Collected thoughts on Google Reader 08 Oct 2005

Having already a Gmail account, it was no trouble to fire up Google Reader, so I thought I’d give it a try. It’s not really worthy of an entire essay, but here are my thoughts for improvement:

Speed

Yes. It’s slow. Lets get that out of the way, assume Google are working it, and analyse the actual potential of the software.

Experiences

I imported flawlessly my feeds from Bloglines, and had a go at labelling them. By the way, if you’re doing any serious feed editing, you want to check out the button with the little downward arrow, just to the left of the filter box. It expands the subscriptions box so that all the feeds fit in them.

Naturally, Google are sticking with their ‘persistent tags’ (labels) metaphor that Gmail pioneered. Good. They’re great. They give the spoteneity of tags a bit of permanence and rigour, while still overcoming the limitations of the file/folder paradigm.

Why can’t OPML be extended to mark up which items I’ve read and which ones I haven’t? I know it’s a generic outline language rather than a specific feed reader format, but that’s what it’s mostly used for. It’s extremely frustrating having to wait for people to publish before i can start using Reader.

Starring

Another feature imported from Gmail are Stars: if you like an entry by someone, you can give it a star. It then stays in your starred list. (Aside: I guess when you star an entry the text is copied to Google’s servers for permanence, otherwise when you star an entry it would just disappear when it drops off the end of that feed).

Stars at the post level are nice, but I’d like to add stars to specific feeds as well. They’re my priority, must-read-when-a-new-item-appears feeds. I’m making do with a ‘favs’ label, but stars should just be extended.

Secondly, this starring feature is a bit like del.icio.us but without the social side (I’ll come to that later). Why not just integrate with del.icio.us, and instead of internally starring an item, share it with the world?

Sociality

Social software is red-hot at the moment and yet I see nothing of it in Reader. I want lots of features here: how many people are subscribed to feed X, what people are labelling it, how many other people starred that entry I just starred, and so on. So much potential here, but Google seem to be turning a complete blind eye. We’ll see.

xmouse restructure 10 Sep 2005

One of my ongoing projects is a planned restructure of my primary weblog, xmouse. The justification is pretty simple: I don’t write to it any more. I haven’t written a proper essay to the site in a really long time, and that was the initial purpose. And so, here’s the plan.

There will be three sections. I’ll keep the old essays around, they’ll keep their URIs (xmouse.ithium.net/essays/blah), as the structure is date-agnostic (desirable for essays, which shouldn’t IMO be organised chronologically), short and generally works pretty well.

The second section will be the asides. I guess these will vary in length from a single sentence to a few paragraphs. I’ll probably organise them by date, and naturally the URI structure will reflect this (I’m thinking xmouse.ithium.net/[year]/[month]/blah). I’ll probably give each a short title, this will just help when going through archives. I may also explore things like syndicating links from my del.icio.us. In terms of implementing the asides, the general consensus seems to be that a dedicated Asides category is the best way to go. Although this has its ugly sides, I’ll give it a try and see where I get.

Thirdly, my wiki. This will be a catch-all service for my half-baked ideas (explore ways of posting mind-maps/outlines?), formal specifications and documentation for my projects. In the long term, I’d like to move everything off my current solution, Mediawiki, and onto WordPress. This has a few advantages: sexier and unified (with my blog) UI, the WordPress plugins, a single DB. I’d like to write a few plugins to make the WordPress Pages feature more wiki-like: I’m thinking an edit-in-place plugin to make Pages really easy to edit (wikis are meant to be fast), and possibly one to show a message along the lines of ‘A page with this name doesn’t exist yet’ instead of a 404 when a non-existant Page is requested.

Thoughts on that structure?