Some of the scienceblogs have been experiencing difficulties (unresponsiveness, sluggish commenting, and some of us can’t even post), and part of the problem can be traced to spammers turning their attention to us and pounding at the door. We’ve been asked to increase the level of security in commenting to suppress some of that activity, so now comments here require a valid email address.
Switching the level of security has an unfortunate side-effect with some browsers—it confused my copy of Safari so much it wouldn’t allow me to comment on my own blog. One solution, though, is to get rid of the old cookies for the scienceblogs.com domain in your browser. If you’re getting a “Comment Submission Error”, try that solution.
If anyone wants to leave other bug reports here, I bet the developer will see them. I know of a few glitches:
- Text encoding incompatibilities—this ISO vs UTF-8 stuff that I barely understand.
- The search page (and others?) don’t seem to be php-enabled, so funny stuff appears.
- Syndication includes only the entry body, and if there is an extended entry, there’s no indication that something is incomplete.
- I want a captcha system for comments. Some spam still gets through that I need to manually delete.
- No Pirate Mode and I need more logos! OK, this is still a very low priority…more serious bugs need to be fixed first.
I know there are lots of these little things floating about, but overall I’m so happy with the improved performance that I’m not too concerned, yet. And of course, the number one priority has to be correcting major bugs like Orac has had.
Hey! If you check your syndication feeds, you’ll see that now not only is there an indicator that there’s more to the post, but there’s a link to and count of the comments. See? Complain here, the developer will notice, and things will get fixed.
If you can read this, I haven’t any bugs to report.
‘Course, if you can’t read it, how are you going to know…?
Valid e-mail address, my ****.
it takes [email protected], is that bad?
Mine are just usability, not function, but they are annoying.
Quoting multiple paragraphs looks like it works correctly, but in reality, only the first paragraph is quoted. All the others revert to standard, which is just asking for misattribution.
Also, back on the old Pharyngula, individual comments had their own reference; here, all the comments are one big ungranular body. Makes interrupting Mr. Raven’s work by referring him to particularly good (or bad) comments less straightforward.
Thanks!
Individual comments are linked from their datestamps.
I think the lack of full entries/”continue reading…” in the RSS feed is pretty high priority, though.
Hmmm…. hopefully removing the cookie will work.
Thank you PZ, that helped.
For the ISO vs. UTF8 and other nastiness — use View Source and look at what’s actually on the page. It should be ASCII. Your “web designer” is probably too young to know about ASCII, but it’s what the Web is supposed to use. See the last line of this for a link to John Walker’s web page and have your web designer use the “Demoroniser” tool there.
Here’s an excerpt from a splendid summary here:
http://www.dotatdot.co.uk/clix/viewtopic.php?t=50&highlight=ascii
QUOTE:
Now to the point: there are only so many characters that are represented in ASCII. More precisely there are 128. That’s it. No more. IBM had their own 256 character ASCII and Apple have had their MacOSRoman, but these are not standards. MacOSRoman is however a single byte character set ….
” …Fancy quotes are dirt. They are NOT ASCII and cannot be saved as such. …. You can’t do much about the strained Unix literature doubled quotes (as opposed to real double quotes) which to my mind look absolutely horrendous, because they’re legit ASCII.”
“John Walker of Autodesk fame wrote a fantastic piece about this. He also wrote a Perl script called the ‘Demoroniser’ because he assesses most of the damage is caused by Microsoft and as he says, ‘when you see something stupid rearing its ugly head, you can be sure Microsoft are lurking right around the corner’. If you visit his site you can find the Demoroniser and his discussion of it. We’ve featured it many times in our newsletters and again just the other week. Like everything at John’s site, it’s a very worthwhile read.”
http://fourmilab.ch/
END QUOTE
Here ya go:
demoroniser – [Perl script to] correct moronic and gratuitously … incompatible HTML generated by Microsoft applications.
http://www.fourmilab.ch/webtools/demoroniser/
Thanks, Todd!
Also, check your site here:
http://validator.w3.org/
If you could get rid of the new background on the post titles (eg. behind ‘Bugs’ on this post), it would make scrolling much faster.
No problem :-)
Since the times ripe for complainingg, something I’d kill for is an RSS Feed to comments. When leaving a comment, adding a feed is the only way to go to see if people leave replies…
I am missing e-mail notification of comments. That saved my life for a few months back when I had to use an ancient computer that could not load Pharyngula as a whole, but could do one post at a time. Thus, by leaving one comment, I got to read all the subsequent comments via e-mail. next day, I just click on “Next post” and repeat the process. I am spoiled, I know – but I forget where I left my comments and I’d like to know if anyone’s added something interesting to the discussion.
“Text encoding incompatibilitiesthis ISO vs UTF-8 stuff that I barely understand.”
I’ve solved that problem with KwickXML on PT and my blog. Standard text formatters on MT don’t handle UTF-8 text properly. You have to do some simple Perl tricks to get it to work.
hank: No, no no. The modern web page is supposed to use whatever the author specifies it to use (and failing that, something sensible like UTF-8).
The Internet is a global place and scientific Internet doubly so. There’s no point in trying to intentionally avoid using standards suitable for that global communication. It’s not just just Japanese and Chinese — I can’t even type my mother’s maiden name in ASCII.
Using only ASCII in the Web is like distributing software on 8″ floppies.
I don’t know what kind of deal you have with ScienceBlogs, but I miss the Pharyngula “brand.” For example, the ScienceBlog favicon is pretty boring, the old Pharyngula one was distinctive. Even with it sitting in the midst of many on my toolbar, I could easily identify the Pharyngula logo and click my way here. Now I need to add text to tell one ScienceBlog blog from another.
I absolutely hate “Click to read the whole post” links in my RSS feed, because what I want is to read the whole post in my RSS feed. My RSS reader replaces my browser for blog reading, it doesn’t supplement it.
I do appreciate the link to comments, though.
When I click on comments for the “Summers is out for school” post, I don’t see any comments. I only see the left and right columns of the blog.
Any suggestions?
if you can see this, the email validity criterion is no more strict than it needs be. if you can’t see this, well, no biggie…
Things are improving. I can click on the comments for the “Summers is out for school” post and find them below the logo for University of Minnesota Morris at the end of the left column. Before it looked like an error message was popping up in the right column but it was truncated. That message is no longer popping up.
I have to admit I’m using Internet Explorer. I wonder if this is the same problem GrrlScientist had last week.
I still have a couple minor concerns with the feed. Right now, every entry has a link – [Read the entire post …]. It would be nice if only posts which had extended copy (not shown in the feed) had this link. Or some other way to indicate that “there is more…”. Or just skip all that and just put the entirety of every post in the feed.
Also, it seems everytime someone leaves a comment, that blog post is marked updated on the feed, which makes it hard to know if the actual post is updated or not.
That came out better as a post that it looked in preview.
Dare I suggest that *not* having the number of comments listed for each post would be better? This means that every time I check your feed (and my feed reader does it quite regularly) and there are new comments, I get your post updated. I’d like to be able to subscribe to the comments on a post I’m interested in, but not get constantly reminded about a post I’ve already read just because there are more comments – it clogs up my feed reader.
Addendum to my above complaint: If you subscribe to the Atom feed instead of the RSS one (substitute “atom.xml” for “index.xml”), the comments aren’t included, so you won’t get spammed. I was wondering why The Intersection wasn’t spamming me ;)
Oh please scienceblogs administrators, don’t change the Atom feed too!
(end gripe)
Antti == yes we all agree it should work.
My point is that it doesn’t work right yet.
As Reed pointed out above, “Standard text formatters on MT don’t handle UTF-8 text properly. You have to do some simple Perl tricks to get it to work.”
We need a way to PASTE UNFORMATTED == just the text == and let the web page manage it. Right now people paste in what looks like quote marks, etc. but is in fact any of a variety of HTML strings, which then get mangled on the way to subsequent readers.