Mr Mist’s Blog

Canon today announced a whole load of new stuff, and heading up the line we see the 1Ds Mk III and the 40D.

Some of the key features are, for the 1Ds -

• Massive 21 megapixel full frame sensor
• Twin Digic III processors
• Live view 3 inch panel
• 5fps shooting

Impressive huh? A mere six thousands pounds will bag you one of those.

And on the slightly more affordable scale, the replacement for the 30D comes in with

• Some kind of weather resistance
• 6.5 fps continuous shooting for a max. burst of 75 JPEGs
• 3 inch live view screen
• New AF system with 9 cross-type sensors

I say “some kind of weather resitance” as, reading around, it’s not entirely obvious how weather-proof the 40D is. It seems to vary from total to partial resistance (partial meaning that just the compartments are proofed.) Still, interesting none-the-less.
The 40D should be available for around 900 quid. Get saving!

To add to the entry I made yesterday commenting on the MT4 v WordPress situation, I spent most of today upgrading Laura’s site from MT 3.3x to MT4. Admittedly it may not have taken that long to have just done an in-place upgrade, but then I doubt that that would have been anywhere near as clean a job. Here’s some observations on the Movable Type 4 product -

• The installer wizard didn’t work
• It’s not obvious how you create your own style rather than using one of the supplied ones
• Publishing a site takes an age, seems much longer than it used to be
• The template editor interface can be slow, and has rendering issues in IE7 with highlighted syntax
• The save templates function is buggy, and lost bits of templates
• The modular approach is much improved on earlier versions
• You can create your own widgets and drag n drop them into sidebars fairly easily
• Default rich text editor doesn’t allow html code
• The interface does look very swish and modern

That’s your summary. In detail, then -

There seemed to be an issue with the installer at the very first step. The install instructions (or at least one set of them, there seems to be mutliple versions each slightly different) said to put the mt-static directory outside of the main movabletype directory - which I did - but then the install wizard produced an error saying that it could not find the directory, and asked me to input the full path to it. Sadly, try as I might, nothing I wrote into that box was going to convince the install wizard that the static directory was actually there. Now, this could be a quirk of the way my hosting is set up, or something, but the fact remains that I couldn’t use the install wizard. I had to manually edit the config file with the relevant values. (Since I had my MT 3 config file this wasn’t hard, but I wouldn’t really want to do it as a novice.)

So, having filled in those details, I got myself in to the system. It created a user for me and a test post and comment. It did not publish the test post and comment though, so at that point there was still no blog to look at. Not particularly user-friendly. I had to hit the publish button myself and build all the files before I could see a working blog, set up in the default style. So, I thought - next step - change the look and feel. I checked out the new style browser functionality, kinda similar to WordPress themes, except there is no functionality exposed within the program to edit the theme itself, you can only edit site templates. Quite why the ability to edit theme styles within the movable type interface is not present escapes me. Unfortunately, given the lack of documentation due to how new the product is, it wasn’t even obvious how you make your own theme. There’s no “create new theme” option or anything like that. I eventually clued in on how the directory structure worked, and found a “themes” area in the mt-static directory that had a few sub directories with names matching the themes. Copying one of those caused a theme to appear twice in the theme browser, and, after editing some comments within the screen.css file of the copied theme, I managed to create my own “theme”. Of course, that means that all the CSS editing for the theme itself takes place outside of MT, unless you fudge it by creating some sort of linked template to the screen.css. Not a big deal, but hardly seemless.

After setting up the theme (and during, to see changes) I published various bits of the site. Wow that takes a while. There are a lot more archive types than I was used to in earlier versions, but I still found myself waiting a while for a full rebuild. Of course, the benefit is that your end-users can then browse the static pages immediately.

Unfortunately, editing the templates is not as pleasant an experience as it should be, either.

Firstly, I had to turn off the syntax-highlighting in the windows. It broke the page rendering in IE7, was very slow indeed, broke copy and paste, and basically wasn’t worth it. At least that change stuck though. Unfortunately, what I couldn’t solve was the entirely random auto-saves, which seem to occasionally interfere with manual saves and cause the loss of half the edited templates. Quite a few times I was faced with an “error” message telling me that my recently saved template was incomplete. Yes, it was, because half of it that was there a second ago had vanished. Pretty sure it’s a save-based bug, due to autosaves taking place right after you manually click the save button, but pretty annoying none the less. Ironically, I didn’t actually lose anything due to this, because it had, neatly “autosaved” the undamaged data beforehand. Pretty annoying though.

On the good side of things, the templates as supplied are much more modular and logical than in previous versions, and the new widget system works pretty well. There’s a good drag and drop system for shifting widgets or code chunks around. It did take me a while to figure out how to get a conditional element to work based on category name, but I think it may have been easier to achieve with more documentation. As it was, it seems that certain MT tags that you can use in, say, a category archive, lose context when you try to use them in a module, even if that module is being ran within the category archive. (That complicated what should have been a trivial task). Overall though the modular approach reduces the amount of repeated code in the templates and keeps things easier.

Lastly, the default editor is a rich-text window. This is as you might expect. However, it seems that it’s a very literal WYSIWYG editor, as any code I entered, expecting it to format my images etc, was happily displayed verbatim as part of the entry. This may be what some people expect and desire to happen, but, personally, if I enter HTML into one of these windows, then I expect it to be treated as HTML unless I’ve put a code block around it, or similar.

I think that it’s only fair to say that the interface itself, dispite the bugs that I came across, does look and feel very modern and classy. Most of the time at least, the context menus direct you to places you want to go. (Some times I find myself wishing for another thing to be there.)

Overall, I’d say that it is an improvement on older versions, but I won’t be going back.

Edit - I noticed just as I was finishing this entry that, when you’re typing in your WordPress entries and it does an auto-save, the manual save buttons get greyed out for the duration. Maybe MT needs such a feature to avoid the corruption that I was experiencing.

Mashable has an entry comparing MovableType4 with WordPress 2.2. Fairly interesting, they draw WordPress as a winner just now but expect close competition. I used MT for quite a few years, and got on with it fine I suppose, but, having made the switch over, I think I would take a lot of convincing to go back. I find WordPress to be much more accessible, and I think that it has a degree of flexibility to it that MT lacks.

The Pages system in WP means that I can integrate things into the rest of my site much more easily than I ever could with MT, and, what’s more, if I want to use any of the blog features outside of the directly controlled pages, all I have to do is a php include to the header files, and the full WP functionality is exposed. *very* simple, plus it’s all written in PHP rather than perl so I actually have a half a chance of understanding it.

I’ll test out MT 4, because we will need to do that anyway for Laura’s site, but I most likely won’t change. I noticed in the mashable article that they mark down WP because its anti-spam solution is not built in. Kinda true, but then that adds flexibility, and the Akismet system is very easy to install as a plugin (in fact the plugin is distributed with the initial download, you just need to get the API key.) And, what’s more, it’s been a lot more effective at stopping me seeing spam than MT ever was.

So there you go. Blog wars. hah.

In other news, today marks the 25th birthday of the CD. Happy birthday CD! Gosh, where would we be without those shiny bits of metal, eh? Vista would sure fill a lot of floppy disks.

Wow. The world of blogging sure has changed since I started this blog back in 2001. Back then - even though that was only a few years ago - it all seemed a lot simpler. People formed links to other people’s site with webrings (essentially just a little text link to the next site in the ring), or you linked to some of your favourite blogs in your links page, or - if you were modern enough - in a sidebar.

Today it seems that all of that has evolved. Now, it’s all about "social networking" and "user powered" rankings and you seem to have to sign yourself up to any number of different ranking sites and odd gadgets that you can click on in order to rate pages.

I’m sure that these sites, like Digg, Myspace, Technorati and so on are wonderful for the 1% of sites that reaches the exponential pinnacle of everything else out there, but I’ve got to admit to having a few issue with them, too -

• Buttons, Buttons, Buttons.
• It seems like the love has gone.
• What about the majority?

Buttons - There’s so many different ones. I’ve seen weblogs with entries with bars at the bottom where there are buttons linking to ten or more different ranking sites, sat there waiting to be clicked on. But how do I, the reader, know which one is worth clicking (if any?). How do I, the content-creator, know which button is worth adding to my entries (if any?). There’s no site out there to say which ranking buttons to use. Or if there is, then it’d be pointless, because everyone else is already using those buttons. What hope has your click got to make any major difference? In short, too many buttons, not enough clarity.

I could just be making it up, but it just seems like the whole thing is very clinical. You add buttons to a site because it’s the way that things work these days, people click buttons on sites because it’s what you should do. The claim is that the sites are user-led, but in reality it seems to be led by group conciousness and trend. Adding a button doesn’t offer with it the same sort of personal touch that used to come from directly-linking to someone else’s site, or from specifically joining a web ring of closely related sites. I suppose the closest you would get to that these days would be to join a "community", but even there, a successful community is massive and generically faceless, and a small community brings with it little of the supposed benefit that is meant to be on offer (since it wouldn’t be necessarily highly promoted or participated in.)

Lastly, I have to look at how these things help the general blogger. People at the top of the pile can, I’m sure, benefit massively, because the exposure from being in the top of these highly visited sites is massive. That’s great, but it’s also operates as these sites’ own worst enemy. To explain, popularity breads popularity. That means that a few sites will, through already-generated popularity, regularly be at the top of the stack, and being at the top of the stack brings more popularity - which keeps them there. This means that the whole concept of the continually-evolving user-generated list is flawed, because it will remain generally static, with the same key sites always sitting near the top. Yes, I expect you’ll get quick wins on some occasions that will produce a momentary change, but in (short) time that will even itself out and revert to form. What all of that means - as far as I can tell - is that, from a content-side, the vast majority of sites don’t benefit (much). From a user-side, I find that I can’t access the niche content, because it’s never going to be ranked up.

If a recommendation site is only telling me about sites that I already know about, what good is it? I’m just not sure. If I’m just starting out - as user or contributor - where do I begin? I’m not sure about that either really. I get something of a feeling that I could be missing the point - but if the point isn’t obvious, then that’s not a great win in my book.

got my k800I today. It’s quite neat.

There are lots of buttons all over. I am currently finding it hard to hold it without pressing one button or another. I imagine that this will change as I get used to it. It is so much smaller than my bulky old P800, but still has enough weight to it to not feel plasticy.

One of the selling points is the 3MP camera, so I’ve taken some quick test snaps with it, the auto focus seems to work fairly well, though is a tad slow. The best pic feature is fairly neat, you can choose from a number of pics snapped around when you press the button. At some point I’ll take some comparison pictures with the phone compared to my DSLR, as with 3MP you should be able to get a reasonable picture.

There’s a USB cable supplied, and I’ve been able to transfer a music file from the computer and play it on the phone, sound quality is fairly good. There’s about 60MB of memory in the phone as supplied to me, you can get memory stick micro to add more.

I cannot discover how to remove the cover for the memory stick micro slot. If anyone knows how it opens, let me know.