SubscribeWhy new features are being delayed
I promised a site redesign, and I’m sure you noticed it’s not there yet. Here’s a short explanation of what’s going on.
I made a list of all the features you asked for. The feeds not always working as expected, new comments are hard to see on the tracking page, and past 20 or so conversations, the tracking page is hard to use. And not surprisingly, people have been asking for tags, and other ways to organize stuff.
I realized that there’s a lot in common to these problems, and it’s easier to come up with one solution that will address them, and make it easier to add more features in the future. It turns out that solution requires some significant changes to the way conversations are retrieved from the database, a change that could easily take a couple of weeks to develop and test.
So I’m working on that. Meanwhile, I’ve also switched my development machine from Windows to Linux. It’s easier to develop the code, test it and deploy it using the same setup as the server. So a few days spent setting up Linux, copying files from one computer to another, getting comfortable with the new machine.
Linux may be great for development, but the hardware support is a bit lacking. So I had to work around issues with the video driver, flakey wifi card and suspend/resume problems. Most of it fixed by now.
So I apologize for the delay, I’m definitely working on these new features, I just need a bit more time.
Posted by Assaf in General on Wednesday, April 5th, 2006. You can follow any responses using co.mments, leave a response, or trackback from your own site.
Eljo
April 5th, 2006 at 7:20 am
No problem here. Take your time to make something nice and we will wait for it. For now It’s manageable in Bloglines with the one rss so don’t rush on my account
Make You Go Hmm: » Comment tracking competition, MAMP and WAMPANY development
April 8th, 2006 at 8:42 am
[…] Developer environments The developer of co.mments has a list of features planned to be added where he also adds that he switched development from Windows to Linux: I’ve also switched my development machine from Windows to Linux. It’s easier to develop the code, test it and deploy it using the same setup as the server. So a few days spent setting up Linux, copying files from one computer to another, getting comfortable with the new machine. […]
Singpolyma
April 11th, 2006 at 2:16 am
I was pleased to see that the co.mments parser recognises comments formatted with the subset of the XOXO Blog Format just fine, but the parser does leave in the ‘body’ label that should really be hidden. An example code snippet would be:
<ul class=”xoxo comments”>
<li id=”c836428″>
Posted on <a href=”http://filess.ning.com/view.php?id=815347#c836428″ title=”1143476310″>2006-03-27 16:18</a>
by <a href=”http://filess.ning.com/?user=singpolyma” class=”author”>singpolyma</a>
<dl>
<dt>body</dt>
<dd>One more test comment :)</dd>
</dl>
</li>
</ul>
the ‘body’ text inside the ‘dt’ tag should not be displayed
Another suggestion I have is a sort of Feed-based API or something of the like where one could call co.mments.com/api/?url=ANY POST URL and have an RSS (or other?) format return of all comments on that page, sorted by reverse date. That shouldn’t be too hard I wouldn’t think since it’s just a different output method for your existing parsing logic.
Assaf
April 11th, 2006 at 2:44 pm
Content should be displayed. There are ways to mark sections of content as having specific meanings, e.g. you can use the class ‘body’ to denote the part of the content that is the comment body, or (as some blogs do), separate comment identification into dt and comment body into dd.
A good place to learn about these concepts and see how they are being applied is microformats.org.
XOXO itself follows those guidelines nicely. I didn’t build XOXO support into co.mments, I wasn’t even aware I should do that. But the fact that it respects semantic content makes it work without any effort.
As for the second question, API is in the works and I’m going to add this to the feature list. I think it’s a great idea.
Singpolyma
April 19th, 2006 at 10:05 am
I’m not entirely sure you understood me as relates to the XOXO stuff… I encode comments on my blogs using XOXO (actually my whole blog is encoded in XOXO) and it’s just a bit annoying to me that when co.mments pulls in comments from my blogs the ‘body’ label text in the <dt> tag (see code example above) is displayed as part of the comment text.
Another thing is, why are links and other formats in comments not rendered on co.mments?
Could co.mments be made to pull in full comment permalinks (ie http://singpolyma-tech.blogspot.com/2006/04/cocomment-jsonp-native.html#c114426420682257372)
Assaf
April 19th, 2006 at 11:09 am
All the contents that comes inside the comment gets treated as part of the comment. For example, a lot of blogs use
in the content of the comment itself. So those elements get preserved to retain as much content as possible.
As for markup inside comments, that gets filtered out to retain consistency and prevent XSS (for likns). I do intend to let some markup and proper links pass through the filter in the future. I think that information is important, it just takes time.