New Blog. Kinda.
[Before reading, let me warn you that this post is technical in nature, and does not contain anything musical. However, if you are a musician wanting to set up/modify a blog, this may be useful.]
Some time ago, I mentioned that I needed to find a new blogging solution. The news is that, I’ve got one already, with some compromise.
I’ll now mention what it is and how it addresses the things that I need and avoid (as I already mentioned). Before that, do you notice anything drastically different? No? Neither do I. The main different thing is the engine that powers this blog. Previously I was using Blogger (which I think is great overall) and now I’m using Wordpress. I think Wordpress is a bit more complicated, but more flexible. In its raw form, the performance of Wordpress in terms of computing power utilisation efficiency is poor, but can be optimised so that it’s on par with my old Blogger blog. More on this later.
Now, what I need are:
- Free. Yup, Wordpress is free. No further explanation needed.
- Customisable template: Down to the HTML+CSS level. The level of flexibility offered by Wordpress is higher than that offered by Blogger SFTP publishing. And if you’re a geek, you should see that this blog is still using the same CSS. I have made so far zero change to the CSS (but may need to do that when I see something really different). [EDIT 4 Feb 2010: I have changed the CSS code for the comments only.]
- Static HTML page rendering: Well, this is one of the trickiest part. In its raw form, Wordpress does a lot of CPU processing and I/Os, and as such, not so efficient. Therefore, I use the WP-Super-Cache plugin so that when a page is accessed for the first time, the resulting HTML is cached to a static file. Given that the content of that page doesn’t change, that file is then served for subsequent requests.
- Keep the current post permalinks. Easy. Wordpress can be configured to use Blogger-style post permalinks, i.e.
year/month/hyphenated-title.htmlwithout any plugins. Individual comment permalinks break, though, but I’m willing to break them anyway. - Manual HTML input. This is also built-in in Wordpress.
- Web-based user interface. Ditto.
- Secure data transmission. Now, this is somewhat tricky. My music still doesn’t sell as well as Pat Metheny’s, so I can’t spend money on SSL yet. As a compromise, I use the Semisecure Login Reimagined plugin.
I’d like to avoid:
- MySQL, but I got no choice here. I hope Dreamhost will support PostgreSQL in the future.
- XHTML. Just edit the template and bye, XHTML. I’m using HTML 5 as much as possible.
Hopefully, nothing impacts the readers of this blog. Enjoy!
13 comments
I only understand point no.1
the rest are… umm.. dark cloud on the grey sky :D
wait, so now after using wordpress, how come your url doesn’t show any ‘wordpress’ ?
like adeishs.wordpress.com or something like that…
aha. :-) i can see where the confusion is. you know, there are some wordpresses. the one you’re using is wordpress.com. the one i’m using is the wordpress software that can be downloaded from wordpress.org. (wordpress.com itself uses wordpress mu, which is slightly different from wordpress.)
so, to answer your question, my url doesn’t show any “wordpress” because i installed the wordpress software in blog.adeishs.com. you can also achieve it by using your own domain with wordpress.com, and last time i checked, you’d need to pay: http://en.wordpress.com/products/.
Interesting, I seem to have a repeated feed of posts since “Keyboard Hunting Continued” (as in there are ‘new’ versions of these in feedreader). Which corresponds to both:
http://blog.adeishs.com/feed/atom
and
http://blog.adeishs.com/feed
Not the latest entry of this particular post, just one copy for this. Was this ‘re-feeding’ deliberate as some sort of context or test?
@kamal: sorry, i can’t follow you. how exactly did you get the repeated posts? i just rechecked both my atom and rss feeds, and i didn’t see any duplicates. perhaps your feedreader uses the
<id>tags and ignores the permalinks entirely? the internal ids of the posts pre- and post-migration are (obviously) different. this may explain why you only one copy for this post, as it was written post-migration.@Ade: Ah, that’s probably what’s happened. Kinda sounds logical though. :P The RSS version has …, the Atom one has … – neither seems to qualify as a ‘permalink’. I knew your feeds do not contain duplicates btw, was just wondering why I got repeated posts. No biggie, was just curious. :)
Grr – comment filtering, I was referring to the differences between the Atom and RSS ‘link’ tags.
@kamal: ?? the rss has
<link>, and the atom has<link>with therelvalue ofalternate. both of them are the permalinks. you know what, this whole thing is a mess.:-pI’m aware of the link differences thanks :P, My point was more that, firstly based on reading the description for ‘alternate’ in the W3C docs:
http://www.w3.org/TR/html4/types.html#h-6.12
Alternate
Designates substitute versions for the document in which the link occurs. When used together with the lang attribute, it implies a translated version of the document. When used together with the media attribute, it implies a version designed for a different medium (or media).
Versus no attribute in the RSS case; neither implies anything permanent or definitive about the document, it’s just a link to an alternative representation of that XML entry.
Whereas, IMV, in theory anyway, an id/guid speaks about the uniqueness of the document within a particular scope. Since the feed or channel – i.e. the scope – (more practically here, URL) has not changed, but a new id is received, it can only assume a new entry in that feed.
I understand that the ‘link’ tag in this case is specifically from the RSS and Atom schemas/DTDs, rather than HTML, but reading both at:
http://cyber.law.harvard.edu/rss/rss.html
http://www.atomenabled.org/developers/syndication/atom-format-spec.php#element.link
seems to imply similar semantics to the link tag from HTML, i.e. just a link, not an identifier :P.
Still I see your point, if they kept a cache of links, then yes they refer to the same alternative representation of the XML document. But then, from a feed reader’s view – it’s about the feed which it can interpret/reformat as it wants, the alternative representation is just metadata for someone who wishes a different view of the feed.
Am I missing something completely obvious?
@kamal: no, you aren’t, afaik. maintaining a guid when switching to different apps is impractical. besides, mere mortals care more that existing links don’t break. in this particular experience of mine switching to a different blogging app, maintaining the look and feel, permalinks, efficiency, etc. is a lot of work already. well, this whole thing is a mess. hee hee…
Agreed – more so from a search engine viewpoint too, you care about your website’s links. Thanks for indulging in the mess of a discussion nevertheless :P. Still I was impressed that the switch managed to maintain the feed URL too.
@kamal: also the blogger-format archive links. i literally tried every link during the migration to make sure all links would be preserved (if possible).
i resorted to apache mod_rewrite to sort out this thing.
@Ade: Hehehe, nice work.
Post a Comment