| Comments: |
|dc||moved the old comments to || |
|PaulO||Are there any plans to improve upon the flash-based grammar quiz (in particular to address the visible area/font sizing issue)? I ask because it's a great learning tool and I think it's fairly unique. I'd help out myself but I'm not familiar with flash programming. OTOH if it's not much different to perl or C++ I'll have a go ...|| |
|eastmael||The site's new look is great for me! :) The lighter color makes it much pleasing to the eyes. Whoever made this made a good job! Keep it up man.. I'll just start looking for points to improve.. :)|| |
Anyway, i quicklyl hacked it to fix that line-length bug (and change the yellow bounce everyone complained about).
It doesnt look -ahead- to the next long item, so sometimes still shows the problem if the prev item is near the end of a line and another long one comes along (at least i think this is what i expect will happen, havent tested it), but much better.
if you (or anyone else) want to hack it, please do. the source files are here:
this also shows how to pull questions out of our database, so theoretically people could whip up their own format flash quizzes too... maybe add some characters + make it into a cool gameshow!
this was the first real flash coding thign i did, so its a bit messy, but there are comments in there! and it only took me about 10 mins to figure out what was going on even now months after i originally made it - so not too bad.
|dc||btw please let me know if it works, did this fix in a rush.|| |
|bi-ru||looking very good !|| |
|PaulO||I'll have a look at that code when I've got an environment to look at the files (they have binary data in them I think). Thanks for this dc. Make sure you license it properly though (I know from experience that someone making money out of your IP can leave a bad taste)!|
There's certainly less incidence of the issue - thanks for fixing this. Where it remains, the problem appears to be excessively long pre- or post-grammar sentences. It might be worth considering allowing up to a maximum example sentence length in grammar entries to prevent the issue (there's a tradeoff I know).
|Ronron||The design is great. Really you have done a real good job. It's exactly what was missing to me. |
|devricha||Wonderful site.. nice, simple design... keep it up|| |
|TBS||The design is fine - but what did you do to the "grammar a day" newsletter? I'm using Outlook 2003 (Japanese version, Japanese Windows) and all I get is "mojibake" in the mail header and the mail itself. The mail only becomes readable when I open the message and manually set the encoding to UTF-8. But this does not affect the header, which still shows a meaningless line of characters.|
I never had any problems with reading/viewing the mails before. Is there a way to restore the old status? Or a trick how to make Outlook 2003 show the correct header when receiving UTF-8 encoded mails?
Would be happy about any suggestions.
I recently switched the emails from Shift JIS to UTF-8 (I had to for various reasons). Unfortunately, it appears that Outlook does some strange things with international languages. As you noted, it isn't too much trouble making it display the body in UTF-8 but I haven't found a way to make it show the subject properly. So far, this is the only email program with this problem that I am aware of (could be others, though). I read somewhere that newer versions of Outlook don't have this problem, but my version is 2002 so I can't verify if that is true. Of course, one would think that any Japanese version would display Japanese properly! Anybody have any ideas/solutions?
|赤毛||I wish to inform Yokoso and DC that the mail corresponding to /JPLT2 is truncated in the version distributed today 三月一日 at 10hUTC.|| |
Thanks for the alert. There was a small error in one of my scripts that hopefully is fixed now. Please do let me know if you notice any more problems in the future.
|TBS||I think it's not a matter of the Outlook version. I think I have the most actual one (Japanese Outlook 200712.0.6212.1000).|
I experimented a little bit with my email account; basically outlook handles also utf-8-encoded e-mails properly. But when I compared the mail headers, I found something interesting. This is your email:
Compare to my testmail, also utf-8 encoded:
Subject: test grammar a day =?UTF-8?B?fuOBq+OBpOOBkeOAgFtuaSB0c3VrZV0g?
In your email the japanese characters are transformed to some hex-format, my mail does something else. Yours appears as mojikabe, mine is displayed correctly in Outlook.
All e-mails sent by you come with this utf-8 Q, my mails (and all other correctly displayed mails) come with utf-8 B. I have no idea what's behind this, but maybe there is someone else out there who is a little bit more kuwashii in these matters?
It really is a mystery. I don't rule out the possibility that the problem rests on my end, but Outlook is the only email program I have heard anybody mention having problems with so far.
As for the Q vs. B issue, my investigation indicates that Q is used when the majority of text is not special characters and B is used in the opposite case. But, to be thorough I modified the mailing script (phplist) to force it to send using B (?UTF-8?B?) and it still wouldn't display properly on my version at least (Outlook 2002).
I've searched quite a bit on the phplist site, Google, etc. and just can't seem to find an answer to this problem. If any encoding/computer science geniuses are out there, I'd love some help.
|TBS||Whatever you did - it now works fine again (at least for me...)! Today's mail (kke) is displayed correctly by my Outlook 2007 from the first to the last line, including the header, without any manual correction necessary.|
|yookoso||Well, I am embarrassed to admit this, but I had set the encoding to UTF8 instead of the officially correct UTF-8. I'm glad to hear that it works for you again. For my Outlook 2002 the subject still won't display properly. I read online that this might be true for earlier versions of Outlook so if anyone else can verify, please do. || |
|赤毛||After adding an example in the entry, Ex # appear before #. Do you understand why?|| |
Sometimes the database isn't sorted by entry number and I wasn't specifically ordering in my selection criteria, but now it should be in order.
|赤毛||The entry [amari-2 + verb in NG] created by Amatuka seems to have disappeared. Do you think you can restore it or shall I create a "amari-nai" entry instead?|| |
That entry (along with amari-2) had some mojibake in their primary database field and also the script wasn't properly encoding the url so these non-typical characters were preventing the entry from being displayed. That should all be working now.
|赤毛||Perhaps it is something personal but I am not able to access [amari-2 + verb in NG] already. Are You?|
Would you mind try to call this entry without so much blank characters?
|yookoso||Well, I double-checked and had no problem accessing amari-2 + verb in NG, but just the same I took your suggestion and renamed it . || |
Now I think everybody can access this entry.
Lately I can't access [stem+dasu] entry with my Mozilla Firefox browser nor my Internet Explorer one. Do you understand what is happening?
The change I made to fix the URL encoding needed to be made in more than one place in the script and I had forgotten to change all those locations. Hopefully I have them all fixed now, but with your ever-sharp eye I look forward to hearing if I did or not.
|赤毛||I didn't get my email "Grammar a Day - Level 2" at the usual timing. Do you have a trouble with your chron table today?|| |
|赤毛||I got at at 17hUTC (usual time is 10hUTC) the same message I had yesterday. I don't think "Grammar by email" is OK already.|| |
|赤毛||御疲れ様でした。"Grammar by email" is OK now!|| |
|yookoso||Occasionally the server used to send out the daily emails goes down and that will usually cause delays (and occasionally bigger problems) with the delivery of the emails. After the server is reset the next day usually will see the emails return to "normal". I am still investigating ways to make the setup more reliable. || |
|赤毛||The two last messages of "Grammar by email", beginning with "?UTF-8?Q?" work worse than the former ones in my computer (Mozilla Thunderbird version 22.214.171.124 (20080421) / Windows Vista Family Edition Premium).|| |
I'm actually a bit confused as I haven't made any changes recently. But, I wonder which email service you are discussing. The "Grammar by email" sends an email directly from this jgram site and then there is the "Grammar-a-Day" daily email list which is sent from a different server. I believe the former is sent in SJIS and the latter in UTF8. So I can be clear, which are you referring to?
|赤毛||I refer to the "Grammar-a-Day". Sorry if my confusion make you loose time investigating in the server which is OK. I made a registration for the JPLT2 last year. Mail is made every day around 10hUTC. The mail sent the 10th of June was OK (Grammar a Day - Level 2 [ たまらない (tamaranai) ]). The 4 mails sent after (for example the one of today 14th of June: ?UTF-8?Q?Grammar_a_Day_-_Level_2_[_=E3=81=AB=E5=BF=9C=E3=81=88=E3=81=A6_?= (nikotaete) ]) have their title wrong. But perhaps it is a release of Vista in my computer which generate this incident.|| |
|yookoso||I haven't made any changes and the email you mentioned looked fine for me using my gmail so it may be the case as you say that it is something with Vista. If so, it will be hard for me to investigate since I am still using XP. What email reader are you using? If anyone else is using Vista can you comment of this issue? || |
|赤毛||I have been checking the beginning of the messages. I found that my provider is using a new server, and I think that this new server have a problem with UTF8. Sorry for the wrong alert.|| |
|赤毛||Bagulloさん said it in the for JPLT1 but I noticed the same trouble in the JPLT2 too: Since August began, the "Grammar-a-Day" daily email is the same every day. Perhaps you should reset the server.|| |
see my comments and explanation on the general feedback page...
|mylz||皆様 お元気？！ |
|pedro_cristian||for WWWJDic links, maybe it would be nice to redirect WWWJDic request on the appropriate mirror according to the profile country|| |
|赤毛||4 times this month we have had 2 following days the same entry of the JPLT2 level of the "Grammar-a-Day":|
- 28/02 and 01/03: karaniha
- 03/03 and 04/03: kotonaku
- 14/03 and 15/03: nokotodakara
- 22/03 and 23/03: nishitemo
Perhaps you should make a check: If the result of my selection is the same that yesterday, I randomize another one.
The current script *could* lead to the problem you are describing, but that is not actually what is happening. Recently, I have been experiencing problems with my server maxing out its resources (mostly due to the growing size of the email lists I suspect) and occasionally the server is down when the script that generates each day's email should run. I can't actually determine the real problem as the server will sometimes go more than a month with no problems and then crash 2 or 3 times in a week. My host hasn't been able to provide any more useful information than suggesting I upgrade to a better package. Most likely I will switch to a different host/server in the future, but I am traveling internationally for the foreseeable future and don't know when I will have time to do that.
|赤毛||I think there is a problem with the "See Also:" section of #. Please check it. I think you should sign the contributions of "See Also:" the same way the author is mentioned in the "meaning", "notes", "examples" and "comments" sections. Friendly yours.|| |
I have removed those spam entries and taken your suggestion and added the contributor to the See Also entries.