At this time, the forum doesn't use that feature. All posts are approved by default. As for the clicking once, you'll probably won't see it but there are additional options for the administrators.
I have finally moved Phorum's configuration data from the files to the database on my development server. Meaning that it is now possible for me to update this forum to Phorum 3.4.8a.
For those who are uninformed as to why that is important, I had wrote: Sourceforge had configured the servers in such a way that scripts cannot create or modify files. This is a big problem because to install Phorum, it needs to be able to create and modify files.
I'll be spending the next couple weeks tweaking the code base, so this forum upgrade may, hopefully, happen on 8th or 9th of April.
Did you know that Phorum 3.x has the ability to use e-mails as a method to receive and send forum posts? I don't know how to use that feature yet and I doubt it would be useful for us, but it's still nice to know. :-)
It would be good if you iluminate us telling briefly the differences between current version 3.3.2a and the new one 3.4.8.a. I assume it must be a very good reason not to update to the latest version (or may I be wrong thinking there are newer versions).
There is a Phorum version 5.1.10 available (there never was a 4.x.x series). We could use that, but I'm guessing that I also have to remove it's dependency on configuration files, just like I did with Phorum 3.4.8a. But feel free to view it in action at http://dev.phorum.org/phorum5/
In any case, I don't see a strong reason to stick with Phorum 3 or move to Phorum 5. Personally, I rather like the current look and feel better than Phorum 5. Perhaps I'm just weird. ;-)
As for the differences, it's significant enough but I'd rather not mention them at this time.
I've taken a look to the phorum linked and I also dislike it's "look". I think you did it right to choose the old style 3.x.x series (I suppose it looks like the current one).
Is the 3.4.8a the last version of the 3.x.x series?
The look is not a problem, you can change it to whatever you want but:
- Migrating the current database to phorum need probably too much resource to be possible.
- Unless you have the evidence that phorum5 can be faster, it's not worthwhile.
<<Did you know that Phorum 3.x has the ability to use e-mails as a method to receive and send forum posts? I don't know how to use that feature yet and I doubt it would be useful for us, but it's still nice to know. :-)>>
I did not know that, but that would be a good feature to my mind.
Does anybody know what is the random parameter in the URLs, in the Sourceforge.net and Chogqed logos on the left, for? I would like to remove them because those random values makes every page unique and unique pages are uncacheable.
Also, those random values forces our browsers to download those logos over and over. For those who know the HTTP 1.0 or 1.1 specs, wouldn't it be better utilise the HTTP 304 status for them? The hit can still be counted but much less bandwidth is used.
No but IMHO You do the work so You can decide what is best.
Besides that: I agree with You: 304 should reduce bandwidth for SF and waiting for us. i assume that is what You intend. (Code 304 - Not Modified -> uses cache).
Hit counters:
Maybe near the download statistic pages?
But i never found any valid statistics like that for forum
- maybe ask SF if they make some?
We are no commercial project where the customer data are needed - we should try to minimize resource use. greetings to MY from DE
I can't be of any help, I know little (or nothing) about it, but, as it looks like you're the best informed here about this, your opinion is the most valuable, so do what you think is best.
You could try to directly remove the random code and "see what happens", or, if possible, why don't you make a duplicate of the web and experiment with it?
Silly me, it took me awhile before I realised that I should remove those random values in here as well (I was only doing it on my development copy).
The Chogqed logo isn't a hit-counter because it links back to a GIF file in our website, so it was quite safe to remove the random value. The Sourceforge logo however had some additional HTTP headers to force the browsers to download a copy of the image every time. I can't do much about that, but I can remove the random value in its URL.
Without those random values, this forum will now properly return a "Not Modified" if the page hasn't been modified since your browser last cached it. You can easily see this if you have a proxy server which shows you the HTTP headers (proxomitron and privoxy can do this). Now you don't have to worry about wasting bandwidth when clicking on topics with 100+ posts. =P
However, the server will still need to generate those 100+ post pages before it can know whether you can use your cached copy or it has to send you a new copy. So please don't click too many times for the fun of it. ^_^;
Turns out that the upgrade isn't going to happen today. But there has been some substantial progress, like the removal of most <table>s for layout in favor of CSS and mostly validating as HTML 4.01 Strict.
The problems I face is mostly with legacy coding style, they had use global variables everywhere and made heavy use of <table>s for layout. It's a headache trying to understand what they do and replacing them. I won't be replacing all of the old code, mostly because it isn't really worth the effort.
I'm also thinking of introducing a third way to view the forum (not yet coded). My current issue with the existing threaded view is that it displays one message at a time. It would be nice if we could view all the messages at once in threaded view. And it would also takes less bandwidth.
I'm going to try (later today) to put up a preview of what I have done at either the backup project site on Sourceforge (which Brian had created) or on my site. I'll be using a small portion of the forum database in either case.
This would be wonderful because we could compare both systems and let everybody say what they think about them, so we could make an agreement about what's better.
Thanks for the update.
I'd suggest testing it on the sourceforge servers so that you can be positive that everything will work the same as it will on the real website.
I definitely prefer your use of CSS to tables, especially considering it means (a little) less bandwidth is used. I've taken a look at the forum code a few times and I would definitely say I commend your efforts as I couldn't seem to make any sense of hardly any of it. The structure seemed messy at best and, as you said, there are global variables everywhere. I know I definitely appreciate what you are doing.
Hopefully there is sufficient warning to inform the casual visitors to don't post in there.
Take note that I haven't modified or tried or tested any of the administrative interface at all. And existing moderators and administrators here should be able to login in there as well. Feel free to test whatever you want that you didn't dare to try here. I can easily restore that forum's database. And please inform me if there were problems.
Also, please ignore the "Threaded View". It's broken, even if it wasn't it won't work for IE, I don't know how to fix it the way I want it to be, and it won't appear in the final build for this forum. I will be reverting it back to the original code for "Threaded View" and work on providing a second Threaded View option that works differently (as talked about in my earlier post).
And please don't do any speed comparison. That forum is using only 1/16 of this forum's database. So it would be completely unfair. However, I may test with the complete database here if sufficiently prodded to do so.
Oh, and that forum's database is from the 27th March backup.
Looks good to me. I've noticed that the fonts looks smaller than in this forum.
I don't know if it's an issue or not but the http links are not working on the experimental forum (see my post on the general forum of the experiment).
The menu fonts are slightly smaller, but the posts are slightly bigger. I used ems to specify the font-size, and they work via relative values and not absolute values.
As for the linking problem, I know what is the cause. Working on a solution at the moment.
That was done with
[ url ]http://kmeleon.sourceforge.net/forum/read.php?f=2&i=4900&t=4900[/ url ]
and
[ url=http://kmeleon.sourceforge.net/forum/read.php?f=2&i=4900&t=4900 ]http://kmeleon.sourceforge.net/forum/read.php?f=2&i=4900&t=4900[/ url ]
It seems that the cause of the problem was a poorly constructed regex. But my fix won't fix the old posts. That regex was used to modify the post data before storing it in the database.