It had been a long time in the coming, but a couple weeks ago we launched the updated version of CScout.com. I’d held off on telling people about it until the bugs had been worked out, and now it’s finally at that point.

Today I made just one more adjustment to the site. The displayed image for the Trend Blog panel in the lower right is dynamic; it pulls the first image in the latest blog post. Initially it was grabbing the full sized image and only displaying its upper right portion, but today I coaxed the images to scale to 300px wide and have any vertical overflow hidden.

Have a look, and let me know what you think.

I worked yesterday on a new feature we’d like to introduce into CScout’s Mobile Trendpool. Right now we have a couple sections on the site that display a list of headlines relevant to people looking at mobile, but it’s a clunky implementation requiring an editor to manually populate the list.

To circumvent the existing updating process, I’ve been working to integrate a river of news type stream from the separate feeds of mobile related websites. This has been a two part process.

First I found a Yahoo Pipe that aggregates content from multiple RSS/Atom feeds, which I duplicated and customized a bit for my own purpose. Basically I just removed everything but the title/headline which is published as a link back to the original post.

Next I grabbed a couple WordPress plugins (linked to on the test page) used for integrating the content. I installed the plugins, then created a page template to hold the code. The Yahoo Pipe outputs an RSS feed, the URL of which I dropped into each plugin. Published, the content is displayed exactly as I’d envisioned.

From the looks of it, the plugins are more than adequate in terms of features/customization for my purpose. Honestly, most of the manipulation happens in Yahoo Pipes. I’ll take another, slightly closer look at both plugins and run with one.

You can see the streams in action on the Aggregator page. What you’ll see are Yahoo News’ science headlines and Google News’ sports headlines. We’re selecting sources for the actual working implementation, but for this test I just wanted a couple feeds that are being updated somewhat regularly.

I showed this system to Rory, who’s also in the office working with Strange Matter. He loved what he saw, and suggested we start building sites that aggregate content similarly to this.

The playing field for this type of thing is occupied, but I don’t think it’s necessarily crowded. What might work would be hyper-local, or at least hyper-specialized streams, meticulously curated.

To take this to the next level we need the ability to parse a feed and add new content to a database. This is another layer of complexity that has to be thought out and implemented. With this next step I’m treading in untested waters, the perfect opportunity to learn something new.

Further out, editors will need the ability to manage their streams through a nice backend at which point the we’ll outgrow the functionality coming from Yahoo Pipes. This is really snowballing…

After working with this more last night and this morning, I’m excited to keep pushing it. Getting into the office, I found Rory rabid about the idea. I think this can work if we implement it right. These days there’s no shortage of projects.

As an addendum, I know I’m kinda giving away the farm here. Thing is, this isn’t new. The value from one of the “accumulators,” as Stan calls them, is in how the feed is curated; no two editors are going to create the same stream.

Yes, there are a lot of other elements required to make it successful, but I’m not going to get that far ahead of myself. Besides, there are others already involved who are addressing some of those aspects.

I found the above panorama of NYC’s Westside on Wikipedia. This is a 6000px wide version, but if you click through you can get at a whopping 22,495px wide image. Loving that the skyline is bookended by the GW Bridge to the north and the Verrazano-Narrows Bridge to the south.

Drafting this post was also a bit of a JavaScript project for myself. I used jQuery and the ScrollTo plugin to animate the view of the panorama. If you’re not seeing the animation you can load the individual post by clicking here.

I’ve worked a bit to tweak the Twitter tweets integration in this site’s sidebar.

I’m using the WordPress feedList Plugin which allows me to funnel in content from RSS feeds. Originally I was piping in the content from my @lightwerk account, but since I decided the @soundnoir feed was pertinent as well, I turned to Yahoo Pipes to combine the two.

I’ve got the Pipe I created for the task grabbing the two RSS feeds from the accounts, but initially the content was doubled up — each tweet was displayed in plain text and as a link at the same time. To fix that I removed the description from the feed.

To edit the RSS feed, I first had a look at the raw contents of the feed using Feed Proxy. With it I was able to pinpoint the “item’s” “description.” Knowing how to address it, I used the Regex module in Yahoo Pipes to find that bit of data and replace it with nothing, essentially deleting it.

Next, I grabbed the newly combined feed’s RSS addy and pasted it into the sidebar.php file of my WordPress theme. It worked correctly, displaying the combination of both feeds in one place.

To highlight the fact that the tweets are coming from two separate accounts, I again turned to the Regex module and searched for the usernames preceding the tweets themselves. By matching ^(lightwerk:) and replacing it with <div>lightwerk:</div> I was able to focus on the usernames and highlight them through their CSS property.

You can see the end result in the image above, and, of course, in the sidebar to the left. I’m pretty happy with it.

One note, since it seems this could be useful for someone else, I’ve published the Yahoo Pipe that drives this functionality. A cool thing about Pipes is that when a given Pipe is published, others can clone and edit it for their own purposes. Feel free to help yourself.

Fluid design work by Italian graphic designer Michele Angelo at superespresso.

On a related note, I’ve been wanting to build an online photo magazine that would employ the same JavaScript pagination you see on superespresso. The first place I saw it was in the piece “Star Wars: A New Heap” on Triple Canopy. At first glance it looks like you’re just paging though images. In actuality the text you see is really text that you can select/copy/paste.

via James Widegren on FB

I’d looked several times before, and it seemed as long as I’d added the track to my Lala account, the files acted playable through their embedded players. Not sure if I was imagining it or not, but the embed code seemed to change before and after adding the track to the account. Before, I’d be given a 30 second preview. After, the whole track was playable. I don’t know when this broke down because I haven’t been hearing any complaints from people. On the contrary, I’ve actually gotten decent responses from readers / listeners.

What to do now?

I first considered sourcing files from YouTube rather than Lala. There are a ton more tracks that I’d be likely to post already on YouTube. The main problem I see is that I’m both dependent on YouTube as a service as well as the user who posted the track. With that kind of mix there’s no guarantee that the track would remain available. So I approached this the same way I’ve recently approached the images I post to this site.

Originally I thought it nice that I could host images on my Flickr account and drop them into posts here. Flickr would then handle the bandwidth for me. Flickr isn’t going anywhere soon, but at the same time there’s really no guarantee of that. I’ve therefore started grabbing my Flickr images and uploading them straight into Limitless Pulse. This is the solution I’ve arrived at concerning music on this site.

Everyone’s familiar with what seems to be the standard flash mp3 player, WordPress Audio Player. I don’t need a ton of functionality, nor do I care to look for anything else, so I’m sticking with it.

To make a long story short(ish) I’m going to replace the embedded tracks with my own mp3 uploads thereby ensuring people can actually listen to them. There’s more of a rush to get the audio replaced as opposed the the Flickr images. The images are safe, but please bear with me as I get sound straightened out.

Images by frank richter

As of late, I’ve been putting a bit of time into Craigslist every day. Generally I’m shopping for furniture and slide projectors. To aid my search I’ve piped RSS feeds — which are created for any search made on Craigslist — into Google Reader so I can process the streams of mostly garbage that people are selling.

The first feeds I imported were for searches of “Eames” and “mid century” in the furniture corner of Craigslist. These brought me tons of over priced knock offs of over priced originals. I learned quickly while thrift store shopping, garage sailing and flea marketing that once an item had the word “antique” attached, it’s price became unreasonable. The same held true for these searches, so I dropped them.

In terms of furniture, the only valuable search I’ve come up with is “west elm.” This feed on its own throws ~40 items my way daily. Some stuff is over priced, but I’ve found a couple amazing deals. Unfortunately when I see the best stuff, other people are also seeing it and end up snatching it before I can make a move.

Similarly, I’m always looking to grow my slide projector collection. A very nice looking Kodak slide projector popped up in Google Reader yesterday evening. It was selling for $40, was basically in the neighborhood, and the post was only 20 minutes old. I emailed the seller while ideas on how I’d use it started bubbling up in my head. I heard back from him about an hour later and could barely finish reading the message through the tears… the projector had already been sold.

OK. If 20 minutes isn’t fast enough, how do I get tipped off to the best posts?

To monitor @lightwerk and @soundnoir Twitter mentions, I’ve been using Pingie to send me an SMS and an email each time it finds activity on those terms. It’s not the fastest service, but has kinda worked in these instances.

The slide projector posts are the most important to me, so I went to feed Pingie the RSS feed created for the local NYC search “kodak slide projector -car.” Car? Yeah, I had to obscure the idiotic car sale posts that some asshole was cluttering up my feed with. When I dropped the URL for my slide projector search into Pingie, I was confronted with an “invalid feed” error. The service was mangling the URL on import. It just wasn’t going to work. I set off to find another option, and with all the talk of FriendFeed launching real-time search last week, I thought it a good place to start.

In the past, I’ve used the invisible friend, or virtual friend, or the whatever it’s called, feature of FriendFeed to keep up with a few friends I couldn’t get to join the service. With it I’d add feeds from friends’ blogs, Twitter accounts, etc. to their virtual accounts then have those updates thrown to me via IM. It’s worked well. When I went to do the same for the slide projector feed, I found the feature had dropped off. The workaround I found functions beautifully.

I created a new private group named “RWW: Critical,” and added the Craigslist feed to it. I then configured the handy notification options to have the service send me an IM and an email whenever a new post pops up in that group.

Cool thing is, I can add any number of feeds to this one group and receive the same notification functionality group wide. I’m looking to add feeds covering my most important friends, artists, brands, etc.

What I’m looking forward to finding out is just how quickly posts get aggregated and piped to me. The virtual friend notifications always seemed to lag behind the actual Twitter tweets. Of course it couldn’t be instantaneous, but FriendFeed’s advertisements of “real time” have me curious to see just how fast it is. Pingie was relatively slow as well, so I’ll most likely move feeds from that service to this FriendFeed group.

Hands off, bitches. That slide projector is mine!

[Update] The real time aspect of FriendFeed applies to searches. Now I’m even more curious how fast or slow these notifications will be.

I’ve worked the last couple days to get a web server up and running. I’ve done some administration working for Flavorpill and CScout, but have never built a server from scratch.

I decided to use Mosso. This decision was in part made based on the cost of hosting. The basic account that I’m using costs just over $10 per month which is significantly cheaper than Amazon. Another factor in the decision was the great documentation in the Slicehost Article Repository. But what finally tipped the scales was finding that I couldn’t host WordPress MU on Dreamhost with the current setup / plan I’ve got there.

I’m looking to use WordPress MU so I can have multiple sites powered by the single install. I’ve been building a lot using WordPress and wanted to collect much of it in one place functioning partly as a portfolio.

Everything about setting up the server was well documented in the article repository. I was quickly able to get an install of Ubuntu up, configured and hardened. Again working with the documentation, I installed and configured Apache and PHP. All that’s left to setup in order to start playing with WordPress is MySQL which I’m turning my attention to now.

The idea of moving my existing sites to this new server is a bit intimidating, but I’m looking forward to learning and implementing the process. As I’ve been doing with the tech and visual aspects of my preoccupations, bits of the process will be documented here.

A cool thing about constraints on selection is that there really aren’t any. Moo Mini Cards come in orders of 100, and out of that 100 one is allowed to have up to 100 different images printed on those cards. I did 28 different iterations, so one option might be to edit out three and print four of each. This could make for cards becoming collectible. Of course nobody would need or care about having every one of them, but it would be great to see people choosing, or even trading, one or another.

It started out with my searching for an email announcement list service last night. I didn’t spend a ton of time looking, because the offering from Campaign Monitor looked so good that after exploring it I called off the hunt. Because we have yet to send an announcement, and as I’m currently the only subscriber, I’ll have to describe in better detail what it’s all about after I’ve had more opportunity to play with it.

After getting an account with Campaign Monitor I quickly added a sign up form to the sidebar at SoundNoir.net. That was easy enough, and as there was still a lot of vacant space in the sidebar, I started filling that out, first with links to podcasts and sites that are in line with us, then by adding syndication links.

I decided that we’d better get setup with FeedBurner before we have anybody subscribing, so I put time into creating feeds for posts and comments within FeedBurner. Through a page in the WordPress codex, I found Feed Icons which I tweaked a bit to match the color palette of the site. Building syndication links will require a bit more time once we get our first podcast cued up, so I’ll be writing a subsequent post detailing the way I craft the feed for consumption through iTunes; I’ve already got a plugin in mind that’ll allow for custom meta data and images for each episode.

While I was spending all this time integrating various services, it seemed appropriate to get Google Analytics dialed. Basic. Not much to speak of there.

Taking a break from tinkering, I took the opportunity to finally write the post announcing Andrey’s win in the Little White Earbuds Mix Competition. This is where things kinda broke down. So you understand the dilemma, take a look at the following screenshots:

keep scrolling…

See what I mean? No, I didn’t think so. When I showed the design I’d implemented to Andrey and Jacob last week, they weren’t feeling the three column layout, with the post’s title and meta data off to the left. It was a quick fix to extend the title and its bottom border across the top of the post, and when I’d done it, things looked decent. I was surprised, and everyone was happy. What happened this morning was I uploaded the first image to the site. The image basically fought the title creating a jumbled mess at the top of the post. Now you see what I mean… you probably don’t care, but these are the little details I like to get lost in.

What I’ve done to build my case is to duplicate the theme and tweak one version back to use the distinct three columns. I’ve uploaded these screenshots to Flickr for illustration, but can easily load both versions in adjacent tabs to prove my point. Pray for me.

I’ve had enough of the site for now. So I’m turning my attention to business cards and laying out the mailer, in that order. We’ve got to start driving people to the site, and without business cards, it’s tough to do at parties where people’s attentions easily wander.