dwlt.thinksOutLoud

I am currently reading Creating Customer Evangelists, in case you were wondering.

All Posts About Tech

Buffet of Death

So, we didn’t win any official awards last night (apparently Twitter didn’t get my text from the event, so sorry if you were waiting for that) at the wrap party for the 48 Hour Film Project, but we did receive a some praise for what we did from the organisers. And I don’t think they were just being nice. Anyway, the team is in no way disappointed, because we had a fantastic time even though we were all terribly broken by the end of it!

The film had to include the following elements: a restauranteur named Robert or Roberta Darling, a CD prop, and the line ”Is that the best you’ve got?”. We randomly drew horror for our genre, but we quickly decided to make it a comedy, and got to work. Amazingly, there were no arguments or lost tempers (at least that I noticed) during the work, and the script made us laugh even if no one else would find it funny. We finished writing about 3am on the Saturday, at which point the director and script supervisor began prepping the shooting script. We were all on set at 8.30am to begin filming and finalise the shooting script. It was the first time I’d been on a shoot, so it was very educational to see just how long each shot takes to set up and get right. The whole thing went relatively smoothly, though our picnic scene was “attacked” twice by dogs. We finished filming at about 1.30am on the Sunday, and then we submitted our final movie with around 10 minutes to spare.

On Tuesday night, the film was screened at the Cameo, and was in fact the first time most of us had seen the final product! We went all out on the marketing front, with t-shirts, free food, chef hats, fake menus and even a dedicated web site. The website probably gives you a sense of what’s going on, but the basic premise was that in the near-future food shortages will be so bad that one restaurant takes advantage of a suicide to serve real meat again. Soon, though, demand outstrips supply… Tagline: ”Because life is a picnic”.

We’re currently checking out what we can do with the finished movie, specifically if we can upload it to YouTube or similar. If we can, then I’ll let everyone know where to see it of course. I imagine it will appear on 48.tv eventually as well, which hosts all the films entered in the project from around the world.

The whole team is keen to work together again, perhaps in less time constrained circumstances (and also perhaps not!), so watch this space!


Kitbag Don't Like It When You Move

I should really have done this a long time ago, but today I finally got around to ordering some new footy shoes from Kitbag. I’ve ordered from them before, and was pleased with the service.

I picked the shoes I wanted, then logged in and tried to update my address (I’ve moved since that last order). However, their system wouldn’t update my details because I’d “made orders from the address”.

Huh?

The system let me place the order anyway, though of course I’ve got no idea if it actually went through or not since what Kitbag thinks my billing address is, is so not what my billing address is. I’ve sent them an email informing them of this incredibly stupid ‘rule’, but I’ve not heard anything back yet.

Update: They’ve now updated my address in the system manually. No word on whether they’ll sort the problem in the first place, though. Ah well.


Take A Deep Breath

Josh Kopelman has a much commented on post about how many – or rather, how few – people are aware of the Web 2.0 darlings such as Flickr, del.icio.us, Digg, or even MySpace. Josh points out that many of these companies, and especially the newer ones, can quickly obtain a user base of between 5,000 and 25,000, primarily driven from the 53,651 readers of the TechCrunch blog. He also guesses that the 10,000 most active users of the top ten Web 2.0 sites are probably all the same people – and there’s always some truth in jest.

Personally, I don’t think there’s any problem with sites having a user base of a few thousand – the trick is how much money you can earn from those users, from whatever method you’ve chosen to do so, and still be a viable site. However, I do agree with the underlying implication that the industry needs to take deep breaths into a giant paper bag.

Flickr is the only site where I’ve personally witnessed take-up and usage by “regular” people – it’s just a really great way to share photos, and it works tremendously well.


One-Minute Vacations

Following on from the 60 second podcast post, you can now take a one minute vacation:

Surely you can spare a minute to clean your ears? Take a one-minute vacation from the life you are living.

One-minute vacations are unedited recordings of somewhere, somewhen. Sixty seconds of something else. Sixty seconds to be someone else.

(Via)


Using FeedBurner

Pete Freitag’s post on guesstimating the number of people reading his feed reminded me that I hadn’t yet posted about me using FeedBurner on dwlt.net and with Tapestry. I first switched on FeedBurner for the Dilbert feed, to provide some information to United Media (it’s all gone quiet on that front, sadly), and it turns out that almost 42,000 people subscribe to that feed! Yow!

Even more surprising was the fact that 160 people subscribe to my own feed here. I figured that maybe six people subscribed, and one of them was me (to make sure it’s all working properly).

Anyway, to the point of this post: setting up FeedBurner was a doddle, but I was wondering if there was a way to set up the redirect transparently. Turns out that there is indeed a piece of mod_rewrite voodoo which I picked up from the FB forums and thought I’d call it out here, since I obviously have a captive audience :)

RewriteEngine on
RewriteCond %{HTTP_USER_AGENT} !FeedBurner
RewriteRule ^dilbert\.xml$ http://feeds.feedburner.com/tapestrydilbert [R,L]

These two lines set things up so that any user agent apart from the FeedBurner bot are redirected to the FeedBurner version of the feed. The FeedBurner bot itself will pick up the original feed for processing. Nifty, huh?


Mark Fletcher on Web Stealth Startups

Stealth Start-Ups Suck

...creating a new web service is not rocket science and does not take a lot of time or money. My rule of thumb is that it should take no more than 3 months to go from conception to launch of a new web service. And that’s being generous.

Mark Fletcher, who founded ONElist (now Yahoo! Groups) and Bloglines, has a short piece about building new web applications. It’s ostensibly a rant against using stealth mode in a web start-up, but is actually some ground rules for getting something up and running quickly and taking it from there.

The one bit I have a slight problem with is his assertion that first mover advantage is important:

Why is first mover advantage important? You get to define the space. Any future competition will be compared to you, which gives you continuous mindshare. Your service will become synonymous with the functionality you provide.

Which is great, but then says that ONElist and Bloglines weren’t first movers with the reasoning that the early competition hadn’t ”gained critical mass”. So why do we keep being obsessed with first mover advantage? Surely we should obsess over first mindshare advantage?

In other words, whoever gets the early mindshare – regardless of whether they were first – benefits from the word-of-mouth and positive feedback.


Tada My Site!

Like many, I’ve been using Tada Lists to keep track of things, and I have one list for this site which you can see on the sidebar. Initially, I wrote a little Ruby script which grabbed the RSS feed of the list I wanted to publish, and converted it into a little bit of JavaScript code which simply wrote a list into the HTML when the page was loaded (exactly the same as Google AdSense code).

Martin liked it, and asked if he could have it as well, so last weekend I turned it into a Rails web application, so now you can include your own Tada lists in static HTML. There’s no styling applied to the HTML which is output, but it is wrapped in a div which has an id of tadalist so you can simply add some rules to your standard CSS file to style it.

Let me know if you use it.


Greasemonkey Self-updates?

I wasn’t aware of this, but I guess that Greasemonkey scripts can update themselves? My del.icio.us tag helper has had some improvements overnight—when I install the script, is it actually just referencing a URI so the script author can republish it and the improvements just filter over to all users automagically?

Anyway, the improvements are nice :-)


Tag Ads Redux

Micro Persuasion: Targeting Through Tagvertising:

As you read this, millions of individuals are working under their own volition to create a new Dewey Decimal System for the internet. In the process—perhaps without even realizing it—they are laying the groundwork for a new contextual online advertising paradigm called “Tagvertising.”

Previously on dwlt.net: Tag Ads. If only I knew how/had time to articulate these ideas… I’m kicking myself for not coining the word “tagvertising” though. <annoyed grunt>


Adobe Buys Macromedia

Adobe to acquire Macromedia.

So does this mean we can expect to see Flash splash screens on PDF documents now?

Russ has a slightly more useful analysis.


Google Ad Links

Hmm. Via Signal vs. Noise:

Basically what they are doing is creating a contextual list of words/tags which, when clicked, take you to — yes, you guessed it — another page of ads!

Somewhat reminiscent of what I wrote about here. The tag bit, anyway.


Tag Ads

Russ on Ads:

There’s no context at all in the Ads. I’m just getting random text ads (or based on my general description of my site when I signed up). I just compared the ads on my iPod Shuffle review page. AdSense had four ads for – of course – iPod Shuffles. Kanoodle had five ads, three of which were for generic Microsoft Office products and the other two were equally as irrelevant. So that’s bad. I think most of my ad revenue (which pays for the server as well as shiny new toys like my Mac mini) comes from search engines, which means context is king. Without that context, I really fear the usefulness of the ads are going to drop tremendously.

Reading this reminded me of a comment I’d posted on Jim’s post on Ads way back in September 2004. Here, I proposed that a blog-oriented ad service could take advantage of tags (as seen in Flickr et al) to do the context matching, rather than scanning the content itself.

You would sign up to the service and tag his blog appropriately (in Russ’ case ‘mobile’, for example). Advertisers would sign up and tag their ads similarly. Voila!

Now, of course this breaks down when you look at the test case of the iPod Shuffle review page. But since many blogging tools allow each post to have a keywords field, or a set of categories, or even in some cases tags attached to individual entries the ad system could pick these up via metadata (either explicit or inferred) in the post. Not that different, I suppose, but perhaps different enough that the blog owner actually feels they have some control over the context, which was the original reason for the AdSense/Kanoodle comparison.

I guess tag bidding could work in a similar way to keyword bidding in AdSense. Or maybe some other way could be invented for this, given the recent how to bring Google down meme.

Yes? No?


Amazon Web Services Blog

Nobody told me about the Amazon Web Services Blog.

Seems they are having some sort of confab today and tomorrow:

For the next two days (January 19th and 20th, 2005), Amazon.com is hosting a number of well-known technologists and luminaries at a private event for our software development team. We (the Amazon Web Services Developer Relations Team) have special permission to blog each of these speakers in near-real-time, so watch this space throughout the conference for further developments. We will be updating each entry one or more times during each talk. Refresh this page (and the linked pages) to stay current.

Damn Small Linux is Damn Fine

Thanks to Damn Small Linux, I’ve successfully recovered my laptop! Happy days!

Basically, I just burned a CD with the 0.8.3 ISO image, and booted directly from it. It detected all the hardware appropriately, including grabbing network connections and my USB keyboard and mouse. Then I typed mount /mnt/hda1 and lo! I was able to access all the program settings, FeedDemon data, bookmark collections, and all my email archives and just tarball everything and scp it to my server box. No idea why Windows wasn’t interested in booting, but I do know that it had been playing up a fair amount recently, and I’d been considering doing a reinstall of Windows. Since it has basically saved my on-line life, I made a donation to the project.

To be fair, Sony have a well designed system on this laptop (a VAIO GR414), whereby they’ve split the hard drive in two, and recommend you put as much as possible on the D: partition. The recovery disk will only wipe out the C: partition (although I backed up D: just in case!). One slightly annoying point was that the recovery disk told me that back up software was not available, and that I should go back into Windows to back up my settings and data; how do you propose I do that, then? Grrr.

So now I’m the proud owner of SP2 as well, and I’m still reinstalling all my favourite software in the latest editions. I also took the opportunity to check out NOD Anti-virus software, which looks very good to me. And is a bit cheaper than some of the other options as well. It seems to run pretty much seamlessly, even for email scanning, unlike the others which take over the settings of your mail program.

Anyway, it’s nice to have a clean desktop for once in my life, as well as a machine that runs noticably faster than it did this time last week :-)

Now I can get on with some work…


Hybrid Devices Considered Harmful

BBC NEWS | Technology | Consumers ‘snub portable video’

The firm said gadget makers should avoid hybrid devices and instead make sure music reproduction was as good as possible.

What did I tell you? :-)


Get That PC Out Of My Living Room!

Mike Krisher, responding to Marc Canter’s post about Media Center, says:

But that they prefer Tivo. What’s the difference? A Tivo is a PC. It may not be the familiar Windows interface, but I’m not so sure that is a bad thing.

Yes, Tivo may be a PC, but it’s not a Windows PC; it’s a diverged device which pivots on the commodity hardware of the PC architecture.

As for the “familiar” interface, Marc hits the nail on the head:

They want interfaces designed for the TV.

And Windows is not that interface.

Looking up info on the net at the same time I was watching a show would be pretty nice.

Yes, but this should be structured, contextual search. Whenever I’m watching TV, I usually only want to look up something on TV Tome which I currently do with my Treo. I don’t want to have to bring up a browser, and I sure as hell don’t want to type anything, I just want a quick route to the info.

And I definitely don’t want any kind of PC, even if “the OEMs have caught on to” the fact that you have to design the boxes differently. OEMs don’t do industrial design.


Amazon Launch Web Services 4.0

Amazon have launched version 4.0 of what is now called the Amazon E-Commerce Service.

Having messed around with the services before, my first impression of the new API is that it is much cleaner than before, and the features they’ve added are most welcome. Reasons for this will become apparent in due course.

Now, what would be really cool is if other retailers such as ThinkGeek, Nerdorama, and CD Wow supported the API. Call it Open Retail or something. There are reasons why that’s unlikely to happen, but a man can wish can’t he?


Context Specific Mirroring

Ben Hammersley has another idea for the LazyWeb: Unleashing memes with Context Specific Mirroring

I just want to treat each individual posting as a single entity and place it in as fertile a set of beds as possible. I want context specific mirroring. I want to be able to choose multiple endpoints for a post, and publish to all of them with a single button click.

Great idea. Wouldn’t it also be cool if you could send book/movie/music/whatever reviews to Amazon via Atom as well?


Continuing The IM Theme

Peter Saint-Andre, the patron saint of Jabber, writes about how there is something rotten in the Jabber community. What’s interesting to me is that some of the things he talks about were all things that crossed my mind as I’ve been reading about Jabber recently. My Gibber Redux post originally had a lot of info about what features I liked about the various IM systems (as well as what I would like and what I didn’t like), especially given that IM is playing an increasing role in my day-to-day life.

Peter pretty much sums up all my thinking with this sentence:

As anyone who talks with Jabber users knows, there are hundreds of Jabber clients but almost all of them are close to useless.

I think that Jabber is a great collection of technology, but which is held back by the clunkiness of the existing clients (especially compared to the pretty immediate experience provided by Yahoo!, MSN, et al) which I’d say is responsible for holding back the building of a network. Indeed, the power of Jabber’s decentralized user model is also one of its key drawbacks: hence my thinking on what might happen if Google did adopt Jabber for an IM service. Suddenly, there would be a massive network of users.

Peter advocates building one brand-new client and one brand-new server, probably in Python, which would act as the de-facto tools for anyone looking to use Jabber services. As he says, “wouldn’t it be cool if you could actually make use of protocols like pubsub, XHTML, and file transfer?

Yes. Yes that would be cool.

Peter ends by saying “let the flames begin”.

I say bring it on – Jabber has been great at producing the specs, but here’s a question: what happens to specifications that nobody can use?


Sun To Acquire Novell? Old News To Me...

Ahem I believe I already suggested this:

‘With our balance sheet, we’re considering all our options,’ Sun Chief Operating Officer Jonathan Schwartz said in an interview Sunday regarding the possibility of acquiring Novell.

I wish people would acknowledge me when they get ideas from me ;-) (via)


Sony Connect To Add Video

Sony say they are going to add video downloads to Connect in the next 12 months. One word springs to mind: “Why”. Haven’t they already learned from the now renamed and gone elsewhere joint venture named Duet? Apparently not, of course, since they’re trying again on the music front with Connect.

I can understand why they would offer PSP games for download, since that’s an entirely proprietary architecture.

But movies will fall into the same problems as music, where there will be massive gaps in the available range because Sony’s competitors won’t licence movies into Connect.

When are these guys going to learn that this kind of integration just doesn’t work? How many more times are they going to try? (via)


Good Sonos, Have A Biscuit

Earlier this month, I chastised Sonos for their browser detection script.

I’m pleased to say that Chris from Sonos left a comment a couple of days ago saying that it was fixed just after my post. Therefore, I now retract my criticism, and praise Sonos for taking the time to respond to my post.


Usability

Matt Croydon pointed to a new document which Nokia released the other day, entitled Tip Of The Month: Usability Is The Heart Of Development v1.0 (I find the version number in the document title somewhat disquieting):

What if, instead, you made one simple change to the familiar development process: implement usability right from the get go.

What the ….? Yeah, just imagine if you did do that… Wait, that is my familiar development process.

I can’t believe that “try setting out to design usable software” is Nokia’s best advice for the whole month of May. I mean, as if you’d set out to do otherwise – you might not get it right, but you wouldn’t set out to create something that people found hard to use, would you? Although, having watched Orange’s portal get steadily worse over the last couple of years, and having seen many shocking examples (mobile and otherwise), I do have to wonder…


Number Ownership

Telepocalypse has an interesting post about direct ownership of telephone numbers, handled in the same manner as domain names. Personally, I think this is a great idea, and one which is deserving of investigation (at the least) by a forward thinking telecoms regulator. Assuming such a thing exists, of course.

My favourite quote:
For the next generation of customers, telephony is a minor feature of an instant messaging client.

I wonder how many incumbents fully understand that one simple fact?


Right-click in Firefox to Subscribe in FeedDemon

“Inspired” isn’t the word.

“Edited” may well be, however :-)

After seeing the NewsGator/Firefox right-click to subscribe plugin by Stuart Hamilton, I got jealous, so I’ve grabbed it and created a new plugin to work with FeedDemon. Stuart, hope you don’t mind.

Torsten Rendelmann also created a generic feed URI extension, but FeedDemon doesn’t support that yet (although it will when Nick is well enough to finalise version 1.1). Am I right in thinking that the feed URI scheme won’t work if you use a service like Bloglines?

Anyway, try it out and let me know if it works; there may be a problem with installing it to a user profile rather than application wide, but that hasn’t been verified. You should just be able to click the above link to trigger the install; if not, save the .xpi file to disk, then drag and drop it onto the Firefox browser window. You’ll need to restart Firefox to enable the extension.


Yahoo! Shopping SmartSort

Diego talks about Yahoo! Search, which integrates RSS and My Yahoo! right into the search results, but via The Shifted Librarian, I noticed Yahoo! Shopping SmartSort, a kind of digital personal shopper.

Go and give it a try, I think that this is the way all specific searches should be done. Google and the new Yahoo! Search are great for general purpose hunts, but sometimes you know what you’re looking for, and this is a great way to find it without the cruft.


Apple & RSS

Yikes! I misrepresented Apple’s support for RSS. Via Joe Wilcox, Apple have many RSS feeds, plus the iTunes RSS Generator. Phew!


Firebird becomes Firefox

So Firebird becomes Firefox with the release of the 0.8 version. Sadly, I don’t see mind control listed as one of the new features.

Thunderbird 0.5 is out too. Upgrades begin.


Wireless Audio Getting Somewhere

NetGear launched their MP101 wireless digital music player, and it looks to be the best one I’ve seen so far. It even makes an attempt at having decent industrial design (it fails of course, but at leasn’t it isn’t totally fugly).

On the downside, it doesn’t support iTunes (not that I can buy anything from that in Europe yet anyway. Grr), and you need to install special ‘music-searching’ software on one PC somewhere in your house.

Good Things: You don’t need to plug it into a TV (the UI is built right into the device; imagine!), and it’s a bargain at just £100! And it’s my birthday soon…


Salesforce.com IPO

News.com reports that Salesforce.com has filed for an IPO, sometime in 2004 (official press release).

I think that this flotation is more important than the impending Google IPO. Why? The Google float is already surrounded by a ridiculous amount of hype, which will only increase as the IPO gets closer, and it will feel like 1997 all over again. The Salesforce.com float will be popular, but I think that because the company is out of the view of the regular public, it won’t be subjected to the same level of media coverage. As such, I think it will provide a more accurate gauge as to the general appetite for tech companies coming to market.

Also, I totally agree with what David Galbraith says on the subject.


Apple News Feeds

Just noticed that Apple have a couple of RSS feeds for their Hot News and Press Releases.


XPlanner

Discovered today: XPlanner, a web-based project planning and tracking tool for eXtreme Programming (XP) teams. It looks mighty useful, and it even has a SOAP interface.


Argh, my eyes!

Extremetech report that Gateway have released a new Media Center PC, which apparently is designed to match and sit nicely with my other hi-fi equipment.

To be quite honest, I wouldn’t put that, that thing in my loft, far less give it pride of place in my living room. I’m quite nauseated just contemplating it…


ESPN RSS Feeds

Seems like ESPN is joining the RSS-fest:


"Wireless Music Done Right"

Following the announcement of the Slim Devices Squeezebox, Creative Labs get in on the act, with a WiFi hi-fi component done right (in other words, without needing to plug in to a TV).

Shame that neither device matches my existing setup…

UPDATE: I just noticed that the remote for the Creative device uses RF rather than IR, so no line-of-sight problems. Nice!


A simple, straightforward test.

Testing C# and XML-RPC.NET:
A simple, straightforward test.

Cool :-)


AOL delivering "original content" for broadband sports

Via Jeremy Alliare: AOL delivering “original content” for broadband sports

Broadband opens up the potential for original programming, but very very few media companies have taken advantage of this, or begun to think creatively about the possibilities.

That last part, about thinking creatively? I feel the same way about mobile, with the dangerous angle that I actually have a chance to try something out…


Digital Music Without The PC

A Gizmodo story reports that some Japanese consumer electronics companies realise that digital audio can work without a PC:
Sony, Sharp, Pioneer, and Kenwood are working together on a new line of stereo systems with Ethernet ports that can download music directly from the Internet without requiring a PC. The joint-venture is called Any Music Planning, and right now the stereos only connect to a Japanse online music service called LabelGate that charges about a $1.90 per track, but there’s probably no reason why these things couldn’t hook up to Apple’s iTunes store or the new Napster 2.0.

Hoorah!


Working at a higher level

Dare wrote a post about the Top 3 Features I Want To Add To RSS Bandit, which all sound good, but there’s a quote at the end I want to pick out:
I’m completely bored with doing the same old shit or even worse trying to reinvent the same old shit when there are more useful things that can be built at a higher level instead of dealing with low level concerns about syntax of formats and the like.

This is important for the guys who are working on the formats without working on the tools because: He who writes the code, makes the rules.

Also, Dare makes another comment here, which I hope means that we will start to see a more streamlined process for picking feeds out of OPML files.


More on Discovering Feed Indices

As my day starts to wind down, everyone else is waking up:

  • Dave Winer feeds back sensibly on comments that I and others left on his original RFC;
  • Sam Ruby clarifies that he wasn’t necessarily proposing a new markup;
  • Dare Obasanjo asks for some real use cases to appear before we get too deep;
  • Chad agrees with me, which is nice;
  • Finally, Diego Doval has some thoughts as well, including nailing things down (especially encodings) more than OPML currently has them.

So perhaps what we’re looking at here is a specific instance of OPML (OPML for Feed Lists) which ties format details down a lot more than the current spec, details the auto-discovery process (a named file in a given directory and the link tag), and lists which attributes are valid on outline tags?

I’ve suggested to both Sam and Dave that Tapestry is used as a real-life use-case, so lets see where things progress from here…


Discovering a List of Feeds

So I woke up this morning to discover that Sam Ruby and Dave Winer are talking about the same thing, in a different manner. Guys, I already wrote about this!

Sam is inventing a new format (FDML; wiki here); Dave proposes using OPML.

In terms of what I think about all this, I think Dave is right that OPML is already understood by all the aggregators, but I don’t agree with his discovery mechanism. I also don’t agree with Sam’s proposal of a brand new markup language to do all this, since OPML can be extended and already supports nesting.

For discovery, I don’t see why we can’t use a system similar to that for discovering RSS feeds, or Atom APIs or what have you. This is what I originally suggested back in August, and is supported by Tapestry right now:

<link rel=”index” type=”text/x-opml” title=”Index” href=”http://dwlt.net/tapestry/tapestry.opml” />

I’m not presuming that I know better than these two guys; I just happen to think that creating more work for the tools and content producers than is strictly necessary isn’t the best approach.

Anyway, whatever happens, you can rest assured that Tapestry will support the result.


Stopping Comment Spam

Via Luke Hutteman: Stop comment spam now!
If you’re running a MT based weblog and have experienced comment-spam, this is the must-have plugin you need to install. Get it now!

Downloading and installing….


Cirrus Logic slap lawsuit on Wolfson

This is a bit depressing: Wolfson blamed in patent dispute: WOLFSON Microelectronics, the Edinburgh-based technology firm preparing to float on the London Stock Exchange next week, was today accused of patent infringement by a rival in the United States.


Close, but no cigar

Gizmodo mentioned this new Samsung Napster device, which claims ‘you can directly connect with the new Napster 2.0 music download service for automatically transferring tracks to the player.’

By ‘directly connect’ however, read ‘plug the device into your PC to connect’. Why not build GPRS or higher into that box so you can skip the PC altogether?

Everyone keeps talking about how the PC will become the home hub, but the PC is so not a consumer device. Consumer electronics tend to be aesthetically pleasing, for a start.

So yes, a TiVo may be fundamentally a PC at heart, but that’s not the point. I don’t want a PC in my living room, thanks very much. I think Dell might be starting to get this; I don’t think Microsoft does; and I’ve already said that I hope Apple does too…


Ch-ch-ch-changes

Martin said that I would write a bit about my experience with email last weekend, so here goes.

When I plugged in on Sunday, I had 1600 emails sitting on my work account, and 60 on my dwlt.net account. Of all these, roughly 3% on each account was actually relevant and useful mail (I’m excluding newsletters which were more than two days old here). That’s just plain crazy.

When chatting about this with Martin, we were both commenting about how we now fire up our aggregators before our mail clients, something that even 6 months ago, I didn’t think would happen.

Now, I know that there’s a whole bunch of talk out there about the death of email, or the re-invention of email, or whatever, but Naval at VentureBlog gets it spot on in this post.

For me, this is Darwin’s theory of evolution (compounded with the rapid leap theory) in action, in Internet time.

For a few years, people have been talking about information overload, and finally something broke. Push services tried to solve this problem back in 1997/98, but maybe there just wasn’t enough spam and junk around back then. 2003 will be seen as a major inflection point in the way that people use the Internet, and particularly the way they retrieve information from sources that they know, like, and/or trust. That method, of course, is all the RSS feeds that are appearing across the cloud.

As more content providers jump on board, and the subscription process becomes ever more streamlined, and the user education levels go up (that last part is seemingly the hardest, but LockerGnome has made a good start), RSS will become ever more prevalent.

But the Web and email will still exist, with all their problems and all their merits.

PS Hope that was OK, Martin!


Subscription Applications

Russell Beattie writes about Apple’s quiet reinvention of the OS purchasing cycle, which got me thinking about how the same could be applied to applications themselves.

Given that the right distribution infrastructure is now in place (more or less), is it time for a reinvention of component-based apps?

As an example, I (like most folks) use MS Word in day-to-day life. However, I only use about 10% of its available feature set (and I feel I’m rounding up here). There are a load of features which just aren’t any use to most users (especially home users), such as revision tools, etc. Is there scope in the market for a bare-bones document processor (let’s call it iWrite or iDocument) where the basics never really need to be changed (ways to edit and format text don’t tend to be reinvented that often), but new features can be added on a paid for basis?

The basic app could be free or minimally priced ($20 or so), and a nice discovery framework would need to be in place so that users could easily and quickly find out about new add-ins that they need, as they need them.

And yes, I know that plug-ins can sometimes make systems flaky, but you could have some sort of approval or signing system to counter that.


Pentax Optio S

Well, I took the plunge and finally bought a digital camera: the Pentax Optio S.

It’s very neat, and has a bunch of cool features. Once I figure out exactly how I want to display and share my pictures, they’ll start to appear on the site. Most likely, this will be in about three or four weeks, since I am going on holiday next week to the far north of Scotland for some peace and quiet… :-)


Using XSL with RSS

That’s weird. I was thinking about this earlier on today, then Russ posts about it:
I’m such a dork sometimes. You’d figure with all the time I’m spending on XML lately, I would have thought of this myself, but I hadn’t. What I’m talking about is putting an XSL Stylesheet header in your RSS documents so that when a browser like Moz or IE6 sees it, it’ll render the XML instead of dumping raw tags in case you click on it.

eBay Development

A while back, I wrote about eBay’s developer program.

Well, it seems that someone was listening (yeah, right): witness the appearance of Jeffrey McManus’ blog. Jeffrey is eBay’s new developer relations pointman, and now their API is indeed free! Nice one.

I’ve no idea what kind of application I want to write, but it might be interesting to just use it to learn C#. Any suggestions, folks?


One thing leads to another...

Yesterday, I had a comment on my post announcing my Ratings plugin for MovableType. The comment asked if you could add a rating to a comment when you posted.

Since the plugin uses the ‘Keywords’ feature of MT entries, the answer is “no, you can’t”. However, in a flash of the blindingly obvious, it suddenly occurred to me that if you did have that ability, then suddenly your bog-standard comments become Amazon-esque user ratings on products (yeah, I’m a bit slow on the uptake, me).

For whatever reason, although I’d been aware that tools such as MT were categorised as ‘Content Management Systems’ (CMS from now on), I’d been shying away from actually thinking about it in that more generic sense. And in fact, I’d been thinking about using it in a new project, but didn’t think it was suited (yeah, I’m also daft). Possibly, I was lead down this dark alley by the seductive power of “that’s all I’ve ever seen it used for”.

Silly old fool me, since in fact MT does suit what I need, albeit with a bit of tweaking (see earlier). And tweaking is one of the great strengths of something like MT, hence the reason my plugin exists. I suppose I’ll really have to learn Perl now, though, if I’m to make this work.

Ah well, such is life.


FeedDemon + RSSlets

Using the Google News Search RSSlet over here, I added a new search tool to FeedDemon in about 1 minute flat.

FeedDemon has overtaken SharpReader as my RSS reader of choice, I think…


Java on Palm

I received this email this morning, and I thought I’d share it with y’all. Following Russell’s post about this, no real surprise to see that IBM are driving this forward, rather than Sun:

You are invited to attend a web event co-hosted by IBM and Palm Solutions Group on Monday, September 8, 2003, (Tuesday, September 9 – Asia Pacific). For specific webcast times and registration, point your browser at:

http://www.ibm.com/software/wireless/wme/palmevent.html The webcast event is related to the previous IBM and Palm Solutions Group announcement (June 9, 2003) that Palm Solutions Group had licensed IBM WebSphere Micro Environment, a J2ME Java Poweredô runtime, and that Java-technology applications would soon run directly on Palm Tungsten handhelds.

Also, this is palmOne, rather than PalmSource, who are driving this. The words ‘tits’, ‘about’ and ‘arse’ spring to mind, and not in any pleasant way ;-)


Motorola Bail Out From Symbian

Via Mobitopia: Nokia and Psion to buy Motorola’s Symbian shares, Motorola to focus Java.

Although this seems a little disappointing on the surface, it may actually be good for Symbian, who I feel may be in a “too many cooks spoil the broth” type situation. As a founding partner, presumably Motorola’s demands for the OS held greater sway than relative newcomers like Siemens and Samsung. Reducing this committee by one may help in the push to standardise Symbian further.

The one thing I really think it should do is settle on a user-interface standard. I read recently (maybe on Mobitopia, in fact) that on Orange, Nokia handset owners generate about 40 text messages per month, whereas Motorola owners generate about 12 messages per month. Now, as an operator, whose handsets are you going to support and subsidise the most? Screen sizes need to be standardised as well.

It’s becoming so complex to develop a base handset these days, that it must be easier to just license a system such as Series 60 so that you have the main featureset (MMS, Java, Bluetooth, etc). But then you still need to differentiate your handset against the others. How? Industrial design, and signature applications. The only manufacturer that really seems to understand the power of this is Nokia, however (Sendo might; we shall see).

You may also enjoy Russell Beattie’s rant.


Allaire on the 'New' Internet

Via Nick Bradbury: Internet Convergence 2.0


The Philosophy of Yes

Mena Trott, of Six Apart (creators of Movable Type), talks about The Philosophy of Yes.

My favourite quote:

This made me laugh, not because of anything that this user had done, but because there is a step 15. No user should have to go through 15 plus steps to get their content out of a system.

This reminds me of the story where one executive is boasting to another (from a different company) that his company’s customer service department has tripled in size over the last 18 months. Not really something you should be boasting about, really…


Java PC

Martin paints a sad picture while trying to buy a train ticket.

This got me thinking: whatever happened to Java PC? I remember some guys at Spektra messing around with a beta of this about 5 years when they build the tuck shop application. It was pretty fast, had built-in support for fellow lost soul the Communications API (useful for the barcode scanner they had bought), and was absolutely perfect for this kind of kiosk application.

Sun’s product introductions remind me of my attempts to play guitar: if I’m not fantastic straight away, I’ll move onto something else. Admittedly, software was never their strong point (with the exception of Java, but even Java’s success has more to do with IBM).

Microsoft, on the other hand, are nothing if not persistent. Witness their smartphone efforts: they know that this will be an important segment, and they don’t want to miss out.


TopStyle

After checking out FeedDemon, for Tapestry purposes, I decided to download TopStyle to help in my quest to understand more about web technologies, and also to help with building future websites. I started off with TopStyle Lite, and this evening downloaded the trial of TopStyle Pro.

I have to say that this is the HTML editor I’ve been looking for, since… forever, I think. It’s literally an IDE for site creation, which for a programmer, is mana from heaven. This post sums it up neatly: Nick Bradbury: Satisfied TopStyle Customer

So, once I’ve messed around with it a bit more, I think I’ll be purchasing this.


SubscriberTrack

Steve Main:

To solve this problem, Iím proposing SubscriberTrack: a simple, XMLRPC-based mechanism to let bloggers know whoís subscribing to their blog.

This sounds very cool, and would be especially helpful for me in understanding the popularity of the various Tapestry feeds.

I would suggest having three tiers of information: with personal info (doesn’t have to be an email address, could just be a date of birth for COPPA and general demographic information); anonymous (the blog still gets pinged on subscribe and unsubscribe); and never (umm, no notification ever).


Lockergnome's RSS Advocacy

Lockergnome’s RSS Resource

Check this out for lots of information and feeds on exactly what RSS is. Chris says that ‘email is polluted’, and he’s right. RSS is definitely the future. Of course, I’m glad that he plugs Tapestry as well… ;-)

I’m on several email lists which try to point out useful information to me, but I’d really rather that they were available as feeds. Why? I could find out about new software releases or relevant information (for example, from Forum Nokia) as they appeared on the site, rather than getting the info two weeks later once they have enough for a newsletter (and sometimes they don’t even have enough after two weeks). Yeah, sure, I could try and remember to visit the site every day, but who has the time or inclination to do that?


Lessig on MP3.com

Via Lessig Blog: [sigh]: mp3.com, we hardly knew you


NVIDIA Acquire MediaQ

Been meaning to write about this since Monday, but this link on infoSync reminded me (official release here).

To me (given my work), this is a lot more interesting than the Novell/Ximian deal, and I don’t know why infoSync say it’s a surprising move. Consider that ATI is NVIDIA’s main competitor, and they already have mobile chipsets. NVIDIA doesn’t have this capability, and they’re going to need it.

MediaQ made some great design wins over the last couple of years, and are in most new PDAs, so its not as though MediaQ weren’t making any headway with their technology (which, from what I know, is pretty good). This is a very smart move by NVIDIA, since their experience in 3D combined with MediaQ’s low-power and SoC knowledge is going to make for some fantastic mobile media devices come 2005 or so (though we’ll start to see them appearing through the next 12-18 months). And they get a great bunch of customers to kick things off: Mitsubishi, Siemens, Palm, Sharp, Philips, Dell and Sony.

Sounds like a sound investment of $70 million.


Nuts

Firebird somehow managed to copy its default bookmarks.html over the one in my profile. It seems this is a fairly common occurence. Fortunately, I had a reasonably recent copy hanging around elsewhere, but am still missing a few links – and of course, I can’t remember what they were :-(

Time to use a scheduler to make back auto backups, I guess…


Novell Buys Ximian

I suppose this might generate a bit of discussion throughout the world: Novell Acquires Ximian to Expand Linux Solutions and Open Source Commitment


Firebird Web Panels

Via Phil Ringnalda: Firebird’s next killer feature

As I understand it, Web Panels will be Firebird’s version of the Mozilla/Netscape Sidebar. If, like me, you mostly associate that with the AOLish push content and stock quotes that shipped as the default Sidebars in Netscape, it’s time to think again.


Seamless

So I’m now fully broadbanded and duly unwired. I have to commend Netgear, D-Link, and Telewest Blueyonder on the ease of use of the products I’m using. The D-Link card just works, Blueyonder has a nice self-care feature which means I can register MAC addresses of other products I want to plug straight into the cable modem (such as the Netgear router, for instance).

Speaking of which, the router also just worked, with little configuration required by me (setting a sensible SSID and setting up the MAC address of my wireless card on an access list). Another nice feature of the router is that it can actually take the MAC address of any device physically attached (ie, with a cable) and present that to the cable modem, so I probably didn’t even need to register that MAC address in the end.

Anyway, I can now wonder freely around the house and back garden (I haven’t fully tested the range yet) and still have a signal strength of ‘Good’ :-D


Digital Shoplifting

via Scott Loftesness:

The AP ran a story about a week ago describing the ways that new cell phone cameras are being put to use in Asia – hardly seen here yet, there 25 million of the devices are already in use in Japan. One of the unintended consequences from the adoption of this technology is digital shoplifting—taking pictures of magazine pages in bookstores, for example.


SharpReader

Luke Hutteman graciously accepts the praise for SharpReader, and here’s some more:

I read some of the blogs I subscribe to offline (since, sadly, I’ve yet to enter the world of broadband), and I found a post with some links which I wanted to flag for later reading, but I wasn’t sure if that was possible. A quick right-click on the entry, and one of the options is “Lock Item”—perfecto! Just what I wanted.

Now, if only I could use the filter to display the complete list of items that I’ve locked…


NEC Mobile Phone TV

Via SmartMobs: I Want My MTV.

I’m not convinced that I want my phone to actually record a TV signal, especially given Martin’s recent story on his P800 deciding to reformat itself without any option. However, if I could send a text message back to my PVR to commence recording…


That's what I said!

[via Mobitopia] This link quotes an article by Ray Ozzie, where he talks about the fact that we are too pre-occupied by inventing new gadgets and technologies, and aren’t taking a step back to think about how to sensibly make use of it all.

But that’s what I said!


w.bloggar & XUL

So I decided to try out v3 of w.bloggar now that it supports MovableType blogs (which is why my posts are getting longer and are posted more frequently!). I have to say that I like the cut of its jib, but there’s a few minor bits that are perhaps only appropriate for MT users that are missing: adding new categories; adding keywords to an entry (I had to login to MT on my site to add the rating for this post); and trackback pings (also suitable for other blog users).

My experience of working on the Thunderbird plugin, which is built on the Mozilla codebase, has kind of spoiled me, I think.

Because Mozilla’s UI is built from XUL (an XML user interface markup language), extensions can easily alter the layout, adding and removing elements to add new features which are required by a minority of users. This is a very powerful mechanism; if w.bloggar supported it, I would probably work on adding some MT specific features as an extension.

But then I thought—do any other apps utilise XUL for their front-end? I haven’t looked at the Mozilla codebase, but presumably the bit that works with XUL is reasonably standalone? Or perhaps not. Has anyone worked on a Java Swing XUL system? That would be a very cool component to have available—if I had more time, perhaps I’d make a start on it if there isn’t already one. I’m not sure if you have to use JavaScript with XUL or not, but since Mozilla have Rhino, a JS engine to embed in Java, that isn’t such a problem. Using JavaScript to program apps might be more of an issue, though ;-) Update: Check out Luxor. You can use Java and Python to build the apps, and embed Rhino if you so desire! I’ve not looked at it in any more detail, but looks like a nice piece of work. That page also links to other XUL toolkit projects.

If more apps worked in this fashion, then you could probably pull out at least 50% of features from a lot of highly monolithic applications which are only required by minority user groups. The main app then becomes more lightweight, with the individual features also becoming lightweight, and as Phil Ringnalda points out, the XP Install system which is used by Mozilla extensions, is quite a powerful thing in itself.

I wonder if the reason XUL hasn’t become more widely adopted is that you don’t really see the full power of it until you start messing around with it a bit?


BitPass

BitPass are the guys who are helping Scott McCloud with his micropayment experiment. It now reminds me of a real-world discussion I had with one or two people last year, after I’d been to the US and driven about for a variety of meetings over a couple of days in Silicon Valley.

I had pre-printed driving directions from MapQuest, which were very, very accurate, and helped me no end in finding my way about. I’ve used their UK site as well since then for the same purpose, and the directions are a lot more practical than other sites I’ve used. Anyway, my point was that these directions were so good, that I would gladly pay $0.20 or $0.30 for each set. Why don’t they, I thought, create some sort of credit account system, so as to mask the credit card transaction fee and then deduct the micropayment from that account?

Which is exactly what BitPass provides of course, but in a far-handier, vendor-neutral fashion. I presume.


Network Aware Apps

I’ve said this a few times to some people, but as usual, Tim O’Reilly is a damn sight more articulate than I…

His assertion that the Unix permission system is the correct one is something I’ve been thinking about for a while in terms of how apps are structured, and the composability of Unix apps is something I’ve been thinking about for the last couple of weeks since someone pointed out that Ant tasks aren’t composable (i.e., the output of one command can act as the input to another command).

Tog’s book Tog on Software Design is what originally got me thinking about network aware apps, and now their time is upon us, thanks to wireless whatevers and full-ahead broadband…


Quick Reply

One of the other projects I’ve been working on is an extension for Thunderbird, along with Erwin Wessells. Designed to replicate the Quick Reply functionality of Opera’s M2 client, we’re churning out new versions as quickly as we can figure the code out, and are getting some great feedback and support from the Thunderbird community. If you use Thunderbird, grab it and give it a whirl.


Little White Lie

I wasn’t telling the truth when I said that I’d had an idea for a Movable Type plugin… What I should have said was ‘I’ve had an idea to port a plugin from another system to MT’.

And that’s what I’ve done: ported Rael Dornfest’s Rating plugin for Blosxom.

You can read more about it