I'm warning you, it's long...)
I'm just glad I'm not the only one who's heard of Lorna Doone, though I may be the only one who's also read it.

I've probably read many more of the children's books and it's just been too long -- I picked up a used young adult novel the other day because I liked the author, and discovered at the table of contents that I had read and loved it before. A rare word in the chapter title gave it away or I never would have known.

If coding for all browsers is like teaching a class of children, Internet Explorer is the special kid who munches paste and thinks 2 + 2 = 5. IE bad. Hate hate hate IE.

I spent a good chunk of yesterday diddling with designy stuff. Coerced IE to not mutilate my latest webpage mockup and fixed some strange errors on a template I had done for someone else. I have a guess that Mozilla slightly altered its rendering engine between the time I first completed the template and the time it began showing misaligned elements, but I can't be sure so I must assume that I didn't adequately browser test last time.

This is a mantra of web design, one of many: It must work in all browsers. Designing for IE is not designing for the entire web. Real webpages work in WebTV, on Tungstens, in Konqueror, and on Nokias, and are at least accessible using Internet Exploiter as well.

Not that I'm, you know, biased or anything. *doodles a blue "e" and stabs it with a pencil*

I think I'm becoming one of those people who goes on endlessly about single updates to their websites. How vexing, particularly given that I haven't even changed much to speak of, not recently. Except that I could list a litany of little things, but nobody would notice if I didn't point 'em out, so meh. If that little RSS bug would just clear up and I could settle on a border colour, I think I'd be ready to upload the final front page (inasmuch as anything online is ever "final").

In other news, Narbonic's five years of archives, normally available to ModernTales subscribers only, are all free temporarily! (I believe they go back to subscribers only on 3 July, but I could be wrong.) There are only two paid webcomics I would actually lay down money for, and this is half of 'em. Mad science is so freaking cute.

But I still don't like IE. *glares at it*

It's probably a very bad sign to be rocking out the Lords of the Rhymes and have your brain auto-correct the song. "I was born Aragorn but you can call me Strider" is wrong; he was born Estel, which means hope.

Hi. I read appendices. They're more interesting than reading about researching. *is soooo not a formal research type*

If somebody out there has an idea how I can get perfect scores on every quiz in a unit but 87% on the test, speak up already. I'm getting peeved.

I just wrote a screenful of text to my world religions class discussion. The topic was whether Christianity had been a good or bad influence on Western civilization. The unanimous opinion was that Christianity is a good thing, because churches have soup kitchens and are generous, or some such drivel.

I have a headache. My temples are attempting to explode with each keystroke. I have been awake and uncomfortable since three this morning. Worst of all, I have a deep dark love of playing devil's advocate.

So I responded at length, and I'm pretty sure I'll spend the rest of the semester utterly vilified. Some small part of my brain is screaming in distant horror, wanting to be friends with everyone instead of driving them off. But the rest of me is satisfied. Perhaps someone will take my questions seriously and start to think about the points I brought up. Just one would be a victory.

In true smartass form, I closed with an apology, not for the content of my post but for the length of my reply. I have the distinct feeling that, once I have slept and somewhat recovered, I may regret what I've done. I hope not. Somebody had to, in the interest of simple fairness, and good may come of it yet.

I finally managed to be on AIM at the same time as my good friend Mutt and we chattered thusly until my brain conked out on me. No more smartness -- all gone.

Mutt finally shooed me off to to my classwork by layering injokes and puns: Callisti! Pope Best-Man awaaaay! I could try to isolate the logic behind each word and how it fits together humorously, but I'm laughing too hard. Giggling seems to have roused my brain into a semblance of productivity, and now I feel loved and bouncy and quite able to finish my assignments. I even happily sang along with a song that normally makes me misty and mopey. Ain't nothin' gonna bring me down now!

I made a silly wish on a shooting star: "I want a kitty!"

Three days later, the subject of this photo was given to me.

Next time I'll have to be more specific.

Question for the day: If I visit a Tabulas site that has failed to use the SITE['HEAD'] template tag, I can still find its feed by editing my own Tabulas feed URL to use the other site's username instead (or using a bookmarklet to automate the replacement process). That is not the question.

If I visit a Tabulas site that is using its own domain name and has failed to use the SITE['HEAD'] template tag, how do I find its feed? For that matter, how do I even know it's a Tabulas site? The default Tabulas templates include a link back to Tabulas.com, but it doesn't seem to be required.

The part that really bothers me is that I am pretty sure already that my only answer is, "You can't." I can't find the information I want (e.g, the site's feed), even though it exists and is public.

That bothers me deeply, though I have difficulty explaining why.

I think it's because years ago, I was used to DiaryLand's templating system, which requires a link back to DiaryLand.com . Fail to include it in your template and the system will reject it. They don't do that with RSS autodiscovery tags (even though they bloody well should), but it strikes me that maybe forcing both the SITE['HEAD'] tag and a link to Tabulas.com would be good, the first so people can easily follow public entries and the other so people can easily discover how to get their own free Tabulas.

Maybe it just sounds like a good idea becase I haven't had any caffeine today, though. I'll let passers-by debate the issue while I get my caffy fix.

GenderAnalyzer thinks I'm a man. Everything of mine that I've pointed it to, it says was written by a man - except for one short story I wrote years ago, which is supposed to sound as though a man wrote it!

Oh, and it thinks all my mopey poetry is girly, but aside from those two exceptions, apparently my writing is mannish.

I want to see its algorithms. I'm betting it looks for a few specific words and ignores the rest. But which words?