I have started to create a list of "Chinese people, one should know" and a first section of the members of the Standing Committee. I have big problems in saving this list. 90% of the time Skritter fails to save the list. First, I just put in the 9 names without any further definition and wanted to save - failed. then I started to put in name by name plus definition and saved after every entry. for the first 3 or 4 it worked, then it failed again. After many tries I could add another successfully, but now again even 5 or 6 consecutive attempts failed.
I have another problem with image-URL. I've put under mnemonics img:plus the URL, but it fails to show the picture. In the URL there is "_" used, is this translated incorrectly?

I can't comment on the first one much, although the app engine did seem to be a little off last night.

You're right about images and the underscores. Just tested it. If you have one _ it's fine, but if there are two with stuff in the middle then the image doesn't load.

Byzanti, thanks for your reply. Do you know, whether this is a Skritter probem, a HTML encoding problem or a general HTML problem. On the wikipedia page, where I got the images from, these pictures are showing up properly and also when I copy - paste the URL into the browser, it also works fine. Only after transfering to Skritter, I got this problem, but I am not so familiar with the HTML encoding of "_".

It's because normally if you put something between underscores, Skritter will make the thing between italic. Eg. _this turns into italic_.

I guess the parsing screws up when it's part of the img tag.

(Incidentally, you can get away with underscores with a space either side though, eg, 我喜欢 _ _ .)

@Roland _ is a normal character in HTML and in URLs; it has no special meaning and can be used "as is" in URLs.

James, that is, what I have input into the mnemonics:

img:http://upload.wikimedia.org/wikipedia/en/thumb/a/ad/Hu_Jintao_Executive_Office.jpeg/245px-Hu_Jintao_Executive_Office.jpeg

That is, what Skritter makes out of it, when I try to open the link (at runtime):

http://upload.wikimedia.org/wikipedia/en/thumb/a/ad/HuJintaoExecutiveOffice.jpeg/245px-HuJintaoExecutiveOffice.jpeg

And this is, what then Google Chrome makes out of it, replacing by %3C:

http://upload.wikimedia.org/wikipedia/en/thumb/a/ad/Hu%3Cem%3EJintao%3C/em%3EExecutive%3Cem%3EOffice.jpeg/245px-Hu%3C/em%3EJintao%3Cem%3EExecutive%3C/em%3EOffice.jpeg

so, I am totally lost, what to do, as I put in the correct URL and cannot influence it any longer. Any help from the skritter team would be highly appreciated.

I get errors when I click both those links, so I'm not surprised they don't work!

You could just download the pictures, rename them without the underscores, upload them to a public dropbox folder or any webspace you have access to and then link there. That would also make sure your pictures don't get lost in case someone changes the wikipedia image.

That's basically what I do, except I also resize them first, and use my own ftp.

It doesn't take long, and isn't much effort, but I was just doubtful whether anyone else could be bothered with it :p.

Thanks for your suggestion. Anyway, I have to wait now,until we can save changes to a list again.

@Byzanti, I have put the image into mnemonics, but it's very small and hardly can't be seen. How are you doing it? Can you give me some tips on that, also how you modify the style sheet (if you do). thanks

The thing is Skritter limits it to 50px height. So if the height of the image is larger than that, the whole image will get shrunk until the height reaches 50px. I've no qualms with that - it stops things getting messy. It does mean however, that if you willy nilly drop images in, they'll end up small and unclear.

Anyway, what I do is I resize the image in photoshop to 140*50. When it's this wide you can actually get a lot of picture in there, and it looks really good. I sort of arbitrarily decided on the width, but 140 seemed reasonable as that's the default thumbnail width on baidu images.

Here's what my Skritter entries look like:
http://www.byzanti.co.uk/1.png
http://www.byzanti.co.uk/2.png

Anyway, the whole process has become a habit now, so from dragging the image to the desktop, resizing in photoshop, and dragging it back into the ftp takes under 30 seconds. However, I am finding that I'm spending longer and longer looking for good pictures :p.

Byzanti, thanks a lot. I'll try it, once I can save again word lists.

Byzanti, thanks a lot! I just tried it out and works fine! Cool feature!

I just used an online website to resize the pic and uploaded it on my own webspace.

I think the original problem with saving lists is now fixed. Let me know if you have any more troubles with it!

I'll also work on the bug where the formatting breaks when there are two underscores in the links.

Thanks, the saving is back to normal.
Nick, the photos in the mnemonics are pretty small with 50 px. Is there any possibility, to have larger pictures in there? Even resizing them, as Byzanti proposed, keeps them still small and difficult to identify.
I am doing these lists now, because I find it annoying, to walk into a DVD shop and I don't know the Chinese names of famous actors and directors, and even if I see their picture on the cover, I don't know, who they are. Same with politicians, if I see them on TV or in a newspaper.
I want to set it up in a proper way, once ready, perhaps after 2 or 3 month, I can make the list public, I think, other people may find it also useful.

Nick, I was thinking about the image size and what possibilities would be there:
- Just allow the mnemonics field to be larger, say 100 to 200 px,
- Allow img: also in the definition field, in this case, I could put the text into the mnemonic,
- Allow a mouse-click on the picture, to enlarge it.
If you want to check, you can go to my "Chinese people, one should know" list, Actresses, Gong Li, this is a 50 px resized image on my own server.

Images are allowed in the definition field. It's where I put mine!

The mnemonics field also expands if you put an image/text in it. Just put things on new lines.

Byzanti, thanks. Just tried it out and it works. However, the image is still limited to 50 px, so hard to recognize the person. Would be great, if bigger sizes would be allowed, too.

It's possible I could up the size limit on the pictures, but the larger they are, the less reliable our formatting gets. People are a particular weak spot, as their aspect ratio is wrong, and they have high detail requirements. If you resize it to just the head, instead of including the shoulders, is it big enough to recognize?

I think that might be a quote worthy of being added to the bottom of the page scroll list:

"People are a particular weak spot, as their aspect ratio is wrong, and they have high detail requirements."

On it's own it could have some interesting interpretations...

The bodies don't matter. Just use their faces. Our brains are hard-wired to recognize and remember faces.